Intel Developer Macs Outperform G5s

bonch writes "Developers working with the new Intel-based, developer-only Macs are impressed with the performance. The machines take as little as 10 seconds to boot from Apple logo to desktop, and apparently run Windows XP at 'blazing speeds.' Rosetta tests demonstrate the PowerPC-native build of Firefox running just as fast as it does on a high-end G5."

Comments

Boot times disk/network bound

[Toby The Economist]

OS boot times are usually disk and network bound.

I don't see how even an order of magnitude increase in CPU power could shorten boot times to the extent described here.

There must be other factors.

--
Toby

Re:Boot times disk/network bound

[antrik]

> OS boot times are usually disk and network bound.

While disk plays *some* role in OS startup, it's usually far from being the decisive factor. In a typical setup, a much larger amount of time is consumed on CPU use; and quite a large amount on various kinds of timeouts, related to networking, but not only -- various kinds of hardware probing etc. are the main reason why OS bootup doesn't even remotely scale with CPU and disk speed improvements.

CPU *does* make a considerable difference, but not an enormous one -- the other hardware in the box (which is also different for Intel Macs) might be quite relevant, too.

Re:Boot times disk/network bound

[hazee]

I've never understood why it takes so long to boot a computer. It takes far longer than the time necessary to transfer the actual amount of data that ends up in memory after the boot sequence.

Why not use something more like a resume from hibernation, where you just restore the contents of memory directly from the disc in one go and be done with it?

Actually using hibernation rather than booting is no good, becuase it only restores you to the state that your computer was in last time you used it, which might not be "clean". For example, if you had been running something with a memory leak, hibernation won't fix that.

But the solution is simple - instead of writing the hibernate data to disc just before you shut down, instead store it just after you've finished booting, so that you've got a "clean" system ready.

This way you get a "clean" system every time at the speed of a restore from hibernation. (And if something goes wrong, you still have the option to do a "full" boot.)

Some might say that you need to go through all the slow processing of a full boot in case anything's changed. Really? Restoring from hibernation seems to cope with that possibility.

More likely, most times, nothing will have changed. And for the times when it has, well, you do the extra configuration necessary after the restore - you're still no worse off.

So why are we still forced to sit through full boot cycles?

Re:Boot times disk/network bound

[imsabbel]

Your logic has a flaw:
A resume form hibernation has to read ALL of your memory from disc, as only saving "used" memory frames would just invide bugs from hell. (think about memory windows from drivers, or what is "clean state" ? which autostarts/services,ect)
And with 1GB, even on a very fast HD it would need 20 seconds... Not faster

Also consider that bootup is usually the time to detect new hardware.

In fact, im quite happy with the 20-30 seconds i get with windows xp.

Re:Boot times disk/network bound

[cbreaker]

That's a good point, too. Not sure if that's a good or bad point for the Mac, though =)

I'm guessing most hardware drivers aren't written by Apple, but the hardware developer (Apple contracts out hardware design like any other company) and some drivers like video drivers are certianly not Apple.

But, Apple can sure QA their drivers a lot more then Microsoft can, so your point does remain.

Re:Boot times disk/network bound

[Skippy_kangaroo]

You might want to clarify that statement - because as it stands it's kind of meaningless. It's almost akin to saying I can do nothing at the speed of light. In fact, I did nothing 50,000 times between that last sentence and this sentence, that is blazingly fast.

At a slightly less ridiculous level, whose fault is it that just installing (but not using?!) software makes the OS performance molasses like? Not being a Windows guy I don't get it. Does this affect OS X as well? I load and open all sorts of software, have multiple users logged in and, unless the process is actually doing something, I don't notice it in OS X - am I missing something?

Re:Boot times disk/network bound

[mabinogi]

that's not true at all.

A PPP connection via a modem is a direct connection to the Internet.
You are _extremely_ vulnerable if your connection is directly through a modem.

I found that out myself the hard way a year or two ago - at home I'm behind NAT, and there's at least two levels of NAT and three firewalls between me and the Internet at work, and I don't use I.E.. So I never really paid much attention to patches and updates, or even anti-virus stuff.

However, one night when I was interstate for work, I dialled connected to the Internet by modem from the hotel room. I had three viruses by the end of the hour.
I'd become so used to always having firewalls in front of me that it never occurred to me to even worry about it.

Now I make sure I patch regularly and that my anti virus software is up to date before exposing myself like ;)

Re:Boot times disk/network bound

[batkiwi]

None of those except norton should effect boot time. Norton, however, is a PoS and should be replaced by something that doesn't try to take over the system.

Re:Boot times disk/network bound

[NardofDoom]

I've been spoiled by my Powerbook. It's usually fully awake before I even get the screen the whole way up. My mother-in-law bought an XP laptop and it takes about five minutes to get back to where it was when you closed the screen.

Even so, I'd love for Apple to create a 'hibernate' feature.

Re:Boot times disk/network bound

[heri0n]

What good is a "blazing fast" system if it has no usuable apps? OS X is blazing fast and it comes with usuable apps.

Re:Dual Boot

[uncle_fausty]

For God's sake, will you please stop beating this issue to death? No, MacOS will not officially run on non-Apple hardware. Yes, l337 h4x0rs will probably find a way to make it happen. No, it will not be the rosy seamless computing experience MacOS provides on controlled hardware. Apple's success in OS development is in no small amount tied to their control of the hardware it runs on; don't expect that to go away anytime soon.

Re:So the G5 were dog slow after all

[raider_red]

It could also mean that firefox's apparent speed is based more on network throughput than code execution. I'd like to see some more robust benchmarks than "it seems just as fast".

We can start by how long it takes to crunch a lot of floating point operations and integer math operations.

64bit and vector code

[Anonymous Coward]

So simple stuff runs as fast or faster...

How does the Mactel box do on floating point, 64bit and/or vector based code? The main reason for getting a G5 was to improve performance of 64bit/floating/vector code like is used in video production and scientific apps.

Since Intel has always been shaky in floating point and probably doesn't really know the meaning of vector I'm wondering how those kinds of apps will fare on the Mactel boxes.

Re:64bit and vector code

[jiushao]

Actually you have it a bit backwards. The P4 is rather back to the roots for Apple, an excellent SIMD performer with a bit so-so performance on linear floating point. It was the initial issue with the P4, a weak x87 implementation with a great focus instead on the excellent performance of the SSE2 (and SSE3 with the prescott).

Compare this to the G4, another weak linear performer that Apple more or less specialized in getting to fly through good use of the excellent Altivec unit. The G5 on the other hand has a somewhat weak Altivec unit but a much beefed up set of single-element FPU units, yielding so-so vector performing but good linear performance. IBM did probably not focus much energy on the Altivec unit but instead threw it in since Apple required it (after all, the single-element FPU performance of the G5 almost puts the Altivec unit to shame).

Some might now be quick to point out that Altivec is a nicer instruction set than SSE2/SSE3, this is by most standards true, but if you are hand-coding assembly you can make do with either. On the flip side Intel has quite impressive auto-vectorization support in their compiler.

So, what does this add up to? The G5 is in a good place for beating the P4 on unoptimized unvectorized code, but the P4 really screams if things are tuned up a bit. Considering Apples history with Altivec I think we can safely assume that they won't be afraid of doing some hand-tuning to get good perfomance.

This all ends up looking quite favorable for the P4, I still don't think we will see a commercial Mac with a P4 derivative in it, but anyone who thinks the P4 is a weak performer has another thing coming. For a bit more on my opinion on the state of the x86 vs. PPC today see my earlier post in the "Apple Switch to Intel Not a Big Loss for IBM" story.

Re:The real question

[chizu]

I don't see how this is a concern. My IBM T41P gets 6-8 hours of battery life with wireless and sound in frequant usage. It's not a slow machine either, 1.7GHz Pentium M, ATI FireGL graphics card. If IBM can do it, I'm sure Apple will have no problem producing a similar laptop.

Re:Dual Boot

[keytoe]

the final product will be tied to an special Intel DRM chip that will prevent it from running on other machines.

I keep seeing this spouted around as fact, but I have yet to see anywhere where this was stated in any official capacity. They may do that. They may not - they haven't said one way or another. Outside of the iTMS, Apple hasn't done much with DRM, and I'm pretty sure that was only to make the labels happy - so I'd say it's still pretty much up in the air.

So stop saying it like it's a fact, please.

Crap article. Period.

[javaxman]

This article is nothing but a vague reference to a couple of un-substantiated percentages of Rosetta performance and some completely subjective, unverified statements.

If I were AppleInsider, I'd be ashamed to print this. Of course, it's not likely that AppleInsider could be ashamed of anything, so there you go ;-)

Really, wake me up when there's an article where someone publishes comparative numbers of PostgeSQL inserts or NSImage composites or timed renders of Safari web pages.

And no, I'm not really interested in Rosetta performance as much as I'm interested in native app performance. I'm interested, don't get me wrong. Just not as interested.

Re:Dual Boot

[drsmithy]

Steve Jobs also said that Apple had no plans to compete in the $500 computer market and then later Apple released the Mac mini.

But by the time they did that, the $500 market had become the $250 market...

Re:So the G5 were dog slow after all

[DurendalMac]

Horseshit. They're running this stuff under Rosetta (oooh, a web browser is snappy on a 3.6ghz processor!), which does NOT run Altivec code. Altivec-enabled apps on a G5 will undoubtedly stomp a native Intel app on one of these developer rigs. That, and this is Appleinsider, so take it with a grain of salt. These guys are way off the mark so often that it isn't even funny.

Re:Impressed

[gabebear]

I am impressed with a technology behind Rosetta. Are ther any open source projects like that?

YES, exactly like Rosetta

QEMU lets you run Linux apps compiled for another CPU architechure. So you can run PowerPC Linux apps on your X86 Linux box. QEMU is nowhere near as seemless or fast as Rosetta, but QEMU does run in two modes: user-emulation where it works like Rosetta, and system-emulation where you can run another whole OS like VMWware.

Re:WinXP on Mac a fluke, Mac OS X Apple H.W only

[R3d M3rcury]

From what I understand from Phil Schiller at Apple, there are no plans to not allow Windows to boot. That said, it doesn't mean that Apple is going to force compatibility (eg, "Oh, we can't use such-and-such technology because then Macs wouldn't be able to run Windows").

For example, from what I understand, the chances are essentially nil that Apple will use whatever PCs use to boot--whatever the PC equivalent to Open Firmware is. Now Apple has pretty much said that they're not going to use Open Firmware on the Intel Macs, but there is some Intel Four-Letter-Acronym that is a leading candidate to be used instead. If that boots Windows, huzzah! If it doesn't, though, I doubt Apple would say, "Gosh. We can't use that because then our users won't be able to boot Windows."

Also, if Apple uses some really cool custom-built chip for doing audio, they will probably not write Windows drivers so that it will work under Windows. Somebody else might--hey, that's cool--and someone could download and install those drivers and everything would work fine. Heck, maybe Microsoft will even include them.

Apple isn't going to go out of their way to support Windows, I agree. But I'm sure some people will figure out a way to run Windows on one.

Re:What about a real computer?

[illumin8]

Intel 3 GHz P4.. kinda in the midrange of PC hardware. I want to know how OS X86 would run on my new home system .. Athlon X2 4400+ SLI mobo.

It won't. Apple will never allow OSX X86 to run on a non-Apple system. Expect to see on-chip Intel DRM enforcing this.

Re:PowerPC 25-30% faster than Intel x86

[NutscrapeSucks]

Well that's the key right there .... The G5 had decent FPU and Altivec of course, but the Integer ("mainstream") performance was never competitive.

Re:Talk about a 180...

[ArbitraryConstant]

"Keep it in context: this is an anecdotal comment from a single person comparing a bleeding edge 2005 pre-release Mac with their old 2003 retail Mac."

You mean if we have objective tests we can admit the Intel chips are faster?

Sweet.

Re:Wrong meaning of "Blazing"

[Anonymous Coward]

'top' takes ~12% cpu, according to itself, on one of my linux boxes....

of course, its 1996 binary running on a ppro 200 and 2.0.x kernel...

Seriously, its not a big deal. these are not applications that need serious optimizing. I'd much rather the developers of such spend their efforts on consistient results reporting than making the app more efficient: for example; in a 4 cpu machine, what does "25%" cpu usage mean actually? one cpu buried or all 4 just idling? NFI how windows does that math, linux; it depends on which versions of (kernel/procps/top/etc) you're using.

With NUMA systems becoming more prevalent, thats getting important, too.

In other words, go ahead and buy the Mac

[SuperKendall]

If you want a Mac, why wait for several moths to a year or more?

Instead just buy whatever Mac you like now and enjoy it - after all you'll have to spend some time learning a new OS anyway. If the newer Intel macs are really a lot more powerful - then sell the current Mac, which is easy to do since used Macs hold value well. And yes PPC macs will hold value just as well as new softwrae will still be comiled for them for several years anyway.

If you want to maximize resale value consider an Apple laptop of some sort, even really old ones fetch quite a lot.

Re:In other words, go ahead and buy the Mac

[argent]

used Macs hold value well

Over the past month, more or less, I've seen used Powermac G4 prices drop significantly. That normally only happens to a model line when Apple releases a version of OS X that doesn't support them (like Beige G3s after Panther, pre-Firewire iMacs after Tiger). I've also seen G5s for sale on the Low End Mac swap list for the first time ever.

This time... Leopard may drop support for some or all G3s and maybe even the Yikes G4, but I don't anticipate AGP G4s being in trouble from Leopard, and in any case it's a year and a half off.

The reason for the lower prices? When it's explained, it's the expected Intel macs. If they're depressing used prices already, a year off, I wouldn't expect your PPC Macs to hold their value the way they have in the past.

Re:I repair and support Macs for a living

[Zobeid]

Who told you "increased performance . . . is what the whole Intel switch is about"?

Oh yeah. Steve Jobs said that, I forgot. :)

The switch from 680x0 to PPC was about increased performance. The switch from Mac OS to Mac OS X was about increased performance. The switch from PPC to X86 is not, it's just a business decision. It's not necessarily a bad business decision, but it's not something Apple's engineers dreamed up as a great way of moving their platform into the future. It's not something the customers were crying out for. It's old fashioned deal-making.

I think it's an understandable move, and one that's likely to pay off in the long haul. But. . . I can't help feeling disappointed that every OS seems destined to someday grow up and become Unix running on a X86 instruction set. It's not the future we all hoped for.

Re:What to do with those slug stickers...?

[evilviper]

so I need some creative ideas as to what I should do with them.

I suggest walking into an Apple store with them, and sticking them on all the new Intel-based Macs.

That should make them a good reminder as to how much Apple has been lying to everyone over the years. Everyone should remember the PPC ads when their next generation of advertising comes out.

Re:Mod down, bs

[Politas]

It doesn't matter whether it's generic hardware that Apple uses, or special proprietry hardware. The important point is that it is a fixed (or at least limited) platform. Apple know exactly which chipset/processor/gpu combinations they need to support, so they can configure the OS to use exactly the features available and avoid the problems. or from another perspective, they choose their hardware to avoid those bottlenecks that are most likely to be noticeable.

Re:Generic PC Hardware and OSX

[Viewsonic]

I dont think your quite understand Apples business model. They make their money off the hardware they sell, not the software. OSX is a great OS, which may sell quite a bit on "generic" hardware, but not nearly enough to compete with selling $3k dual processor machines that only cost them peanuts to make.

You will see Apple selling OSX to "generic" hardware the day Dell creates and licenses their own OS to make more profits. If you think that will happen, you're nuts.