Ars Technica Reviews Intel iMacs

Milton Waddams writes "Ars kick off what I'm sure will be a torrent of reviews of the of the new Intel iMac. Overall it looks like it's a bit faster than the iMac G5 and a bit slower than the PowerMac G5 dual core. I'm sure it will surprise many slashdotters to find out that Jobs' statements about the new iMac being twice as fast as the iMac G5 as being slightly over optimistic. AND it doesn't run Windows...yet..." I'm still waiting for the most important benchmark: frames per second in molten core combat.

Comments

Benchmarks, accuracy, and choice

[Space cowboy]

To be fair, Steve's statements were absolutely 100% accurate (assuming the figures are accurate, which I expect them to be). For that benchmark, the intel machine is 2x-3x faster. If anyone really expected them to provide not-the-best-benchmark-results, can I have some of what you're smoking ? And I have several bridges to sell you too...

My point is that the story write-up makes it sound like SJ is lying, and he's not. He's just presenting the best set of benchmarks he can, which is pretty much what I expect from the CEO of the company...

As for the multimedia-style benchmarks presented in the review, I think you can expect those to improve as Apple gets its collective head around SSE3. I would have thought the G5/G4 implementations would have been altivec'd to hell and back, and SSE doesn't have the immensely useful 'permute' operation, so the transform operation will have to be rewritten to SSE's strengths - I doubt that has happened yet...

Simon.

Re:Benchmarks, accuracy, and choice

[conteXXt]

not to mention the intel GMA900 graphics processor isn't exactly the speediest thing around.

1000fps in glxgears? I can beat that by a good 50% with a 4 year old NV GE440 go in my compaq laptop

watch what happens when there are nvidia drivers and ATI drivers available.

P.S. ATI 800 series cards do work (fully accelerated) on the development platform.

I'd post a link but the lawyers are loose.

Re:Benchmarks, accuracy, and choice

[tpgp]

To be fair, Steve's statements were absolutely 100% accurate (assuming the figures are accurate , which I expect them to be). For that benchmark, the intel machine is 2x-3x faster. [emphasis mine]

You've really just shown your bias haven't you? Absolutely 100% accurate, oh, unless they're not accurate.

Steve Jobs may not have been lying, but he was most certainly being deliberately deceitful.

I don't see such a huge moral gap between the two.

Re:Benchmarks, accuracy, and choice

[conteXXt]

UPDATE:

sorry I can beat that by a lot more.

1908 fps in glxgears.

That being said people using osx on an intel D915GAG mobo (very close to the imac and FULLY supported)
report that the gui is faster than xp on the same hardware.

I am fairly sure that the apple boxes feel even slicker.

And I am NO apple fanboy (my girlie won't even let me buy the intel board cpu combo :-(

and yes I am whipped.

Re:Benchmarks, accuracy, and choice

[AKAImBatman]

You've really just shown your bias haven't you? Absolutely 100% accurate, oh, unless they're not accurate.

I think you've missed his point. This is a common industry practice used for just about every piece of hardware and software on the market. To single Steve Jobs out for this practice rather than accepting it as the "norm" shows a distinct anti-Mac bias.

Steve Jobs may not have been lying, but he was most certainly being deliberately deceitful.

It's hard to be deceitful when it comes to something as nebulous as benchmarks. Every benchmark you run will tell a different story. The result is that you can pull a variety of different conclusions from the benchmarks depending on how you spin it. Given that Steve Jobs is the CEO of Apple, we can expect that he will spin the benchmarks positively. On the flip-side, we can expect that the Mac haters will spin the benchmarks negatively. The ones to really listen to are the moderates who tell us whether we're generally being delivered what we're promised or not.

Re:Benchmarks, accuracy, and choice

[kfg]

When you are selling something it is always the best policy to lie with the truth.

KFG

Re:Benchmarks, accuracy, and choice

[Andy Dodd]

"This is a common industry practice used for just about every piece of hardware and software on the market."

Not necessarily. When AMD moved to their new speed rating system for the Athlon XPs, they *usually* performed equivalently to a P4 clocked at the "speed rating" on average. It did NOT reflect the peak performance of the CPU compared to a P4, but the average instead.

Re:Benchmarks, accuracy, and choice

[EvilTwinSkippy]

Well Duh. They used the average rather than the peak because that rating is what made their stuff look better than Intel's. What, are they going to publish a stat they makes their stuff look bad? No you just keep changing the test parameters until your stuff looks better.

Re:Benchmarks, accuracy, and choice

[The Lerneaen Hydra]

I don't think you understand what parent said, what was meant was that AMD intentionally set their ratings conservatively, showing the average performance instead of the maximal under a certain benchmark.

Re:Benchmarks, accuracy, and choice

[poot_rootbeer]

1908 fps in glxgears.

So? Your laptop's LCD is never going to display more than 60 frames per second...

Re:Benchmarks, accuracy, and choice

[EvilTwinSkippy]

I undersood perfectly what the grand(^n)parent was talking about. The fact of the matter is that several intel products (the P4 anyone?) were notorious running benchmarks and not much else. Odds are the AMD chip would have been at a disadvantage on the quarter mile, so they decided to run a distance match.

Considering I was there when all this history was going down, and I've heard comments from execs of both companies state this in so many words, I'd say the issue is pretty much settled.

(I worked in the semiconductor industry back in the 90s.)

Re:Benchmarks, accuracy, and choice

[et764]

To Apple's credit, I've been impressed that they advertise the MacBook as 4x faster on their web site, but the charts they show there show that on one of the tests it's actually over 5x faster. I would have expected most companies to advertise the 5x more than the 4x.

Re:Benchmarks, accuracy, and choice

[squiggleslash]

Steve Jobs wasn't "singled out" for this, he just happened to be the subject when the issue was raised.

The fact is that lies and deception are (a) common and (b) shouldn't be. It is completely legitimate to (a) criticise liars and decievers, even if you haven't done so and (b) (as the submitter did) highlight lies and deceptions to reduce their impact. If Joe Blow tells me that Blow Inc's Widgets are "99% plutonium free", by which he means that 1% of each widget is plutonium, rather than that they contain 100 times less plutonium than his competitors as would be the most likely initial reading, I'll tell people. I'll not sit there and say "Hey, there are probably a lot of people who think that his widgets contain less plutonium than everyone else's" simply because of some misplaced "moral code" that says I shouldn't be seen to be apparently critical of someone for doing something that a lot of other people do.

The Intel iMac isn't significantly faster than the iMac G5, despite Job's attempts to imply otherwise. Indeed, if the iMac G5 had undergone the same revisions that the PowerMac line had a few months ago, the chances are they'd be faster than the Pentium equivalents. I've written in my journal that I really don't understand why people reacted with such enthusiasm to the Stevenote this year: read between the lines Jobs spoke and you find that Apple actually was struggling to come up with convincing evidence that justifies the Intel switch. Nonetheless, Jobs used wording that implied massive (2-4X!) improvements in performance, which, surprise surprise, are clearly massive exaggerations and, in the context of comparing this year's Pentiums to last year's G5s, are actually covers for practical failure on their side.

If you're going to use wording that implies massive (and 2-4X really is massive) improvements in performance, don't expect your honesty and virtue to be unquestioned when we subsequently find that no such improvements have actually occurred. If Michael Dell does the same thing, we'll criticise him too. If Intel's CEO does, we'll do the same thing. Today it's Steve Jobs.

Re:Benchmarks, accuracy, and choice

[Tim C]

The fact of the matter is that several intel products (the P4 anyone?) were notorious running benchmarks and not much else.

That can be clearly seen in the mother of all CPU charts on Tom's Hardware. Generally, and especially in the gaming benchmarks, the AMD chips take most of the top spots - apart from some of the synthetic benchmarks (3DMark05 Futuremark especially), in which the first dozen or so places go to Intel chips.

Se for example the 3dMark Futuremark, which Intel utterly owns, and the Far Cry or UT 2004 benchmarks, which AMD dominate. (The Doom 3 benchmarks are a little more neck-and-neck)

Re:Benchmarks, accuracy, and choice

[timeOday]

Is your point really that Steve isn't lying, or rather that we shouldn't expect the truth?

Re:Benchmarks, accuracy, and choice

[tpgp]

I think you've missed his point. This is a common industry practice used for just about every piece of hardware and software on the market. To single Steve Jobs out for this practice rather than accepting it as the "norm" shows a distinct anti-Mac bias.

Horse crap. Common industry practice or not, I think most slashdotters will call bullshit to these sort of claims whether it comes from Steve Ballmer, Steve Jobs or Linus Torvalds.

It's hard to be deceitful when it comes to something as nebulous as benchmarks

Well I don't know about that - seems pretty easy to be deceitful and called for it if you ask me.

Re:Benchmarks, accuracy, and choice

[AKAImBatman]

EvilTwinSkippy is correct. AMD is practicing just as much deception with this scheme as the rest of the industy. I'd remind you that it was AMD who used started the MHz war by consistently clocking their CPUs higher than Intel's. Intel responded with the Pentium IV which was able to clock higher than the AMD chips even if the performance wasn't better. AMD responded with their "speed rating system" which is intended to trick consumers into believing that they are getting a faster clock rate than they are actually getting.

For example, an Athlon XP-M 2400+ is clocked at 1800 MHz, not 2400 as one would expect. The AMD Athlon XP-M 3000+ is clocked at 2200 MHz rather than 3000 MHz. Even more deceivingly, there is another chip called Athlon XP-M 3000+ ("Dublin") which is only clocked at 1600 MHz! How am I, as a consumer, supposed to sort all this nonsense out?

So you can keep your comments about "conservative" ratings. The industry is all about deceptive practices, and (most likely) always will be. Learn to deal with it.

Re:Benchmarks, accuracy, and choice

[ozydingo]

Please think about that. If they showed an average, and that average was better than the peak, then....wait, does that make an sense?

I'm sure the average was better than some specific benchmark runs. But that necessarily implies that it was worse than some other runs. Which means there would have been a way to skew the results more in their favor by picking and choosing which data to consider relevant. Which is what Apple did, and should not be accepted as "ok" just because many companies do it too.

Re:Benchmarks, accuracy, and choice

[Andy Dodd]

AMD set about dispelling the "myth of the megahertz", but they did it in a reasonable fashion. Let's face it - clock rate isn't everything (the Pentium 4 proves that without a doubt), and the public needs to stop thinking that the clockrate of the CPU is important. Yes, the AMDs are clocked lower. Despite that, they routinely kill P4s clocked at over 1.5x their clock rate in nearly all applications. Cases where Intel wins are the rare exception and not the norm. Hell, even Intel has had to move away from publishing actual clock rate in preparation for the Netburst architecture's imminent demise.

Average performance on a wide variety of applications is an excellent performance indicator. Raw clock rate and peak performance on a single app (the former being a favorite of Intel and the latter being classic Apple) are both crappy methods of measuring performance.

Re:Benchmarks, accuracy, and choice

[Tibor the Hun]

I think what Steve said is accurate in its context, but their advertising is very misleading. The 2x and 4x are taken out of their context and applied to broader terms of overall performance. They do not mention that 2x and 4x are related to specific benchmarks which do not directly translate to performance.

This is similar to MS claiming, in their install screens, that Windows XP will allow you to be more efficient, and that it is faster and more secure.
They don't say what it is faster and more secure than. (Except "more than ever".)

Those stats are taken out of their context and are misleading.

Now, who can correctly classify me?
I love PowerPC macs for their Firewire support (Booting, integration, etc.) and the multimedia capabilities of PowerPC.
A quad G5 will be quite a beast for quite some time, in multimedia circles, I believe.

Re:Benchmarks, accuracy, and choice

[Basehart]

"I don't think you understand what parent said"

Who cares what the parent said! I never listen to them anyway (especially now I'm in my 40's :-)

Re:Benchmarks, accuracy, and choice

[Otter]

glxgears is a benchmark for graphics acceleration. You don't (unless you're insane) sit there and watch it spin at 1908 fps.

Re:Benchmarks, accuracy, and choice

[Cipster]

100% accurate yes but 100% deceptive too.

Re:Benchmarks, accuracy, and choice

[pi42]

This isn't entirely accurate -- most people can tell the difference between 25fps and 60fps in a computer game, while they can't on a film or video. I believe 3dfx released an infamous demo back in the day with a spinning cube at 30fps and 60fps and you could always tell the difference.

What's the difference? Video and film have motion blur, which makes for smooth transitions between frames whereas games display things in discrete frames with no blur whatsoever.

Ever tried waving your hand underneath a strobe light going at 30 cycles/sec? That's 30fps yet the motion still looks strange, since it's like you're seeing discrete frames and not continous motion burred between frames.

Re:Benchmarks, accuracy, and choice

[Yahweh Doesn't Exist]

>Many people notice no difference over about 25 fps.

I find that in pool games where you have balls moving in one direction and spinning in another it's easy to tell the difference between 40fps and 60fps

Re:Benchmarks, accuracy, and choice

[Thrudheim]

Actually, if you listen to the Macworld keynote, Jobs said something to the effect that "of course, the other components like the drives, memory, etc. are not 2x-3x faster." He was clear that many other things will factor into real-world performance. Before you pillory him, put his comments in context.

There are stronger grounds for critique in the kind of benchmarks that Jobs used: ones where dual-core chips naturally do much better. On the other hand, the presence of dual-core chips does produce major speed increases for certain tasks, as the Ars Technica review shows, such as ripping a CD (which the Intel iMac finished in less than half the time of the iMac G5).

Re:Benchmarks, accuracy, and choice

[Milton Waddams]

Nonetheless, Jobs used wording that implied massive (2-4X!) improvements in performance, which, surprise surprise, are clearly massive exaggerations and, in the context of comparing this year's Pentiums to last year's G5s, are actually covers for practical failure on their side.

Of course, the big boon that the Intel chips will bring will be to the Apple laptops and Mac Mini. If these benchmarks are anything to go by, benchmarks comparing the G4 powered PowerBook to the Intel Core Duo powered MacBook Pro should show a huge difference. Although, we'll have to wait and see.

Re:Benchmarks, accuracy, and choice

[AKAImBatman]

AMD set about dispelling the "myth of the megahertz", but they did it in a reasonable fashion.

AMD set about misleading the public based on the "myth of the megahertz" that they helped create. Whether you find it "reasonable" or not is irrelevant. It's still a deceptive practice used to combat another deceptive practice that was a response to the original deceptive practice that AMD started.

That being said, I appreciate the AMD64 performance lead quite a bit. However, I can accept the bad with good without a need to rationalize their marketing department. (Is such a thing possible, anyway? I thought marketing departments were inherently irrational? ;-))

Re:Benchmarks, accuracy, and choice

[dborod]

According to this inforworld article http://weblog.infoworld.com/enterprisemac/archives /2006/01/the_mac_perform.html Apple chose to use benchmarks optimized for dual core CPUs and then compared the new Mac's results against single core G5 Macs... seems kind a fishy to me.

Re:Benchmarks, accuracy, and choice

[rwhiffen]

I think it gets a little more fuzzy than you make it out to be. The Steve was also talking about the 20" iMac, which runs at 2Ghz, not the 1.83Ghz Ars tested. So unless you know that the 1.8Ghz G5 vs 2.0ghz G5 difference is the same as the 1.83Ghz intel vs 2.0Ghz intel, it's difficult to make any claims in absence of facts.

Re:Benchmarks, accuracy, and choice

[djohnsto]

While the dev systems uses intel GMA900 integrated graphics, both the MacBook and new iMac use ATI Radeon X1600 chips.

The iBook and mini may use integrated graphics, but they will probably use newer chipsets with graphics faster than the GMA900.

Re:Benchmarks, accuracy, and choice

[Thrudheim]

"Apple zealots don't like thinking . . . "

Well, there's a perfectly inane comment. Try some thinking yourself.

You really think Jobs made up the whole 3 GHz thing years ago just to have an excuse to move to Intel? Right. It's always a great idea for a CEO to might a high-profile pledge and fail to deliver for two years running.

Look, the G5 is a nice chip. There are two of them in the box on my desk and I am one happy camper. They are friggin' hot, however. Speed increases required the addition of liquid cooling systems. These things are not going into laptops, period, and it was laptops that drove this whole move. Apple is too small a player for IBM to devote the R&D necessary to keep up development. The game consoles are where it is at for IBM right now. The PowerMacs will be the last to move to Intel for a good reason: they are served very well by the G5 for the moment.

In the x86 world, Apple can count on being able to run the same chips that every other machine is running (including AMD). That assurance of supply relieves a major worry for Apple.

Re:Benchmarks, accuracy, and choice

[Heembo]

How do you benchmark a laptop that has almost no native software released for it yet? This is all very theoretical. I'm waiting for a real test with side-by-side machines running the same apps.

Re:Benchmarks, accuracy, and choice

[spectral]

http://slashdot.org/comments.pl?sid=173521&cid=144 36559

???

Re:Benchmarks, accuracy, and choice

[digitallife]

Hehe. I don't like the incredibly rampant lieing which our society has embraced either, but by now you must realize that pretty much anything a representative of a corporation says is at least partially a lie. If you look at advertising, the lie rate rises closer to 100%. Fun.

Re:Benchmarks, accuracy, and choice

[tkrw]

Not only were they 100% accurate, he also qualified them. Had /. members actually watched the keynote, which, BTW, is available on their website, he indicated that those were benchmarks on the processor and wouldn't necessarily in fact "double the computers speed."

Heresay is evil. Investigate before you hate!

Re:Benchmarks, accuracy, and choice

[wasabifan]

I know the linked article doesn't say so, but for anybody that saw or remember what Jobs actually said in the keynote was that the benchmark showed the cpu was 2 to 4 x faster. He said actual performace would not be that drastic as the processor is only a part of the real world speed equation. The new computers are still limitted by slower disk access and even RAM to some extent. It was only misleading to people who don't understand want SPEC (int and float) is.

To be honest, I thought Jobs was going to be more deceptive, because they always used to tout how PPC beat Intel easily in floating point. The first thing he showed this time was integer (where Intel always beat PPC anyway). I was impressed that float actually was about 2x faster on the new Intel than PPC. Now granted that is most likely due to the Intel chip being dual core, but the fact that float speeds per GHz, per core are about even now, goes to show the work that went into this new Intel chip. That is definately better than the old Intel days.

Re:Benchmarks, accuracy, and choice

[MrCam]

Also the G5 had 1GB of memory and the intel had 512MB

Re:Benchmarks, accuracy, and choice

[bnenning]

The Intel iMac isn't significantly faster than the iMac G5

There's not nearly enough evidence to reach a conclusion either way. QuickTime export is one of Altivec's strongest areas, and Xbench scores are notoriously bad at having any relationship to reality. Let's wait and see how they do in real life; perhaps you'll find Apple really does have a clue.

Indeed, if the iMac G5 had undergone the same revisions that the PowerMac line had a few months ago, the chances are they'd be faster than the Pentium equivalents.

Yeah, all they had to do was put quad G5s in the iMac and Powerbook enclosures...

I really don't understand why people reacted with such enthusiasm to the Stevenote this year:

Because we get faster Macs for the same cost, and in the case of Powerbooks, *much* faster. Desktop PPC is dead; deal with it. Yes, it's a fundamentally better architecture, but IBM and Freescale have repeatedly shown they have no interest in developing products suitable for Apple unless they're bribed heavily. I'd much rather have Apple focus on their software.

Re:Benchmarks, accuracy, and choice

[Minstrel Boy]

And in fact, the memory, or at least access to it, *is* 2x as fast - the FSB speed doubled.

KeS

Re:Benchmarks, accuracy, and choice

[diegocgteleline.es]

"Apple zealots don't like thinking . . . "

Well, there's a perfectly inane comment. Try some thinking yourself.

Hey, that was a perfect definition. Zealots, by definition, _are_ stupid. I never said "all the apple users" or something like that...

Re:Benchmarks, accuracy, and choice

[Andy Dodd]

How, pray tell, did AMD invent the "Myth of the Megahertz"?

Consistently, since even the early 486 and Pentium days, AMD (and in fact also Cyrix) CPUs routinely beat Intel CPUs running at somewhat higher clock rates. With a small exception near the end of the Pentium III's lifetime and prior to the introduction of the Pentium 4, AMD CPUs have almost never been available with equivalent or higher raw clock rates than Intel's finest. They HAVE been available with performance matching or beating Intel's finest in many cases, but AMD has rarely ever been the winner in terms of raw clock rate. The one time they were, Intel's Netburst (Pentium 4) architecture was already well along in the development phase, Netburst was NOT a direct response to raw clock rates from AMD.

Intel CPUs have consistently been near the bottom of the barrel in terms of performance per clock cycle, with the exception of specialized low-cost/low-power CPUs.

Re:Benchmarks, accuracy, and choice

[trevor-ds]

Important differences between the keynote benchmarks and those shown by ArsTechnica:

• The keynote benchmarks were compiled using Intel's C compiler, which tends to produce much faster code than gcc. To be fair, Apple compiled the G5 code with the IBM compiler. I assume xBench was compiled with gcc on both platforms.
• The QuickTime test almost certainly uses hand-tuned Altivec instructions on the G4/G5. There's no evidence yet that the Intel code is tuned for SSE.

Even though ArsTechnica likely doesn't have access to the Intel compiler, I was surprised to see that they didn't try compiling any code themselves to benchmark with. That would have been the best way to know exactly what variables are under control here. There are plenty of applications (encryption, rendering, etc.) that have optimized open source implementations; these seem like the best place to start.

How can it be manipulative to quote performance numbers based on the SPEC benchmarks, which are the industry standard?

Re:Benchmarks, accuracy, and choice

[destinationmoon]

Yes, but the Intel iMac actually uses an ATI Mobility Radeon X1600...

http://www.apple.com/imac/whatsinside.html

Re:Benchmarks, accuracy, and choice

[Halo1]

The QuickTime test almost certainly uses hand-tuned Altivec instructions on the G4/G5. There's no evidence yet that the Intel code is tuned for SSE.

It is. It already was when Quicktime/x86 only existed for Windows. And the OS' core libraries (like the Accelerate framework) are also already optimized with SSE3 wherever possible. Source: Apple's performance optimization mailing list.

You actually missed the only really important difference: Steve's benchmark, namely spec_rate, is a benchmark which automatically spawns as many instances of itself as you have cpu's/cores, and then adds up the scores. So if you compare a single core G5 to a dual core Yonah and the latter obtains twice the score of the former...

At best that demonstrated that Yonah is speed-competitive with the G5, although possibly while using less power.

Re:Benchmarks, accuracy, and choice

[AKAImBatman]

Consistently, since even the early 486 and Pentium days, AMD (and in fact also Cyrix) CPUs routinely beat Intel CPUs running at somewhat higher clock rates.

Too far back in history, chief. In the 486 days, the AMD and Cyrixes were nothing more than a "cheap" upgrade for a 386. They simply didn't compete. Consumers were thus able to somewhat trust the MHz rating of Intel chips as a general performance indicator.

When the Pentium era arrived, AMD was still not a competitor, but they did manage to produce chips that were "good enough" to be considered cheap alternatives to the Intel processors. As a result, the PC industry did start producing AMD-based machines for their "low cost" product lines.

When the Athlon and PIII came to market, however, is when things got interesting. For the first time, AMD managed to put out an extremely capable chip in comparison to Intel's offerings. But far more interesting was that AMD started ramping their chips to exceptionally high MHz levels to pass up Intel's chips. This practice gave the AMD chips a reputation for high performance, but extreme heat levels and lower reliability. This left Intel with the server market as their chips proved more reliable over the long haul, and performed just as well in most non-gaming situations.

Long story short, Intel and AMD traded various blows in performance, each one gaining a slight lead for a short time only to be quickly shown up by their competitor. AMD, however, managed to keep the MHz trophy the entire time.

Intel got the bright idea to beat AMD at its own game and thus produced the Pentium IV chip. Now the PIV isn't a very good chip, but it can be ramped up in MHz in ways that AMD's Athlons couldn't. AMD struggled for awhile, but quickly realized that they could no longer win the MHz trophy. So they came up with a Jedi Marketing Trick to make consumers think that AMD's chips were running at the same clock levels. That trick was to assign an arbitrarily created "speed rating" that placed the direct competitors to Intel's chips right next to the Intel MHz rating. That way reviewers would pick up the slower AMD chip and compare it with the Intel chip, rather than look at the real MHz.

AMD got a lot of bad press for the decision, but it did eventually pay off. Consumers accepted the PR rating as the "real speed" of the chip, and Intel again started losing market share. AMD finally ripped the market away from Intel with their AMD64 platform, which proved to be much more popular with consumers than Intel's own Itanium line.

So to summarize, AMD started competing with Intel. Knowing that Intel customers used MHz to judge performance, they ramped up their chips to extreme levels. Intel responded with their PIV Northwood core and took the MHz trophy back. AMD got smart and skewed the market by printing a number on the box that wasn't actually a MHz rating, while convincing many consumers that it was.

Clear as mud?

Re:Benchmarks, accuracy, and choice

[DrXym]

My point is that the story write-up makes it sound like SJ is lying, and he's not. He's just presenting the best set of benchmarks he can, which is pretty much what I expect from the CEO of the company...

Proudly stating in large graphics all over the spiele that your new machines is "2x faster" when it isn't *is* lying. Or does Steve Jobs now think all his customers are so tech-literate as to read the small-print where the 2x faster boast actually only applies when running a specific benchmark test and not indicative at all of its actual real-world performance?

Re:Benchmarks, accuracy, and choice

[squiggleslash]

Yes and no. I'm still underwhelmed. My journal explains why, but in essense, the Powerbook G5 does not exist because IBM had no reason to develop a low power version after Apple's announcement of a switch to ix86. We have absolutely no way of knowing what the performance of a Powerbook based upon a modern processor would have been.

And to continue the thread of dishonesty and misrepresentation, this entire issue was another example. Jobs went up on stage with a graph that supposedly showed how crap the G5 was in a laptop. His graph compared the low power version of the G4, the low-power laptop-oriented Pentiums used in the MacBooks, and... the desktop version of the G5.

And gosh, didn't the desktop G5 look awful compared to the laptop processors. What a big fucking surprise! As if to add insult to injury, the graph was bullshit. I mean, seriously bullshit. One axis had "Performance per Watt" as an axis (no units, no actual mathematical or physical mapping to the real world), and it had numbers on it. And no, none of the numbers were "1".

So yeah, I'd imagine the MacBook will be faster than the Powerbook G4. Hey, I bet it's going to be better than the Mac 128k too! Whoop!

That, in some ways is a secondary issue. Yes, the MacBook almost certainly represents a huge improvement on the existing Powerbook range. But I wonder if it really is "4X", and I have my doubts, severe doubts, that this in any way justifies the switch.

Re:Benchmarks, accuracy, and choice

[evilviper]

I'd remind you that it was AMD who used started the MHz war by consistently clocking their CPUs higher than Intel's.

You've got to be joking...

You're saying that AMD, having a superior product, started the MHz war (just a few years ago) forcing Intel to introduce the underperforming mess that is the P4?

First of all, The MHz numbers for AMD CPUs have never been over-inflated as far as I can remember, as Intel did with the P4. Athlons had a higher MHz rating because they were legitimately faster (although at the expense of being hotter).

But besides all of that, the MHz wars didn't start when AMD finally started to compete... They've been going back through the introduction of 500MHz Alphas (when Pentiums were at 100MHz), at least since the start of Cyrix, and probably even pre-dating that.

AMD responded with their "speed rating system" which is intended to trick consumers into believing that they are getting a faster clock rate than they are actually getting.

No, actually it's intended to provide a reasonable comparison to Pentium 4s, despite the disparity in MHz that Intel caused. What Intel did seems to be the attempt to cause confusion.

AMD could have been dishonest with their ratings, and used numbers much larger than the MHz of comparable Intel CPUs, but they didn't do that... They've actually kept their ratings quite honest.

How am I, as a consumer, supposed to sort all this nonsense out?

What is there to sort-out? It's pretty straight-forward. If they just listed the MHz rating of their CPUs, then you'd really have to work to figure out a comparison between AMD and Intel CPUs.

Re:Benchmarks, accuracy, and choice

[ottffssent]

Truthful, yes. Honest, no.

Not that anyone expected anything different. To be fair, nobody's completely honest with benchmarks, but Apple's among the worst offenders. That said, the Intelmacs are only hideously slow in comparison to the PPCmacs on CPU-bound emulated code, which is to be expected, and is the case with the best chance for improvement in the future. And they actually manage to win a few of the benchmarks, which is nice.

I'd like to see linux running on the new machines. OSX is going to be slow on X86 if for no other reason than the Apple programmers aren't used to the architecture yet. Linux should be a lot closer to best-case speed, so it would be useful to see approximately how much faster OSX can get on the same hardware. I'm not a compiler guy, but my understanding was that GCC still has improvements to be made with regard to newer CPUs and multicore CPUs which should help OSX on Intelmacs as well.

Based on my completely unsubstantiated memory of the M68K -> PPC switchover, Apple's learned some things. It looks like this processor switch is going to go a lot more smoothly than the last one did, which can only mean good things.

Re:Benchmarks, accuracy, and choice

[Glock27]

You actually missed the only really important difference: Steve's benchmark, namely spec_rate, is a benchmark which automatically spawns as many instances of itself as you have cpu's/cores, and then adds up the scores. So if you compare a single core G5 to a dual core Yonah and the latter obtains twice the score of the former...

At best that demonstrated that Yonah is speed-competitive with the G5, although possibly while using less power.

No, at best that demonstrated that each Yonah core is speed-competitive with the G5. Since the G5 has one core, and Yonah has two, that makes Yonah...twice as "fast". (Really only twice as fast on perfectly parallel code, but hey...)

Also, the dual-core Yonah uses something like 1/4 to 1/2 the power of the single-core G5. Pretty nice!

I'm looking forward to getting a Macbook Pro as soon as possible.

Re:Benchmarks, accuracy, and choice

[Bun]

not to mention the intel GMA900 graphics processor isn't exactly the speediest thing around.

Don't mention it, then. Especially since the iMac Core Duo uses a PCI-Express ATI Radeon X1600.

Re:Benchmarks, accuracy, and choice

[Thrudheim]

Dimwitted and condescending, eh? That's a good one. Perhaps you can enlighten us, but it would help if you would start by using terms like "faint praise" correctly.

Yes, the G5 is "nice" for lack of a more technical term. I am very pleased with how much I can throw at my G5 dualie without it needing to strain. So, no, that is not faint praise at all. Not sure how that is condescending, dimwitted, or pathetic either. What is your point? Just to call out insults?

Re:Benchmarks, accuracy, and choice

[Anonymous Coward]

You're fabricating all of this.
Let's see. AMD creates the K5, and coins the PR rating precisely to fight the "MHz Myth." The K5 while having a superior, more modern design with better integer performance had an inferior FPU than the Pentium. This would plague AMD until the K7. The K6 line (K6-2, K6-3) carried with it the same superior integer performance, dropped the PR rating (there was a lot of backlash at the PR rating because of the "MHz Myth") in favor of its clockspeed, that never surpassed the Pentium 2. Then came the K7, and the K7 was clock-per-clock faster than the Pentium 3 up through Coppermine. AMD's clockrate outpaced Intel's simply because their architecture was better. With NetBurst Intel wanted to have high clockrates (not for the purposes of the MHz Myth, but that certainly helped) but for streaming performance. The NetBurst architecture was designed under the idea that Intel could produce 5GHz+ Pentium 4 processors with high-bandwidth RDRAM. The sacked FPU performance, created a huge pipeline, and cranked up clock speeds. Intel took the MHz Myth to the next dimension while trying to maximize throughput. This forced the K7 back into using PR ratings for Athlon XP and forward. When Intel couldn't increase their clockrate any further, what did they do? They increased the pipeline length and increased the clock speed. Making the "MHz Myth" all the worse. Ironically Intel has fallen victim to its own failed architecture and opted using PR ratings for their own new processors.

Re:Benchmarks, accuracy, and choice

[Sebastopol]

Yes, but Nvidia or ATI components would require more cooling.

Re:Benchmarks, accuracy, and choice

[ParisTG]

AMD was never really in a position where their chips were the clear MHz winners. Up until the K6/K6-2/K6-3 series, they were always lagging in MHz. The Althon changed that, and for the first time they could out perform the Intel equivalents.

There were brief moments where AMD did have the MHz advantage with the Athlon, but the advantage swapped often between Intel and AMD. Once the P4 was released, that MHz advantage disappeared forever... however the AMD chips could STILL compete in performance, in most cases, even at a lower frequency.

This timeline should clarify this somewhat.

Re:Benchmarks, accuracy, and choice

[tkrw]

Keynote address, 1:08:32, quoting:

"In the most important benchmarks of performance (Integer and Floating Point), 2 to 3x. Now everything isn't going to run 2 to 3x, you know the disks aren't 2 to 3x faster, etc.... But in the most important benchmarks, 2 to 3x."

Instead of doing the research for this post, you instead have posted hearsay trying to get the whole group up in arms over a story based on half-truths. Did you even watch the keynote? What is "muddy" about what he said? Next time, before you post this bait, at least have your story complete so you know where you stand when trying to defend it.

Re:Benchmarks, accuracy, and choice

[znu]

You can tell the difference with film or video, it's just a little more subtle. A lot of people actually prefer the 24 fps look for narrative works. There are all sorts of theories about why this is the case. Some people claim it looks less realistic, which causes the audience to suspend disbelief more easily. My guess is it's just that people have come to associate the 24 fps look with movies (usually high-quality production), and the 60 fps look with television (often 'extruded consumer product'). So, they see the 24 fps look and immediately think "Hey, this must be good."

Of course, you do need the motion blur, or 24 fps becomes unwatchable pretty fast. But modern video cards support motion blur. Sooner or later (maybe after we've got somewhat more photorealistic rendering), I'm sure someone is going to make a game that deliberately runs at 24 fps (with motion blur) to try to get that 'cinematic look'.

Re:Benchmarks, accuracy, and choice

[Achromatic1978]

It's hard to be deceitful when it comes to something as nebulous as benchmarks. Every benchmark you run will tell a different story. The result is that you can pull a variety of different conclusions from the benchmarks depending on how you spin it. Given that Steve Jobs is the CEO of Apple, we can expect that he will spin the benchmarks positively.

And others get bitchslapped for it, so why should Steve be immune?

Re:Benchmarks, accuracy, and choice

[Thrudheim]

But people were waiting for a PowerBook G5 for well over a year before Apple announced the switch to Intel. It seems much more likely that Apple pushed IBM hard about getting a low-power G5 and it just did not materialize. IBM had weak incentives to develop the low-power G5 because Apple is too small a customer to justify the R&D spending necessary. The game console market, in contrast, appeared much more lucrative.

See the comments from Freescale's CEO. He was at IBM and sold the G5 to Apple. As he says " . . . then IBM decided not to take the G5 into the laptop and decided to really focus its chip business on the game consoles."

http://news.com.com/Is+the+PowerPC+due+for+a+secon d+wind+-+page+2/2008-1006_3-5983157-2.html?tag=st. next

Re:Benchmarks, accuracy, and choice

[sgt scrub]

I'd remind you that it was AMD who used started the MHz war by consistently clocking their CPUs higher than Intel's.

Not true. The MHz wars were going when AMD was making chips for Intel and Intel was competing with DEC.

How am I, as a consumer, supposed to sort all this nonsense out?

Toms Hardware, HardOC, ExtremeTech, AnandTech, and FiringSquad have good comparison articles. I'm partial to Tom's and FireingSquad but...

Re:Benchmarks, accuracy, and choice

[Midnight Thunder]

This is only the first generation of Intel Macs. The Intel chip is faster for some things and slower for others. One thing that the PowerPC chip was the Altivec unit which did help it in certain multimedia applications. I wonder whether Apple has the rights to get it adapted to the Intel chip. Something else that is likely to change is the performance of Rosetta. Apple has a history of starting off with slowish products and then trying everything to improve the performance.

Re:Benchmarks, accuracy, and choice

[poopdeville]

Please think about that. If they showed an average, and that average was better than the peak, then....wait, does that make an sense?

Uhm... you misread that. Benchmarks are used for comparisons. The GP's point was that AMD's average benchmark was better than Intel's. Hence AMD chose to use the average benchmark in its marketing.

Re:Benchmarks, accuracy, and choice

[gnasher719]

'' There's not nearly enough evidence to reach a conclusion either way. QuickTime export is one of Altivec's strongest areas, and Xbench scores are notoriously bad at having any relationship to reality. Let's wait and see how they do in real life; perhaps you'll find Apple really does have a clue. ''

One of the XBench tests where the iMac x86 gets "slaughtered" is the "User Interface" test.

It turns out that the iMac x86 runs this test at 67 frames per second. Which is quite consistent with some newer Apple technotes that tell you that screen updates are now coupled with the monitor's refresh rate. If you draw more frames per second then the monitor can display, you are just wasting your time. Seems that the other Macs tested run this test at several hundred frames per second.

Re:Benchmarks, accuracy, and choice

[Millenniumman]

That's simply not true. If that person knows that the fps is 60 then they will imagine it to be a lot better than 25 fps. Unless they are under the impression that less fps is good, in which case it will be opposite. And someone who thinks it is 1900 fps will imagine it to be really good.

Re:Benchmarks, accuracy, and choice

[drsmithy]

You're saying that AMD, having a superior product, started the MHz war (just a few years ago) forcing Intel to introduce the underperforming mess that is the P4?

For most of its life, the P4 wasn't "underperforming", it was as fast - if not faster - than the Athlon.

First of all, The MHz numbers for AMD CPUs have never been over-inflated as far as I can remember, as Intel did with the P4.

Uh, you can't "over inflate" a clock speed. It is what it is. "PR" numbers, OTOH...

Athlons had a higher MHz rating because they were legitimately faster (although at the expense of being hotter).

P4s were as fast - if not faster than Athlons. The fact that they used a higher clock speed to do this is irrelevant.

It always amazes me the number of people who say "clock speed doesn't matter" then immediately turn around (often in the same sentence !) and say "$CPU is faster than the Pentium 4 at the same Mhz".

No, actually it's intended to provide a reasonable comparison to Pentium 4s, despite the disparity in MHz that Intel caused.

The PR rating is a number that relates to the original Athlon, not the P4.

What Intel did seems to be the attempt to cause confusion.

Huh ? Intel labelled their chips with the clock speed and processor type, AMD uses an approximate number relating a chips performance to the original Athlon, but it's *Intel* attempting to cause confusion ?

Re:Benchmarks, accuracy, and choice

[GaryPatterson]

Jobs can't win.

He never claimed the new iMacs were 2-4 times as fast. Watch the keynote. He claimed that on the SPEC scores, which he said were key indicators of performance, they were 2-4 times as fast.

He then went on to say that the speed improvements won't be across the whole system, because other components (he singled out hard drives) aren't improved over the G5 models.

I say that he can't win because for years he put up Photoshop numbers, and many people around here said "show us the SPEC numbers!"

Now he puts up SPEC numbers and people miss the bit where he said that they weren't the whole story.

He didn't lie. He told the full picture, but so many people haven't watched the keynote and are rushing to judgement based on second-hand information.

Re:Benchmarks, accuracy, and choice

[SteeldrivingJon]

"the Powerbook G5 does not exist because IBM had no reason to develop a low power version after Apple's announcement of a switch to ix86."

So, basically, what you're saying is that as of mid-2005, IBM hadn't even pulled their thumb out of their butt and started *developing* a laptop G5? That would imply that a G5 powerbook would be at least 18 months out, maybe two years. IBM isn't going to pop out a laptop-capable G5 in a few months of 'development'.

If true, that would be pathetic, and Apple would be pathetic for remaining with IBM after such a failure to deliver.

However, the problem is actually that IBM was unable and unwilling to develop a low-power version in *2004*, not mid-2005. For IBM to start development in mid-05 would be a textbook case of too little, too late.

Re:Benchmarks, accuracy, and choice

[SteeldrivingJon]

"It turns out that the iMac x86 runs this test at 67 frames per second. Which is quite consistent with some newer Apple technotes that tell you that screen updates are now coupled with the monitor's refresh rate. If you draw more frames per second then the monitor can display, you are just wasting your time. Seems that the other Macs tested run this test at several hundred frames per second."

That was added in 10.4, on PPC too, so it should effect both architectures - UNLESS the PPC version of the benchmark was built using 10.3, rather than being built on 10.4. Applications linked with versions 10.4 don't get the updated behavior.

The benchmark ought to be run again on Intel after using QuartzDebug to "Disable Beam Synchronization".

Something does seem to be screwy about that benchmark. Subjective reports about the Intel Macs have typically described the UI as being quite fast.

Re:Benchmarks, accuracy, and choice

[SteeldrivingJon]

Whoops, "linked with versions 10.4" should be "linked with versions before 10.4"

Re:Benchmarks, accuracy, and choice

[mdwh2]

To be fair, Steve's statements were absolutely 100% accurate (assuming the figures are accurate, which I expect them to be). For that benchmark, the intel machine is 2x-3x faster.

And I remember the days when they hand-picked the benchmarks to show the PowerPC being 2x faster than Intel...

Re:Benchmarks, accuracy, and choice

[Scudsucker]

For most of its life, the P4 wasn't "underperforming", it was as fast - if not faster - than the Athlon.

Maybe if your benchmark is a P4 friendly application such as Quake 3. Otherwise, no. The P4 made a big tradeoff between clock-for-clock speed in exchange for ghz. A 1.3 ghz P3 was faster than a 1.4 ghz P4, which is why Intel waited until faster P4's were available for release the last P3.

Re:Benchmarks, accuracy, and choice

[Space cowboy]

[sigh] It would be foolish beyond belief to quote non-reproducible benchmarks that are part of a standard set. That's why I believe the figures. I've never had occasion to disbelieve the figures on SPEC benchmarks from any of the major computer manufacturers. You can bet that MS will duplicate the tests and would cry foul if they were markedly different...

So, no, he wasn't lying, and he wasn't being deliberately deceitful (that part was where - immediately after showing the figures - he said "you may not see this in real-world results"). I would say that was remarkably honest from a company CEO, actually...

It ain't me who's biased/blinkered here...

Simon.

Re:Benchmarks, accuracy, and choice

[Halo1]

Also, the dual-core Yonah uses something like 1/4 to 1/2 the power of the single-core G5. Pretty nice!

According to Anandtech, a 2GHz Yonah consumes between 92 watts (idle) and 108 watts (fully loaded). That's about double of the original 130nm single core G5's and 4 times as much as the 90nm single core G5's.

A dual core G5 at 1.6GHz consumes 16 watts (typical usage). Given that power consumption goes up quadratically in relation to the frequency, this comes to about 23 watts for 2.0 GHz (typically, max consumption can be (much) higher). It seems to me rather that the G5 consumes 1/2 to 1/4 of Yonah, instead of the other way round.

Don't ask me why Steve kept saying the G5 requires so much power, I don't know... Maybe it's one of the G5's chipsets that requires so much power (I do notice that the U3 chipset gets a lot hotter than either of my first generation 18 GHz G5's -- but afaik that's an Apple and not an IBM design).

No, at best that demonstrated that each Yonah core is speed-competitive with the G5. Since the G5 has one core, and Yonah has two, that makes Yonah...twice as "fast". (Really only twice as fast on perfectly parallel code, but hey...)

Oh come on, like the G5 was twice as fast as a single core P4? It doesn't work like that in practice.

I'm looking forward to getting a Macbook Pro as soon as possible.

I'm definitely going to wait until the second or third generation and until someone can tell me the battery life, because for the first time Apple does not mention battery life on its tech specs page. This suggests to me that battery life has become worse, even though they use a higher capacity battery for the MacBook Pro.

Re:Benchmarks, accuracy, and choice

[Halo1]

Small corrections:

18 GHz G5's

1.8 GHz G5's, obviously (it's a dual 1.8 GHz G5, with the G5's fabbed at 130 nm).

Oh come on, like the G5 was twice as fast as a single core P4?

... like the dual G5 was ...

Re:Benchmarks, accuracy, and choice

[steevc]

Ever tried waving your hand underneath a strobe light going at 30 cycles/sec? That's 30fps yet the motion still looks strange, since it's like you're seeing discrete frames and not continous motion burred between frames.

Who needs a strobe. Just wave you hand in front of your monitor. Just realised this only works on CRTs. I have an LCD on my desk as well that I checked with.

Re:Benchmarks, accuracy, and choice

[HuguesT]

The G5 is a cut-down version of the IBM POWER5 architecture, which IBM puts in their mainframe. The POWER5 is great and has been multi-core for a long time, but eats up power like there's no tomorrow (hence its name?). It's never been made to fit into a laptop. It was easy for IBM to come up with the G5 for desktop use, but very hard for them to design a laptop version.

On top of that IBM has a series of word-class fabs, but lately they've had serious problem converting to a 65nm process. They are still on the more power-hungry 90nm process.

Notice that Intel was *never* able to put together a serious laptop P4 version. All the P4 laptops were desktop replacements with a 45-minute battery life. Notice further that the same problem is befalling AMD. Even though the Athlon64 is not a power-hungry processor compared to the P4, the laptop versions don't compete well enough with the pentium-M, let alone the new Core-Duo.

To pull the Core-Duo feat, Intel essentially went one step back to the P-III and added all the recent power step-down features, the multi-core stuff and built it with the best fab in the business. Even then they can't produce it in volume yet.

To summarise, G5 essentially only sold to Apple. Apple is the only large entity interested in a laptop version. IBM crunched their numbers and came to the conclusion that developing a serious G5 laptop version would cost them dearly and return little. Embedded systems builders are still happy with the G3 and G4s, a G5 would be overkill at this point, essentially since embedded developers are not tied to any architecture.

IBM is retreating on that front, but notice they have scored a number of important hits with games consoles lately. The next round might be different. Which will be the processor to have in a laptop in 3 years' time? In all likelyhood IBM is preparing their POWER6, and that might have interesting power-consumption features. People in compute centres are tired of paying for air-conditionning, and large centres like Google's have repeatedly stated that they are not interested in raw compute numbers, but in electricity consumption.

Re:Benchmarks, accuracy, and choice

[aminorex]

If you spend your time in Photoshop, you'll want a PowerMac G5. It runs 5 to 10 times faster than the Intel system.

Re:Benchmarks, accuracy, and choice

[evilviper]

For most of its life, the P4 wasn't "underperforming", it was as fast - if not faster - than the Athlon.

Not true, except on a very, very narrow range of applications.

Uh, you can't "over inflate" a clock speed. It is what it is.

No, you can very easily over inflate a clock-speed. Just increase the pipeline, then the MHz are meaningless, because you then need many more of them to get the same performance.

"PR" numbers, OTOH...

You're somewhat correct. PR numbers could EASILY have been over inflated. That's why it's so significant that AMD has never done so, despite having nothing to stop them.

It always amazes me the number of people who say "clock speed doesn't matter" then immediately turn around (often in the same sentence !) and say "$CPU is faster than the Pentium 4 at the same Mhz".

That's funny, because I never said anything even remotely close to that... Not even REMOTELY.

The PR rating is a number that relates to the original Athlon, not the P4.

Of course that's all they can officially say... It's just an amazing, almost magical coincidence that they coincide so PERFECTLY with the P4's MHz ratings, and were introduced at just about the exact time that Pentium 4s came out... I hope someone told AMD about this...

http://www.amd.com/us-en/assets/content_type/white _papers_and_tech_docs/25426C_WP_FINAL.pdf
With the advent of the AMD AthlonTM processor and the Intel Pentium® 4 processor, the design architectures of these two companies fundamentally diverged. This design divergence has resulted in a difference in work done per clock cycle. Thus, microprocessors operating at identical frequencies may offer dramatically different levels of performance. Consequently, frequency is no longer the most meaningful metric for judging relative microprocessor performance. Today's end users need a better approach for comparing relative processor performance.

Re:Benchmarks, accuracy, and choice

[drsmithy]

The P4 made a big tradeoff between clock-for-clock speed in exchange for ghz. A 1.3 ghz P3 was faster than a 1.4 ghz P4, which is why Intel waited until faster P4's were available for release the last P3.

As I keep trying to say, clock-for-clock performance comparisons are meaningless.

Re:Benchmarks, accuracy, and choice

[drsmithy]

Not true, except on a very, very narrow range of applications.

I don't care enough to go chasing down dozens of benchmarks, but some highlights from a big timeline of processors here.:

October 2001, AMD introduces the Athlon XP 1800+. Fastest P4 at the time: 2Ghz.

March 2002, Athlon XP 2100+. P4: 2.2Ghz

June 2002, Athlon XP 2200+. P4: 2.53Ghz

November 2002, Athlon XP 2800+. P4: 3Ghz (HT)

June 2003, Athlon XP 3200+. P4: 3.2Ghz

The point here is that it's been a game of leapfrogging for most of the last few years.

No, you can very easily over inflate a clock-speed. Just increase the pipeline, then the MHz are meaningless, because you then need many more of them to get the same performance.

Mhz are always meaningless on their own. Something everyone seems to say, but no-one seems to understand.

You're somewhat correct. PR numbers could EASILY have been over inflated. That's why it's so significant that AMD has never done so, despite having nothing to stop them.

That's not strictly true. The PR numbers used for the old K5 chips were ...optimisitic... Which was why AMD coppsed so much flack for going back to it with the Athlon - people had bad memories of the K5 days.

That's funny, because I never said anything even remotely close to that... Not even REMOTELY.

Certainly seems to me you're implying P4s are slower when clock speeds are kept constant (or, as you prefer to put it, their "Mhz ratings").

Clock speeds cannot be compared between chips of different architectures. This means it's just as wrong to say an Athlon at 2Ghz is faster than a P4 at 2Ghz as it is to say a 2.4Ghz P4 is faster than a 2Ghz Athlon.

Re:Benchmarks, accuracy, and choice

[admactanium]

You can tell the difference with film or video, it's just a little more subtle. A lot of people actually prefer the 24 fps look for narrative works. There are all sorts of theories about why this is the case. Some people claim it looks less realistic, which causes the audience to suspend disbelief more easily. My guess is it's just that people have come to associate the 24 fps look with movies (usually high-quality production), and the 60 fps look with television (often 'extruded consumer product'). So, they see the 24 fps look and immediately think "Hey, this must be good."

i think it has more to do with the fact that 24fps is closer to how the human eye perceives motion. video, at 30fps, is rather unnatural-looking because it's TOO sharp. it's not a matter of what the human eye can resolve but more a function of how the brain perceives that information. the brain perceives a lot of motion blur where if you concentrated on the subject you could likely see it much more clearly.

i've worked on a number of commercials. some we chose to shoot at 30fps in film and most at 24fps (30 fps is useful in effects work because you avoid having to deal with pulldown frames). the 30fps spots, even shot on film, always looked a little too "video-y" and the parallax looks somewhat unnatural. a number of tv shows are going to 24fps since video cameras are now being produced with 24p.

Re:Benchmarks, accuracy, and choice

[evilviper]

The point here is that it's been a game of leapfrogging for most of the last few years.

I'm afraid you are mistaken. AMD hasn't gained so much market share by maintaining pairity, they did it by being consistently faster, as well as lower power and cheaper.

Mhz are always meaningless on their own.

Of course they are, but that doesn't change the fact that Intel was heavily marketing MHz, and designed the P4 just so that it could get a very high MHz rating, even without particularly good performance.

Certainly seems to me you're implying P4s are slower when clock speeds are kept constant

P4s are slower compared to AMD CPUs on the market at similar points in time.

Re:Benchmarks, accuracy, and choice

[drsmithy]

I'm afraid you are mistaken. AMD hasn't gained so much market share by maintaining pairity, they did it by being consistently faster, as well as lower power and cheaper.

I think you'll find "cheaper" is the one you're after.

Of course they are, but that doesn't change the fact that Intel was heavily marketing MHz, and designed the P4 just so that it could get a very high MHz rating, even without particularly good performance.

Except they *did* have good - or at least equal - performance.

P4s are slower compared to AMD CPUs on the market at similar points in time.

Typically, the company with the fastest CPU at any given point in time will be the one who most recently updated their product line. The page I linked to previously supports this.

Re:Benchmarks, accuracy, and choice

[P3NIS_CLEAVER]

In 'saving private ryan' they used high speed film in daylight. The strobe effect on the action was quite noticable.

Re:Benchmarks, accuracy, and choice

[FuzzyDaddy]

Hey, now THAT's cool. To think of all the money I wasted on acid as a teenager.

Re:Benchmarks, accuracy, and choice

[Paul Jakma]

glxgears is a benchmark for graphics acceleration.

glx gears is NOT a benchmark, the numbers it produces are meaningless for anything other than 'how fast does glxgears' run. If you don't believe me, maybe you'll believe this ATi-er saying glxgears is not a benchmark?.

Re:Benchmarks, accuracy, and choice

[Otter]

The only thing I use glxgears for is when I'm trying to get acceleration turned on in Linux and I check to see if glxgears has started running 100x faster. Obviously a 5% difference in its speed on two different systems is meaningless.

It's still a benchmark, though! It's certainly not a game, which is what the comment I responded to and the lengthy discussion spinning off it seemed to think. The debate of whether >60 fps in glxgears is meaningful is ridiculous -- nobody sits and watches glxgears.

Re:Benchmarks, accuracy, and choice

[wayward_son]

Not quite.

AMD first beat Intel in 1997 when the new AMD K6 beat the then top-of-the-line Pentium MMX.

Two months later, Intel took the crown back easily with the Pentium II. However, AMD had established itself as an alternative to Intel on the medium to low end. It is no coincidence that this when when PC prices started falling rapidly.

Re:Benchmarks, accuracy, and choice

[ryanov]

My point is that the story write-up makes it sound like SJ is lying, and he's not. He's just presenting the best set of benchmarks he can, which is pretty much what I expect from the CEO of the company...

So, misrepresentation is how much better than lying?

Re:Benchmarks, accuracy, and choice

[EnVisiCrypt]

>Ever tried waving your hand underneath a strobe light going at 30 cycles/sec? That's 30fps yet the motion still looks strange, since it's like you're seeing discrete frames and not continous motion burred between frames.

When you watch TV, you are seeing discrete frames with no motion blur. The reason that you perceive a continuous image is due to the phenomenon known as persistence of vision. The reason motion looks so strange under a 30 flash per second strobe light is because the strobe actually circumvents persistence of vision.

Think of it this way: If you ran normal video at 60 frames per second, and 30 of those frames were black, motion blurred or not, would you think it looked normal?

Re:Benchmarks, accuracy, and choice

[Paul Jakma]

No that makes it a test.

FWIW, some of the highest-end OpenGL machines would have their display inter-locked with the display's sync rate, so (say) 60fps would be the most you'd get, regardless.

Re:Benchmarks, accuracy, and choice

[cookd]

There is actually some value to getting a game to run at a framerate higher than VSync. The framerate is an average rate, so the actual framerate goes up and down depending on the scene complexity. With an average framerate of 60 fps, you are going to hate life when you walk into a room with something actually going on and your frame rate drops to 15 fps. On the other hand, if your average framerate is 100 fps, things are still ok when the things get busy and the rate drops down to 60 fps.

Comment removed

[account_deleted]

Comment removed based on user account deletion

What about...

[TooMuchEspressoGuy]

"I'm still waiting for the most important benchmark: frames per second in molten core combat."

What about frames per second in, say, Alterac Valley, or a town raid? PvP is much harder for a system to handle than PvE.

Notable, regarding Windows

[daveschroeder]

http://arstechnica.com/reviews/hardware/imac-cored uo.ars/7

I tried to boot from a Windows XP installer CD. No dice. I then tried booting from a Vista installer DVD (Build 5270). Again, no dice. When holding down the Option key, the only icon that appeared was for the iMac's internal hard drive. Holding down the D key to try to force booting off of the optical drive failed as well. With the Vista DVD, the optical drive churned a bit and the iMac hesitated as though it were contemplating whether it wanted to boot the foreign OS. Soon afterwards, the familiar gray Apple logo appeared on screen and Mac OS X finished booting.

The new Intel Macs don't have an EFI shell, so there's no way to directly get at the EFI. Someone is ultimately going to have to write and/or use an existing EFI shell to tell the EFI to boot from alternate media to get things going. Naturally, running Windows under virtualization, with technologies like Intel's VT/Vanderpool, which the Core Duo in the new Macs does support, are going to be the way to go for most users anyway.

Re:Notable, regarding Windows

[wild_berry]

Would GRUB do it? I hear there is an EFI-compatible edition around.

Re:Notable, regarding Windows

[IntergalacticWalrus]

Windows needs BIOS interupts to bootstrap itself. Having EFI GRUB running wouldn't make any difference.

Watch boot video here.

[gasmonso]

YouTube.com has a video of both systems booting. So if you're in to computer drag racing here ya go: http://www.youtube.com/?v=zmaAZwkhYeQ

http://religiousfreaks.com/

Re:Watch boot video here.

[shippo]

There's obviously something seriously wrong with that G5 if it is taking that amount of time to boot. My G5 boots in roughly the same amount of time as that Intel iMac.

Re:Watch boot video here.

[falcon5768]

something had to be wrong with the G5 iMac. My g4 iBook runs faster than that shit booting up.

Re:Watch boot video here.

[Phillup]

I think they "prepped" it by completing the setup and then unplugging it.

Which means it is running a file system check...

Sure looks like it, anyway.

Re:Watch boot video here.

[solios]

My workstation (dual g5) behaved a lot like the machine on the right, until I replaced bad memory and defragged the hard drive (diskwarrior will make ANY well-aged OS X install visibily snappier, I swear).

I've never seen OS X boot that slowly unless there's hardware problems.

Re:Watch boot video here.

[i_finally_got_an_acc]

Once again, digg.com to the rescue.

Comparison of G5 and Intel boot times.

Re:Watch boot video here.

[akhomerun]

there is definatly something wrong. my G4 ibook boots faster than that. same with my G4 iMac

Re:Watch boot video here.

[LanMan04]

I thought OS X did defragging on-the-fly, or am I incorrect? Man, maybe I should drag out DiskWarrior....

Re:Watch boot video here.

[solios]

It allegedly does, but in my experience that's been anything but true. I've noticed speed improvements with every machine I've degragged, though none of them are "fresh" installations by any stretch of the term.

Re:Watch boot video here.

[Ineffable 27]

DiskWarrior doesn't do 'defragging' of the hard drive. It rebuilds a feature of HFS disks called the 'directory' -- the listing of where files are on the disk. Analogous to defragging the disk's contents, but not the same thing at all. Yes, ever since Panther, files (under some conditions) are defragged on-the-fly.

Re:Watch boot video here.

[g0at]

No kidding. And such an opaque video doesn't tell us anything about why. The least they could have done is start them up in Verbose mode (holding cmd-V) to watch the console messages. I bet we would have seen some messages about fsck or similar taking care of some issues on the right-hand Mac.

-b

Re:Watch boot video here.

[sasha328]

I'd have to agree with the above statement. There is something seriously wrong with the iMac G5. My G4 400 boots faster than that on 10.4
I can imaging the Intel iMac to be faster, but not by a minute.

Re:But does it run...

[AKAImBatman]

Support for System 6 is slated to be released at the same time as the latest BeBox. You'd better hurry and pre-order your system!

The G5 is still quite the chip

[maynard]

It doesn't surprise me that it still competes with Intel's latest offering. I wonder if it makes sense for Apple to continue supporting both x86 and PPC platforms long term. I'm sure Intel will -- in time -- crush the G5 in performance. But if Apple wants to dominate the HDTV editing workstation market, Cell looks like the most appropriate processor for that task. Are fat binaries really so obnoxious as to prevent permanent multi-arch support over the long term?

Re:The G5 is still quite the chip

[EvilTwinSkippy]

From what I understand, Apple has no intentions of supporting the PPC for very long. By the end of this year they will no longer sell machines that run on a PPC chip. They will probably continue to support Rosetta for backward compadibility, but gradually let PPC binary support die off. Apple practically gives away Xcode to software producers, and x86 native support only requires a recompile. It's not like the move from MacOS to X where the architecture had changed. Artsy types are going to buy the new Macs, and if vendors don't want their stuff to run like crap they are going to release a FAT version sooner than later.

Re:The G5 is still quite the chip

[dasil003]

I'm planning on buying the fastest G5 tower that Apple ships just before they shift completely to Intel. Although the G5 towers are mammoth, they are possibly (IMO) the most beautiful towers, inside and out, ever to be created. And in terms of performance I feel they will hold out for a very long time. Eventually I will probably retire Mac OS X in favor of some flavor of Linux on it, and run it as a server into the distant future. If I ever have any spare time I can tinker with the compiler and explore the Power architecture as a hobby.

Re:The G5 is still quite the chip

[chipset]

The fat binaries (or Universal) has been around for a very long time. NeXT first introduced them in this exact same format. Knowing that Mac OS X is the current incarnation of NextStep (the OS for NeXT machines), I assumed the Intel move would be relatively easy, if the had been maintaining the code.

Now, in retrospect, it looks like they have for Mac OS X, but maybe not for all the other applications (iLife, FCP, etc).

Now, given that the OS has a long history of multi-platform support, it is only a piece of the puzzle.

Application level changes are a bit harder, especially in relying upon functions specific to a chip. Which, for some applications, is the case. Others should be able to do a direct recompile, if the application is still around in source form, the author is interested, etc.

Back when I had acess to NeXT Cubes, I didn't have to worry about it. However, when I later bought NeXTStep 486, I had to. There were lots of applications for the 680x0 systems, I sometimes had to search for those 486 applications. I assume we are headed back into that world.

So, can it happen? Yes. However, I suspect that Apple will move on with the Intel architecture. I assume the PowerMac G5 will be a well respected machine in the meantime, as it does great for video editing, something Widnows machines still work hard to do poorly.

I suspect it might be like the Amiga. While the Amiga didn't get a lot of respect, those in the video editing world used it much longer than people antipicated.

But, in the end, the new macs will be Intel. As a side note, I just sold my G5 DP to someone looking to do video editing with FCP. Even with them knowing the Intel systems were coming out, they still wanted it.

Re:The G5 is still quite the chip

[frodo from middle ea]

I too am considering buying a used G5 Power Mac.

I am waiting for some early adopters of intel iMac to sell their G5 Towers for much cheaper than what apple is selling.

Another advantage is that most of these systems are loaded with RAM and Software. I have already started seeing some on craigslist.org.

Here is hoping there are a lot of early adopters ...

Re:The G5 is still quite the chip

[nanter]

Well, when you say for not very long, what exactly do you mean by that? I think it would not be the wisest move for Apple to leave those of us that made the switch to Apple recently on the PPC platform high and dry. Leave us with unsupported hardware in the near future and you may alienate that entire segment.

Re:The G5 is still quite the chip

[Anonymous Coward]

Apple practically gives away Xcode to software producers

Practically? Apple literally gives Xcode away. It's free. It even comes with every new Mac, right in the box. And anybody who wants to sign up can get updates and SDKs at no charge.

Re:The G5 is still quite the chip

[be-fan]

I strain to see how you get "the G5 is still quite the chip" from those benchmarks. Overall, the benchmarks suck. The xbench results are inconsistent (look how low the UI result is for the Intel Mac, yet read the review which says the Intel Mac is snappier than the G5 iMac). The Rosetta results are irrelevent, since they test a 512MB iMac against a 4.5 GB PowerMac. 512MB is already marginal for OS X, Rosetta (being a JIT) is going to eat up a lot of RAM on top of that. So what's left for Photoshop? The results are completely meaningless.

The only two decent results are the QuickTime and iTunes results. The Intel Mac makes quite a strong showing, even considering the fact that Quicktime is a piece of software that has been highly optimized for PPC/AltiVec over the years.

Re:The G5 is still quite the chip

[sharrestom]

Pretty sure that it was Jobs, at Apple, that pushed IBM, hard, to add vector processors to the PPC core, which, as we now see, are the basis of all those pretty machines you are talking about, including the cell. I expect that Jobs, at Apple, will push Intel to accelerate development of SSE, not that Intel needs any asking.

Re:The G5 is still quite the chip

[timeOday]

It doesn't surprise me that it still competes with Intel's latest offering.

But they didn't compare power usage. I'd like to see you cram dual watercooled G5's into a laptop! Yet as I understand it, the new iMac CPU is just the same Centrino Duo you can get with 4 hours of battery life.

Re:The G5 is still quite the chip

[DGregory]

Apple doesn't s upport anything for very long. I'm still running Panther and can't even download the new JRE because they ONLY made it for Tiger. Piss me off.

Re:The G5 is still quite the chip

[Mantus]

Quite the chip? 2x dual core G5 @2.5GHz perform a bit better than 1x dual core Intel mobile cpu @2.0GHz and you think the G5 sounds good?
Mac/PPC fanboys never cease to amaze me.

Re:The G5 is still quite the chip

[n8_f]

No, they will be supporting the PowerPC for quite awhile. They will still be selling G5 boxes well into 2006 and they will primarily be selling them to their professional users. If they burn their large professional customers (who are very slow to move; some are still pissed off about Apple completely dropping OS 9 boot support) by dropping support, a large number of those customers will simply switch to Windows PCs.

For example, we just picked up 12 PowerMac G5s and we will probably pick up another bunch of G5s when they announce the Intel-based PowerMacs. If we could buy Intel-based PowerMacs today, we wouldn't and neither would any of their other professional users. Intel support isn't simply a recompile. It is only easy if you are using Xcode and even then it takes at least a little work. Obviously Apple's pro apps (Final Cut, etc.) are developed in Xcode and obviously they are having to do quite a bit of work to transition them. If it were a simple recompile, don't you think they would have announced support when they introduced the first Intel Macs rather than a couple months from now? And a bunch of the other large apps aren't using Xcode. Adobe was using Codewarrior exclusively and are switching primarily because of this transition (they would have had to anyway as Codewarrior is dead, but it would have been a lot slower). Microsoft is another example and there are a ton of apps in vertical markets that are going to be very slow to transition over (some of them still run in OS 9!).

Re:The G5 is still quite the chip

[cmdrbuzz]

The fat binaries (or Universal) has been around for a very long time. NeXT first introduced them in this exact same format.

I thought that NeXT applications had separate files under the .app folder for each of the different architectures. Universal Binaries seem to have both PPC and x86 code combined into one file.

Re:The G5 is still quite the chip

[maynard]

I had a NeXT slab on my desk back in the early '90s. Loved it. And certainly Intel is where Apple is migrating toward for their general purpose users and the scientific imaging community. The only market I think Cell is worth considering would be for the film/video sector - and only because of its silly single precision FP and vector capabilities. It's about using the right technology for the job. And as you point out, Intel just doesn't have the performance for good HD editing yet.

Re:The G5 is still quite the chip

[maynard]

The G5 may lag in integer performance, but it's still very competetive for vector and FP. Where I work there are plenty of folks who prefer a dual G5 over a PC for Mathematica and Matlab compute due to their reliance on FP tasks. Though dual opterons running Linux are begining to make inroads here too.

Re:The G5 is still quite the chip

[maynard]

Yeah. We have a bunch of folks with G5s too, and nobody here wants to run Mathematica or Matlab through an emulation layer. We'll be supporting PPC in the lab for at least another two or three years. Hell, one of our professors just took delivery on a quad G5 last week. He didn't buy it to run Word. If Apple wants they have an opportunity to support both PPC and x86 permanently. The only reason for this is if they think it's worth the money. I'm just speculating that Cell would be appropriate for one of their target markets. Is it worth it? Dunno.

Re:The G5 is still quite the chip

[be-fan]

I've got a dual G5, and I use Matlab a lot, and I prefer to use it on my dual core amd64. Matlab code often contains a very large integer component (since Matlab itself is an interpreted language), and for most tasks an Opteron is faster. Just try the built-in "bench" utility on a G5 versus an Opteron. The G5's FPU performance also suffers heavily if the data is in complex data structures in memory. The PowerMac's memory controller is rather high latency, which means accessing big structured data sets is slow on the G5. The G5 is just fine if your code is low-level and your data is simple arrays (eg: signal processing), but there is a lot of scientific and especially engineering code that doesn't fit that mold.

Re:The G5 is still quite the chip

[jcr]

I thought that NeXT applications had separate files under the .app folder for each of the different architectures.

No, the executable was a Mach-O file that contained code for each architecture, just like today.

-jcr

Re:The G5 is still quite the chip

[javaxman]

maybe not for all the other applications (iLife, FCP, etc).

Actually, aren't nearly all of the iLife programs now Universal Binary? With maybe the exception of GarageBand?

The Pro video apps are exactly what we shouldn't expect to be ported quickly, because they must rely heavily on optimized Altivec code... that's the segment of apps that will be most difficult to port.

There were lots of applications for the 680x0 systems, I sometimes had to search for those 486 applications. I assume we are headed back into that world.

Not on your life. There's a world of difference between NeXT in the early 90's and Apple today. Forget about the fact that Apple's market share and install base is something NeXT never dreamed of. The other important differences are :

- NeXT wasn't making it's own hardware. It was relying on OEMs to bundle NeXTStep with compatable Intel systems. Some users were buying the NeXTStep package and finding their hardware unsupported, as NeXT couldn't possibly support every hardware configuration used by PCs of the time.

- Microsoft was entering into licensing agreements with OEMs which included exclusive contracts ( later ruled to be in violation of anti-trust laws ) preventing them from marketing NeXTStep and other operating systems.

- Microsoft was not developing any products ( notably Word and Excel ) for NeXTStep, it's releasing Office on OS X ( for the next 5 years, even ).

- Developing for NeXTStep was actually quite expensive, if I remember... well, not horribly expensive, but you weren't going to pick up the developer kit if you were a student or just checking it out or building non-commercial software, that's for sure. Apple's developer kit? It's free.

- Microsoft had just released "Windows NT"... people were really, really excited about that, and NeXTStep Intel went up against that with a tiny install base, little OEM support, and a small developer base. Are people as excited about Vista after their experience with XP? Probably not.

I'm sure there are other key differences, these just seem like the more important ones to me. Another key difference: there are already a good number of Universal Binary applications. A partial list can be found on this MacRumors page, but it's *very* far from complete, as a search on Apple's "Downloads" web page using the term "Universal Binary" reveals, there are a good number of apps shipping with Universal Binaries today which aren't on the MacRumors list. Maybe there is a better list somewhere, I didn't look very hard.

As a side note, I just sold my G5 DP to someone looking to do video editing with FCP. Even with them knowing the Intel systems were coming out, they still wanted it.

If you want to do some work today, you use the applications and technology that are available today. That DP G5 is a great machine fully capable of just about any video editing task you'd want to throw at it, and it has the native software to do that work *today*, so it's an obvious choice. They'll be wishing they had the PowerMac replacement when it comes along, probably, and I'd expect the Pro apps to be ready when the PowerMac replacement is ready... but that's not today. Today, PowerMacs are what you want if you're editing video or audio... unlike the Amiga ( or the NeXT ), it's not because of some special-purpose vector-processing chip ( though you could make the case for Altivec ), it's because of the software. That same software will be available natively on the Intel versions, and the other problems that plagued NeXTStep on Intel aren't there for Apple.

There's no problem with hardware compatability- OS X is only supported on Apple hardware. There's no problem with OEMs- Apple is the only OEM. There's no problem with Apps- there are plenty Intel Native already, and non-Altivec apps largely just recompile. There is a large and healthy community of Apple developers, which includes even Microsoft

Re:The G5 is still quite the chip

[SteeldrivingJon]

"Developing for NeXTStep was actually quite expensive, if I remember... well, not horribly expensive"

Oh, it was quite expensive. The operating system cost $795, and the developer tools cost around $2000 or $3000. Per machine.

Intel NeXT software availability is a bit weird, though. There was software that wasn't available for x86, but for the most part that was because the developer had left the NeXT market completely, some before NeXT even stopped doing hardware.

The stalwarts who kept going after NeXT moved to Intel were pretty good about putting out fat binaries.

Re:The G5 is still quite the chip

[javaxman]

Oh, it was quite expensive. The operating system cost $795, and the developer tools cost around $2000 or $3000. Per machine.

Wow, I thought it was a lot less than that ( funny I don't know since I worked at NeXT at the time ), but it appears you are correct. gee, I wonder why that didn't catch on. All this time I've been blaming MSFT anti-compete OEM licenses.

Re:The G5 is still quite the chip

[SteeldrivingJon]

"gee, I wonder why that didn't catch on. All this time I've been blaming MSFT anti-compete OEM licenses."

Well, considering how (if I'm not mistaken) Be tried giving their OS away at one point and that didn't work either, I'm not sure price was the whole problem. NeXTSTEP's hardware requirements were rather high compared to the median machine in the early years of the Intel port, which didn't help either. Before fast hardware could get cheap enough, Java came out and pretty much sucked all the air out of the Object-Oriented development space.

The $2995 for the development tools was in the ballpark for high-end corporate development tools of the time. I think Borland's Delphi Enterprise package cost $1995.

FireWire 800

[sg3000]

The benchmarks from the article are useful.

It sounds like from the review that Apple's pro apps aren't well suited for the Intel-based Macs until they have the Universal Binary versions (suggested to be in late March). Maybe that's why they left FireWire 800 off the initial MacBook Pro -- if you need FireWire 800, you're probably doing pro work. So Apple left it out to reduce costs until they have a complete system for pros.

Re:FireWire 800

[MBCook]

I have a PB G4 (early 2005) and one of the things that makes me cautious about wanting a new MacBook Pro is the lack of FW800.

My understanding is that it was not put on because none of Intel's chipsets (which Apple is currently using) supports FW800. Thus they would have had to add a FW800 card or chip onto the motherboard to add the port which isn't usefull to most people. These are home computers after all. I would expect they would keep the FW800 on whatever repalces the dual G5s.

Hopefully they'll bring it back to things like the MacBook and iMac in a later revision.

And hopefully this doesn't mean Apple has to wait on Intel to implement things, this is a one time only thing.

Re:FireWire 800

[TERdON]

Photoshop hasn't been ported, yet at least. It runs crap slow because it is being emulated through Rosetta. Considering that, the Intel iMac really didn't do THAT bad. And next time, RTFA. Don't just look at the pictures and assume stuff!

Re:Why?

[TooMuchEspressoGuy]

But what about if someone wants to run games (and perhaps other software that runs only on Windows) *and* still get the superior OS/apps found in Macs? Why should they be forced to buy two computers just so that they can preserve their "entire Mac experience"?

Out of topic but somewhat in topic though

[Arthur B.]

Have you seen the new apple commercial? "...for years it's been trapped in a PC, dull little boxes doing dull little tasks"... Honestly I felt down the whole evening... How many people know the new macs actually ARE pc's? Of course they aren't, cuz there are macs and pcs right?... macs have this apple thingie on the top of the screen and ya now the pc's come with this colored flag and it says windows. Of course in the public mind, mac and pc's are opposed on the operating system side... this is really Mac OS Versus Windows, no one gives a shit about the internal architecture... but people don't make the difference between a machine and an OS... so Apple is doing a good job, bashing PC's, it's still there concurrent...but it really makes me sad that this commercial will deepen public confusion... shame shame shame on them. Shame on people for being easy marketting prays. Fortunately, porting OS X to standard PC will maybe awaken some puzzling among the general audience.

Re:Out of topic but somewhat in topic though

[dwayner79]

I saw it last night during 24... good stuff. If anyone finds a link or a torrent, please post.

Re:Out of topic but somewhat in topic though

[Arthur B.]

http://www.apple.com/intel/ads/

Re:Out of topic but somewhat in topic though

[MindStalker]

Think their point are that modern PCs are very powerful but their potential is not tapped by the Windows operating system.

Re:Out of topic but somewhat in topic though

[Bazzalisk]

Well, they are PCs in that they are Personal Computers. They aren't really "PC Compatible" as such - as evidenced by the fact that windows and x86/Linux boot CDs don't load on them. There's a bit more to an architecture than the CPU and chipset.

Re:Out of topic but somewhat in topic though

[Arthur B.]

Yes but come on, it's not like it's a different architecture used here, it's just incompatibility built-in for the sole purpose of being incompatible...

Re:Out of topic but somewhat in topic though

[EvilTwinSkippy]

Apples are really PC's?

PC's today really aren't PC's. PC's haven't been PC's since the introduction of USB and the PCI bus. Haven't you wondered what all these "Legacy Free" PC's were selling?

Re:Out of topic but somewhat in topic though

[Anonymous Cumshot]

How many people know the new macs actually ARE pc's?

A "PC" is not defined by its processor. The mac/pc differentiation is one between a designed-to-be-complete system (that is, OS running for specific hardware and running well) and a hackish put-it-all-together-and-hope-it-works one.

"PC" is a generic term for any personal computer. "Mac" is a particular term for a specific type of computer.

Re:Out of topic but somewhat in topic though

[dadragon]

Did it occur to you that they may have ditched some of the cruft in the x86 architecture? For instance, the BIOS? Or maybe you don't have to ask the keyboard controller to enable gate A20 on these new machines? Or maybe there's no Intel 8259 interrupt controller emulation, and you need ACPI support?

Leaving out these things isn't building in incompatibility. It's getting rid of cruft that should have been removed with the introduction of Windows 2000.

Re:Out of topic but somewhat in topic though

[Bazzalisk]

ACtualy I doubt that. I think they just decided to use the best available technology without consideration of what other PC manufacturers were using ... and as a result got something incompatible.

Re:Out of topic but somewhat in topic though

[poot_rootbeer]

How many people know the new macs actually ARE pc's?

How do you define a PC? Is it the CPU architecture? Is it the manufacturer of the CPU? Is it the company that wrote the prevalent operating system? Is is the company that first called its product a 'PC', 25 years ago?

Why make things so hard when the answer is right in front of us?

PC = personal computer. All Macs ever produced have been PC's. For that matter, so were the Apple 8-bit computers.

Re:Out of topic but somewhat in topic though

[poot_rootbeer]

it's just incompatibility built-in for the sole purpose of being incompatible...

No, it's incompatibility because EFI is a far superior firmware solution to the 25-year-old cruft known as PC BIOS. The better technology won out.

Re:Out of topic but somewhat in topic though

[Firehed]

I think by definition, "PC" has included Macs since day one. Or around day one anyways. Last I knew, an iMac was still a Personal Computer, regardless of the OS it uses. There's just this general assumption that PC=Windows, as if you can't actaully own a Mac and therefore it can't be a "personal" computer. Hell, handhelds and technically even calculators are PCs, but the Windows machines in the school library aren't as they're not personal.

It's honestly stupid. Mac's "aren't" PCs, but have used the PowerPC architecture for who knows how long. Mac is a SUBSET of PC, not an adjacent but seperate category. You have Dell and HP as Mac's adjacent categories within PC, as they're hardware manufacturers. Of course now we're getting this bizarre misshapen diagrams because it's no longer Mac=PPC=OSx and [OEM]=x86=Windows (I'd include *nix, but good luck finding that as an OS choice off of Dell Home). Think of the heierarchy of numbers: you have real and imaginary as subsets of numbers, then you have "i" as a subset of imaginary and "rational" and "irrational" as subsets of real. It's just ironic that imaginary, and thus i, would be akin to Mac, simply due to how the number of subsets works out.

Time to rethink our definitions. And this isn't the only case... it seems to me that "dictionary conservative" is more or less equal to "politial liberal" and vice-versa. But that's a seperate discussion.

Re:Out of topic but somewhat in topic though

[product byproduct]

Oh come on, do you really say things like "what personal computer did you buy?", "do you play doom3 on your personal computer?"

The term "computer" is adequate.

And there's this large class of computer hardware that is compatible with the IBM PC of 1981. How should we call those? They don't deserve a name?

Re:Out of topic but somewhat in topic though

[AKAImBatman]

Hrumph. You young whipper-snappers don't know anything. Back in my day, the Apple Macintosh was referred to as a "home computer" or (more technically) a "microcommputer". The IBM "Personal Computer" was nothing more than an invention of IBM's marketing department. And now everyone seems to think that "Personal Computer" is a better definition than "microcomputer" just because IBM says so. Would one of you young'uns care to explain to me exactly how my rack mounted servers are "Personel Computers"? What's so "Personal" about a computer that serves hundreds to thousands of users each day? Huh? Huh? Answer me that, future-boy!

Bah humbug! </grumpy-old-man>

:-P

Re:Out of topic but somewhat in topic though

[drinkypoo]

They're called IBM PC clones. This has been shortened to "PC Clones" or "Clones" or "Clone PCs". HTH.

Re:Out of topic but somewhat in topic though

[bloodmusic]

Since the IBM Personal Computer popularized the term, I think it's meaning since has been tied to descendants of that original machine. By that rationale, the Mac is, in fact, not a PC. Just a thought.

Re:Out of topic but somewhat in topic though

[linguae]

Since the IBM PC first came out, the acronym PC has become equivalent to a x86 box instead of just any personal computer. It doesn't matter what OS I'm running on an x86 box; DOS, Windows, Linux, BSD, OS X, Plan 9; it is still just a PC. Macs have always been personal computers, and Apple has always referred to them as such (but never with the acronym PC), but they never architecturally became plain PCs until the new Intel Macs came out. Now the Macs are just plain, boring x86 PCs now, but of course Apple will never come out and say, "We now make PCs like everybody else now." Instead, they still want people to feel that there is still something special about the insides of the Mac (and for the umpteenth time, the operating system doesn't count), that demands paying the Apple premium.

I still maintain that Apple has made a mistake with its switch to Intel. It is only a matter of time before customers realize that a Mac is now just another PC (the only difference is the operating system), and it is only a matter of time before somebody with a lot of time on his/her hands gets OS X running on their $299 Dell Celeron machine. As soon as people add two and two together and figure that it is cheaper to buy a 64-bit Sempron box at $399 ($349 with rebate + $129 for OS X = $478) compared to a $499 Mac Mini that will most likely have a 32-bit Core Solo processor with a case that isn't as expandable, and with somewhat underpowered stats (the current Mac Mini only has a 40GB hard drive, whereas the eMachines box has a 100GB hard drive), Apple will have a harder time justifying its prices and specifications, unless Apple diversifies its offerings and sells its offerings at competitive prices.

Don't get me wrong, I still like Apple and the Mac, but I still mourn the loss of the PowerPC, especially the G5. I am personally not a fan of the x86, and I wish that crufty architecture will just die and lead to new innovations in other architectures, such as the Cell processor. Sure, the new Intel chips are nice and run cooler, but I was waiting for something that was truly compelling, like a Cell computer or in 2008 or so a PowerBook G5. Oh well, c'est la vie. Guess I'll save up for a dual dual-core Power Mac G5 while they still exist....

Re:Out of topic but somewhat in topic though

[TeknoHog]

Since the IBM PC first came out, the acronym PC has become equivalent to a x86 box instead of just any personal computer. It doesn't matter what OS I'm running on an x86 box; DOS, Windows, Linux, BSD, OS X, Plan 9; it is still just a PC. Macs have always been personal computers, and Apple has always referred to them as such (but never with the acronym PC), but they never architecturally became plain PCs until the new Intel Macs came out.

IMHO, these x86 Macs cannot be called PCs in the usual sense, because it would imply original IBM PC compatibility, which includes BIOS.

Personally, I try to avoid the term 'PC', simply because of the confusion between 'personal computer' and 'IBM PC'. I also dislike the term because it implies the use of Windows, as in 'PC gaming'.

Re:Out of topic but somewhat in topic though

[xouumalperxe]

Well, they're still not exactly PCs, seeing as how PCs don't yet have EFI. But they're close, yeah. Still, I bought an iBook (my first mac) under 2 months ago, and it was because I wanted the user experience. I bought the bloody thing for the 1-button mouse (yes, I like it! And I'm proud of it!) and the operating system, processor be damned!

The difference between a PC and a Mac is one of philosophy, not one of processor. I believe the slogan "What is an intel processor doing in my Mac? A lot more than in any PC!" to be quite the accurate description (if you're a fan of the philosophy). The Mac experience is all about ease of use and general quality of product. Truth be told, the only path forwards in laptop performance seemed to be this way. The iMac is at a point where the internals are closer to a laptop than a desktop -- and the new hardware further shows this. All in all they changed to where the quality lies right now, and for the predictable future (in laptop processors).

All in all, I don't think the general idea of saying that the Mac is different from the PC even if the entrails are the same doesn't really bother me much

Re:Out of topic but somewhat in topic though

[BlueStraggler]

PC = personal computer. All Macs ever produced have been PC's. For that matter, so were the Apple 8-bit computers.

Since you brought up 1980s terminology, you're confusing "PC" with "microcomputer". A micro is a computer that fits on your desk. PC refers to the IBM PC, and infers compatibility with that architecture.

PC is to microcomputers as Kleenex is to tissues: a brand name often misused as a generic noun. People used to keep it straight by saying "PC-compatible", but that has been contracted to just "PC" now, and the meaning has blurred correspondingly.

Re:Out of topic but somewhat in topic though

[NutscrapeSucks]

Or maybe you don't have to ask the keyboard controller to enable gate A20 on these new machines? Or maybe there's no Intel 8259 interrupt controller emulation,

Hardware-wise, the chipset supports these. So, nothing has been left-behind. In fact, Apple is subsidizing the remaining DOS users that need these features.

Re:Out of topic but somewhat in topic though

[dadragon]

Hardware-wise, the chipset supports these. So, nothing has been left-behind. In fact, Apple is subsidizing the remaining DOS users that need these features.

I did not know that. I don't generally follow chipset features, I figured those were still seperate. The BIOS, however, is definitaly not a part of the chipset, as they tend to be made by Phoenix. Apparently not having one is enough to keep Windows XP from booting. Shame.

Shut up!

[Anonymous Coward]

Haven't you heard? It's cool to hate Apple now. It makes you '1337.

Anybody who says anything remotely positive about Apple, or especially about Steve Jobs, is a "fanboy." You don't want to be called a fanbody, do you? Then get with the program. Talk about how cheaply you can get a Gateway that's just as good as the new iMac or something, and insist that Woz is the only person who ever had anything to do with Apple worthy of any respect at all.

Oh... and maybe Tog, if you are a UI nerd.

Re:Shut up!

[NanoGator]

"Haven't you heard? It's cool to hate Apple now. It makes you '1337."

[Warning: This is an OT rant, no hard feelings if modded down.]

Wish I had known that before I made a not-so-nice comment about Apple which resulted in several mods going well out of their way to mod me down until I couldn't post on Slashdot for a couple of weeks. (From a certain IP, anyway. At least now you understand the origins of my sig.)

If it has suddenly become a little too cool to hate Apple now, I blame extremist mods for it. Over the years I've made silly little quips about Apple that nobody on Earth should have taken too seriously and have been mod-bombed over it. I wouldn't be the least bit surprised if, out of anger, they were finally M2'd out and the replacements came in to even up the balance by shifting over to the extreme opposite view. (i.e. over-modding anti-Apple sentiment.) Too much zealotry will always lead to people with too much opposition to your view.

This has already happened with regards to Microsoft. Go back a few years and ANY comment ridiculing or insulting MS would be modded up, but polite criticisms of Linux would be modded down. Even uninformed posts (i.e. there still seems to be some impression that Win2K was built on the same kernel that Windows 98 was) would get modded up. 2K is nearly 6 years old now, XP is 4, and the BSOD is virtually gone. Yet, the blue screen jokes STILL fly with full karma around here. The result? People stand up and say "uh, you guys need to get with the 21st century." People whinge about MS fanboys flooding Slashdot. Sorry, can't see that from my point of view. Fire is being fought with fire. My advice? Don't give Apple praise for being wrong or Microsoft scorn for being right.

No, I'm not pro-MS or anti-Apple, I'm just tired of these karma-fueled battles happening every year. I appreciate Taco's desire to keep Slashdot 'democratic', but it's irritating that ordinary Homer Simpson'ish people are allowed to be cops.

Re:Shut up!

[Sique]

I appreciate Taco's desire to keep Slashdot 'democratic', but it's irritating that ordinary Homer Simpson'ish people are allowed to be cops.

And there was me thinking that one point of democracy was to empower normal Homer Simpson people to make the rules... Silly me.

Re:Shut up!

[level_headed_midwest]

You're only irritated that ordinary Homer Simpson'ish people are allowed to be cops because they get free donuts. Mmmmm...do-nuts.....garrrrrrrrrllllll...

Re:Shut up!

[MobileTatsu-NJG]

"And there was me thinking that one point of democracy was to empower normal Homer Simpson people to make the rules... Silly me." Cops, not lawmakers. You don't want Homer Simpsons being vigilantes.

Re:Shut up!

[nathanh]

Wish I had known that before I made a not-so-nice comment about Apple which resulted in several mods going well out of their way to mod me down until I couldn't post on Slashdot for a couple of weeks.

You hit the nail on the head, Nanogator.

I have also noticed extremist Apple fanboy moderation around here lately. My Mac credentials extend back to the late 80s on System 6 and I've owned a half dozen Macs over the years. I'm even typing this from a Powerbook (running Linux admittedly). I'm a strong supporter of Apple and I love to read books about their history. Yet even the most mild criticism of Apple or MacOS on /. will result in my comments being moderated down as Flamebait, Troll and Overrated. I never get similar mistreatment for negative comments about Linux or Windows. It seems Apple fanboys have no qualms abusing the moderation system to ensure that only positive Apple comments are seen.

Unfortunately this isn't new behaviour for Apple fanboys. As far back as I can remember - including the glory days of Usenet - the Apple fanboys have been the most intolerant, the least receptive to criticism, the most judgemental and often the least educated of all the enthusiast groups. The negative moderation of any criticism of the latest Macs is yet another example of this behaviour. Anybody who thinks Linux fanatics can be over the top has never seen an Apple fanboy in full swing. Even the Amiga users were never so extreme. That sort of stupid fanatacism is what led to one of my earlier sigs: "I love Apple hardware but goddamn I hate Apple users".

The example at the start of this thread epitomizes everything I hate about Apple fanboyism. Steve said something that deservedly should be called out for being deceitful bullshit. If any other CEO - Gates, McNealy, Ellison - had said something similar we'd have people throwing figures around and using datasheets to prove that the CEO was a lying bastard. Even when a relative nobody from GNOME or Xorg attempts to massage the figures there will be 100s of /. comments crying "Bullshit". Yet when Steve does the same thing the Apple fanboys are rallying behind him, providing him with excuses, apologising for his behaviour, rationalising the lies, and moderating or shouting down anybody who points out that the emperor has no clothes. Apple gets "special treatment" and I find that despicable.

Re:Shut up!

[idsofmarch]

How many negative mods does it take to get one banned? It seems like a lot, and with that in mind, I would suggest that maybe your posts have been more inflammatory than insightful, more trollish than interesting. You might have been mod-bombed into oblivion, but I wonder just how many times you incited the ire you received. Your sig seems indicative of a crude way of expressing yourself at best.

As for Apple, IMHO it's become best to either be a hater or a zealot and it's being a moderate that gets one into trouble. The BSOD joke is lame, like the Russian joke, the profit joke, but contains more truth than the oftly repeated "Gore invented the internet" meme. Slashdot, you will realize, is dumb. The collective IQ, as evidenced by comments and moderation, shows a clear case for the stupidity of the average.

Try not to take it too personally.

Re:Shut up!

[entrylevel]

Just read at -1

Re:Shut up!

[poopdeville]

It takes around 5-10 negative mods to get banned. But a positive mod won't counter-act a negative mod with regards to banning. So if a post you make gets moderated up and down a lot, you can get banned and still have a positive net moderation. This has happened to me more than once. And the period of time you're banned for seems to double each time it happens.

Re:Shut up!

[swv3752]

Dude, get with the times, XP still blue screens, the Vista beta blue screens, hell we have seen screenshots of the Xbox360 blue screen (though weren't they actualy grey?), so trying to sweep MS's flaws under the rug is still BS and I will call you on it. I don't care if they are trying harder with patches, they stil have a horrible track record. Billy G may donate money but his company still operates immorrally.

And yes Steve was a bit dishonest with benchmarks, but then benchmarks are just a form of statistics. How did that quote about lies and statistics go? But then anyone with half a brain realized the Jobs was just in another snit with IBM. He did the same thing with ATI.

Re:Shut up!

[NanoGator]

"...but I wonder just how many times you incited the ire you received."

It's a matter of perspective. In some ways, I deserve some of it. I let out a few harsh words. I am in no way claiming to be 100% innocent, only that I was treated unjustly. I On the other hand, several of my unrelated posts were modded down. I could nost post from a certain IP address for 3 weeks as a result of it. There's no way in my mind I deserved that. (I also recieved numerous positive mods on the topic...) There have been other times over the years where I made a small quip about Apple and was modded down severely for it. For example: Somebody put a PC inside of their iMac case. And I said something like "It's the first time GTA was ever seen on a Mac!" Flamebait.

"Your sig seems indicative of a crude way of expressing yourself at best."

It's funny how one can always find information to support their point of view. Heh. Anyway, this sig is the result, not the cause. When I had been modded down enough times to get banned, the only way to get my message out was to change my sig. So, that's what I did. Mature? No. Disarming? No. Intelligent? No. Says anything about what I deserved? No.

This is not the first time I have run into this problem, rather it's the first time I was actually banned for it. Worse, it was malicious. Pinning it solely on me is dangerous.

Re:Shut up!

[Smurf]

Even though I'm a MacOS X addict, a definitely agree with your feelings about irrational moderators. I feel that Apple fan-boyism is not only annoying but in the end it also hurts the interests of those who, like me, honestly feel that the options we chose (OS, software, etc) are worth considering by others (if only to prevent those options from going the way of the dodo as has happened in the past) but who respect those who dissent.

So allow me to apologize to you in the name of the numerous (but evidently not as vocal) Apple users who abhor irrational fan-boys.

On the other hand, I would anyway like to decide on my own if you are really a victim or of the negative mods were deserved. I see that in your recent history all comments that are rated under 2 are unrelated to Apple, so I guess that you are referring to older comments. (I'm not a subscriber because I'm poor, so I can't look at your full history). Can you please point me (us) to one or two comments that you consider were unfairly modded negatively by Apple fanatics?

Anyway, I feel that we should start to meta-moderate more. That's supposed to be the mechanism that should prevent zealots from getting too many moderator points.

Re:Shut up!

[NanoGator]

"Dude, get with the times, XP still blue screens, the Vista beta blue screens, hell we have seen screenshots of the Xbox360 blue screen (though weren't they actualy grey?), so trying to sweep MS's flaws under the rug is still BS and I will call you on it."

Go ahead. You'll just hear the testimony of a bunch of people telling you that Microsoft virtually killed the BSOD with the release of 2k. Ignorance of this fact is part of the reason ppl keep rushing to Microsoft's defense. It's not because everybody loves BillyG.

"And yes Steve was a bit dishonest with benchmarks, but then benchmarks are just a form of statistics."

Heh. "... trying to sweep Apple's flaws under the rug is still BS and I will call you on it." Funny.

Re:Shut up!

[strikethree]

hm. I just deleted a long rant that I had typed in. Let's make this more concise:

I metamod twice a day, every day. I metamod as unfair many negative mods, one or more of which, you have been a victim of.

I get mod points only about once every two weeks nowadays. I have had one of my moderations overturned. I have only negatively mod'ed goatse and gnaa posts. All other moderations I perform are positive mods. I have probably positively mod'ed you when you have levelled reasonable criticism against Apple.

The moderations system should NOT be used to further someones personal agenda. I dislike Microsoft and most of their products intensely. I am strong enough to not attempt to prevent others from displaying their enjoyment of Microsoft products. I have a 17 inch Powerbook. I like it a lot, but I am secure enough in myself to allow others to expose the warts that do exist in that Apple product.

Honestly, you sound like a reasonable person. I am writing this so you will feel comfortable knowing that their are other reasonable people who are working within the moderation system. If you have metamoderated, you should see that the majority of mods are reasonable as well.

The intense amount of noise created by the small minority who are haters (of whatever camp) is incredibly annoying. It gets so loud, that it can persuade you to believe that almost everyone is a hater. Don't fall for it. Be positive and I will be positive with you. Be negative, and you will be negative by yourself.

respect.

strike

Re:Shut up!

[NanoGator]

"The intense amount of noise created by the small minority who are haters (of whatever camp) is incredibly annoying. It gets so loud, that it can persuade you to believe that almost everyone is a hater. Don't fall for it. Be positive and I will be positive with you. Be negative, and you will be negative by yourself."

I appreciate the comment. You're right.

I'd say more but it'd be just a basic restating of what you've already said. hehe. Thanks. :)

Re:Shut up!

[strikethree]

You are quite welcome. :)

strike

Re:Shut up!

[IntlHarvester]

> I metamod twice a day, every day.

I appreciate your effort, but the M2 system has a nice fat loophole called "overrated". Furthermore, the system operates largely out of context, so what could just be an over the top statement with wry wit could easily seem like "flamebait" (another fav Apple mod).

I will give the Apple moderators some credit, they are very willing to dig through four day old discussions to moderate down opinions that nobody will ever read again.

Re:Shut up!

[strikethree]

It is true that "overrated" can not be metamod'ed. I agree that it is a loophole.

To your other point, in the long rant that I mentioned that I had deleted, I described how I thoroughly investigate each negative moderation. I read parents, children, and sibling posts to get a feel for what is going on. I frequently metamod as unfair posts that are mod'ed as troll or flamebait. I imagine quite a few mods hate me because of it, but I try to allow the widest lattitude in expression. Many times, when it is clear that a comment was meant to insightful but failed and was mod'ed falmebait, I just refuse to metamod it. I also mod as fair when the post was clearly flamebait or trollish.

What I am trying to get at is that many posts get negative mod points when the person making the comment was not trying to troll or otherwise be negative. Some mods get offended at the harsh way things are worded. Others, at a sentence that they disagree with. Regardless, I see through these things and metamod (or mod!) to reverse their unfairness.

In conclusion (as ungraceful as my post has been), negative moderations are investigated thoroughly by myself and "wry wit" and other such things get evaluated IN CONTEXT.

Positive moderations are not investigated as thoroughly by me though so some insightful or interesting mods get rated as fair even though they may not be. Informative is an easy moderation to metamod though. :)

strike

Re:Shut up!

[mdwh2]

The fact that this gets modded to 5 Insightful shows that actually, nothing's changed: It's still cool to love Apple...

Re:Shut up!

[idsofmarch]

I guess I think of sigs as a persistent thought, so if yours is always incendiary, then maybe your posts are the same way. I read your available history and I didn't see anything real obnoxious, so your sig seems out of place, but I was allowing for the possibility that you've been immature and were modded correctly. I also take into account that mod-bombing is inherently unfair and goes against how the system should work.

Re:Shut up!

[Geoffreyerffoeg]

Yet even the most mild criticism of Apple or MacOS on /. will result in my comments being moderated down as Flamebait, Troll and Overrated.

I've been thinking about this..and I think it's that there's a single moderation system. There should be two: one which controls post visibility, and one which controls karma (and karma bonus). That way it's easier to M2 a censor as opposed to a moderator, and people like GP don't lose posting privileges for unpopular but not trollish comments. If it costs a mod point to either hide a post or to hurt his karma, fanboy moderators are more likely to use the former - as are vandal trolls, but enough mods will find it at -1 already to hurt their karma. (And of course for good posts you can always use both, but it costs 2 mod points.)

Re:Shut up!

[damsa]

Those silly little quips are what people call trolling. Stop trolling and you won't be modded down.

Re:Shut up!

[guet]

I have also noticed extremist Apple fanboy... from a Powerbook (running Linux admittedly)... It seems Apple fanboys...abusing...Unfortunately this isn't new behaviour for Apple fanboys... As far back as I can remember...most intolerant, the least receptive to criticism, the most judgemental and often the least educated.....never seen an Apple fanboy in full swing. Even the Amiga users were never so extreme. That sort of stupid fanatacism...goddamn I hate Apple users.

everything I hate about Apple fanboyism...Steve...deceitful bullshit...a lying bastard...Yet when Steve does...the Apple fanboys are...rationalising the lies, and moderating or shouting down...anybody who points out that the emperor has no clothes. Apple gets "special treatment" and I find that despicable.

Maybe there is a cabal of Apple users (or fanboys if you prefer) all moderating you down, or maybe you're just gratuitously inflammatory?

As for the speculation that hating Apple is the new badge of honour round here, seems like you got modded up for your rant on Apple fanboys, so really QED.

Re:Shut up!

[IntlHarvester]

Metamoderation used to have a great feature where if you disagreed with more than 2 moderations, it subtracted karma from the meta-moderator. And then Taco would brag about positive M2 stats as proof that the moderation system was working. I would be curious if any biased moderator was ever removed because M2 -- it probably catches the GNAA guys and that's it.

I think any illusion that Slashdot Moderation is designed to provoke a balanced discussion is absolutely wrong. In fact the opposite is the case -- it's designed to promote extremism and silence 'moderate' views. It's the nature of the place, so you pretty much have to lower your thresholds and roll with it. The real battles take place in the comments, not the moderation.

Re:Shut up!

[strikethree]

Metamoderation used to have a great feature where if you disagreed with more than 2 moderations, it subtracted karma from the meta-moderator.

I disagree. I think that is a negative feature. There have been many times that I have seen more than two abuses of the moderation system in a single metamod session. Taking my karma for attempting to repair those abuses is a bad thing.

I think any illusion that Slashdot Moderation is designed to provoke a balanced discussion is absolutely wrong.

The moderation system is useful in that it promotes interesting talking points. I read those interesting talking points and when I think there might be interesting replies, I follow those at -1. I get to read lots of interesting discussions without wading through tons of crap.

I can not tell you how many times I have read some other forum and wished there were a moderation system. Honestly, I believe that the moderation system is far from perfect but it is far far better than nothing at all.

strike

Re:Shut up!

[IntlHarvester]

"Great" was used sarcastically there, BTW. I was trying to argue why M2 is largely useless at 'moderating' the discussion.

I will agree that, while it pretty fails totally on the micro-level, Moderation was a genius idea on the macro-level. If you only want to read 50 or 100 posts and get the gist of what's going on, the option is there. It lets everyone get into their petty arguments, but it also lets the readership 'scale' above that. At least on a good day.

Probably the only thing I would really change would be the ability to mark a post off-topic without changing the karma-level.

Re:Shut up!

[CapnGib]

You had me until... ...the BSOD is virtually gone...

It's a big job making Windows stable, compatible and secure, and it has gotten progressively better, but the BSOD is far from extinct, its not even endangered.

Re:Why?

[AKAImBatman]

Why is there all this talk about running windows on it? Honestly, it's like buying a new Porsche and ripping out the engine and putting a Dodge Neon engine in it.

Indeed. I'm personally waiting for WINE to be ported to the Mactel platform. That should provide pretty much everything people want, as well as provide an easy migration path for video game producers. Or in other words:

1. Compile game against WINELIB
2. Release game for Macintosh
3. Profit!!!

Re:Why?

[MindStalker]

Dual (or tri) booting is always desirable for a tech. If one can run Mac ,Linux (or some unix varient), and MS Windows natively all on one machine they are achieved something dreamt about by the great geek high council from the beginning. It is THE holy triniety.

Re:Why?

[Golias]

Why keep a second computer around for games if you can make the Mac dual-boot?

Of course, the only non-console game I still play is World of Warcraft, so it's a non-issue for me.

Re:But does it run...

[Anonymous Coward]

vMac Mini, a open source Plus emulator, can run System 6 (which you can get from Apple for Free (you can get up to 7.5.5 for free)

It's Called A 'Lie'

[Anonymous Coward]

"being slightly over optimistic"

In the real world that is called a Lie.

Jobs doesn't look like he even remotely concerned about even making -plausible- performance claims about the Intel stuff.

Looks like Jobs is going to be doing a 'optimistic' spinning this year with the mess Intel's Roadmap(tm) looks to be in. He should have been less of a pain in the ass to IBM he wouldn't be in the mess Apple is in with their hardware.

Re:It's Called A 'Lie'

[Durandal64]

What the hell is so unbelievable about Jobs' claims? The old PowerBooks sat on a 167 MHz bus with a processor architecture that was barely on life support. So now you've got a dual-core processor on a PCIe bus architecture and a 667 MHz FSB. 3x faster is certainly not an implausible claim. Same goes for the iMac. If a single-core Yonah is approximately comparable to a single-core G5, then 2x faster for a Core Duo over the single-core G5 isn't terribly unbelievable either. The upper limits of 4x faster and 3x faster might be stretching it a bit, but I'm sure there are benchmarks out there where this is the case. When you give a range like that, you're giving exactly that: a range.

And by the way, Apple's now trashing their own product now. As far as I'm concerned, it's a victimless crime. :)

Re:It's Called A 'Lie'

[Jewdass]

Who here actually watched the keynote? Show of hands? I know I did. Let's all go to www.apple.com/quicktime/qtv/mwsf06/ and load the keynote up to 1:07:00. Steve Jobs is completely up front about which testsproduced the numbers (SPECint_rate2000 and SPEC_fp2000) and outright says "Now, everything is not going to run 2-3x, the discs aren't 2-3x faster, etc." He makes it very clear that his numbers are based off of these two benchmarks. He claims they are the most important benchmarks of performance, which is debatable, but they are certainly a fair test of raw cpu power. Other than the chip and motherboard, the only other significant component that has changed is the GPU, going from a Radeon x600 to an x1600. Does anyone disagree that this is in the 2-3x faster range? All in all, people are making a mountain out of a molehill rather than checking the source of the numbers. god bless the internet. -justinb

Re:It's Called A 'Lie'

[Durandal64]

Watch the god damn keynote. He explicitly says, "This isn't going to be true for all benchmarks".

Can't Boot Windows

[EvilTwinSkippy]

Er, did they try using the "Startup Disk" control panel to tell it to boot off the optical? How about holding down the right key, namely "C"? (The Option Key... it does nothing...)

Re:Can't Boot Windows

[oberondarksoul]

Holding down Option when booting does do something - on Open Firmware-based Macs, it presents a graphical list of the volumes you can boot from. (For example, with my iBook with the Mac OS X DVD in the drive, it shows I can boot from the hard disc, from the install partition on the DVD, or the Apple Hardware Test partition on the DVD). As the review says, it did a similar thing on the new EFI iMac, only it only showed his hard disc as bootable.

Also, apparently on the new Intel-based Macs, one holds 'D' instead of 'C' to boot from the optical drive - presumably because Mac OS X ships on DVDs by default.

OC?

[TubeSteak]

Can it be OCed?

And by OC, I don't mean "The O.C."
I mean Over Clocked.

I realize it isn't in exactly the best form factor to start pushing out extra heat, but someone's going to try it.

Re:OC?

[cnettel]

The reviewer mentioned that he didn't want to void his warranty on this piece of hardware. I guess we'll have to wait for someone not so careful.

Re:OC?

[Tumbleweed]

No, they didn't try overclocking it, but then again, the iMac is hardly the right platform for that kind of thing. No room for 4" high copper-finned monsters. For that kind of thing, wait for:

the replacement for the PowerMac - lots of space inside

or

the replacement for the Mac mini - cheap enough to rip the guts out and mount in a larger case with room for real harddrives, big HSFs, etc. Also, put in a regular dual core version of the processor (assuming it comes with a single core), etc. This should be the modders dream Mac.

Re:OC?

[johnpaul191]

i am sure it could be overclocked and i bet somebody will try really soon. it is worth noting that people report the chip is in a socket, so it could be swapped out in the future. the G5 iMac chips were soldered down.

over the years Apple machines have bounced between sockets, daughter cards and the chips being soldered down. it would be great if they are all in sockets, but i assume the specific design will call for that. some of this may have been waiting till the last second to have the chips delivered and installed?

Re:Why?

[Qubit]

I am a geek. I work on computers. A more flexible computer is better for me.

During the day I work on a multimedia engine that is currently Windows-only, but will soon be cross-platform. At night I hack on my linux boxes, surf the web on whatever webbrowser is on my couch, and laugh along with my friends at the Flash animations they show me. Generally speaking, it doesn't matter what OS I'm running as long as I can browse the web and ssh places.

But when someone asks me a question about OSX, I don't have a test machine to poke at. My lappy can dual-boot Linux and Windows, so if I need to I can switch from one to the other and poke at things. If I could just run one OS and emulate the others on top of it, that would be awesome, but I can't make that work (fast enough) on a laptop today (at least not one I could buy on my budget). So what if I had a triple-boot machine? That would be cool. Of course, being able to run OSX on stock hardware would be even cooler, but maybe Apple just can't handle that much coolness right now. Maybe soon... soon they will be able to be that awesome.

Though I still am wary of Apple's power cords. Too much breakage/sparking of them in the past to forgive and forget this soon. :-(

Best Features of the iMac

[Luke PiWalker]

1) Doubles as reading lamp
2) Automatically emails fan letter to Steve Jobs during start up
3) If you cup your palms over the domed base, your hair will rise in air
4) Sprouts set of cybernetic insectoid legs and scutters away when threatened
5) Perfectly matches the iBlouse
6) Screen is flat, which is good for some reason
7) Special drool tray catches saliva from enthralled technogeeks
8) Communications directly with human pineal gland by firing information-rich beam of pink light
9) Wuvs you

Stolen from The Onion of about 4 years ago but still true today.

Re:Best Features of the iMac

[poot_rootbeer]

Stolen from The Onion of about 4 years ago but still true today.

Except for the part about the domed base, which iMacs no longer have.

Re:No AMD macs?

[aSiTiC]

Leapfrog straight to the Turion?? Hmmm... how did oversee that opportunity??

Waiting for the second generation

[G4from128k]

Look at the history of Apple's processor switches. The first generation PPC machines (6100/7100/8100) were nice, but the second generation PPC machines (7500/8500/9500) were much better. The 2nd gen PPC machines had PCI instead of NuBus, a faster interleaved memory architecture, and a much improved dual-SCSI bus. With the first Macintel, it's obvious that Apple worked very quickly to put Intel Inside and I'm sure that some parts of the design represent a borrowing from PPC designs. I bet that second generation Intel machines are both faster, less likely to have flaws, and more likely to enjoy longer-term OS upgrades.

I know its ungeek of me not to want to be on the bleeding edge, but I'm waiting for the second generation machines.

Re:Waiting for the second generation

[djohnsto]

By all accounts, the new machine's insides are practically designed by Intel. Intel CPU's, Intel chipsets, probably Intel motherboards. While the SW for Intel will definitely get better over time, I doubt the HW will be less bug-prone (it's already very solid).

The only big change on the horizon is the switch to Merom/Conroe/Woodcrest in the second half. This will bring the eventual switch to a fully 64-bit OS X.

Re:Waiting for the second generation

[Moby Cock]

Of course the second generation of Intel Macs are going to bet better than this generation. The third generation will be better than the second. Maybe you should wait for them instead...

Re:Waiting for the second generation

[geoff2]

I think the 68k-PowerPC transition is not quite apt. You are correct that the first generation of PPC boxes were not nearly as nice as the second generation boxes, which had much more extensive changes than just the processor. But I think there was a lot more room for improvement in the last 68K boxes versus the PC state of the art at the time than there is in the last PPC boxes versus the PC state of the art. The NuBus expansion card bus versus the PCI bus is just one example. Both the last PPC boxes and the first Intel boxes have the latest and greatest interfaces such as USB2 and PCI Express.

In addition, the first Intel box is not a motherboard that Apple slapped together on its own, like it did for the first PPC boxes. It is a state-of-the-art Intel motherboard with all the latest doodads. Sure, Apple could stick in a fastre graphics card or hard drive, but the motherboard support chips are all modern. I think the next rev of the MacBook Pro will include Firewire 800, which I assume Apple couldn't integrate into the MacBook Pro motherboard in time to meet their ship date, but that's more of an incremental change than was included in the second-generation PPC boxes. (And don't forget the problems with the then-new PCI-bus Macs. Networking was so broken -- remember the Open Transport fiasco? The brand-new networking architecture that wasn't ready at the time the 7200/7500/8500 were released and which those new machines relied on, MacTCP having been deliberately obsoleted? It took several months after the boxes were shipped for *any* PPP dialing software to work at all with the Mac, and it took more than a year after that until most of the more significant networking bugs to be quashed.)

Re:Waiting for the second generation

[hritcu]

Better wait even more and buy a second generation Intel-powered Apple from eBay for 300$, or even less.

Re:Waiting for the second generation

[cnettel]

The 64-bit switch is exactly the kind of "WTF, why should we support gen 1" thinking that could be going on, stopping future updates. The same would possibly apply to everything; Office, Photoshop, whatever -- if quartz and other UI components are also made 64-bit, 32-bit Intel Mac applications will be nothing more than a fad.

Considering that 64-bit UI isn't present on current PowerPC Macs, this doesn't have to happen, but I can understand the worries against joining the 1st-gen bandwagon right now. In this situation, even a PowerPC Mac might be supported in a more consistent manner in a few years time.

Re:Waiting for the second generation

[podperson]

There's a huge difference here ... Apple designed the original PowerPCs on their own. This is just a state-of-the-art PC in an iMac box with a little hardware magic to handshake with the OS.

It's also worth nothing that when the first generation PowerPCs came out, they were just dandy. It's not like contemporary PCs had a better expansion bus (Apple was pretty close to the leading edge in adopting PCI) or graphics or whatever. For that matter the 8100 et al went through two generations before being supplanted by the 604-based boxes.

Re:Waiting for the second generation

[Ulrich Hobelmann]

It's not really a new architecture, like the first PPCs were. PCs with Intel CPUs have been standard for decades, and Intel did a lot of the design work (mainboards probably), so I'm quite positive that there won't likely be any big problems or surprises.

Re:Waiting for the second generation

[DECS]

The notable difference:

68k to PPC:
1) Last generation of 68k Macs was custom-weird-NuBus.
2) First generation of PPC Macs was the same + PPC
3) Second generation of PPC Macs was PPC + Old World PCI
4) G3 PPC Macs were PPC + New World PCI

ie: Apple first put PPC into their old Mac design, then fixed the bus, then brought everything else up to date.

Result:
Users who bought the last generation of 68k Macs ended up with rapidly obsolete, dead end crap.
Users who bought the first generation of PPC Macs are ended up with new wine in an old wineskin.
Users who waited for the second or third generation of PPC Macs were far better off (although it was a LONG wait)

PPC to Intel
1) Second to last generation of PPC Macs was PPC + PCI-X
2) Last generation of PPC hardware was PPC + PCIe
3) First generation of Intel = nearly identical to existing PPC, but with Core Duo

ie: this time around, Apple completed their transition to a modern bus & hardware prior to introducing the Intel chips.

Result:
Users who buy the last generation of G5's are not really screwed with old crap.
Users who buy the first generation of Intel Macs are not screwed with old crap or handicapped potential performance.
Users who wait for the second generation of Intel Macs will not benefit as much.

Certainly nobody is under the illusion that PCs in general are ever immune to rapid obsolescence, but I think the transition planning this time around is far better executed.

Re:Same Good looks

[oberondarksoul]

It's not a Power Mac, nor has it ever tried to be - one wouldn't have bought the original bondi blue iMac when it first came out if they needed the power of the Power Mac G3 then, either. Also, Jobs admitted that Rosetta wasn't really professional-level yet - Photoshop and the like are usable, but professionals will want to wait for the x86-native releases. They'll come, and so will an Intel-based Power Mac - until then, the G5 Quad will be more than sufficient and will last a long time to come.

Why is it using 945GM?

[denjin]

Just wondering. I checked Intel's pricing (in 1000 quantities), and it costs like $4 more to get a 945GM+Yonah than it does to get a 945PM+Yonah. On top of it, the iMac doesn't use GMA950 anyway. 945PM Chipset shortages?

Re:Why is it using 945GM?

[cnettel]

Or a volume deal where the GM is used in some of the other models.

Cell isn't a desktop processor

[Andy Dodd]

It has insane floating point throughput capability which will help on some apps, but for most desktop apps the Cell is extremely slow. It was designed for a very specific set of tasks.

Existing PPC binaries won't run fast on the Cell. In fact, they most likely won't run at all.

There is no way we'll see a general purpose desktop system based on the Cell - it's just not designed for that kind of purpose. We might see some sort of Cell coprocessor board become available though.

Re:Cell isn't a desktop processor

[drinkypoo]

IIRC the Cell's SPE is a 64 bit PPC. I don't remember it having Altivec, though. Thus [if I am correct] any non-altivec PPC code should run fine, but get no boost out of the Cell PPEs, and altivec-requiring PPC code should fail.

Re:Cell isn't a desktop processor

[maynard]

Yeah, I agree, A Cell based workstation would be terrible for Word and Excel. However, for editing HDTV the Cell processor has significant advantages over every other general purpose processor. Since Apple has targeted the film/video workstation market it would seem to make sense that they consider the Cell for that application. *shrug*

Re:Cell isn't a desktop processor

[Andy Dodd]

How is saying that the Cell is a totally unsuitable CPU for a desktop machine "Mac fanaticism"???

Re:Cell isn't a desktop processor

[Andy Dodd]

It's a very, VERY stripped down PPC. Same basic instruction set, but missing a lot of features. (For example, I don't believe it has any instruction reordering capabilities, just like the Xbox 360's CPU is missing many features needed for good "general purpose" performance but not needed for specialized systems such as a gaming console. Those missing features might cause unmodified PPC binaries to fail miserably despite having an instruction set that is mostly identical.

Either way, there's no way we'll be seeing a general-purpose Apple desktop relying solely on a Cell CPU as its processor. Possibly a dedicated Apple media center device, or a coprocessor, but not as a desktop main CPU.

Re:Cell isn't a desktop processor

[maynard]

It has Alivec, using the IBM name: VMX. Same thing though. What the other guy said about it lacking instruction reordering is true though. It also lacks cache locking instructions (that the Xenon has) due to it being a single core. It would be plenty fast enough on its own as long as one optimized for reduced branching though. And, of course, there are those SPEs - which are silly fast as long as you have a parallel single precision FP problem, easily vectorized. Like HD video editing, for example! :)

Re:Cell isn't a desktop processor

[Andy Dodd]

From what I recall, the coprocessors on the Cell were optimized for floating point calculations.

Most video codecs use integer calculations. The Cell won't really be that much of a benefit.

It would kick ass for a software defined radio for HD reception, but not the actual video encoding/decoding.

Re:Cell isn't a desktop processor

[maynard]

Well, each SPE vector unit supports single and double precision FP, and ints. Though if you can't vectorize your code you'll be wasting a good deal of its potential. However, nonlinear editing is well suited to vector code. Each SPE also operates independently of each other and the PPE (the PPC core). Data transfers to/from each SPE happen through independent DMA channels. One can move data in and out of main RAM (slow) or down a set of chained of SPEs (very fast) for processing. Based on what I've read at the IBM developer site I think the Cell is very well suited for nonlinear editing.

This is not to say that it's a panacea chip. With no out of order execution unit developers will have to optimize out as much branching as possible. A Cell target for GCC will help somewhat I suppose. But the PPE alone is no speed demon, for sure. Which makes it a pretty anemic desktop CPU.

From a gaming standpoint, I think MS may have made the better choice with Xenon. Especially for RPG, real time strategy, and game AI. But if you have highly vectorized SP FP code, Cell spanks everything else out there (if the benchmarks IBM claims are true).

Re:Why?

[TheGuano]

It's more like taking a Mercedes (Porsche? Let's go with something that implicates experience/luxury rather than blazing speed) and and putting a Chevy pickup truck chassis on it. It may not be pretty, but sometimes you just need to haul stuff.

Seriously...

[Moby Cock]

I'm still waiting for the most important benchmark: frames per second in molten core combat.

Does anyone relly have any info on this? Even anecdotal evidence would be appreciated.

Re:Seriously...

[nappingcracker]

Mooolten KombaaaaT!

Jobs Wins

Reality

Re:Why?

[yoyosmuggler]

i thought cheap and dell dont agree with eachother.

Re:Why?

[tgd]

Thats a bad analogy. Its more like buying a race car and deciding you really want to have turn signals, seat belts, emergency brake and other stuff you need to drive your vehicle once in a while on the same roads everyone else is on.

molten core

[poot_rootbeer]

the most important benchmark: frames per second in molten core combat.

I'm pretty sure that if you overclock your Dual Core to the point where it becomes "molten", your FPS rate is going to be Zero.

32-bits?

[Nom du Keyboard]

Is this a 2x32-bit processor? The Intel PDF's on it (search Intel for 'core duo') don't seem to want to decode for me today.

I can't believe Intel is building anything new these days that isn't AMD64, but I've already had a couple people tell me I'm wrong about Core Duo.

If so, why would anybody buy it at these prices?

Re:32-bits?

[MBCook]

Yes, they are 32-bit.

I agree, I find that a bit dissapointing. But when it comes time to replace the G5 towers, they will HAVE to go with something with x86-64. After all, 64-bit computing was one of their big selling points for moving to the G5 processor.

Re:32-bits?

[MarcQuadra]

Alright, Lets get a few things straight:

          These machines don't have a need for more than 4GB RAM, they only fit 2GB, so why go 64-bit when it's not going to be used to address more RAM? Also, Intel's AMD64 instruction set is SLOWER than their IA32 instruction set. The machines would be slower if they were Intel 64-bit machines.
          AMD's (and IBM's) chips are optimized for 64-bit operation while retaining full 32-bit support, Intel's chips are optimized for 32-bit use and provide the ability to execute 64-bit instructions.
          Trust me on this, I've got a Pentium D under my desk running Linux, and it's slightly faster running 32-bit top-to-bottom than it is with 64-bit optimized builds. The story would be different if I had an Athlon64, and it doesn't matter a hill of beans anyway unless I needed that huge address space (which I don't, I'm quite content with 1.5GB RAM).

Re:32-bits?

[nagora]

I can't believe Intel is building anything new these days that isn't AMD64,

Well, I can't believe people are still putting Intel chips into their machines, so there you go!

TWW

Re:32-bits?

[argent]

Intel's AMD64 instruction set is SLOWER than their IA32 instruction set.

You're pulling our legs. Please tell us you're pulling our legs. No?

Intel managed to make an AMD64-compatible processor that couldn't take advantage of the larger register file?

Remind us why Apple decided these clowns were the bees knees again?

Re:32-bits?

[argent]

Since the Power architecture doesn't have any really crippling shortcomings like the x86 small register file, the 64-bitness of the G5 is more a marketing advantage than a practical one... the number of people who need and can use large address space on the G5 can probably be counted on the fingers of one foot.

Re:32-bits?

[drinkypoo]

Intel has spent many years dealing with register starvation and they make excellent use of register renaming to deal with context switching.

Re:32-bits?

[Guy Harris]

Intel's AMD64 instruction set is SLOWER than their IA32 instruction set.

Really? A 64-bit Pentium D is slower than a 386SX?

Oh, what you meant was "Intel's current chips that implement AMD64 run slower in 64-bit mode than in 32-bit mode". That doesn't necessarily mean future implementations will be slower in 64-bit mode. The chips using the "Next Generation Microarchitecture" might not have that problem.

(Well, technically, that might be "implement EM64T". The Wikipedia article on EM64T, at least, claims there are some differences, although, if it's correct, some of them went away in newer chips, and others are just standard AMD vs. Intel differences, e.g. the lack of 3DNow! instructions in EM64T.)

Re:32-bits?

[MarcQuadra]

You're pulling our legs. Please tell us you're pulling our legs.

        Nope. From my understanding, which is very abstract and rudimentary, Intel's bottom-up 64-bit platform, the IA64 has all but failed, so they took the existing P4 'netburst' core and added instructions to the microcode that let it break-down and process AMD's 64-bit instructions. The core of the CPU changed little if at all for Intel to process 64-bit instructions, but the way they do it is a stopgap measure until they can build a real 64-bit core.
        Remember that the x86 series from the i686 on has been an odd RISC/CISC hybrid, there's (oversimplified) a very small fast CPU at the core, with lots of logic around it to decode and optimize x86 instructions for it. It's essentially a RISC chip emulating a CISC x86, but less of an emulation and more of a hybridized design. Intel had similar issues when they debuted the PentiumPro and Pentium II, 16-bit apps were much slower on the new chips than the original Pentium because the chip was designed to handle 32-bit instructions, 16-bit code had to be 'pulled apart' more to feed efficiently to the new core. This is the same problem but in reverse.

Re:32-bits?

[argent]

"so they took the existing P4 'netburst' core and added instructions to the microcode that let it break-down and process AMD's 64-bit instructions"

OK, so this comment has nothing to do with Yonah/Core Duo/any future Intel chips, since there's absolutely no similarity between the P4 and PIII cores and none of the changes in the P4 core will be applicable to the PIII-based chips they'll be coming out with in the future.

Re:32-bits?

[argent]

Intel has spent many years dealing with register starvation and they make excellent use of register renaming to deal with context switching.

I'm sure they're doing the very best they can with their crippled architecture, but they can't do much about the laws of physics. Despite spending massive amounts of time and money, not to mention transistors and chip area, their processors don't display the kinds of performance they should be able to manage with that kind of budget.

As for IA64, Intel has a long history of being hopeless incompetants at instruction set design, a grand tradition that goes far beyond the dysfunctional 80*86 family and includes true-blue eccentrics like the 4004, the iApx432 and the i860, and of course today's IA64. The closest thing to a decent modern chip they've done is the i960... which they promptly caponised and relegated to the embedded market lest it compete with their precious x86.

So... no matter how well Intel has done, they could hardly help but do even better if they could leave at least some of the handicaps of the x86 behind. The question is... will they?

Re:32-bits?

[gnasher719]

'' Intel has spent many years dealing with register starvation and they make excellent use of register renaming to deal with context switching.
  ''

Intel processors can execute up to three operations per cycle. Every time a program has to spill a register to memory and load it back later, it costs four operations (one for address, one for load or store). There is no way to get these four wasted operations back.

Add 1 to register = one operation. Load from memory, add one, store to memory = 5 operations.

Steve doesn't lie

[palad1]

He just momentarily bends reality to his will.

He's Q, with a turtleneck and a pair of jeans.

Re:Why?

[hungrygrue]

It isn't flame bait. If you have applications that you can't live without that only run under Windows, run Windows under a VM for those rare occasions that you will need it until you can find replacements. And if your problem is games, then go get a playstation/xbox/whatever - the pain that Windows will cause is not justified by the need to play Slaughterfest 3 or Unreal Shooter. Better yet, read a book, get out of the house, talk to real people that don't have to be rendered on your screen. Dual booting Windows is just dumb, if too many people start doing it, how long do you think it will be before there are viruses which infect your OS X partition while you are running Windows? Windows is not secure enough and should not be given access to the actual system - at least with a VM, you can back up the disk image and simply revert to a clean install on a regular basis without it having the ability or opportunity to access your actual hard drive.

Re:No AMD macs?

[man_of_mr_e]

Actually, the new Yonah processor is significantly faster than the Turion. Check out these benchmarks:

http://www.tomshardware.com/2006/01/16/will_core_d uo_notebooks_trade_battery_life_for_quicker_respon se/

the specs will not change drastically

[asv108]

That was my thought initially too, wait for the second generation. Its probably still good advice, but I think they fact they are using a widely implemented Intel product will increase the reliability of these first generation Intel macs. My first generation pentium-m is still a good performer today. Apple will be using the standard processors, boards, and components of the Centrino duo, which will have a long a lifespan. There is not going to be some earth shattering change in the Macbook components a year from now. Apple will be using the same Intel specs that every other major pc laptop manufacturer will be using, so we can probably through all the first-gen history out the window.

Re:Why?

[dantheman82]

I'd definitely consider biting if I knew this could be the case. There is VS.NET which I use at work (C# primarily) which I really need to run on the machine at times. However in 80%+ of cases, I'd just stick with the OS X. I could even move from Outlook to Apple's Mail app as it's likely at least as good and less of a memory hog (I hope). I'd like a fast machine to crunch VS.NET though... Ah well, wishful thinking.

Re:No AMD macs? Excuse me!!

[Nom du Keyboard]

MD can't produce the volume of chips that Apple need.

Excuse me, but while Apple is big on noise, they're not big on production. I'm sure AMD could have given them all the chips they need. They might not have been so forthcoming with the Marketing Money however.

For Intel, getting Apple is a coup worth paying enough for that even if they never make a cent from Steve Jobs, they've still silenced the biggest critic of the i86 architecture.

Their problem right now is keeping Dell/HP happy, both of whom sell a lot more systems than Apple will, and who aren't very pleased about Apple being allowed to announce the newest, latest and greatest systems first.

Re:Molten Core Combat?

[bajan_on_ice]

Pretty sure he is talking about World of Warcraft. There is a dungeon there called Molten Core which is infamous for the game lag it produces...

Re:Molten Core Combat?

[Theaetetus]

Molten Core Combat? Is that another one of those games that came out for the PC 6 years ago and just got ported to the Mac?

It's an inside joke, referring to an area in the most popular MMORPG around, World of Warcraft. That's by this company called "Blizzard", and it runs on things called "computers". It comes on a couple of "CDs", which is short for "Compact Disc". Someday, when you get electricity in your cave, you might be able to try it too!

To get it in the door.

[Shawn Parr]

A lot of people don't seem to see why you would really want to do this. Many businesses and universities are very PC only, even though they are curious about using/integrating Macs.

When I worked for a satellite campus of Indiana University, I was directly in charge of 2 Mac labs and 6 PC labs. One of the departments seriously was discussing moving their 3 PC only labs to Intel Mac if they could dual boot.

Why?

Because then they could still have all the software they already use, and the faculty and students could reboot into Mac OS to play around and see what was up with it. Surprisingly it was the most hard core PC only faculty that was really excited about this option. A chance for them to work in the same room(s) they always work in, but try out the other side.

Many of the students we encouraged to work on both ended up purchasing a Mac, but that is only because they had the ability to get exposed to them. In many universities this isn't the case anymore.

Most of the students where using Adobe and Macromedia products, but the fastest to consider switching where the Maya students. Showing them a lower spec'ed Mac easily outperforming our PC's (Opterons BTW) was very interesting. The only way we got better performance was on the PC's under Linux, which after trying that most students found they did not like dealing with Linux (sorry, I'll probably get mod'ed down for that, but it was truthfully what I saw).

Apple not hindering the ability to run windows, which they have stated they haven't done, and I have no reason to believe they have, in some cases could actually net them quite a few sales they would not have been able to get otherwise.

I disagree

[bogie]

Look at it this way. OS X has been running in parallel on X86 for years. X86 is a well known platform that 90% of the computing world uses. What kind of "hardware flaws" are you expecting to see?

If Apple had moved to say xyz platform that was not being widely used for computing then IMHO you would have a point. But that isn't the case so I don't think you theory about there being something wrong with this generation of Intel hardware is accuarte. The "next" gen Mactels won't be 2nd gen, they will be 15th gen or whatever of a stable and widely used platform.

I do expect OS bugs to crop up simply due to a wider sample of users, but I also expect next gen Mactel hardware to look just like next gen PC hardware. Afterall Apple now essentially makes "IBM PCs".

Re:No AMD macs?

[neildiamond]

This may have been the old AMD, but didn't they just outsell Intel in quantity for two months straight (at least in the US market)? (Intel remains more profitable, of course.)

Google Link

Re:Why?

[Shivetya]

Because OS/X doesn't fulfill all the needs some of us have? I would not mind have OS/X around but damn if I am going to have a machine just for a OS.

I want OS/X, XP, and Linux on one machine. The benefit of buying the Mac is that is should be easier to support multiple OSes on it as the hardware is pretty much guaranteed.

As for the experience, if OS/X was as beautiful and stable as all the people claim a couple of my friends would not have jobs supporting people who use it.

I have yet to find a really stable OS outside of midrange computers, anything on a PC type of hardware seems to come with instability built in

Yes, 32-bit...

[SuperKendall]

The 64-bit version are out later this year.

Not that many people actually need 64-bit capabilitity, mostly for programs that need very large memory access. iMac users certainly will not, Macbook Pro users is more questionable - my guess is they will upgrade that line with the 64-bit chip at the same time they release the Intel Powermac equivilent.

Hmm, that leads me to wonder what the new name for the Powermac will be... MacMac?

Re:Yes, 32-bit...

[DavidpFitz]

Hmm, that leads me to wonder what the new name for the Powermac will be... MacMac?

BigMac?

Re:Yes, 32-bit...

[argent]

Not that many people actually need 64-bit capabilitity, mostly for programs that need very large memory access.

The main advantage of AMD64 is not the 64 bit address space, it's the larger register file. The small register file of the x86 family really cripples it: register renaming can be used to hide some dependancies and avoid bubbles in the pipeline, but the renaming has to operate in real-time over a tiny instruction window, and the compiler still has to generate (and the CPU still has to execute) code to shuffle data into and out of memory... and hitting even L1 cache has a cost.

Normally you would expect a processor with a real 32-bit and 64-bit mode to run faster in the 32-bit mode, because it's moving less data. The only exceptions to this are when the 32-bit mode is somehow crippled. For example, on the Alpha 32-bit code still has to wait on 64-bit operations and memory transfers behind the scenes. And for Intel processors, the x86 instruction set is crippled by the small register file.

And THAT is why people want 64-bit modes in their Intel processors, because they're faster and more efficient... even on otherwise 32-bit code... for reasons having nothing to do with their larger address space.

Re:Yes, 32-bit...

[Senjutsu]

Hmm, that leads me to wonder what the new name for the Powermac will be... MacMac?

Based on the MacBook Pro name, the most obvious candidate is Mac Pro.

Re:Why?

[treerex]

I would expect that Apple is putting pressure on Microsoft to get the next release of VirtualPC working on the MacIntel machines so that you get performance similar to that of the existing Windows version when running X86 code. In other words, instead of emulating the processor (as it does now on the PPC) it uses the native CPU as much as possible. Then you can boot XP or Win98 inside Virtual PC.

Re:Why?

[slashdotnickname]

Honestly, it's like buying a new Porsche and ripping out the engine and putting a Dodge Neon engine in it.

Jesus Christ man, where are your nerdly priorities?

Sure, the Porsche might get me some hot babes... but the Dodge Neon will allow me to pwn teh n00bs!

Spanning

[ronanbear]

The intel iMac supports spanning! I'm surprised Steve didn't make a big deal about this. There goes one more major reason for people to buy a powermac. Kudos for Ars for mentioning that on the first page.

Re:Spanning

[Blakey Rat]

What the heck is spanning? Spanning between what and what?

Re:Spanning

[Scudsucker]

Spanning between what and what?

Monitors. He's talking about "spanning" your desktop across multiple monitors, something not allowed with iBooks and older iMacs. They had video ports, but only for mirroring your display. This was not a limitation of the graphics hardware, but a limitation created by Apple to provide a reason to get a Powerbook or G5 tower. There are firmware hacks around to renable this ability; I hacked my G3 iBook and it can use a second monitor at up to 1600x1200.

Re:Spanning

[rthille]

Even more important (since you can hack previous iMacs for spanning) is that it's got DVI out.

Re:Spanning

[777film]

The intel iMac supports spanning! I'm surprised Steve didn't make a big deal about this. There goes one more major reason for people to buy a powermac. Kudos for Ars for mentioning that on the first page.

I could be wrong, but I bet it only supports spanning because they didn't have time (or it wasn't worth the trouble) to cripple it.

I'd lay odds on the next generation of iMacs losing that feature. Which may explain why the Apple publicity machine is mum about it...

Re:Console

[wangmaster]

Yeah, and we saw how well this method worked on Linux. I'm still waiting for increase marketshare to get us games :)

Console gaming is starting to get there, but for a sizeable chunk of gamers, it's not there yet. FPSes with a console controller suck. Simulation games suck on a console due to lack of input options. With newer consoles able to support USB (and usb keyboards/mice) this is bound to get better, but gaming on a PC for many people still beats consoles hands down.

Re:Why?

[MSFanBoi2]

There are LOTS of other similar hardware PC's out there for less money. Apple has claim to nothing other than a limited upgradabiltiy all in one unit.

Re:Same Good looks

[HTH NE1]

The concern of myself and many other mac users is the Photoshop/Illustrator/Final Cut/Shake performance.

Regarding Final Cut:

Some reportedly won't launch at all, like Final Cut Pro 5 and Logic 7. I tried launching FCP 4.5 and it wouldn't run, saying that I needed an AGP card. I tried to tell it that PCIe video cards are much better, but it wouldn't listen.

There are ways to make it install and allow itself to be run, but I wouldn't be confident it would run smoothly until a Universal Binary is built. Reports have it that not everything will run under Rosetta.

Re:Why?

[timeOday]

It's not just games. A lot of people including me have been tied to Windows for a long time. I've considered a "switch," but I'm not going to do it bellyflop-style, all at once. Getting a MacBook and dual (or triple) booting will be a safe way to test the OSX waters (assuming Windows runs smoothly).

Apple should not just allow Windows to run on them, they should support it. It can only grow the market for their hardware.

Thankfully...

[SuperKendall]

Console gaming is starting to get there, but for a sizeable chunk of gamers, it's not there yet. FPSes with a console controller suck.

Well I agree with you there, thankfully FPS's are one area where the Mac is not as far behind in ports as other areas.

"forced to buy two computers"??

[sczimme]

Why should they be forced to buy two computers just so that they can preserve their "entire Mac experience"?

No one is being "forced" to do anything in this context. The choice - and it is a choice - is left to the individual user. Please don't be so melodramatic.

No way, that's a myth.

[cbreaker]

That's entirely untrue. You can easily spot the difference in smoothness between video captured in 24FPS and video captured in 60iFPS.

Try playing a video game at 25FPS, and then at 60FPS. Can't tell the difference? If you can't, you've got to be full of it.

Maybe your brain can't absorb all the information on each frame past a certian point, but *anyone* sure as hell can see the difference when it comes to smoothness and fluidness of movement.

And a note about GLXGears - the higher the number, the better chances of getting more complex objects on the screen at a decent frame rate. If you haven't noticed, games are a little more detailed then GLXGears. So while you can spin a few objects at 2000FPS, you might only see 20FPS in the latest game title. But if you get 10,000FPS in GLXGears, you'll probably see much higher performance in the game. It's a BENCHMARK. Seriously.

And what does this mean: "a lot of resources are wasted computing and rendering"? Explain to me what else you want the computer doing when you're running a graphics benchmark? I want mine running the damned benchmark, what else? It's not like everyone's machines are attempting to cure cancer and we should let that happen at all costs. I buy fast computers because I want to use all of the speed, not have an abundance unused in the background.

Re:No way, that's a myth.

[TheRaven64]

Try playing a video game at 25FPS, and then at 60FPS. Can't tell the difference? If you can't, you've got to be full of it.

When you are seeing a game at 25fps, you are likely to be seeing it at 6fps for a fraction of a second, then 50fps for a fraction of a second, etc. The reason you need 60fps average in a game is so you get 25fps or so in the worst case. Record a game being played at 100fps, then create movies of it at 50fps and 25fps constant frame rates, and you will see what I mean - there won't be much difference in visual quality.

Re:No way, that's a myth.

[mjg59]

glxgears is not a benchmark. Differences in the way the drivers handle synchronisation between frames mean that the limiting factor may be the number of frames per second, not the complexity of the objects drawn. Going from Xorg 6.7 to 6.8, my glxgears FPS values dropped. My UT ones went up. This ought to tell you something.

Re:No way, that's a myth.

[ZiakII]

When you are seeing a game at 25fps, you are likely to be seeing it at 6fps for a fraction of a second, then 50fps for a fraction of a second, etc. The reason you need 60fps average in a game is so you get 25fps or so in the worst case. Record a game being played at 100fps, then create movies of it at 50fps and 25fps constant frame rates, and you will see what I mean - there won't be much difference in visual quality.

Not true at all launch good ol' half-life and go into the console and type in max_fps 25, and tell me you don't see any diffrence .... you would be lying and there is no way my 3.0 ghz ati x600 can't run the game strait with any framerate droping below 25 MAX fps.

Re:No way, that's a myth.

[LetterRip]

[QUOTE]That's entirely untrue. You can easily spot the difference in smoothness between video captured in 24FPS and video captured in 60iFPS.

Try playing a video game at 25FPS, and then at 60FPS. Can't tell the difference? If you can't, you've got to be full of it.[/QUOTE]

You have confused video and video game - videos will tend to have good (or real) motion blur and hence can be percieved as smooth even at lower frame rates. Games tend to have poor if any motion blur, and thus need a much higher frame rate to appear smooth.

LetterRip

Re:No way, that's a myth.

[cbreaker]

It certainly is a benchmark - anything that's used to compare is a benchmark.

Obviously, glxgears isn't a comprehensive benchmark as it only utilizes a small set of GL functions - but everyone uses it to see if their video drivers are performing up to where they would expect basic functions to run.

Re:No way, that's a myth.

[cbreaker]

I'm not confusing them at all - there's no real difference between the two. You certainly notice low frame rates in video games more then a video, however - and it's not just motion blur. Some of it is response time and the fact that you control it, some of it is the fact that you sit very close to a screen when you play video games as a general rule. Another factor is that video games are still not life-like enough to pass as "real" and smoother video helps offset the lack of realism in graphics by adding more immersion.

Next time you watch a movie, pay attention to pans across landscapes and such. Usually a DVD is sourced from 24FPS film, so it applies here too. You can usually easily see the jerkiness of the video when it pans. Then, watch some panning video from a home camcorder which usually records at 60 interlaced frames a second. The difference is immediately noticeable.

My point is, the human eye is perfectly capable of perceiving well over 25FPS. 24FPS is the standard for movie film, and it's really the minimum you can use and still have it seem fluid enough. Any lower and it's distracting. Any higher and it looks strange because we're so conditioned to 24/25FPS. That's why home video tends to look like exactly that (cheesy) - it's a much higher frame rate.

Video games exasperate the issue, and frame rates mean even more. 60FPS is smooth enough for most people that it seems perfectly fluid, which is why the industry has pretty much standardized on it as a base-line.

Re:No way, that's a myth.

[mjg59]

And everyone's wrong. The only thing that glxgears tells you is how fast glxgears runs. It's astonishingly unrelated to the speed of almost any other 3D application you can run, with the possible exception of the screensaver that's based on the glxgears code.

In the real world, useful applications don't run at 1000FPS. As a result, there's no real need to worry about per-frame overhead. When you're running a game at 30FPS, it's negligable in comparison to the drawing time. When you're running a piece of crappy code that does nothing other than draw spinning gears, that per-frame overhead is suddenly significant and may even dwarf the time spent drawing the application. Does this tell you anything about the speed of your card and drivers? No. Does this let you make comparisons between driver releases? No. Does this let you make comparisons between different drivers? No. It just results in people making unfounded statements about how Y's drivers are 3 times faster than Z's under Linux, when the information provided says nothing of the sort.

Re:No way, that's a myth.

[cbreaker]

In the "real world" applications also don't run at 3FPS, which is what some parts of 3DMark give you even on fast machines. Does that make the benchmark useless too?

Your argument here is that you personally don't believe glxgears is an accurate measurement of overall performance. I agree with you. But it's still a benchmark. GLXGears does tell you how fast glxgears runs: on your hardware, with your software. I can't control what other people do with glxgears data, and I don't personally tout glxgears scores as a measurement of how good my system is. I won't say it's not a benchmark just to stop these people, though.

It's still useful. If I boot up a linux box and run GLXGears, and I see 40FPS, I know that A) The system is severely mis-configured, B) There's no GL accelerator, C) Maybe I'm using it over the network. If I run glxgears and see 3000FPS on my workstation, I know I'm using the OSS driver, not the nVidia one where I'd see 30000.

Re:No way, that's a myth.

[Strolls]

Try playing a video game at 25FPS, and then at 60FPS. Can't tell the difference? If you can't, you've got to be full of it.

Ha! Empirical evidence! I knew my Mom was full of it!

Re:No way, that's a myth.

[McTaggart]

That's not the point; he was using it as a benchmark.

Re:No way, that's a myth.

[Thorgal]

Why is then necessary to pass "-iacknowledgethatthisisnotabenchmark" as an argument in order for it to display framerate values? ;)

Re:No way, that's a myth.

[Paul Jakma]

Obviously, glxgears isn't a comprehensive benchmark as it only utilizes a small set of GL functions - but everyone uses it to see if their video drivers are performing up to where they would expect basic functions to run.

So that makes glxgears a unit test, not a benchmark. You're deluded if you think the fps of glxgears is in anyway generally meaningful.

Re:Why?

[cypherz]

"...it's like buying a new Porsche and ripping out the engine and putting a Dodge Neon engine in it."

Not arguing your point about the usefulness of Winders on a MacTel machine, but my Dodge SRT-4
(essentially a Neon) will whip the crap out of a Porsche Boxter (and LOTS of other more prestigious cars). Putting a Neon engine in a Porsche ISN'T necessarily a bad idea! :-p

Putting Winders on a MacBook... I wish there was a VMware solution for this. I don't want to have to permanently install Windows on ANYTHING anymore. The only OS I want my macs booting is OS X.

VMware are you listening?

Re:No AMD macs? Excuse me!!

[Anonymous Coward]

At something on the order of 80 million chips per year (counting dual-proc, dual-core machines as two, obviously) Apple would instantly become AMD's largest customer, by far. There's no way AMD could hope to keep the pipeline full, and Apple's biggest constraint to growth has always been supply-chain issues.

Re:Why?

[DGregory]

I do a lot of work with embroidery and so have embroidery software as well as an embroidery d-card reader/writer and not only do they not run on my Mac, but if I did run virtual PC, I'm not sure that it would find the reader/writer but it requires some hefty computing power and a virtual PC that's already slow would be REALLY DOG SLOW if I tried it. So I run it on my husband's computer and use my Powerbook for pretty much everything else. There's really no choice for embroidery software, the one program out there that runs on a Mac costs like $4000.

Having a dual-boot computer would definitely be ideal.

benchmarks schmenchmarks...

[capsteve]

benchmarks are such bullshit. that's like saying you can determine the performance of any car by just looking at the tachometer and speedometer while driving on a race track... i say show us the real world tests, man!

i suspect that performance in areas such as gaussian blur and other filtered transformations in photoshop will show that the g-5 dual core is as spritely as the intel dual core, if not more so... and quartz composer would be another area that the g-5 dually would compete well.

i'm not nay-saying the core-duo, i'm just being cautiously optimistic, and i suspect that the g-5 proc will be in apple products for a while, even after the transition... why not maintain OS code for both processors, apple has been doing it for years now, and if their major software partners are committing to UB, then having a few ppc options trickling for a couple more years will be tolerated...

Re:benchmarks schmenchmarks...

[podperson]

Quartz Compositor mainly runs on the GPU so the CPU shouldn't make a significant difference.

Re:Why?

[John Jamieson]

Insightful?

Not all Porsche's had guts. In fact, some of the VW powerplants were anemic. I know a Porsche owner that wishes he had the power of a Dodge Neon SRT-4. That is 230 HP stock, and with a couple of mods, 310 HP on 92 octane.

I think you need a new analogy. Next time try Lada (or Yugo for those in the U.S.)

you forgot...

[Run4yourlives]

...it's played in seemingly neverending, all encompassing, life sucking fashion by these people called "loosers".

Re:you forgot...

[Senzei]

"loosers".

You keep using that word. I do not think it means what you think it means.

Re:you forgot...

[nightrain_tg]

the opposite of which would be the "tighters", eh?

Re:No AMD macs?

[Lars T.]

There are reports that Intel can not produce the volume of chips that Apple need.

"My source said there was some stuff, "some very, very cool stuff," that Jobs couldn't unveil because of "supply issues."

"They can't get enough Core Duo (chips)," said my source.

And then there are reports that this is despite Apple getting top priority in shipments.

"CORPORATE BUYERS of notebooks from Dell and HP are fuming that people can buy Duo Core machines from Apple now, while they will have to wait for weeks for Yonah based notebooks."

Re:But does it run...

[Jim Hall]

Along a serious note: Does the new Intel Mac run Linux? That new MacBookPro is looking like a sweet Linux laptop. It is also very cute.

Re:But does it run...

[johnpaul191]

geeze, everyone knows System 7.x is the peak of the classic OS. the last system that will boot off a floppy drive because it did not have the "bloat" (features?) of 8 and 9.

Fat Binaries

[Arandir]

Question from a developer to all you Mac fans out there. Does the Intel Mac systems signal a return to fat binaries? Will developers need two Macs, one PPC and one x86, to develop Mac applications?

Re:Fat Binaries

[Carthag]

Jobs cleared that up when he made the announcement last year. XCode can make universal ("fat") binaries outta the box. You don't need two computers to do it, only if you want to run extensive testing.

Re:Fat Binaries

[ronanbear]

Fat binaries alright. Xcode 2.1 supports them and was released last June on the day that Apple announced the Intel switch.

Re:Fat Binaries

[99BottlesOfBeerInMyF]

Yes.

Re:Fat Binaries

[Guy Harris]

They can compile for all architectures from either a PPC or x86 Mac.

But they can test for all architectures only on an x86 Mac, as there's no flavor of Rosetta to run x86 binaries on PPC machines - and it's not possible to run all x86 code in Rosetta; no kernel loadable modules (kexts), and I think some userland plugin modules might not also work; I don't know whether you can force arbitrary fat binaries to run under Rosetta ("arbitrary" means "if I download the source to GNU Hello and compile it fat or for PPC only, can I run the resulting executable image under Rosetta?", not "can I tell Launch Services or whatever does this to launch an arbitrary .app under Rosetta?").

Re:Fat Binaries

[Guy Harris]

and it's not possible to run all x86 code in Rosetta

Ow. That should be "...all PowerPC code..." (and, yes, I previewed, I just didn't read it carefully enough).

Re:Fat Binaries

[Arandir]

So you're required to use XCode to get a fat binary?

Re:Fat Binaries

[Carthag]

I don't think it's required, I can imagine a scenario where you homecook the .app bundle, but it just seems like a boatload of work. Of course, if that boatload is lesser than the one required to make your app XCode compliant (or vice versa), it might be preferable.

Re:Fat Binaries

[Arandir]

The "boatload" of work isn't much. I'm using Qt as my cross platform toolkit. qmake creates a minimal app bundle, which I "homecook" by adding icons, full Plist, resources, etc. It's more complicated than a one-click build, but it ain't rocket science.

I'm assuming that Qt's qmake will eventually support universal binaries. I'm just wondering what to do in the meantime. Xcode is just a frontend, so I need to figure out what it's doing on the backend.

Re:Fat Binaries

[gnasher719]

'' Question from a developer to all you Mac fans out there. Does the Intel Mac systems signal a return to fat binaries? Will developers need two Macs, one PPC and one x86, to develop Mac applications? ''

You can build fat binaries on an Intel machine, then test the x86 version, then test the PowerPC version running under Rosetta. I would say that Rosetta isn't likely to hide bugs, but more likely to introduce bugs, so if your application runs fine under Rosetta, it should run fine on a PowerPC.

I am also told that nowadays gdb can debug PowerPC applications on an x86 Macintosh, just in case you have bugs that don't happen in the x86 version.

Re:Fat Binaries

[Guy Harris]

I'm not sure I understand the distinction you're making...

The distinction between making an arbitrary fat Mach-O executable run under Rosetta when you invoke it from the command line and between making a fat Mach-O executable inside an application bundle run under Rosetta when you, for example, double-click on the app.

perhaps the Forcing an Application to Run Translated section of Apple's Universal Binary Programming Guidelines has the answer to your question though.

Ultimately, yes.

"Make a Setting in the Info Window" didn't apply - there was no such setting in the Info window for the binary in question.

"Modify the Property List" didn't apply because it's an executable image, not an application bundle, and has no Info.plist file.

"Use Terminal" does, in a sense, but it achieves the goal by sending the executable to the fat farm - the ditto command in question forces it to run under Rosetta by making a PPC-only copy; the PPC-only copy runs under Rosetta on x86 because it has to. It handles the "compile it ... for PPC only" part of my question, but not the "compile it fat" portion.

"Use sysctlbyname" does; I whipped up a small program with that sysctl that should run an arbitrary fat executable under Rosetta on x86.

The easiest way to run a fat binary under Rosetta is to open the Get Info panel on it and check the "Open using Rosetta" checkbox.

You meant to say "the easiest way to run a fat application...", with "application" meaning "item delivered in the form of an application bundle". There ain't no such check box on a fat binary, just on the top-level ".app" directory under which the binary (and other stuff) is stored.

Re:Fat Binaries

[Dahan]

OK, I get the distinction now, but I don't get why it matters. IIRC, this thread is discussing the feasibility of testing universal apps without having a PPC Mac, so why the focus on getting the bare binary to run under Rosetta, rather than the .app bundle?

Re:Fat Binaries

[Guy Harris]

so why the focus on getting the bare binary to run under Rosetta, rather than the .app bundle?

Because some software for OS X might not have all its executables as apps run from .app bundle. It might include daemons, or it might include command-line apps.

(One of the apps I'm involved with has a command-line app and an X11-based GUI app, and has, in the past, had code contributed to it that broke when run on big-endian machines. Being able to test both the big-endian and little-endian versions on a MacIntel would be useful. That might be useful for some commercial apps, too.)

Re:Same Good looks

[Lars T.]

The real question is: how would it compare to a dual-core G5 iMac?

"D" key?

[jscotta44]

Using the "D" key to force a boot from an internal optical device on a Mac is new to me. I have been using the machines for many years and never heard of that, nor I can get my current Macs to boot from the optical by holding down the "D".

The "C" key, is a horse of another color. That is the traditional key and it works fine under every version of the Mac operating system that I have ever used since the advent of optical drives for computers up through now OS X.4.4.

Interesting that the author didn't mention the "C" key. And no, I did not read the story. Just looking at your quote. :-)

Re:"D" key?

[daveschroeder]

The "D" key is the new key for booting optical media on Intel-based Macs:

http://docs.info.apple.com/article.html?artnum=303 081

Re:"D" key?

[bedouin]

Why did they change it to D after all these years? D for DVD-ROM? Kind of confusing.

Re:"D" key?

[coldcup]

No, it is not.

Holding the D key is for booting Apple Hardware Test only. You cannot boot from a CD normally in this way.

C is still the correct way of booting from CD on intel.

Re:"D" key?

[commodoresloat]

"D" is the new "C".

Re:"D" key?

[NatasRevol]

Try the option key. It lets you choose what bootable device to boot from, regardless of it's type - optical, HD, firewire.

Re:"D" key?

[anothermortal]

I think specifically, the D key looks for a diagnostic partition. C looks for bootable media on standard partitions such as the OS X Install CD, or a FW attached HD with OS X installed on it. I'm still waiting for an Intel iMac to pass through my shop to verify.

Re:"D" key?

[daveschroeder]

Right you are! I just verified myself with Apple Enterprise Support:

D - diagnostic partitions only (Apple Hardware Test)

C - optical media

I stand corrected. Thank you.

Re:"D" key?

[drew]

The article did mention the C key. The GP post must have edited it out:

Holding down the D key (according to the Core Duo iMac User's Guide, Apple has changed it from the C key, perhaps because OS X now comes on DVDs instead of CDs. Who knows?) to try to force booting off of the optical drive failed as well.

Memory Anyone?

[puppetluva]

These benchmarks don't seem entirely objective.

The older imac was sporting twice the memory, and the g5 desktop had 9 times the memory.

Clearly the memory disparity was a factor in many of the tests.

I would give more credence to a test where all three machines had the same amount of memory so that paging/swapping/caching would be more at parity.

Re:Memory Anyone?

[2megs]

Clearly the memory disparity was a factor in many of the tests.

Why?

Seriously, what makes it so clear to you that this was a major factor? If all the tests run could fit in 512 MB without swapping, going to 1 GB wouldn't gain anything, right? Is there something about current Mac platforms that I don't know?

There were many differences between the machines. I'd be more inclined to point out that a significant minority of the benchmarks tested the graphics chips more than the CPU.

Re:Memory Anyone?

[snuf23]

OS X runs best in 1GB+ of RAM. All the Macs we have with 512MB or less are significantly slower than those with more. OS X is not a lightweight operating system.
The Quicktime test in particular was a bit dubious. Especially if you note how well the iMac performed in the MP3 compression test, where memory wouldn't have been a constraint. Then again it could be that Quicktime is not as optimized for Intel just yet.

Re:Memory Anyone?

[ottffssent]

Really?

You've run xbench and it consumes more than about 300MB of RAM? What about quicktime dealing with a 50MB movie? And iTunes - we all know that sucker benefits from having 4GB of RAM to play with.

The quicktime test is streaming - it will be bound by CPU and memory performance. Ditto CD ripping. Ditto Xbench, except for the graphics-bound tests there. I'm not familiar with photoshop, so that might have wanted a bit more memory to work on a 50MB file, but the actual tests run should all be CPU bound as well. After all, that's the whole point of the article - to see what difference the change in CPU / chipset makes.

Would I rather have seen the new imac with a gig of RAM? Sure. But I'd be surprised if it made more than a 5-10% difference in any one test, or more than 2-3% overall. The benchmarks being run just aren't memory-hungry enough.

Re:Memory Anyone?

[2megs]

Then again it could be that Quicktime is not as optimized for Intel just yet.

From an architectural standpoint, Altivec is miles ahead of SSE3. Given carefully tuned code for each processor and a task like video encoding that can make heavy use of their respective vector units, GHz for GHz the PowerPC should come out well ahead.

On the other had, is Quicktime even 100% x86 code yet? Including all the codecs?

There's so much more to benchmarking than bar charts, and normally Ars Technica is better than this.

Re:Memory Anyone?

[snuf23]

I agree - I think this article was a bit sloppy. So theoretically Altivec should be faster, but I would like to see an even comparison rather than the RAM difference when doing encoding. Some background on how Quicktime is optimized would be nice as well.
What about the MP3 encoding? Is that something that can be significantly sped up via Altivec? I just found it interesting that the MP3 encoding was much faster than the G5 iMac, almost matching the PowerMac.

Re:Memory Anyone?

[Kenshin]

Have you ever USED Photoshop?

I don't mean for making a little web graphic. I mean for editing something high-res.

RAM is a BIG factor.

Re:Why?

[Penguin Programmer]

Actually, to a true geek, booting any more than one OS is a sign of shame. A true geek just boots Linux (or BSD) and uses emulators to use the two or three Windows programs he needs.

Rosetta in same thread as app or not?

[yabos]

The guy seems to be a bit confused in what he writes.

"Rosetta runs in the same thread as the application, and translates blocks of code as they come up. "
Then
"...That allows the translation to run on one core while the application thread executes on the other core, meaning that the translated code will have a short distance to travel."

So, which is it? Does Rosetta run in a separate thread or not? Maybe he meant it runs in the same process, I don't know.

Comment removed

[account_deleted]

Comment removed based on user account deletion

we know photoshop is a little pokey

[johnpaul191]

Jobs ran photoshop in the keynote, and Adobe has not yet released a "universal" binary (built for intel and PPC chips) of anything so it is run through rosetta(PPC chip emulation). it works, but it is not zippy. i guess some of it was that it took a little time to start up (also loading rosetta), maybe how classic would load in OS X. i know the keynote was done on Intel chipped machines, but i am not sure what kind of machine? these user reports are done on iMacs, and as decent as they seem, they are still considered consumer level machines.

i do agree though, it would be interesting to see some side by side tests of what the iMac is made for, on software that is universal. like an iLife06 shootout of a new iMac versus what they were selling at Christmas. that would be pretty real world comparisons for the average iMac user.

Re:we know photoshop is a little pokey

[shentzu]

"i know the keynote was done on Intel chipped machines, but i am not sure what kind of machine?"

imacs.

Re:But does it run...

[filesiteguy]

...no but by SE FDHD does. :)

Darn thing, though, I can't see where to plug in the ethernet cable nor where the browser is located...

Multi-Monitor Spanning

[Betelgeuse on Ice]

One note I haven't seen anywhere else that was mentioned in the Ars review is that the new Intel iMac supports monitor spanning, and has a mini-DVI connector. AFAIK, the iMac G5 only supported mini-VGA and mirroring. (Well, there is that hack to enable spanning, but with only an analog video-out, it isn't that that useful for me...) It always seemed like a trivial crippling of iMac to force users who want/need desktop spanning to upgrade to a PowerMac.

Re:Why?

[Clover_Kicker]

How exactly does Apple put pressure on Microsoft?

Re:Why?

[Khyber]

To be fair, your SRT-4 essentially whips a Porsche boxter because a good deal of your car is plastic and fiberglass, whereas most Porches are solid metal, all over. You're pushing significantly less weight.

Throw that Porsche engine in your Dodge and do an actual comparison test. I'll bet that Porsche engine, thrown in your SRT-4, will tear your normal SRT-4 up.

Re:Console

[LurkerXXX]

That's funny, I play a lot of games, right after they are released, and I've *never* spent $300 or more on a video card, and I've never replace my video card after only one year. I hate to tell you, but millions of people play games all the time on middle or lower end video cards and have a good time doing it. I've spent less on gaming hardware in the past three years than any console owner I know.

Don't people WTFK??? Speed claims were qualified!

[Jewdass]

Who here actually watched the keynote? Show of hands? I know I did.

Let's all go to www.apple.com/quicktime/qtv/mwsf06/ and load the keynote up to 1:07:00.

Steve Jobs is completely up front about which testsproduced the numbers (SPECint_rate2000 and SPEC_fp2000) and outright says "Now, everything is not going to run 2-3x, the discs aren't 2-3x faster, etc." He makes it very clear that his numbers are based off of these two benchmarks. He claims they are the most important benchmarks of performance, which is debatable, but they are certainly a fair test of raw cpu power. Other than the chip and motherboard, the only other significant component that has changed is the GPU, going from a Radeon x600 to an x1600. Does anyone disagree that this is in the 2-3x faster range?

All in all, people are making a mountain out of a molehill rather than checking the source of the numbers. god bless the internet.

-justinb

That's a fallacy carried over from film

[OS24Ever]

I used to think this way myself in that with film you can't really see more than 30 fps.

However the difference in FPS measured in a game vs. measured in a film is that there is persistance of vision in that our brain fills in the gaps as a person moves an arm or jumps or moves or whatever.

With the video games the FPS means that the arm moves in it's range of motion in 24 steps at 24 fps, or in 60 steps at 60 fps. that's 150% more data being presented on the screen providing a much smoother range of motion for the object.

It's pretty easy to see by watching a film at half speed, or 12 fps. Watcha computer game at 12 fps. The film just looks like the guy is moving slow, the computer game the arm moves from by his side to straight up to your fragged.

Re:That's a fallacy carried over from film

[Heian-794]

These comments have been very instructive. Never thought about motion blur before!

I remember reading long ago that supposedly humans could only see 24 fps, and accepted that. But when watching film that had a minutes:seconds-frames clock at the bottom moving upward from 00:00-00 to 00:00-29 and then to 00:01-00, I could clearly see the numbers increasing, which meant that I was "seeing" 30 fps.

Applying this idea to a digital stopwatch running at 100 "frames" per second resulted in a blur on the last digit, so obviously my vision wasn't so superhuman as to be able to see 100 fps.

Someone with more ability than me should set up a website or small application where you enter a number N, and the computer displays numbers from 00 to N at N frames per second. If you're seeing discrete digits, you can see N fps. If you're seeing a blur, enter a smaller N.

I wonder what the average human score would be.

Re:That's a fallacy carried over from film

[donbenot]

Someone with more ability than me should set up a website or small application where you enter a number N, and the computer displays numbers from 00 to N at N frames per second. If you're seeing discrete digits, you can see N fps. If you're seeing a blur, enter a smaller N. I wonder what the average human score would be.

If done on a website, the average human score would be a lie, probably by at least 20%.

Re:That's a fallacy carried over from film

[ls -la]

The average human can see at 60 fps. What happens is that your eyes average what they see over 1/60 of a second, so if you start the 60 fps in time with your eyes, you could theoretically make out all 60 frames. What's likely to happen is that you see a blur at 60 fps because the show is out of time with your eyes. At 30fps, you have reliable vision of every frame. At any frame rate above 60, whether your monitor can show it or not (unlikely, and I haven't seen above 90) you will not be able to see each frame.

Thanks...

[jscotta44]

Again, that is news to me. Thank you for the update and the reference link. I appreciate it.

Re:Why?

[filesiteguy]

Watch out - you're gonna get slapped as a troll for that kind of remark.

As one who agrees with the OP - why indeed? I am seriously considereing this new Mac (aka Unix) platform for my home PC as an upgrade to my Win2K desktop. I can't imagine dual-booting to something as lame as XP or that future hog, Vista.

For those of use who need to test stuff under Wintendo - such as Java apps or PHP scripts, there's a lot to be said for Virtual PC or VM Ware. Even on my Linux laptop I only run Windows apps in a Crossover Office (Wine) session when I need to test compatibility.

Re:an oldie but a goodie

[daviddennis]

I hate this troll. Someone wrote it many years ago with the exact same words only substituting SGI and SGI applications, and I've seen it popping up in random places on Slashdot ever since.

He did not do what he claims.

D

Re:Don't people WTFK??? Speed claims were qualifie

[Jewdass]

correction: SPECfp_rate2000

Apple tried being honest, it didn't work

[podperson]

For about five years Apple insisted on, for example, referring to its monitor sizes by the size of the image rather than the picture tube, so Apple's 14" monitor corresponded to everyone else's 15" monitors. Its 25MHz 68040 was called a 25MHz 68040.

At some point they must have gotten tired of explaining that really the Sony trinitron they sold as a 16" monitor was really just as big as the trinitrons everone else was selling as a 17" monitor.

Re:Why? (my 2nd try, it seems first didn't work)

[John Jamieson]

Insightful???

If the owner meant to imply that a Neon engine was a huge downgrade, he needs a new analogy.

I believe my dealer had a 2004 Boxster until recently. It had an optimistically rated 228 HP, VS a Neon SRT's 230 HP. Now the 2005 Boxter has 240 HP and a slightly modified Neon engine delivers 310 HP on 92 Octane. The neon engine is actually not bad.

If the author really wants to use Automotive analogies, try puting the engine from a stock VW Beatle in the Porsche. (Oh ya, Porsche almost did that already, I forgot. lol)

amount of ram in benchmark

[DietFluffy]

What is the significance of arstechnica benchmarking the 3 macs with the following ram configurations:

iMac Core Duo: 512MB
iMac G5: 1GB
PowerMac G5: 4.5GB

Wouldn't such a large difference in the ammount of ram have a significant impact on benchmarks?

Re:amount of ram in benchmark

[ottffssent]

That would depend on the working set size for the given app.

512MB may be slightly cramping the style of the new imac, but it didn't look like any of Ars' benchmarks would need much more than that. Certainly 4.5GB isn't going to make any difference, and if you've been following Ars' articles, you'd know why that particular machine is so loaded. The CPU-bound, disk-bound, or graphics-bound benchmarks aren't going to notice the change in RAM amount. The photoshop test, being done on a fairly large image, might have seen some impact from the difference in available memory.

Given how heterogenous the systems are already, I'm not too concerned with a slight difference in memory size. Given the different instruction sets, execution hardware, cache layout, and memory controller, I think having only 512MB rather than a gig is unlikely to show up in the benchmarks or in most users' usage.

Re:amount of ram in benchmark

[Steve Cowan]

I have to disagree. OS X 10.4, as wonderful an OS as it is, uses LOTS of memory. Dashboard is a notorious hog, as is the core OS.

My G4/1.4 GHz is definitely much snappier at everything with a gig than 512.

I think these benchmarks are a little off, because nobody in their right mind would leave this machine at 512 MB in this day and age.

Re:amount of ram in benchmark

[tribentwrks]

And isn't just basic scientific practice to control all the variables that you can, so that your benchmarks can be considered valid? I mean, how hard could it have been to add a little ram, or take some out, just for this test?

Re:Why?

[n6mod]

You just described my daily driver.

Oh, we were talking about Macs, weren't we.

Same story here. My next machine is a MacBook if it can boot Windows... or if VMWare ports to MacTel...either way.

Re:Molten Core Combat?

[TobascoKid]

A shame really, when I first read "Molten Core Combat", I thought it was some form of benchmarking that pits various machines against each other, overclocking them to hell and seeing which one lasts longest (hence, "molten core" combat).

Apple marketing

[Savage-Rabbit]

...How many people know the new macs actually ARE pc's? Of course they aren't, cuz there are macs and pcs right? ... Fortunately, porting OS X to standard PC will maybe awaken some puzzling among the general audience.

Abstracting the Mac from the concept of a PC to the point where the consumers don't realize that the only really significant difference between the MacPC and the WinPC is that the former has (or used to have) a PPC processor and the latter has a crappy OS has to be one of the greater triumphs of Apple's marketing department. I suppose that even though one is sometimes tempted to think so not all marketing people are morons.

Re:No Windows?

[vux984]

Me.

I would gladly pay Apples prices to dual boot MacOS X and Windows.

I need to use both OSes. 1 MacBook no matter how expensive it is, will be cheaper, lighter, and more convenient, than 1 MacBook *and* a PC laptop.

Mod parent up (Informative)!

[argent]

Whether this guy or the grandparent poster is correct, this should be modded "Informative".

Why SHOULD the public be aware of what is inside?

[flaming-opus]

Why do we want the public to be aware of what type of CPU is inside the box? Doesn't that go against the last 15 years of developments in the personal computer evolution? Isn't it good that the consumer doesn't need to be bothered with this sort of thing? I think it's a possitive thing that buying a PC, except for the enthusiast or yuppie-gamer, is more and more like buying a stereo, or a microwave.

Apple has done a good job of telling the consumer: You aren't buying a central processing unit linked to a primary memory unit, linked to a magnetic and optical secondary storage controller, and a 3-dimensional rasterizing coprocessor. Rather, you are buying a mac.
You are not a systems administrator buying a piece of industrial equipment, you are a consumer buying a gadget.

I don't think it is our role, as computer people, to make the rest of the world more computer savvy, as we shall fail in this. Rather it is our job to make computers more usable, such that the consumer need not be more savvy. Apple has done a good job of creating usable software, and it's only natural that their marketing should mirror this philosophy. Should we question outrageous claims of what a mac can do, that other PCs can't, or unrealistic claims of how a mac is better than a PC? Sure we should, as we should be skeptic consumers of all marketing. But I think it's good that Apple sells computers the same way that Levi's sells Jeans.

Re:No AMD macs?

[99BottlesOfBeerInMyF]

Apple chose the same processor that Dell so heavily rely on. Of all the reasons, I just don't believe AMD can't manufacture enough chips. I think Steve Jobs always wanted to use Dell as the model to follow whether his mouth admits it or not.

Are you trolling or just slow in the head? Apple went with Intel for laptops. They needed a fast portable. AMD has nothing useful for laptops right now. their top chip uses 15-60% more power and is slightly slower than the Intel Duo. It uses more power idle than the Intel does at 100%. Choosing between going from a 6 hour battery with the g4 to a 3 hour battery with the Intel or a 2 hour battery with the AMD. Gee, tough choice. Apple may very well ship AMD chips some day, but not in portables or all-in-ones until they get their power consumption under control (AMD 65nm is due Q4). As for business models, Dell is about cheap, cheap, cheap with little inventory and interchangable supply. Apple is about grabbing the high end with innovative tech as a differentiator. The business models are very different.

Re:No AMD macs? Excuse me!!

[flaming-opus]

There are no AMD macs for a reason, a good reason.
Apple had a fine-and-dandy desktop processor. The G5 works very well in their workstation-class machines. The real problem was the laptops. Half+ of the macs sold are laptops.

AMD competes very well in the desktop processor arena, and pretty well in the desktop-replacement notebook segment. However, they don't have a great answer in the thin-n'-light laptop segment. This is an area dominated by intel. Choosing to make AMD-based macs would not have improved Apple's position relative to the power architecture. They would still have good desktop systems, and still would struggle with laptops, and minis.

While I'm sure the marketing dollars are nice, I think that's not the primary reason for Apple to choose Intel over AMD.

Re:Why?

[kadathseeker]

You want to run Windows and play your precious games and whatnot? Buy a cheap-o Dell.

What kind of games do you play? If I'm $2000 in the hole buying a Mac, I sure as hell want to not have to spend another $1000-2000 building a machine capable of playing BF2, F.E.A.R., Stalker: Shadow of Chernobyl (when it comes out), and UT2k7 in a year. Macs are really nice, but a nice PC dual booting Windows and Linux will have to suffice for me, for now. Unless you want to loan me $2000...

Re:Why?

[drinkypoo]

Well, the Boxster S has about 50hp more than the SRT4, but it's also got a flat motor that wouldn't fit in the neon :)

By all accounts the Boxster (even the S model) is something of a dog. Sure, it's more solid, but really it's just a toy car. The SRT4 Neon is a different kind of toy - yes, it is a complete shitbox! However, it is a fast shitbox. On the down side, the new Neon got the worst side impact crash rating for an American car made in the last twenty years, so I would stay the fuck away from them myself.

MOD PARENT FUNNY

[cttforsale]

Gold baby! Gold!

Isn't that the way it is now?

[IANAAC]

Why should they be forced to buy two computers just so that they can preserve their "entire Mac experience"?

So what? That's how it's done currently. Most people I know buy a Mac for fairly specific purposes, games not being one or them. Honestly, if you want to run Windows software, buy a machine with a "Designed for Microsoft Windows XP" sticker. You'll be much happier.

Re:But does it run...

[A. Bosch]

Microsoft Flight Simulator? The WINTEL version of Microsoft Flight Simulator? I miss being able to put those Boeing 747 wings on a Piper Cub.

Re:Why?

[gstoddart]

But what about if someone wants to run games (and perhaps other software that runs only on Windows) *and* still get the superior OS/apps found in Macs? Why should they be forced to buy two computers just so that they can preserve their "entire Mac experience"?

IT's not Apple's job to help you run Windows software. Least of all on their hardware.

As the OP said, if I bought a new Mac, the last thing I'd want to do is try to figure out how to run software for Windows on it. Period.

Nobody is forcing you to buy a second machine to do anything. You can do without that software, buy a second machine, or (possibly) void your warranty by trying to get Windows to run on it. That doesn't mean you should expect Apple to roll over and give you a machine which it is easy to make run both OS's. They want to give you a good user experience if you bought their stuff.

If I buy a Honda Accord, is it reasonable to expect Honda to ensure that the turbo-kit I got for my Ford Escort runs on that Honda? Of course not. What does Honda care? And it's not about "the full Honda experience", I'll tell you that.

Apple would probably prefer you leave them out of the equation when it comes to running your Windows games. Specifically so they don't get calls from people who have either bodged their systems together from spare parts, or generally done stupid things with them.

You have complete freedom to buy, or not buy an Apple computer, and all that implies. Whining about being 'forced' to own a second computer to be able to have another platform is a completely specious argument in my opinion -- how is this any different than from when the computers were on completely different platforms?

Re:Shut up! (Not the only one)

[mkiwi]

I criticized CmdrTaco the other day and I got blasted for it.... so much for that +5 informative I got earlier in the month about a very similar subject, which happens to be my research field at my college- something few people on slashdot even understand.

--
I'm suprised Steve lets the mods talk with their mouth full.

I think I am on sig revision 5 now, still trying to get something that is intellectually provocative yet appealing to slashdot moderators. I never post as an AC, maybe that is my problem.

Security or Something Like It?

[Kartik3]

Am I the only person who is wants to know about how the new Intel iMacs and MacBookpros implement Trusted (Treacherous) Computing and other "security" measures? I really would like to know if I have the option to turn trusted computing features off in the BIOS (or EFI) and I was hoping Ars would be able to tell me. I'm really pretty surprised Ars didn't mention much about trusted computing when it seems as something of a hot-topic with the new Intel-based Macs.

Re:Security or Something Like It?

[Kombinat]

As I wrote in another Apple thread before, there is a Infenion TPM chip onboard and its activated. And it seems it can't be switched off...

Re:No AMD macs? Excuse me!!

[Overly Critical Guy]

Apple's decision to go with Intel was likely based on their future processor roadmap, including Merom and Conroe.

Re:Watch boot video & Incompetent Benchmarking

[DECS]

MacWorld demo people, although trained not to demonstrate or allow rebooting the new Macs on display, did remark that they booted up really fast.

In the video, the G5 likely had more RAM installed, which would make it POST considerably slower. The boot time, however, is probably very representative of how much faster the Intel iMac is at booting. Other reasons the Core Duo may have booted so fast compared to the G5:

- Two processor cores!

- Mac OS X is expressly designed to boot fast by bringing up as much as possible in parallel. That's part of the point of launchd: to identify dependancies and kickstart multiple things at once. This is also why Apple gave up on displaying what was being booted in 10.4, and now just shows a progress bar (which is unrelated to what's actually happening, and only timed to match the previous boot time as a relative indicator). Reporting what servers are being launched would take longer than actually starting them. This parallelism would clearly benefit from multiple processor cores in the Core Duo.

- the G5 may have been booting for the first time, or they may have deleted the cache in an attempt to make the test "fair," not realizing that the cache has a huge impact on boot times. Among other things, Mac OS X caches the kernel extensions so that the next boot only stops to numerate which kexts to load if something in hardware has changed. If you wipe your cache files (/Library/Cache, ~/Library/Cache and System/Library/Cache), the next boot will take a lot longer while boot performance caching is rebuilt.

- other hardware may have been unfairly compared: how fast was the G5's drive? was something wrong with it? was the G5's drive full, and struggling to find space for cache files? was it bound to a directory server, and stalling on boot while looking for the server? was it full of 3rd party software, kexts, startup items, etc?

The video doesn't reveal anything about the demonstrators competence at setting up fair comparison, or their motivation, so we don't know.

Recall the comparison of database servers running on OS X server vs Linux, where they intended to be fair but their assumptions about how to do so were actually really bad?

Or look at the Ars review and benchmarks of the new iMac Core Duo vs the iMac G5. He does an array of benchmarks where the G5 has 1 GB of RAM, and the Intel iMac has 512MB! Sorry Ars, but that's just plain incompetent. Your benchmarks are WORTHLESS to even skim over. How about benchmarking the G5 iMac with 512 and 1 GB installed, and reporting if that makes any difference?!

Reliance on XBench

[Overly Critical Guy]

The reviewer and submitter relied on XBench results, but XBench's page says it is not optimized yet for the Intel iMacs and to take the results with a grain of salt. So now there's all this bitching about Steve Jobs "lying" about performance without recognizing that Intel XBench support is experimental according to its own website.

Re:Missing 1 piece of information

[level_headed_midwest]

All of Intel's Core chips are 32-bit. The *next* generation (a.k.a. Merom) will be 64-bits and comes out in September. I hope this helps.

Re:Why?

[TooMuchEspressoGuy]

"the pain that Windows will cause is not justified by the need to play Slaughterfest 3 or Unreal Shooter."

Jeez, you make it sound like running Windows is like getting a root canal.

It may not be the perfect OS, but running it certainly isn't painful (most of the time.) The small-moderate amount of hassle that I need to go through to maintain a Windows box is more than made up for by the range of good PC games I can run on it.

RE: PowerMac naming

[Anm]

If MacBook is the laptop, and iMac is the consumer desktop, then I'm betting on something like MacStation (a la workstation) to replace the PowerMacs.

Anm

Apple do use AMD chips!

[CdBee]

not in a mac though. the Airport Extreme contains an AMD CPU to run its embedded firmware

They didn't use fresh installs!

[vmardian]

From the article...

Lastly, the Core Duo outperformed both the iMac G5 and the Power Macintosh G5 on the disk test. This could be due to fragmentation, as the Core Duo was the only system with a "fresh" OS install.

So, uhhh, why didn't they do a "fresh" install on the older Macs? All the other benchmarks are probably affected too. What a useless review.

Re:32-bits? - Uh, How about this reason?

[Nom du Keyboard]

These machines don't have a need for more than 4GB RAM, they only fit 2GB, so why go 64-bit when it's not going to be used to address more RAM?

Uh, how about so that Apple doesn't waste their limited resources supporting both 32-bit and 64-bit Intel OSX?

And then there's, I'd like to run Windows on my system sometime maybe, and that may be all 64-bit soon.

Followed by the worry of, just how long will Apple keep 32-bit Intel OSX as an upgradeable option when all their new machines are 64-bit?

Putting it altogether, Is the 32-bit Intel Macintosh going to quickly become an orphan that was only produced for 6 months before 64-bit processors take over everywhere, and shoved out solely so that Steve Jobs could make good on his promise 7 months ago that we're switching to Intel as quickly as possible?

I sure won't be buying a 32-bit Apple anything. And I guess this means that the Universal Binary will be getting even fatter (fatter -- get it?) when it has to include 32-bit G4, 64-bit G5, 32-bit Intel, and 64-bit Intel. Apple may be the next adopter of BluRay just for software distributions.

Why buy windows?

[dancpsu]

When you can run wine?

Re:Why buy windows?

[Mr. Underbridge]

Because lots of programs don't run correctly in Wine?

Re:Molten Core Combat?

[codeonezero]

Noobs..don't you mean test out the FPS in Blackwing Lair?

Definitely something fishy...

[RedBear]

Same here. I have experience with a 350MHz G3 iMac that boots Panther faster than that G5 iMac. That video is totally bogus. The day that a G5 iMac is slower than a G3 iMac, something is wrong.

Re:Definitely something fishy...

[Halo1]

Well, there are some weird things with Mac OS X boot times (possible linked to open firmware). My iBook G4 1.2GHz (with 768MB ram) also boots Mac OS X quite a bit more quickly than my dual G5/1.8GHz. I can't imagine that 256MB less makes that much of a difference to test...

Re:Why?

[fpillet]

I'm a Mac guy, primarily, but I develop on both Mac and Windows. If I can use one machine for both (preferrably a laptop) the better for me. I can carry my laptop with me in client meetings and still only need one machine for both kinds of projects.

Running both Windows and OS X on the same machine is more useful than most people think. Especially when you're a techie. In my case, I do both Palm OS (on Mac) and Pocket PC (on Windows) development, plus desktop software, some of which is fully cross-platform. I can't imagine a better world than one where my Mac runs both OSes at the same time, and both at full processor speed!

Experimental Error?

[pyite]

I get so annoyed everytime someone supposedly benchmarks something on a PC and includes no experimental error figures, no mean, no standard deviation. Maybe that's because when you only test things once, the sample standard deviation is infinite! Doing this in an engineering or scientific paper would get you laughed out of the journal or conference. Reading the following in the Ars discussion forum just reinforced my thoughts:

XBench is not great for benchmarking unless you repeat it's tests about 10 times or more each... its results vary too much (even from one run to another on the same machine, never mind when comparing two different ones).

Come on people, do many tests, compute the data, adjust with Student's t-distribution. This is elementary stuff yet no one does it.

Re:Experimental Error?

[Superfarstucker]

Well to be fair most benchmarking isn't exactly a rigorous procedure, so 'error' computations are largely meaningless. It is meant to be more of a relative than absolute metric.

Re:Why? (totally off topic now)

[cypherz]

Shitbox? Hmmm, my feelings are hurt! I guess I have to go out and humiliate a few Porsches and Mustangs on my way home!
Horsepower isn't that much different between Boxter S and SRT-4. Most SRT-4's dyno out to about 227 - 235 *at the wheels*. They're rated by Dodge as 230 hp. So the engine HP (measured at the engine) will be close to the Boxter S numbers, which IIRC is 258. FYI: With Mopar's stage 3 kit, an SRT-4 makes about 365 hp on pump gas and just shy of 400 on 100 octane! Yikes! Now THATS a fast shitbox! I submit that the SRT-4 is about the most fun you can have (behind the wheel) for 22 kilodollars. :-p

Re:Why?

[MindStalker]

I know I know.. :( I realized this fact halfway into writing this. And decided a coverup of this fact was best. Currently the fact is locked in a small closet.

Re:No AMD macs? Excuse me!!

[Nom du Keyboard]

At something on the order of 80 million chips per year (counting dual-proc, dual-core machines as two, obviously) Apple would instantly become AMD's largest customer

I don't know where you get your figure of 80M chips per year, but sites like this one list Macintosh sales for 2004 at approximately 3.5M units. Even if all were dual processors (they're not), you are overstating processors by more than an order of magnitude.

Even at current sales Apple might have become AMD's biggest customer, but not overwhelm them at all like you suggest.

That lesson might be for laptops especially

[ianscot]

Apple's first PPC laptap, the 5300, was a disaster. They'd had a huge lapse with the wait from the old 040 chip laptops -- "Duos" being the last of that breed and the first docking laptops around IIRC -- and the new PPC machines, and then the 5300 was a lemon.

As a test of the difference between Steve Jobs and Gil Amelio, the MacBook is interesting. We've already had the slight delay -- you know Jobs wanted to announce the MacBook was orderable during his keynote too. The name is lame, in a way that's decidedly not Apple's style.

I don't know. Seems like one to wait on, but I'm tempted. So tempted. RDF... taking... control... must... fight... the power...

Re:Why?

[alanQuatermain]

Why should they be forced to buy two computers just so that they can preserve their "entire Mac experience"?

I wouldn't say forced exactly, but having the games on a separate computer makes it really easy to browse GameFAQs' and Gamespot's walkthroughs without having to exit the game altogether.

.
.
.
.
.

...I'll get my coat...

-Q

Re:Why?

[cypherz]

Power to weight Baby! Under 3000 lbs. 230 HP (at the wheels!). But the Porsche engine isn't that much more powerful (if any at all). See my post below.

Xbench Problems

[Richard5mith]

The Xbench scores aren't a good indicator. From MacRumors... "While Xbench is a Universal Binary, it is not entirely multiprocessor aware. As a result, it does not generally test both cores of the Dual Core Intel processor."

That's why it scored so low in the CPU tests.

Re:Why?

[UnknowingFool]

But what about if someone wants to run games (and perhaps other software that runs only on Windows) *and* still get the superior OS/apps found in Macs? Why should they be forced to buy two computers just so that they can preserve their "entire Mac experience"?

Then they should buy two computers. Never did Apple say that you could run Windows on this machine. In fact they have said the opposite. That there exists the slightest possibility is not the fault or credit of Apple. They have only designed the new machines to use an Intel processor. You might be able to get Windows to run on it, but you are own your own.

To offer the analogy, it's like buying a Ford and trying to install a Chevy engine only to realize that the engine might not fit exactly. Then blaming Ford because the Chevy engine won't fit.

Re:Why? (totally off topic now)

[drinkypoo]

I'd rather buy a used rice burner (rather than a bad imitation of a rice burner) and trick it out. I need not go into what I can get out of a cheapo nisan for $22,000... Granted the Neon is fast (like I said) but it's still a shitbox, and front wheel drive sucks :)

You obviously know nothing of SpecRate

[Dan Berlin]

Guess what happens if you run SPEC rate on a two processor machine?

You get twice the SPEC rate of a single processor machine!

So if it's only twice as fast as a *single processor* iMac G5, then it's ...
NOT FASTER.

SPEC rate is a useless benchmark when comparing a different number of CPU's.

The cinebench stuff or whatever is more valid.

Re:You obviously know nothing of SpecRate

[Palshife]

So...it's not fair since the processor has two cores? That's ridiculous. What was stopping IBM from designing the G5 with multiple cores? If Intel can fit two cores on one die and still have a near 2x speedup due to parallelization, then they win. We're comparing processor to processor, not core to core.

The comparison was from a single G5 to a single Intel Core Duo. They're both single processors with different guts. While a single core of the Intel processor is slower than the G5, the Core Duo is a faster processor overall.

Re:32-bits? - Uh, How about this reason?

[argent]

I don't think Apple will ever ship a pure 64-bit G5 OS. It would be slower than the 32-bit version, and since it's a lame duck now there's not really any point.

Favorable review

[gh]

The summary would have you believe that Ars Technica was dismissing the Intel iMac when actually the review was overall favorable toward the new machines. But, of course, I only know this because I actually read the review.

It's 32 bit. Wait for the 64- bit machines.

[Animats]

It's really a bit late to be introducing a new product line based on IA-32. Especially coming from a chipset that was 64 bit. It was a real surprise to see Apple taking this route. Clearly, they're aiming at the "cool laptop for web browsing" crowd, not the "graphic designer using Photoshop" crowd, which used to be Apple's target market.

Heavy photoshop users will say "You expect me to go back to a system that has to swap?"

(Vaguely relevant thought: "Web 2.0" runs down the battery faster.)

Re:It's 32 bit. Wait for the 64- bit machines.

[captaineo]

From a developer's standpoint I agree it is a strange decision not to jump immediately to EMT64. Apple had a fine opportunity to eliminate all the compatibility headaches of supporting both 32-bit and 64-bit code in one system, and they chose not to take it. They are going to end up with the same DLL compatibility problems as Microsoft has with 64-bit Windows.

Perhaps Apple figured that most developers will need to support both PowerPC and Intel for some time to come, so it's no big deal to add a third platform? And they trust their "fat binary" system to take care of all the compatibility issues? There have got to be a lot of corner cases like shared-memory IPC and driver ioctls that can't be handled automatically... (e.g. Linux's emulation of the 32-bit system call interface on EMT64 is generally great, but there are plenty of device drivers that don't have properly implemented 32-bit interfaces).

Regarding performance - my experience has been that Intel chips do not perform much faster on 64-bit code as opposed to 32-bit. However, AMD chips gain a lot from running in 64-bit mode. I presume Intel just hasn't had enough time to optimize their EMT64 implementation as well as AMD has.

One bit of relief: the release of Intel Macs finally explains why there was never any effort towards a 64-bit user-mode for OSX/PowerPC :).

Re:Why?

[commanderfoxtrot]

Porsches quite heavy cars- but there's a certain thrill from driving a car weighing virtually nothing and lots of power through the rear wheels. Nothing wrong with glass-fibre.

I bought a TVR Chimaera a little while ago- 350bhp and it weighs something like 1060kg. 0-60 in 4.1secs - lots of fun!

Four years ago the lampshade model was out...

[ianscot]

but still true today.

The "desk lamp" iMac design hasn't been around for a long while now. Yeah, boy, that flat screen sure is novel; nobody sees the advantages of that...

The G5 model has no tray to catch drool in, even. Slot loading drive, on the side.

Re:Why?

[un1xl0ser]

So somehow adding safety gear is a better analogy for installing Windows? WOW. Maybe an analogy is not what we need here.The truth is that you are talking about the same hardware, just running different software. It is so simple, any analogy just clutters the simplicity of the concept with no benefit.

It's more like you have a piece of hardware that can run two different programs and you want to switch between them.

PowerMac Replacement?

[fupeg]

This comparison really makes me wonder about the Intel-based PowerMac replacement. What kind of processor are they going to put into that? The logical choice would seem to be the Conroe. There are rumors of a 3.3 GHz dual-cores being sent out later this month. Intel claims that Conroe will outperform Core Duo 2-1 on a performance-per-watt basis. So a 3.3 GHz Conroe might be as much as 3.6 times as fast (pure performance, it's ok for a desktop chip to consume twice as much juice as a laptop, right?) as a Core Duo. So if you take the iMac comparisons against the current G5s and extrapolate... well a PowerMac based on a Conroe could be a mighty beast. Give it 4 GB of RAM like the PowerMac in the comparison, and it should easily outperform the PowerMac (at least on non-Altivec tasks, but that's a different story.) Of course it's still going to have the stupid front side bus, albeit running at 1.0+ GHz with 4 MB of L2 cache.

Another possibility would be for Apple to wait for the Extreme Edition of the Conroe, the Kentsfield. That would give them four cores, like the current PowerMacs. It won't be out until 2007, and Apple seems anxious to switch everything over ASAP. So they could go with Woodcrest, basically Conroe for servers. This might let them put together a dual-cpu/dual-core setup like they have with the current PowerMacs. This kind of setup was demonstrated by Intel last fall. There were also rumors last year of Apple pressuring Intel to give them Woodcrest chips ahead of schedule.

And of course there's the more mundane question of what will they call the PowerMac replacement? They seem to want to get away from the Power prefix, while stressing the Pro tie-in to their Pro apps. So maybe Mac Pro? Seems too short. Maybe bring out the whole name, Macintosh Pro. Whatever it is, can it make people as upset as "MacBook" did?

Re:PowerMac Replacement?

[ZombieRoboNinja]

I'm guessing they changed the "PowerBook" designation for two reasons: first, because "Power" is somewhat associated with IBM (PowerPCs, etc), and second, because "PowerBook" doesn't make it clear to the non-geek that it's a Mac. Even on Slashdot, I've had people tell me that I should reboot my PowerBook more often if I don't want Windows XP to get laggy... that name confusion won't be a problem with the MacBook.

If they want to change the name of the line, now's the best time available. I can understand why they didn't change the iMac brand, though - the less geeky Mac users might be distressed if they could no longer buy an "iMac," regardless of what's under the hood. I imagine that the Intel iBook won't get a name change for the same reason.

Now, "PowerMac" does still have that IBM connotation, but it's also got the "Mac" right there in the name. I'd guess that they WILL change the designation, because (a) "power"-users are probably more likely to embrace change as a positive thing, and (b) screw IBM with their Power-whatsits.

SO, a new name. Maybe MacDesk Pro? Mac Pro, like you said? MacMac? Lucy 2? Heck, I dunno.

iMacs performance is irrelevant

[frostilicus2]

I doubt that Apple's move to Intel had a great deal to do with performance, and I dislike this fact being used as a key selling point. If you refer to the "definitive" G5 vs. everyone else benchmarks at http://www.anandtech.com/mac/showdoc.aspx?i=2436 it is apparent that the G5 is largely comparable to offerings from AMD and Intel (admitedly the new Intel Core Duo is not benchmarked) and although the G5 is, in many cases, not the fastest chip, it is similar. The increases of 2-3x in performance between the G5 and MacIntel iMac are a consequence of having a dual core chip (and being a generation ahead of the G5) besides, Apple could have feasibly used the dual-core G5 chips that they've had at their disposal for a while now. Any Mac zealot will argue that their PowerPC Mac is "just" as fast as an intel based system, but performance is NOT the issue. This is why the iMac was updated first, it is a consumer product, supporting Apple's fledgling attempts to enter the living room (consider front row) - it desperately needs Intel's brand name associated with its hardware.

The significance of this new product is long term and cannot be underestimated.

Apple finanlly has penetrated the consumer electronics market with the iPod, and their brand recognition and image could not be better. Apple has shoehorned its way into the psyche of the common man. It now has to bring its key product, the mac, to the masses. Consumers will be attracted from a design perspective and because it shares the same logo as their iPod, the OS is a little different to windows, but now at least you have the reasurrance of dual booting into windows (I'd like proof of this concept, but I'm sure it will come) and the processor gives the security of a well recognised brand name (consider brand strengh of Intel vs. AMD).

In the future, I doubt that IBM's die shrunk Power chips will share the low power consumption that I expect Intel will bring, and many concepts for great products will never be realised. I'll be interested to see if the new Intel chips can match up to the PowerPC altivec-ised vDSP FFT's, but in a way I don't care. It is an exciting time to be a Mac user, as more people join the fantastic experience that we have had for so long, and new software and hardware comes our way. Either way, they're finally here and it will be interesting to see what the future holds.

Re:Why?

[xouumalperxe]

I've been using apple mail for the last 2 months, now. It is quite good, in an outlook express sort of category (NOT in an outlook express sort of WAY). It's just a simple, functional mail program. iCal and Address Book make the whole PIM suite come together quite nicely, and the way it integrates into spotlight is just beautiful. All in all, I don't regret leaving Outlook (the big one) and Evolution behind for Maill.App and iCal.

Yep.

[jscotta44]

I knew that. Just didn't know that they had changed the "C" to a "D". It makes sense, just didn't know they had done that.

Dual processor not dual core

[lukket]

Oh well. Just found an error in the original post. I've got to the point in TFA where he's listing the systems, and it's not a "Dual core" G5 he compares it to but a dual processor one.

Steve's statements

[Stan Vassilev]

"I'm sure it will surprise many slashdotters to find out that Jobs' statements were a bit too optimistic"

Think about it, how can anyone be surprised when the same happens every year at MacWorld.

Though can you blaim him? He sells to artists and casual users. Most of them will dispute "2x to 4x" as long as it's white shiny box with the Apple logo on it.

At least half the value a Mac has is design and perceived value (former of course for a reason, while the latter is rarely based on reality).

Re:Steve's statements

[99BottlesOfBeerInMyF]

He sells to artists and casual users.

Lets see, who do I know that uses a mac? Well, I know a couple of graphic artists and a few professional writers. Yup they are artists. And then there are about fifty computer programmers working on various software. There are a few computer security experts and a couple of artificial intelligence researchers. There are a few chemists and biologists. I know a few geologists. There are a lot of teachers and a few lawyers. Hmm, from my personal survey, most mac users are computer geeks, not artists or casual users.

At least half the value a Mac has is design and perceived value (former of course for a reason, while the latter is rarely based on reality).

Or perhaps those that can perceive the value of the OS and accompanying software as it impacts them and their daily work are just more perceptive than you. Or maybe all those people I mentioned are just stupid whereas you are some sort of intellectual giant. Get a clue.

Re:Steve's statements

[Stan Vassilev]

"Get a clue."

Sure, I'd like to get a clue on how the accompanying Mac software which represent mediocre family album organiser, DVD video burner and a calendar (basically) impacts the daily work of artificial intelligence researchers, security experts and so on.

Re:Steve's statements

[99BottlesOfBeerInMyF]

Sure, I'd like to get a clue on how the accompanying Mac software which represent mediocre family album organiser, DVD video burner and a calendar (basically) impacts the daily work of artificial intelligence researchers, security experts and so on.

The main thing is, of course, the OS and its capabilities. Instantaneous searching of files, including the contents of many common binary formats and a plug-in architecture that lets you add any missing ones. It helps to be able to instantly search your PDFs, proprietary format prototyping files, .doc, open office, and e-mails all at once.

And then there is the ability to run all the staple Linux/UNIX software and most major commercial software and all the mac specific software. The ability to use functionality across applications, like adding a common text script and having it available for all text in the vast majority of applications. And then there is the old Volkswagen mantra. They wanted to build a car that you don't have to think about. Something that was a tool that never got in your way. That is what I feel Apple has managed. I don't have to install an application, I just drag it where I want it. Need it on another machine; just drag it onto the mounted drive. Need to configure some obscure setting for the computer itself, just open the system preferences and the whole thing is searchable.

Multitasking. It actually works and you can do multiple CPU intensive tasks at once. I can compile something and browse the Web. Amazing! Some day Windows will catch up.

Finally, your characterization of the software that comes with macs is misguided. You forgot a complete dev environment, e-mail, browser, address book, disk utilities, encryption tools, GUI scripting linking software, etc., etc. As for the calendar, it imports my bugzilla items as a to-do list.

Basically it is no one thing. It is just that everything works better and easier and I don't end up waiting for that crashed application to restart or the mouse to start responding again or spend 15 minutes searching for where they hid the preference I need in this version. Ease of Use is not just for newbies. Maybe that is why more than half of my office now uses powerbooks. Been to a Blackhat or Defcon conference lately? There sure are a lot of macs there for some reason. The same goes for a whole slew of scientific fields. Decent hardware and a great OS make it hard to resist for people who just want to get shit done.

Re:Steve's statements

[Stan Vassilev]

"Multitasking. It actually works and you can do multiple CPU intensive tasks at once. I can compile something and browse the Web. Amazing! Some day Windows will catch up."

The day Windows caught up was the day NT was released, i.e. dunnno, like 20 years ago? And even if it wasn't the case back then, now NT based OS is in common usage (2000, XP and 2003).

Or can you actually name one specific instance where Windows fails and OSX excels at multitasking? I never stop my mp3 player and sometimes even watch DVD's on my second screen, while coding Flash / doing 3D / audio editing and the most I ever needed if I really run CPU intensive tasks was bump the priority of the video player up a notch.

I also run MySQL/PHP/Apache for development purposes and never felt performance degradation because of it.

What is this magical multitasking enhancement OSx has that we don't?

Re:Steve's statements

[Stan Vassilev]

"You forgot a complete dev environment, e-mail, browser, address book, disk utilities, encryption tools, GUI scripting linking software, etc., etc."

Are we talking a Ubuntu here. Because the description is pretty accurate. Except well that Ubuntu is free, and runs on any PC.

Re:Steve's statements

[99BottlesOfBeerInMyF]

Or can you actually name one specific instance where Windows fails and OSX excels at multitasking?

Certainly i can. Using both Win2K and Windows XP I have attempted to use the same workflow as on the OS X machine sitting next to it. Both have the same amount of RAM and the CPU in the Mac is about 25% slower. Common tasks include:

• Create a large PDF from InDesign while browsing Web pages.
• Run a perl script on a large XML file while reading e-mail.
• Edit photos in Photoshop, while performing search and replace operations in a word processor.
• Export a large .doc file to .rtf while updating items in CVS.

In every one of these tasks Windows has choked, using three separate machines. As of Win2K doing something in one application while others sit idle has been acceptable, but you still can't actually do two CPU intensive operations simultaneously and expect the GUI to actually respond. Attempting some of the tasks above has left me waiting as long as ten minutes for the mouse to start responding again. All of these tasks work just fine in OS X. This alone has saved me so much time at work that it makes up for posting on Slashdot all the time.

Note, Linux handles multiple operations as well or better yet. Pretending this flaw does not exist in Windows, however, is just fooling yourself. Talk to any PC gamer. How many of them leave other programs running in the background while playing a CPU intensive game? How many would leave a big, resource intensive application running? If I'm gaming on Windows, everything else needs to be shut down. On the mac, I don't bother.

Re:Steve's statements

[99BottlesOfBeerInMyF]

Are we talking a Ubuntu here. Because the description is pretty accurate. Except well that Ubuntu is free, and runs on any PC.

Ubuntu is a fine distro, but it does not compare well for functionality or ease of use with OS X. I work at a company that makes network security devices. Three years ago there were a couple of powerbooks here. Most everyone was running Linux or a BSD. Today powerbooks dominate and only one regularly runs anything but OS X. That is because a lot of people switch, but basically no one ever switches back. Ubunto is great if it does what you want and need. Do not, however, believe it can do the same things as OS X, as easily until you try both.

Re:Why?

[777film]

Why is there all this talk about running windows on it? Honestly, it's like buying a new Porsche and ripping out the engine and putting a Dodge Neon engine in it.

That's a tired analogy, and it also rings false... An Intel Mac and a Windows machine have essentially the same engine, an Intel processor.

As for why someone would want to run Windows, just because such a thing seems unnecessary in the world you've carved out for yourself doesn't mean it wouldn't be a boon to others (otherwise Virtual PC wouldn't be as successsful as it is, and Virtual PC ain't that great.) It goes beyond games, there are a lot of Windows-only programs that have and will never make it to the Mac, and some of us are forced to use them because of our employers.

Re:Why?

[tgd]

Mine too, which is why it was a good analogy, only most of the time my turn signals don't, in fact, work, I don't have seatbelts, just the harnesses and by any reasonable estimate it probably shouldn't be street legal.

Its always good to know the right inspection shops...

Re:Why?

[Dhrakar]

Exactly! Plus, I figure that my new iMac 20" should be even easier to carry to BF2 LAN parties (which I will finally be able to participate in) as those fancy, shmancy Shuttle systems. And as for speed, I'd be wiling to bet that even with Rosetta, it is still way faster than my 800Mhz G4 flat-panel iMac :-)

Re:Why?

[n6mod]

It's always good to know the right inspection shops...

Or live in CA. It *smogs* fine.

Off to put on a fresh set of A032s :)

Re:Shut up! (Not the only one)

[sp0rk173]

I think I am on sig revision 5 now, still trying to get something that is intellectually provocative yet appealing to slashdot moderators.

How about no sig? It works for me. I'd rather have people focus on the issue i'm spouting on about, rather than pidgeonhole me ideologically based on my sig (Which people can and will do).

Er - did you read the slides ?

[Space cowboy]

The title was the benchmark-name. The introduction was about the industry-standard nature of the benchmark, and the disclaimer was that real-world performance may not reflect that. The idea was to show the isolated CPU gains.

Seems pretty straight-forward to me.

Simon

Re:Er - did you read the slides ?

[DrXym]

It is straightforward if notice all the places on the apple site touting the new Mac as "2x faster" without a mention of benchmarks or their likely real world performance which will be nowhere near that. It's outright lying. Apple got in trouble in the UK for deceptive advertising when they claimed the tower G5 to be the fastest personal computer when it wasn't. I expect they'll do so again.

Re:Er - did you read the slides ?

[Space cowboy]

Go to the home page for the imac - where you get taken whenever you click on an imac on the site.

Clearly next to the 2x image are the specific benchmark names and values.

You're just trying to find fault. For the record, I don't think you have to be tech-literate to read small-print. Contracts have had small-print long long before we had computers...

Simon

Re:No AMD macs? Excuse me!!

[evilviper]

AMD competes very well in the desktop processor arena, and pretty well in the desktop-replacement notebook segment. However, they don't have a great answer in the thin-n'-light laptop segment.

I'm afraid that your information is several years out-of-date...

AMD's mobile CPUs are now commonly lower power than even the best of the Pentium-Ms to-date. That's in-addition to being cheaper, and higher performance at the same time.

For example:
    mSempron 2800+ (1.6GHz) 25W
    mTurion MT-34 (1.8GHz) 25W
    mTurion ML-37 (2.0GHz) 35W
    mAthlon 64-3000+ (2.0GHz) 35W
vs.
    mCel-2.5G 35W
    Solo-T1300 (1.66GHz) 37W
    Pentium M-780 (2.26GHz) 27W

And that's as fair of a comparison as possible.

While I'm sure the marketing dollars are nice, I think that's not the primary reason for Apple to choose Intel over AMD.

Since we've debunked the notebook justification, what alternatives are there?

Re:Why? (totally off topic now)

[cypherz]

Cada loco tiene su tema. I've done this math too. I got more than I could with a Jap car. The trans, differential etc is usually the weak point in building up a japanese car. Try running more than 10 lbs of boost in a Civic and see what happens. You'll not be driving it for long. The cool thing about the Dodge SRT-4 is that the car was built to be hot-rodded. The tranmission is heavy duty and has a torque biasing differential already in the driveline. By the time you replace the transmission and the differential and retrofit a turbocharger etc, you'll likely spend more than just buying the Neon. Plus, the SRT-4 can run 15-20 lbs of boost with no work needing to be done to the engine's bottom end. Also, the stock clutch is already upgraded. SRT-4 only needs a new clutch if you go to the Mopar stage 3 setup (or similar). In other words the stock drive train is good up to about 300 HP. Take a Civic Si for example, (I love the Civic Si) and try getting 350 HP out of it. You can do it maybe, but not run it very far unless you upgrade the *entire* drive train. The SRT-4 is perfectly happy with 350 - 400 HP and you can order one that way from the factory! (They'll ship Stage 3 motors in a crate though cause its not "approved" for street use - must assemble).

And what exactly makes it a "bad imitation of a rice burner"? Its not an imitation of anything! So go build your rice burner and I'll see you on the track! :-)

Re:SSE

[Space cowboy]

I was referring more to the change from Altivec to SSE(x) rather than SSE2 to SSE3. One of the big "wins" for Altivec is the permute instruction, and there's no equivalent in SSE (yet, as far as I know). It takes ~60 clocks in SSE to do what permute does if you code it in SSE, and that's a lose.

The solution, then, is to redesign the algorithm so it doesn't use permute too much, and that's a lot more than a recompile. That's why I think it'll get better...

Simon

Re:No AMD macs?

[evilviper]

Are you trolling or just slow in the head?

Such an ironic statement. Considering that you're the one with your facts wrong, I have to ask you the same thing.

AMD has nothing useful for laptops right now. their top chip uses 15-60% more power and is slightly slower than the Intel Duo. It uses more power idle than the Intel does at 100%.

I have a hard time imagining you could be so far wrong by accident.

Since I don't want to duplicate my last post, here's a link: http://apple.slashdot.org/comments.pl?sid=174203&c id=14493705

Feel free to point me to any of your sources showing an Intel chip using less power at 100% than the lowest-power AMD chips. (Note: Idiots on other web forums, and talking dogs don't count.)

Re:No AMD macs?

[zoomzit]

This just flat out isn't true. When Intel and Apple did their press release regarding the upcoming Intel Macs, AMD's official response was pretty much, "why didn't they talk to us? Perhaps we can talk to Apple in the future to make AMD based Macs."

Oh, and volume certainly wouldn't be an issue. Apple has only 4% of market share, while AMD has recently been outselling Intels.

Re:No AMD macs? Excuse me!!

[theAtomicFireball]

I don't know where you get your figure of 80M chips per year, but sites like this one list Macintosh sales for 2004 at approximately 3.5M units. Even if all were dual processors (they're not), you are overstating processors by more than an order of magnitude.

80M sounds awfully high to me also. A modern motherboard has many chips on it, some of which might also come from Intel, but even at the most generous, I'm having trouble seeing 80M.

Your number of 3.5M units appears to be about a million low, though. 4.5 Million is what they shipped in 2005. Granted, that figure is a lot closer to your 3.5M than the OP's 80M, but you're still shorting Apple by roughly 25% on units shipped and even if only 10% of Macs were dual processor/core, and 5% were quad processor, you're still bumping up against 5 million chips, and considering that many previously single-chip machines are going to now be dual-core, 2006 could very well be even higher. AMD might be able to handle that amount, but it'd be a risk, and supply chain is one area where Apple is going to play it very conservatively because of all the problems they've had historically.

Re:Why? (totally off topic now)

[drinkypoo]

If I had $22,000 I'd probably put a down payment on a house.

With that said; I can get a cherry '98 S14 240SX (note: rear wheel drive, with a multilink) for around $8000, a S13 SR20DET for $2500 or so, and definitely get 400HP for around $5000 (priced the stuff before.) Over 400HP you have to replace rockers, but let's go ahead and aim for 400. That's about $15,500 leaving me plenty of wiggle room, money for suspension parts and wheels, et cetera. When I'm done, I'll have a car with about the same weight and power as your 400hp modified neon, but a vastly superior suspension and with the rear wheels providing the power. The transmission is just about bulletproof and is close-ratio in the bargain. Another $500 will get me a 200mm limited slip (like the one in my S13 240SX.)

And on top of all that, I won't have the worse side impact protection in America :P

The Neon is designed to be a no-frills econobox, and it shows. It most certainly is an imitation of a Mitsubishi, and a not particularly good one at that IMO, when the lancer evo is finally alive and well.

Oh yeah, and the 240SX has near 50:50 weight distribution, and 50:50 is trivial to reach. So let's see, RWD with 50:50 weight means extremely predictable power oversteer on tap. Rear wheel drive means quicker launches. Close ratio transmission means harder acceleration, and it comes stock. Faster on the straightaway and faster on the turns? Sounds like a win-win to me... But then, the integra and the neon are pretty much the natural enemies of the 240SX so I'm biased. (I would include the civic but it's even more of an econobox.)

Well....

[snuf23]

The point is that in real world usage many applications run on a single thread. Therefore the only benefit in speed boost to a single threaded application is that the other processor core can handle background tasks or OS related threads, potentially freeing up a bit of speed maybe. A single threaded application will not run twice as fast on a dual core processor, although a benchmark designed to run on both cores will come close.
The biggest benefit for most users in a dual core system will be from multitasking and not having the OS lose the snappy feel when an app is running.

Re:Well....

[dangitman]

Therefore the only benefit in speed boost to a single threaded application is that the other processor core can handle background tasks or OS related threads, potentially freeing up a bit of speed maybe.

That alone is not an insignificant benefit. But with so many Mac applications supporting multiple processors, I'm not sure what your point is. What Mac user is using processor-intensive applications, that do not also support multiprocessing? It's not like dual-processor Macs and applications are a new thing.

Re:Well....

[snuf23]

The point is don't expect MS Word to run twice as fast.
I'm sure apps like Photoshop and Final Cut will take advantage of two processors but I doubt a lot of commonly used non pro apps will.
I'd like to see a benchmark break this down and show multithreaded performance in an application on a Core Duo versus a single G5 iMac.

Re:Well....

[ngm]

And if MS Word ran twice as fast, what would I do get my papers done faster?

Taco's Chicken!

[everphilski]

cause Taco is a wimp and plays on a PvE server :)

Re:Why? (totally off topic now)

[cypherz]

True. I didn't do something similar because:
1) I have no place to work on cars.
2) I wanted something reliable. I wouldn't have confidence in the car you described as a daily driver (if _I_ built it that is). Now maybe if you built it, it might be reliable.
3) I am being perverse. I LIKE driving the underdog. (when the underdog is as fast as the SRT-4).
4) I got a new car loan for the SRT-4.

Re:But does it run...

[Yjerkle]

Not to mention System 7 got rid of the option to turn off MultiFinder. Nothing sucks up resources like having programs open when you're not using them!

this

[everphilski]

Yonah

Comparable to a Athlon 64 X2 (that's a desktop chip) with way less power draw (both idle and peak load).

Other factors exist too... AMD used to have a reputation for poor QA on the line, and while they seem to have overcome it, hey history is a stinger when you are dealing with companies like Apple.

Re:this

[evilviper]

Comparable to a Athlon 64 X2 (that's a desktop chip) with way less power draw (both idle and peak load).

Of course if you compare a desktop chip to a mobile chip you'll see a huge disparity, but that doesn't make it a fair comparison.

In other words: Bullshit factor too high!

Re:this

[everphilski]

Yeah, disparity in power at the same processing power. Different power in, same amount of processing done. Mobile chips don't magically use less power. That is innovation folks. The intel chip is a lot more efficient. AMD didn't have a mobile chip to compare to.

Re:this

[evilviper]

Mobile chips don't magically use less power. That is innovation folks.

No, it's called undervolting. You build the chip to a somewhat higher standard, and they will run stably at much lower voltages, which drastically decreases power consuption and heat output.

In fact, AMD mobile chips have generally just been desktop chips that tested to a much higher tolerance.

AMD didn't have a mobile chip to compare to.

True, but that doesn't make the comparisons any more fair. Intel just beat them out of the gate this time. People weren't lambasting Dell for using P4 CPUs just because Intel didn't have a dual core chip available as soon as AMD did. I don't see how this particular twist is getting billed as Intel being better for notebooks, or being a reason Apple chose Intel over AMD.

I have no doubt, if Apple had chosen to use AMD processors instead, AMD would have gone out of their way to rush an effecient dual-core notebook chip to market as well.

Re:this

[everphilski]

No, it's called undervolting. You build the chip to a somewhat higher standard, and they will run stably at much lower voltages, which drastically decreases power consuption and heat output.

And yet, at that undervolting, it can still kick an AMD chip's ass at a processing task.

I have no doubt, if Apple had chosen to use AMD processors instead, AMD would have gone out of their way to rush an effecient dual-core notebook chip to market as well.

and been late to market... Yonah wasn't made for Apple, Yonah was there, and Apple ordered a bunch...

Re:No AMD macs?

[99BottlesOfBeerInMyF]

I have a hard time imagining you could be so far wrong by accident. Since I don't want to duplicate my last post, here's a link... Feel free to point me to any of your sources showing an Intel chip using less power at 100% than the lowest-power AMD chips. (Note: Idiots on other web forums, and talking dogs don't count.)

How about next time you feel like posting numbers you actually look up the chip we're talking about. The Intel Duo is not a Pentium-M, like the ones you listed in your previous post. You're the one who is out of date. Intel Core Duo (known as the Yonah before the official name was released) is the 65nm chip and has been out for a while now. Here's a link to a review and a pertinent quote, "In fact, a 2.0GHz Yonah under 100% load consumes less power than an Athlon 64 X2 3800+ at idle." These are comparisons of Intel's latest versus AMDs current offerings, that compare similarly for performance. Please feel free to show me any comparisons of AMD chips that outperform the Intel Core Duo, or even come close, but I don't expect to see anything in that range until AMD brings their 65nm solution to market in Q4.

I wish macs came in black

[Xoth]

Intel move was not a surprise. OS X is just BSD UNIX with a pretty graphical desktop. Easier to run on CISC chip. Apple is in a quandry, makes good software and consumer electronics but macs bought only by people who dont know better. Its a slippery slope as we move along and people realize thats is merely a choice between gnome, kde and OS X. Maybe some compromise and convergence will happen in the future. We'll get a pretty graphical desktop, good software but on linux on intel. Hey you can still run it on a computer that looks like a lamp if you want :)

Re:I wish macs came in black

[Xoth]

My post said I preferred OS X over X Windows so dont burn an apple on my lawn. And that was my point, to see OS X for what it really is so more people can use it. Dont get infatuated with cute hardware. I can run X Windows on a mac, but I cant run OS X natively on my FreeBSD PC. Its even more easier now to run OS X on something other than a mac but the only thing stopping me is Apple. Stop putting XP CD's into intel macs. Start putting OS X CD's into PC's. Its the same old debate but this time Steve Jobs might take the leap.

Re:No AMD macs?

[justsomebody]

Actually, the new Yonah processor is significantly faster than the Turion. Check out these benchmarks:

Actualy, test done were 32-bit. AMD has 8 extra 64-bit registers usable only when running 64-bit mode (which makes them unusable to 32-bit appliaction in 64-bit environment too). Performance gain is more than significant. At the same time power consumption should be lower because of shorter powering up of CPU (valid only for real life use (shorter time of workload, longer time of idle), this should not be valid for benchmarks since benchmarks do tend to stress cpu to get results).

I know there's no difference for Windows users. But, some of us do run 64-bit environment. So it will be probably Turion dualcore for as soon after it comes out. I would prefer even Turion over Duo. But since it is just few months until dualcore...

btw. Vista will be 64-bit. Duo won't support that. But in case 32-bit release will pop out too, I'm wrong

Re:Why?

[ClamIAm]

void your warranty by trying to get Windows to run on it

Next you're going to be telling me that installing third-party applications will void my warranty. Or ripping CDs. Or using Capslock! Please, try and remain sane while posting.

If I buy a Honda Accord, is it reasonable to expect Honda to ensure that the turbo-kit I got for my Ford Escort runs on that Honda? Of course not.

Nice try, but this is a really flawed analogy. Changing out the engine is like putting a new CPU into the system, not like booting an OS that didn't come preinstalled.

You have complete freedom to buy, or not buy an Apple computer, and all that implies.

You have complete freedom to do whatever you want with things that you have bought (pretty much). Saying that it is some sort of sacrilege to install something not blessed by St. Jobs is a completely ludicrous argument. And you aren't called by divine providence to defend Apple's right to a "clean" world, or something. Get over yourself.

Real benchmark

[Macgrrl]

It's not the number of frames in MC that's important (though the imps used to crash out my PowerBook), it's the framerate at Vael that's the best benchmark....

WHY WOULD ANYONE WANT TO RUN WINDOWS?

[Wakko Warner]

Seriously. Buy the damn thing to run MacOS. Why did the author of the article even fucking bother trying to install Windows? Give us some information that's useful, not "well I couldn't run something nobody would want to try running on it anyway".

Re:WHY WOULD ANYONE WANT TO RUN WINDOWS?

[imthesponge]

They want to run Windows apps while still having Mac OS on the same machine. I'm sure that if they could run Mac OS on their PC, they would do that instead.

Re:WHY WOULD ANYONE WANT TO RUN WINDOWS?

[metallic]

This information is important to me. I do a lot of development using ASP.NET and for that I need access to a machine running at least Windows XP Professional. If I'm away from an internet connection, it would be very important for me to be able to boot into Windows and have a full featured development environment. And at $2k for the Apple laptop that I want, buying a second laptop just to run Windows is unpractical.

Megahertz Myth?

[defunc]

I remember Apple kicking out a campaign regarding the megahertz myth to counteract Intels' marketing blitz about their chip crossing the 3.0 GHz barrier, where is that ad again?

Re:Why?

[sgt scrub]

Honestly, it's like buying a new Porsche and ripping out the engine and putting a Dodge Neon engine in it.

But, then you could brag to your friends about having a MOPAR in your porche!

running windows

[heatdeath]

The part of TFA about trying to run windows reminded me of the scene from zoolander where they're spoofing space odyssey 2001. He tried mashing all sorts of button combinations to try and get the windows CD to boot, and concluded after all his button-mashing that "I think it's going to be a matter of time before Windows is running on the iMac, especially Vista."

Re:Why?

[Anonymous Slacker]

I find the 'safety gear on a race car' analogy to be a good comparison, so long as the issue of computer security isn't involved.
If I think about how my station wagon is properly equipped to drive the local roads and interstates with only regular maintenance it serves me well as a daily driver taking me anywhere the general public is likely to go, but I'd probably get beat down if I were to try to run at a race track.
But if I had a Formula 1 race car in my garage, I could tear things up at the track all day long, but it doesn't have all the right features required by law to drive to the grocery store (it can get there, but the road choices are limited to where the police aren't patrolling, etc.)
Now the MacBook and new iMac come out. This is like making a race car with a rotary engine. Yeah, it's different, but it works as a race car, but it's still not street legal. I can also buy a Mazda RX-8 with a rotary engine, but it's not a race car.
There's more to making a race car street legal than changing the engine, just like there's more to running Windows on a computer than changing the processor. Especially one that was built for a different purpose. Sure, it's theoretically possible in both cases, but some technical expertise is required for both, and in either case don't expect the manufacturer to go out of their way for a few end-users who want to do something that wasn't supported.

Re:No AMD macs?

[man_of_mr_e]

64 bit doesn't improve perfomance by that much, certainly not the 25% or so that Yonah has over the Turion. At best, you get 3-6% improvement from 64 bit optimized code.

I've never seen any benchmarks for 64 bit code over 32 bit (even under Linux) that are anywhere near the 25% performance gain that Yonah has over Turion.

Re:No AMD macs?

[justsomebody]

64 bit doesn't improve perfomance by that much, certainly not the 25% or so that Yonah has over the Turion.

Personally, I can't say about 25% in 64 over 32, but I can say very, very, very noticeable (from a bit slugish desktop to completely snappy as never before). Didn't do benchmarks, just accidently installed 32-bit version, figured that out in about one hour. Reinstalled 64-bit, forgot benchmarks. Next time I plan to do that without accident and actualy run benchmarks.

Here is a benchmark on Windows though (OS 32-bit against 64-bit on the same architecture, didn't found any test on Linux)
http://www.pcstats.com/articleview.cfm?articleid=1 665&page=1
While 64-bit lagged on few (streams only, and that doesn't account for 20% of tests) it was sometimes even more than 3x faster on others. So, registers do mean a LOT.

Sad, but true. Yonah is dualcore, Turion in test was singlecore. Turion dualcore is yet to come out in few months. I pointed out that too. And said it is worth to wait. But I never said that Turion (that one used in tests) is faster than duo. This is why I prefer Turion over duo (price/performance and the fact I run 64-bit Linux, duo is 32-bit).

Dualcore over Singlecore having performance gain of less than 25% (now here I'm counting 64, 32 bit diff)? SAD. And Turion dualcore comming out in first six months of 2006:) Guess which one I'm buying. Name definietly won't start with I... or be named after a fruit

true that - and another thing

[nobodyman]

Parent is absolutely right. I'm certain that the vast majority of viewers could tell the difference between a video captured at 24fps vs 60 fps.

And the claim that the blurryness of video offsets ths framerate is also debateable. I'd argue the opposite, in fact - 60fps video is much, much more sharp than 24fps because the motion blur obscures the detail(you only notice it on the edges, but it affects the entire frame).

Re:true that - and another thing

[geenome]

I read somewhere that they tried shooting the second season of the original Twilight Zone in 60 fps video, over 24 fps film, in order to save money. The experiment lasted six episodes, the result being so bad that they eventually ditched the idea and went back to 24 fps. It depends entirely on what mood the movie tries to create. Yes, frame rate matters. No, one isn't always better than the other.

Re:No AMD macs?

[man_of_mr_e]

I should have been a bit clearer. Yes, if you're doing large math, like PKI encryption/decryption, then you're going to see very good results with 64 bit, but that's not a common task for most people.

I really doubt you would see that great an improvement in responsiveness between 32 and 64 bit, unless the 32 bit version were optimized for 386 rather than 686 (which is pretty common). If you're going to benchmark, you should at least make sure you're using a kernel that's optimized for the 32 bit version of the CPU you're using.

One reason I doubt the improvement is that you also have a tradeoff of nearly doubling your memory load. 64 bit pointers are all twice as large, and ints will be 64 bit by default, which will mean the average unit of storage is also double.

Re:No AMD macs?

[Anonymous Coward]

Those power figures are from an entire _desktop_ systems, not just the CPUs as you imply. Now, the idle load of a desktop is very different from that of a laptop system --- so those numbers don't support your claim.

Re:Why?

[Rebelgecko]

Their new magnetic power cords supposedly solve that problem (Breaking)

Re:No AMD macs?

[justsomebody]

I should have been a bit clearer. Yes, if you're doing large math, like PKI encryption/decryption, then you're going to see very good results with 64 bit, but that's not a common task for most people.

Image manipulation, video, audio, archiving... all rely on large math. Or do you find nowadays computers not being responsive in notepad? So, yes it is common. Look at standards, cheap cameras now ship with 8MPixels. It is not really conservative. DV cameras, cheap scanner scanning high resolutions, cheap printers with higher resolution. Face it, multimedia needs grow in drastic steps. More and more relying on heavy math because no op can still be simple, it is just geting larger and more intensive

I really doubt you would see that great an improvement in responsiveness between 32 and 64 bit, unless the 32 bit version were optimized for 386 rather than 686 (which is pretty common). If you're going to benchmark, you should at least make sure you're using a kernel that's optimized for the 32 bit version of the CPU you're using.

There are two kernels amd and amd64 optimized. You can't find i386 kernel practicaly in any distro. So, yes after seeing that diff I become almost religious 64-bit user. I don't go to museum. And gain is more than noticeable, yes. It is realy enjoyable.

Tests I pointed you were done on Windows. What are Windows 32-bit optimized for? If you say i386 than you are just discrediting any benchmark until now where Intel and AMD64 were tested because in that case 64-bit were always severly crippled.

This difference is not common for PPC arch. PPC already had enough regs from before 64-bit.

One reason I doubt the improvement is that you also have a tradeoff of nearly doubling your memory load. 64 bit pointers are all twice as large, and ints will be 64 bit by default, which will mean the average unit of storage is also double.

Very, very bad arguments. It almost seems like you lost all your sane arguments

1. Memory is cheap. Very cheap. And you only need to double (or better percentage gets higher with segmentation) it when data is segmented, meaning 100% data consisting of only pointers or ints. Images, videos, audio, (any larger format), etc are not segmented, mostly few block allocations, which means pointer is 64-bit (taking whole sum of 8 bytes instead of 4), data inside memory can be 8,16...-bit. RGB values when opening picture are the same size in 32 and 64 bit system
2. Hard disks are cheap. Even cheaper than RAM. Again does not count for any standard format like audio, video... If it would count, you couldn't load for example mp3 made on 64-bit when using 32-bit arch. It only counts for some internal formats where you wrote int to output file, but those are very rare.
3. You forgot dualcore, singlecore diff:) Now that is not really fair of me:)
4. In measuring performance, factor of your hope does not count;)

Re:No AMD macs?

[lostchicken]

AMD has only been outselling Intel on the retail market, which means that more people who are building their own computers are using AMD than Intel, and it's still pretty close. The OEM market, which is where the vast majority of CPUs go is still absolutely dominated by Intel. Yes, AMD likely would be able to supply Apple in terms of percentage of production, but I'm not sure they could supply the same kind of flow that Intel could for mobile processors.

Re:No AMD macs?

[man_of_mr_e]

Image manipulation, video, audio, archiving... all rely on large math.

No, not really. These tasks are typically handled by SSE instructions, which haven't change between 32 and 64 bits. Large integer operations are far less common.

You can't find i386 kernel practicaly in any distro.

i386 kernels are still VERY common. I just installed Kubuntu 5.10 and it defaulted to a 386 kernel, for instance.

Tests I pointed you were done on Windows. What are Windows 32-bit optimized for?

I believe that Windows has been optimized for 586 for a number of years. There's not a huge difference between 586 and 686 in terms of performance optimization. Windows is NOT optimized for AMD, though, that's for sure.

The only tests that showed any major improvements were large integer math tests, mostly the PKI stuff, which according to the article was hand optimized for AMD, which is really quite a bit different from general optimizations.

Very, very bad arguments. It almost seems like you lost all your sane arguments

You seem to completley misunderstand. I said nothing about how MUCH memory or disk space. I was talking about memory bandwidth, and the amount of throughput. When you double the data without doubling your throughput, you lose performance. Now, granted, the AMD's integrated memory controller really reduces memory latency, but currently that's limited to DDR400. Intel systems can run at DDR2-667 speeds (though Yonah is significantly below that right now, being 32 bit).

Comment removed

[account_deleted]

Comment removed based on user account deletion

Re:Why?

[Overly Critical Guy]

Don't just "seriously consider" it, man. Get one! I got a Mac February of last year and was blown away, and already ordered the new iMac. I'm hooked.

Re:No Windows? INT WTF?

[davidsyes]

This should at LEAST be funny. Maybe 3 funny. The question IS valid: Who the hell's gonna pay Apple's crazy prices? Those who can afford to, or who receive donor money (schools do, right?) to do so.

OTOH, I sure as hell wouldn't try to run Mac s/w on a PC. I might ooh-ahh over some Apple/Mac stuff, but if I REALLY want a Mac, I'll save up for one. Currently, I have no need to. I like my Mandriva the way is its, flawed or not, so long as I get stuff done.

But, I can sure answer this (my own question posed): "Who the hell will run windoze 98 inside Linux?" with: ME!

I do so because I am hopelessly and helplessly hooked on Lotus SmartSuite. I cannot get WINE to run for SHIT, and the folks at the various emulations/host software companies brag about NOTES and ORIFICE, but won't take the time to certify SMARTSUITE. If they did, I'd buy their stuff just to avoid kernel patching and dependency issues, at least when Win4Lin kernel patching was usually what I ended up doing. Now, when I get money, I'll buy W4L's software, just as I've been doing since around 2000. It WORKS. (Why SmartSuite: Usability. Hands down it beats orifice, and it has stuff that OO.o and SO.o won't have for YEARS to come: WordPro and Approach. Those two products alone--with their features-- are worth $200 of my money, even tho SO is $68 and OO.o is "free"...)

Now, as for running windoze on a Mac... I could see labs and schools wanting to save money, but as stated elsewhere, it's not as if Apple should be accommodating a perceived or actually inferior hegemon's warez on Apple hardware. If someone GAVE me choice of a Mac w/ OS X or a $5,000 windoze-based rig with *doze foisted upon it, I'd take the Mac, mainly because I don't care to learn ANY MORE stuff from ms. I haven't had Mac experience in years, and they have INTERESTING stuff. I just can't afford a Mac, I'm not an artist in the Mac sense, and the GUI paradigm is a bit different for me, tho I WILL learn it if I had a Mac for free. I have been impressed with the various Mac desktops and laptops a friend of mine owns. I feel jealous. Mac stuff is functional ANNNNDDDD cool. A nice blend for something you're gonna look at. Most PCs are hideous enough to be sequestered to beneath the desk. Converse for most Macs (most, not all of them: I don't like that blue egg, the pink and orange and other colors.... If I want an Easter Basket, I'll go to Rite-Aid when I pick up some pumpkin-flavored ice cream, thank you very much...

As for Jobs being Q with a turtleneck and a pair of jeans... nice. I equate m-shaft's core officers as Starling, Braxton, the Vidiians, the Kazon. I'd rather BE a Borg than one of the other ilk... And, while it looks funny to put gate's head on a Borg suit, I respect the Borg. I feel about mshaft the way I felt when Seven of 9 responded to Neelix's "Are there any Kazon among the Borg?" with "The Kazon are unWORTHY of assimilation."

And, there you have it, if I can suggest another answer to why Apple is not required nor desired (by me, at least) to "assimilate" mshaft. Just as the Borg wouldn't assimilate the Vidiians, either, especially with all that slacking, tearing skin...

Re:No AMD macs? Excuse me!!

[MojoStan]

AMD's mobile CPUs are now commonly lower power than even the best of the Pentium-Ms to-date. That's in-addition to being cheaper, and higher performance at the same time...

And that's as fair of a comparison as possible.

Your comparison is unfair, inaccurate, and incomplete. You didn't include any links to where you got your numbers, overstated the Core Solo's power, and didn't include the Celeron M (which is based on the Pentium M core) or Core Duo. You also neglected to include the low power versions of the Intel CPUs.

Here's my attempt to make your comparison more fair and complete:

AMD (W = Thermal Design Power)

• LP mSempron 3300+ (2.0GHz) 25W
• Turion MT-40 (2.2GHz) 25W
• Turion ML-44 (2.4GHz) 35W
• LP mAthlon 64 3000+ (2.0GHz) 35W

vs Intel (W = Thermal Guideline)

• Celeron M LV 383 (1.00GHz) 5.5W
• Celeron M 360 (1.4GHz) 21W
• Celeron M 390 (1.7GHz) 27W
• Pentium M LV 778 (1.6GHz) 10W
• Pentium M 765 (2.10GHz) 21W
• Pentium M 780 (2.26GHz) 27W
• Core Solo T1300 (1.66GHz) 27W
• Core Duo T2600 (2.16GHz) 31W

Re:But does it run...

[gmhowell]

Sorry, Charlie, but 6.0.8 is the ne plus ultra of Mac System software. If you don't believe me, I'll dig out my Mac Classic and beat you in the head with it.

Provided I can find an ADB mouse around here...

Re:Why?

[filesiteguy]

It is all just a matter of money. If I can afford it I will. I just have to keep in mind that I can get a PC loaded with SUSE 10 and all the trimmings for around $800, when compared to a similarly equipped Mac. However, since this is for my wife, I am really looking into the option.

Thanks!

Re:No AMD macs? Excuse me!!

[Scudsucker]

How do you figure this:

I'm sure AMD could have given them all the chips they need. They might not have been so forthcoming with the Marketing Money however.

With this?

For Intel, getting Apple is a coup worth paying enough for that even if they never make a cent from Steve Jobs, they've still silenced the biggest critic of the i86 architecture.

I think AMD would happily be "forthcoming with the Marketing Money" if they got Apple, because they would have a show pony for their technology, as opposed to higher producing but boring, supplier-of-the-week companies like Dell or HP.

Re:Why?

[Dekator]

That analogy doesn't hold water either. You can ride the same roads as far as (funtionally identical) apps go. Also, more importantly, I've never seen a racing car without seat belts and wouldn't recommend using one either...

Re:No Windows?

[Dekator]

On the other hand, who'd buy a cheap PC to run MacOS X ?

Re:No AMD macs?

[superpulpsicle]

You said nothing about why Intel is more innovative than AMD. The only thing you did was post as anonymous coward and say Apple wants to be anti-oem. No shit. You're just pissed that I might be right about your godly Steve Jobs using Dell as a guide to a better business model.

Was done on iMac

[Dekator]

As Steve said, the keynote was done on an Intel iMac. Also, he said Photoshop was running on Rosetta and that it wouldn't be fast enough for pros, but good enough for occasional home-use. Doesn't anybody actually watch the stuff they comment on ?

Re:Was done on iMac

[johnpaul191]

i never watched the keynote (because it wasn't live and i know how it ends?) but i read the live play by play feeds on Mac news sites. i remembered SJ said the keynote was done on an Intel machine, but i was not actually looking at it for 90 minutes, so i forgot it was an iMac.

Never said you have to

[SuperKendall]

I didn't say you had to spend $300 a year. It's just that a lot of people do... hence the "reason to drop $300" wording.

I agree that does not mean you really have to, but PC's have other costs as well and really using a consle as a primary gaming device is just so much more tranquil. Then you always have access to some game even if the PC is acting up.

I like that guess

[SuperKendall]

Whatever they have cooked up, I hope they drop in favor of "Macstation". That sounds pretty cool.

Complete Trash Post

[Nazmun]

Pardon me for sounding angry (most likely because I am). But the parents post above is absolute trash. Theres a reason amd got the market back and it's not soley because they had a decent performance rating system so as not to confuse customers. The new intel naming system is utterly ridiculous.

The athlon naming system is based on benchmark performance on how well their old athlon tbird would perform at that respective mhz. For example, a 2800+ would theoretically perform as well as a 2.8ghz tbird. This is not an attempt by amd to skew data.

What consumers want in the end is performance. That is why amd has been gaining marketshare. They've been providing processors that perform well and use less heat then their intel desktop counterparts. AMD's prices have also been a far greater value on everything but the highest end processors where they are similarily priced to intel.

If amd simply rated slower processors with higher numbers then they'd be finished a long time ago. Even if the mhz doesn't increase in many cases the core improvements are enough for a ratings increase and realworld benchmarks support it.

I can argue on much of your "history" too but it's late and i'm tired.

More correction

[Halo1]

Damn, I missed that the Anandtech page was talking about total system power consumption. CNet mentions the cpu's power consumption: maximum between 25 and 49 watts. So it seems it's about the same as the G5's.

Re:More correction

[Glock27]

Damn, I missed that the Anandtech page was talking about total system power consumption. CNet mentions the cpu's power consumption: maximum between 25 and 49 watts. So it seems it's about the same as the G5's.

I don't understand this at all (and I did look at IBM's site, they list power consumption for a 2 GHz 970 at around 50W). How is it that a G5 Powerbook was such a problem?

The only thing I can think is that there are no power conservation features on the chip, and/or the support chips take quite a bit of power.

At any rate, Core Duo laptops look to be good for 5.5 hours of moderate use...that's quite good for a single core system, let alone dual core. :-)

Re:More correction

[Halo1]

I don't understand this at all (and I did look at IBM's site, they list power consumption for a 2 GHz 970 at around 50W). How is it that a G5 Powerbook was such a problem?

The only thing I can think is that there are no power conservation features on the chip, and/or the support chips take quite a bit of power.

The latest G5's have many power conservation features on the chip. The only thing I can think of, as mentioned earlier, are indeed possibly the support chips (doesn't Yonah have an on-die memory controller?). Or Steve may have just been fed up with the PPC partners for failing to deliver on their promises time and time again, and used fake excuses to the public...

At any rate, Core Duo laptops look to be good for 5.5 hours of moderate use...that's quite good for a single core system, let alone dual core. :-)

Where did you get that data? All I found was "probably around 3.5-4 hours", based on a comparison with an Acer that also uses this processor (the Travelmate 8100). Of course an Acer is incomparable to an Apple as far as design is concerned, but I don't know how much different power consumption will be.

And I really wonder why on earth Apple doesn't mention it on its website, after it has done so for all these years...

Not quite true

[Andy Dodd]

The "baseline comparison" CPU was the Pentium 4, not the Athlon Thunderbird.

The XP was slightly more efficient clock-for-clock than the Tbird, but not by very much. It WAS significantly more clock-for-clock efficient than the Pentium 4.

Like the XPs, if the Tbirds had used the same rating system, they would have been rated on average around 1.4 to 1.5 times their clock speed.

Other than that, your post was accurate.

One person bashed AMD's performance rating system for being confusing (for example, multiple CPUs rated at 3200+) - It may be somewhat confusing, but it's less confusing than trying to figure out which AMD CPU with a base clock rate of 2.0 GHz is going to perform better. (The Athlon XP 2800+ in my system is clocked at 2.0 GHz. The 512kb cache Venice-core Athlon 64s clocked at the same rate garner a 3200+ performance rating due to a much better memory subsystem than the XP. Not sure what the 1M cache version rates.)

Intel has gone to a performance rating system that is even more confusing. They seem to have a different rating system for every CPU, rather than a common baseline comparison unit.

Re:Not quite true

[Nazmun]

They may have changed it i guess... but i was pretty sure they said in comparison to the tbird way back in the day of the first athlon xp's.

Re:No AMD macs?

[99BottlesOfBeerInMyF]

Those power figures are from an entire _desktop_ systems, not just the CPUs as you imply. Now, the idle load of a desktop is very different from that of a laptop system --- so those numbers don't support your claim.

True, but they are very indicative of power usage in real world applications. I'm quite willing to read any comparisons you can show to the contrary, or for that matter any technical specs on the chips themselves. Everything, and I mean everything, I've read by a reputable journalist (not some guy in a forum like you) has shown the Intel Core Duo to outperform every portable AMD in regard to power consumption and performance. This is expected as it is a fairly new chip, using a smaller die than other portables. When AMD comes out with their new chip it may very well be better yet. To flame multiple people in a forum, however, based entirely upon your lack of knowledge (first mistaking which chips were being discussed and then claiming an article that does not fundamentally prove a chip is better supports your position that it is obviously inferior), well that is arrogant and rude. You are an AMD fanatic, unwilling to objectively look at information. You only seem interested in information that might support the opinion you already hold. I don't see much point in having a discussion with you.

Re:No AMD macs?

[justsomebody]

No, not really. These tasks are typically handled by SSE instructions, which haven't change between 32 and 64 bits. Large integer operations are far less common.

But still depending on the number of registers when more complex math is involved.

386 kernels are still VERY common. I just installed Kubuntu 5.10 and it defaulted to a 386 kernel, for instance.

I can't even remember when I saw last one.

You seem to completley misunderstand. I said nothing about how MUCH memory or disk space. I was talking about memory bandwidth, and the amount of throughput. When you double the data without doubling your throughput, you lose performance. Now, granted, the AMD's integrated memory controller really reduces memory latency, but currently that's limited to DDR400. Intel systems can run at DDR2-667 speeds (though Yonah is significantly below that right now, being 32 bit).

Fact is that you're more guessing by the papers than relying on actual tests

p.s. Just to rub it in once more. Singlecore, dualcore? Low diff?

Re:No AMD macs?

[man_of_mr_e]

As far as I know, Turion isn't dual core. You might be implying that Turion single core would beat out Yonah single core, but that's really beside the point. Apple has to ship a product, and can't wait around for AMD to catch up.

Re:Why?

[tgd]

Heh, my car wouldn't smog fine. No cats, and runs so rich I blow flames out the exhaust when I upshift.

Compilers

[miller701]

No one seems to have jumped on the Steve's statement about complier choices. They used Intels' for the new iMac and IBMs for the iMac G5. As far as I know neither of those are used in XCode (GCC).

As others have pointed out, the RAM issue may have an effect on some of the tests, especially the Power PC binary programs.

Re:No AMD macs?

[evilviper]

Here's a link to a review and a pertinent quote, "In fact, a 2.0GHz Yonah under 100% load consumes less power than an Athlon 64 X2 3800+ at idle."

Completely out of context quote. First of all, you are comparing a DESKTOP AMD chip with a MOBILE Intel chip. Secondly, those are power consumption figures for the WHOLE SYSTEM, not just the CPU, as you very clearly implied in your last post.

These are comparisons of Intel's latest versus AMDs current offerings, that compare similarly for performance.

You're glossing over it, but that's an important point. Not only are thes comparisons incredibly unfair, but they're also vastly unfair due to the timing. You're comparing Intel's nice new chip that was just recently available, to AMD's older chip, that has been on the market for a very (relatively) long time now.

As I have to keep saying on this Intel-loving thread: Bullshit factor too high!

Re:No AMD macs? Excuse me!!

[evilviper]

Your comparison is unfair, inaccurate, and incomplete.

Bullshit, bullshit, and bullshit.

You didn't include any links to where you got your numbers,

A minor oversight: http://users.erols.com/chare/elec.htm

overstated the Core Solo's power,

Not according to my source, which I trust to provide accurate numbers. Considering that Intel uses a different power rating standard than AMD, it's not fair to compare AMD's spec sheet with Intel's spec sheet. Intel was the one to switch to "Thermal Design Power" when AMD was still listing "Max Power Output", and now that AMD has switched to "Thermal Design Power" Intel changes once again, seemingly trying to make sure direct comparisons are never fair.

and didn't include the Celeron M

What the hell do you think "mCel-2.5G 35W" means?

You also neglected to include the low power versions of the Intel CPUs.

That's because the "low power versions" are so vastly slower that it would be completely unfair to throw them in this comparison... Obviously, being incredibly unfair doesn't phase you.

Celeron M LV 383 (1.00GHz) 5.5W
        * Celeron M 360 (1.4GHz) 21W
        * Celeron M 390 (1.7GHz) 27W
        * Pentium M LV 778 (1.6GHz) 10W

All of which are so much slower than the 2+GHz AMD CPUs that it is terribly unfair to compare them.

As I have to keep saying: Bullshit factor too high!

Re:No AMD macs?

[justsomebody]

Seems you missed whole point of my comments. Some of us don't need to cripple Turion or Opteron to use it. For people like me your benchmarks are near to irrelevant. They are just singleminded benchmarks which come very valid for Windows users, nothing else. As I pointed out in my first posts. Some of us don't need to cripple our computer to use software we need. And if you don't superior can become something completely else. Correct benchmark would be some OSS tools (benchmarks, whatever), run on Win-32 or OSX-Duo and same OSS tools run on Linux-AMD64.

As far as I know, Turion isn't dual core.

Yes, it is not. I tried to point out that few times already to you. It just confirms my saying that Duo is far from being superior tech as Apple would like you to believe it is. On one side you're perfectly allright with crippling AMD, in fact you actualy enforce crippling AMD while keeping as optimistic as possible benchmarks for Duo.

You might be implying that Turion single core would beat out Yonah single core, but that's really beside the point.

As I said, I'm willing to wait dualcore. I won't buy Apple. I even said I would prefer current Turion over Duo for the 64-bit sake.

Apple has to ship a product, and can't wait around for AMD to catch up.

What has Apple to do with the CPU benchmarks?

chip on your shoulder, eh?

[flaming-opus]

My information is out of date? Well, that's also the data Apple was using when deciding what CPU to use in their newest offerings. While the slashdork crowd might use up-to-the-minute reviews and speculation to drive their purchasing decissions, Apple takes a more measured route. While we think of Apple as being very daring compared to other computer companies, it's also in the business of making money, and switching architectures is a major commitment, which has to pan out over several years.

Lets look at the example of Dell. Dell is nothing if not good at making money. We can make fun of Dell all we want for not using opteron, a clearly superior server chip, as compared to Xeon. However, Dell is not in the business of selling processors. They are selling server solutions, and they are selling a brand. Apple is less conservative than Dell, but they want a decission that will drive their products for at least 2 generations of iMacs. They can reexamine AMD at a later date, but they obviously feel that Intel can offer them a compelling processors for very light laptops, for at least 4 years. Intel has a strong history here.

Furthermore, all of the benchmarks I've seen, seem to indicate that CoreDou is a very nice processor.

Re:chip on your shoulder, eh?

[evilviper]

My information is out of date? Well, that's also the data Apple was using when deciding what CPU to use in their newest offerings.

That's extremely unlikely. Apple is a big enough company that they very easily get the inside information from both AMD and Intel (which they did). It's extremely hard to believe they looked at the specs for current Intel/AMD processors (and about a year and a half before they made the announcement) and completely ignored inside information from both companies, press releases on up-and-comming processors, review sites, etc.

They can reexamine AMD at a later date, but they obviously feel that Intel can offer them a compelling processors for very light laptops, for at least 4 years.

No, that's not obvious, actually, unless you jump to the conclusion that there is no other justification. It's been very well covered that Dell is sticking with Intel at their own peril largely because of the deep, deep discounts Intel is offering. AMD has an anti-trust lawsuit in the works for this very reason. So while this may possibly be the case, it's certainly not a foregone conclusion, and doesn't even seem the most likely explanation.

Re:No AMD macs?

[99BottlesOfBeerInMyF]

Not only are thes comparisons incredibly unfair, but they're also vastly unfair due to the timing. You're comparing Intel's nice new chip that was just recently available, to AMD's older chip, that has been on the market for a very (relatively) long time now.

This was in response to someone claiming Apple made the wrong choice for a chip, they are idiots, and they should be using some (unspecified) AMD chip. Apple obviously needed something on the market now, or very soon. AMD will not putting a new mobile chip out till Q4 according to most estimates. How is comparing what chips are available to put in their machine unfair? You think they should consider delaying going to market for a year rather than using the best available chip now and considering new chips as they become available?

Here's a great way to change my opinion. Find an AMD chip better suited for Apple's all-in-one and pro laptop and then show me a comparison of the performance and power to the Intel Duo they are using now. Otherwise, you're just an AMD fanatic who did not bother to read the context of this thread and whose knee is jerking faster than light. Do tell, what do you think is the best chip available for this application right now and why?

Re:No AMD macs?

[evilviper]

Do tell, what do you think is the best chip available for this application right now and why?

Obviously a mobile Opteron. Very good performance, high clock-rates, low latency, low power, and 64-bit.

Dual cores is slightly nice, but not worth the increased power requirements, particularly since very few applications can take advantage of both cores.

Re:No AMD macs? Excuse me!!

[Nom du Keyboard]

Your number of 3.5M units appears to be about a million low, though. 4.5 Million is what they shipped in 2005.

If you'll look, I was quoting 2004 shipment figures since I did not have 2005 accurate figures quickly at hand. 2004 is a reasonable benchmark because Apple did not suddenly catch fire in 2005 and increase their shipments of Macs many fold in the process. I have no trouble accepting your 2005 figures because they still make my point, so please don't criticize my 2004 figures since they are well within the ballpark.

Re:No AMD macs?

[99BottlesOfBeerInMyF]

Obviously a mobile Opteron. Very good performance, high clock-rates, low latency, low power, and 64-bit.

AMD does not list a mobile Opteron on their Web site. They provide Embedded, Workstation, and Server versions only. If you want me to take you seriously actually provide a real processor name/number preferably with a link to the specifications such as performance and power consumption and ideally with a comparison.

Dual cores is slightly nice, but not worth the increased power requirements, particularly since very few applications can take advantage of both cores.

This is for a mac, not a Windows machine. Dual cores don't provide double the processing but they do provide very significant increases in power and given the number of threads running on an average machine will be used the majority of the time a machine is being used.

The "best" mobile AMD chip I see is the Opteron Mobile 4000 which seems to have less performance and higher power requirements than an Intel Duo based upon indirect comparisons. It is a 130nm fab, accounting for the larger power requirements versus Intel's 65nm chip. Do you have any reason at all to think that this chip would be a better choice or are you just trying to find facts to justify the opinion you hold?

Re:Why?

[n6mod]

RX-7?

Re:Why?

[tgd]

Classic 911 with a race-spec engine.

Thankfully emissions exempt!

Re:No AMD macs?

[man_of_mr_e]

This entire story was about Intel Mac's, and I was responding to someone that was asking why Apple didn't just "leapfrog" Intel and go AMD. My point was, that Yonah currently beats any Turion out there, so it wouldn't be a 'leapfrog", and we're talking about MacOS X on Yonah, not Windows.

Re:No AMD macs?

[justsomebody]

and I was responding to someone that was asking why Apple didn't just "leapfrog" Intel and go AMD. My point was, that Yonah currently beats any Turion out there, so it wouldn't be a 'leapfrog", and we're talking about MacOS X on Yonah, not Windows.

Woooow, my bad. I must've overlooked that post. Either was rated to low, i missed it or it got crossreferenced with the wrong post in another tab by accident.

If that was the question, you're right as much as 120%. I was relating it to other post, zealoting how Intel now rules and Apple Duo CPU is the best Duo ever, iMac rules and beats any PC CPU for the next decade.

Now to add on the real question:
Second reason (why Intel and not AMD;) is that Intel is a win-win situation, AMD is a win-loose. In case Apple does not succed with computers because of that move, they know for sure Intel (which looses ground against AMD and this would be a very good PR stunt) would buy them. AMD wouldn't do that, not enough money and no interests.

Again, my sincere apologies on missing the parent post. Mea culpa:)

On the other hand I still think benchmarks you posted are valid for Windows and OSX users only as I said (if not else whole SW is 32-bit). In fact in my first post I was mentioning 64-bit environments only and how some of us aren't forced to use 32, which mekes benchmarks you posted rigged in Intels favour for us. But maybe it would help if you would point out the parent post sooner, since I was completely OT.

Re:No AMD macs?

[99BottlesOfBeerInMyF]

The "best" mobile AMD chip I see is the Opteron Mobile 4000...

Whoops, I meant Athalon Mobile 4000. Pardon the typo.

If I may be so bold...

[Jerry Smith]

AMD has been the performance/MHz performance/watt leader for more than a year, and they didn't need to make 3 Ghz chip to do that (right now I think that AMD has not even released a 3 Ghz chip, just a 2.8 one) In other words, CPus didn't need to have more than 3 Ghz to remain competitive, only the crappy presscot Intel chips, and Jobs (who has spent years telling people that MHz are not important) knows this very well. He knew that G5 didn't need to go beyond 3 ghz to perform well

Problem with IBM, as well as Motorola (or the daughtercompany that took care of the laptopchips) and AMD (and probably Sun as well) is that they could not provide Apple with enough and fast processors. I might be a little tired right now but as respectable as above mentioned businesses are, Intel is a BIG player, capable of providing Apple with everything they need at the moment, times two. I would've loved to see Apple go with Sun, go Sparc, but that would just not be wise.

Re:No AMD macs? Excuse me!!

[MojoStan]

Bullshit, bullshit, and bullshit...

As I have to keep saying: Bullshit factor too high!

Settle down, Beavis. I'm not disagreeing with your point (I think) that AMD makes competitive CPUs for "thin and light" notebooks. However, I still think your comparison (which you call "as fair as possible") is inaccurate, incomplete, and (therefore) unfair.

You didn't include any links to where you got your numbers,

A minor oversight: http://users.erols.com/chare/elec.htm

That's a nice source. I just bookmarked it. It doesn't list all of the newest CPUs, though.

overstated the Core Solo's power,

Not according to my source, which I trust to provide accurate numbers. Considering that Intel uses a different power rating standard than AMD, it's not fair to compare AMD's spec sheet with Intel's spec sheet. Intel was the one to switch to "Thermal Design Power" when AMD was still listing "Max Power Output", and now that AMD has switched to "Thermal Design Power" Intel changes once again, seemingly trying to make sure direct comparisons are never fair.

Your source apparently made a typo for Core Solo. It's an easy typo to make (37 instead of 27). For the other Intel CPUs in our comparisons, your source's "Thermal Design Power" numbers match Intel's "Thermal Guideline" numbers. I'm sure "Thermal Guideline" has the same meaning as "Thermal Design Power." I don't know why the heck Intel would change the term if it means the same thing. Maybe they think the work "guideline" is clearer than "design power." BTW, "Thermal Design Power" is defined in the "glossary of terms" for Intel's Spec Finder tool as:

The maximum amount of heat which a thermal solution must be able to dissipate from the processor so that the processor will operate under normal operating conditions.

Unfortunately, "Thermal Guideline" is not defined in the glossary (ugh). However, that definition sounds like a "guideline" to me. Your source seems to agree.

and didn't include the Celeron M

What the hell do you think "mCel-2.5G 35W" means?

(Hell) "mCel-2.5G 35W" means Mobile Celeron, a discontinued line of low-cost mobile CPUs based on desktop CPU cores (Pentium III and 4). These CPUs sucked ass and were inappropriate for "thin and light" notebooks. The model you listed is based on the Northwood P4 core, but was "crippled" by halving the L2 cache (down to 256K) and lowering the bus to 400MHz.

"Celeron M" means Celeron M, the current line of low-cost mobile CPUs based on the Pentium M core. These CPUs kick ass and are hardly "crippled" by its halved cache (1MB) and reduced bus (400MHz). Even without SpeedStep, the Celeron M is a high-performing, low-power, and cool CPU that's good for "thin and light" notebooks.

You also neglected to include the low power versions of the Intel CPUs.

That's because the "low power versions" are so vastly slower that it would be completely unfair to throw them in this comparison... Obviously, being incredibly unfair doesn't phase you.

I included LV versions and "lower power" versions of Pentium M and Celeron M for two reasons: (1) AMD apparently doesn't make true "low voltage" (LV) versions, and (2) you included a "lower power" version of T

Re:Why?

[tgd]

You haven't been in very many race cars then. Most sanctioning bodies explicitly disallow them, as they can be dangerous in a crash if the unclipped buckle hits you.

And where harnesses are concerned, as far as I know there is only one >3 pt harness that is legal to operate on the public roads and FIA-legal in the US, because a DOT legal restraint must have a push-button release and not a camlock and FIA hasn't approved any other 3" harnesses with push button releases.

this stinks

[a984]

really. They claim significant improvements in GUI responsiveness and admit to not knowing how much MAc users were suffering before which is kind of stupid because Macs were awfully slow for anybody even half objective. I don't think this has much to go with GHz, Apple tends to overabstract their system software which has tremendous consequences performance wise. Take their text handling framework, nicely abstracted, robust but slow like molasses. Same with everything else. Unless they start hand tuning code on system level with view toward efficiency not abstraction don't expect any realistic CPU to be able to handle their stuff competently.

Re:Why?

[Jearil]

But they're not really obligated to support unsupported software such as windows.

I mean if you buy a Dell machine, will you expect them to support installing an intel version of OS X on it? I mean I have this great cheap fast Dell machine that can play games wonderfully, but damnit, Dell won't support iLife or the Pro apps.

That's the same sort of argument you could give to apple not supporting windows software on their apple hardware. They're different platforms, both in OS and actual hardware. Granted they both use x86 chips now, but the processor is not the only piece of a computer. As we see from the (currently) failed "Windows on Mac" experiments, the bios also plays an important role.

Once you buy the hardware, you can do as you like. It's your hardware, so install what OS you can figure out to install. But Apple shouldn't be under any requirement to help you out in your experimental ventures to install things never meant to be installed. Gateway won't help you with Linux, Sun isn't going to work with you to get Windows going on their Solaris machine.. It seems a bit silly to expect such things from Apple.

Disgusting

[pele_smk]

It's disgusting to see all of the pro Apple people not being subjective here. Back in the "day" all Apple people would say was how great the video was on his/her computer opposed to that of a pc. Not long after, quietly under the table nobody seems to be willing to attack Apple. So, now we know Apple laptop screens BLOW (just take a look at a Sony and you'll see who really owns the laptop screen.), Apple isn't the fastest computer, so no more of the b/s line about "oh, it's not an Intel based processor so an Apple running at 500mhz is actually equal to your 1ghz" comments that flooded my mind for the last ten years. And last but not least Apple doesn't have the best sound card for my home recording studio. It's all mediocre crap stuffed into a pretty case running Unix.

The real genius behind Apple is Unix. Will a pc manufacture hurry and come out with laptops that look the part as well as perform the part? Enough of these $500 64bit power desktop replacements. Lets think about the toys to go along with the laptop. Give me a magnetic power supply detachment, a clean and handy wall adapter, a thinner notebook, better headphones, and I guess that's it. If electronic manufactures put more time into appearance, we'd have a world full of Apples.

Yes, I blame Steve Jobs(not gonna give him the uber cool nick SJ, because I don't know him, we're not boys, thugs, hommies, or whatever we call people these days) for over rating his new Intel based products. We would all rape Nvidia, ATI, Intel, and AMD if they told us the wrong specs. If Apple is going to take five years before the processor reaches its advertised speed, don't be a salesman. Be an honest person and tell me the current speed. You're not working at a frickin' car dealership. We're not buying your computer for CPU speed, we're buying it for Unix and its sleek appearance.

But in the end, come on, who realy believed the advertised speed? And who didn't just say, "I want one" before we had any idea about the CPU speed?

Re:Why?

[ClamIAm]

I like how you're trying to argue all the angles. It would be amusing, if it weren't so hard to follow.

Oh, I see

[Dekator]

That explains it. Well, I guess I'm a sucker for keynotes and yes, sometimes, you get some interesting detail out of them. I mean, watching them yourself. If you haven't watched it yourself, you wouldn't have heard Steve note it was done on an iMac. So, excuse my sharp reply, I was just astonished.

And BTW...

[Dekator]

Could somebody please explain the 'scoring' system here? Why is johnpaul getting 2 points for stating that he didn't actually watch the keynote, while I get only a score 1 for actually providing clarifying information ???

Re:And BTW...

[johnpaul191]

that 2 has nothing to do with the quality of this specific post.
it is because i have posted enough stuff over time that got moderated positively so anything posted starts at a 2. i guess it goes back to having an account long enough to prove you are not a troll or spammer? notice it just said score: 2 and not something like "insightful" or "funny" after it. the moderation system has been tweaked in the time i have had an account, and i am sure is well documented somewhere on here.

I'll go on record...

[BiOFH]

I'm a staunch Mac evangelist. I totally marched for the whole G5 performance parade.
But when Jobs said twice as fast, etc. I knew it was bullshit.
It's spin and it's no better than the crap we've get from Intel or MS in the past.

I'm just relieved that it didn't turn out that the G5 was worse than I'd believed.
I think Jobs needs to come clean on this one.