Slashdot Mirror
PC World: Apple G5 Gets Trounced By Athlon 64
StewedSquirrel writes "PC World magazine has published an article comparing the AMD Athlon 64 and Opteron versus Apple's G5 processor, both 64-bit contenders for the title of 'fastest desktop processor.' Apple has made many claims to be the first, fastest and only 64-bit processor for the desktop and workstation market, but (not mentioning the fact that Opteron beat the G5 to market by over 4 months) the benchmarks should speak for themselves. Of note is the 3.2GHz Pentium 4, coming in competitive with the G5, but significantly behind the Opteron and Athlon 64 systems."
we would like to thank AMD for their contributions to this story and our bank accounts
http://www.thebesttrek.net/forum/index.php - visit my FORUM
last time i checked the operon was to be the server class amd64 cpu, where as the athlon64 was to be the desktop version.
if you're going to compare workstation class chips, compare the freaking workstation class chips...
From the article:
"But upgrading to XP 64 could mean giving up functionality without getting much in return. In fact, XP 64 looks like a throwback to Windows past: Its interface mirrors that of Windows 2000 or even Win 98. Microsoft has not disclosed what else will be in the OS, so it is possible that you'll still get most of XP's other features.
XP 64 won't have the 32-bit XP's support for DOS apps at all, nor will it run 16-bit apps (but it should have no trouble with 32-bit software). More important, 64-bit drivers for common hardware, such as printers, will be scarce when the OS debuts."
In moving from a Dual 1GHz G4 (Quicksilver 2002) to a Dual 2GHz G5, I have yet to find any software incompatibilities - everything works just fine.
This may change once my copy of Panther shows up, but my printer and other hardware continue to work for now.
Specialization is for insects. - R.A.H.
What did the Intel exec. say? "64 bit is not ready for the desktop" ?
Haa haa
Comment removed based on user account deletion
I've never seen Apple claim to have the first 64 bit workstation. First 64 bit desktop maybe but not workstation. No idea how they distinguish the two, but please don't go putting words into Apple's mouth.
It seems to me that none of these aps were specifically compiled and optimised for either of the processors. While this coresponds to the real world (sometimes), it leaves me wondering how the numbers change in ideal conditions.
It might still get owned, but redoing the tests with the OS that the G5 was meant to run on will be a better comparison. What can it hurt, it's only 9 days away from release.
Single disk versus RAID array...how much RAM etc etc (thanks for listing the gfx ram, that was great[end of sarcasm])...
Thats like comparing Apples and oranges (pun intended)
Next Week: the National Grid versus two half sucked lemons with electrodes (in parallel folks!!!)
G5, Athlon64...any way you go it's an alternative to Intel. I think the importance isn't which is quicker but that they both offer serious alternative solutions to Intel which forces everyone to innovate. Both companies deserve credit for working toward better solutions for customers.
Of course we torture people, we need the information --Gen. Pinochet
Just as long as it's not Intel, I'm still quite happy.
Luck favors the prepared, darling.
It's still the fastest desktop processor, because there is no desktop OS that runs on the Opteron until Microsoft releases the XP version in 2004. And no, Linux is not a desktop OS - ie something regular people can use (yet).
I don't know why Apple shoots them selves in the foot with this speed BS anyway. Seriously I like my iBook for many reasons, but speed isn't one of them (because it's slow - although seems as fast as many PC laptops for some reason), but I'm willing to put up with a little drag to have a cool computer. They just aren't going to win the speed race and they need to realize that. They need to focus on the value of the overall computer where Apple is indeed ahead in many respects.
After dealing with various versions of Unix, Linux, Windows, and Apple boxen over the last 15 years, whenever a novice asks me what kind of box to get I tell 'em: "Get a Mac. The up-front costs are a bit higher, but it's stable, it doesn't have massive security problems, the hardware you can buy for it works pretty much right out of the box, and it's easy to use".
And besides, what novice needs a 64-bit Athlon at home? Because without Winbloat a 16MHz 386sx is probably fast enough for a Mac to read emails and surf AOL via a 28.8 dialup connection...
...you'll wake the Mac zealots!
Get paid to code OSS
Now let's see how long it takes the software community
to catch up. Not long, I'd wager, before we're seeing
ridiculously torqued up system requirements for the
next version of Office. "A terrabyte of RAM? let's
render clippy on the fly in 12 dimensions and map
out all the quantum placeholders too!"
Mod me down and I will become more powerful than you can possibly imagine...
NetInfo connection failed for server 127.0.0.1/local
So they compare 32-bit apps running on a 64-bit AMD chip to 32-bit apps running on a 64-bit G5 and conclude that the AMD chip is much faster than the G5.
This does nothing to benchmark the capabilities of the chips -- just the capability of the chips to run non-native apps.
Go back to your lives, citizens, nothing to see here...
Life is short: void the warranty.
Funny this, it seems that Apple always provides by far the fastest hardware to run OS X. Amazing.
I was surprised that a single CPU Opteron could be in the same performance ballpark as a dual CPU G5. Does the Opteron do more per clock cycle than the G5? Are applications not taking advantage of the second processor? Is there some other performance bottleneck, such as the memory subsystem? I look forward to finding out...
Enable 3D printed prosthetics!
SUN Microsystems did the same thing.
After their $40k servers were proven to run slower than a $3k AMD boxen, they started pretending to be OS vendors and went on saying that Solaris is what's *really cool* about sparcs.
That's a statement on the road to dying...
Exactly, that puts the Athlon even further ahead!
Since everybody is talking about how Apple should have moved to a x86 processor, what if Apple finally makes this decision, and uses AMD 64?
I mean, it would make sense: Darwin could (is) be compatible with x86 (all is needed is a true 64 OS from Apple, which does not exist for the G5 - yet), all you would need is to recompile the higher layers: Acqua, QuickTime (which is AFAIK much simpler than porting lower parts of the OS) and Software (easy too)
After all, G5 also uses Hypertransport...
how long until
Er... "and for the real mac apps" ofcourse :)
I thought everything that we've heard said that the G5 was good in parallel and was generally mediocre for integer performance...
> OS X
OSX only runs on the slow hardware. Try again.
_You_ shouldn't.
Great!? This seems more like 8 pages of AMD advertising than anything else.
Not even close!?! More like: Not even Scientific!
-B
Chicks dig 'em.
"If you're thinking what I'm thinking, you're right." -
"Apple has made many claims to be the first, fastest and only 64-bit processor for the desktop and workstation market"
No, it never has.
It has made claims that it has the first desktop with a 64 bit processor, and that it's has the fastest desktop out there.
Since when has a AMD or even Intel put out a desktop? Dell and HP do, I know that. But AMD and Intel don't.
Jason Lotito
And why should I even want OS X?
"Fighting terrorists with millitary might is like killing a mosquitor on your Dad's forehead with a rifle."
I'll take a slight speed hit (oh no! only 294 frames per second!) if it means not having to use an OS which finds a way to annoy me every 20 seconds (Windows), or an OS straight out of 1997 (Linux, etc). OS X is a revolution in usability.
"Our test suite, PC WorldBench 4, cannot run on Macs. The new Macs aren't great values either, as the top-of-the-line G5 ($3549 as configured) costs about $200 more than the similarly configured Alienware Aurora.
The dual-G5 sparkled in one main area: our Photoshop test, which it completed in 18 seconds, or about 17 percent faster than the Aurora's 21 seconds. The 1.8-GHz single-chip G5 ($2999) trailed at 27 seconds.
Elsewhere, the Alienware earned top marks, performing particularly well in the Premiere QuickTime test."
======
>>$200.00 is nothing and no direct testing comparision is funny.... This is pure marketing hype.
I think the jury's still out. We haven't seen OS's or applications optimized for either platform. However, both systems are still pretty damn fast. I think it's going to come down to what you like best. Personally, I like OSX better than Windows or Linux on the desktop. OSX gives me all the power and stability of Linux, and it's easier to use and prettier than Windows, and it runs Photoshop. I'm a photographer, so that's pretty important to me. I still run Linux on my servers, though...those Mac servers are ridiculously expensive.
We don't have a state-run media we have a media-run state.
I quote;
"The Power Mac G5 is the world's fastest personal computer and the first with a 64-bit processor"
They are not claiming the "the first, fastest and only 64-bit processor for the desktop and workstation market" as StewedSquirrel barfs.
The key word in Apple's claim is *personal* and looking at the prices at the time of that claim to 64-bit systems I think you'd be hard pressed to find a single 64-bit CPU solution in the same price bracket let alone dual processor.
Learn to read.
File this one under "duh"
And as for the dept:
from the os-x-doesn't-run-so-hot-on-athlon dept.
should be:
from the os-x-doesn't-run-on-athlon-at-all-dept.
#!/
If Apple were ever to release OS X for x86, Billy G. would call up Apple: "Bad little lap doggy! No more Office for Mac!"
Hmm...G5 systems with standard ATA hard drives and 128mb of graphics memory being compared to systems with RAID and 256mb of graphics memory? Those quicktime renders and search/replace operations are heavy on the disk access, I wonder why the G5 might possibly lag behind....
If you prefer a more reputable source that actually bothered to configure their systems similarly: PC Magazine
The new Macs are fast enough to even inspire debate, which is a shocking development; if you want to buy a Mac, you no longer have to be ashamed of throwing money away. As for platform bashing...sheesh, aren't people on this site always begging for a reasonable alternative to Wintel? You should all be cheering for Apple....
Because you've got more money than brains.
You mean like the different flavors of Solaris, Linux, and BSD I have running on the machines here in my lab?
Unfortunately, convincing "management" to let me run anything but Windows on a "company" machine is an exercise in futility.
Specialization is for insects. - R.A.H.
that the software choices except for Photoshop are not designed we for OS X (premier runs in Classic for christ sake)
these bench marks are useless since they used software that runs like crap on Macs anyway.
I am the Alpha and the Omega-3
Plus going with a mac will take care of that annoying "money" problem. no doubt you'll want to get an iPod and other sleek accessories to go with your powerbook. You'll be money-free in no time!
Conserve Oil, Recycle, Boycott Walmart
where they cannot even format an html table?
BTW here's the full story without clicking through 8 pages.
Learn to format PC Mag (aka pull your head out).
The KEY is: does either the Athlon or Pentium run OS X?
If that is the key, then why isn't THAT what Apple is trumpeting when they make performance claims ? If that is the key, why isn't THAT what Apple zealots have been trumpeting recently when talking about G5 performance ?
I used to work for people like that so I left and got a better paid job with someone else.
Stick Men
Someone should point out that the Mac was running on 2 CPUS not 1 like the Athlon 64 and Athlon FX. Even so, both CPUs still trounced it in most of the tests. Does it matter if AthlonFX is different from Athlon 64? Both are desktop CPUs.
My 286 9Mhz CPU is also a GREAT CPU. Its good for my musium of computer history. Unfortunately it is not GREAT where speed is concern. So lets stick to speed when the article is talking about speed alright!
"Fighting terrorists with millitary might is like killing a mosquitor on your Dad's forehead with a rifle."
And why should I even want OS X?
You shouldn't. It's not for everyone. Stay on your side of the railroad tracks, thanks.
I'm personally not too concerned that my G5 is not the fastest desktop in the world. The speed of my processor defines who I am as much as my car does. Mac OS X is just really spiffy.
Well, you say they're "all great CPU's," but there seems to be quite a large margin between these machines in most of the tests (Photoshop excluded, heh). But it remains fairly obvious to the outside viewer that any of the Intel / AMD options will be faster than the comparable G5, and cheaper as well.
And from the standpoint, the only reason anyone would pick the Mac over the PC would be OS X.
Ehm. Because the link you are showing is a damn ripoff of apple site/look. same goes for the laptop there.
And as others pointed out: OSX.
Don't get me wrong. I work as a Linux developer. Haven't looked back into Windows camp in a long time, but OSX is just too nice to pass up. Run it at home for my primary OS, and let me tell you. Nothing, Windows or Linux can muster up will ever be as user friendly as OSX.
Have you seen their API's too? They are nice (here's the developer in me talking)...
Because on the desktop side of the story Linux is to OSX as SCO Unix is to... any other Unix out there.
Freedom is the freedom to say 2+2=4, everything else follows...
http://saveie6.com/
Oh hell. Did you look at those benchmarks? I hate this sort of performance comparison because it's really testing system and application performance. (e.g.: My apple is bigger than your orange, hence better.)
Besides the fact that this is an application test (e.g.: render times and frame rates), many of the included tests are not even on equivalent hardware (e.g.: did anyone else noteice that more than half of the Atlon "benchmarks" had twice as much RAM and a RAID system running where the Apple didn't?).
Give us a real test, or shove over, pcworld. Cruft like this is why I stopped reading Ziff-Davis years ago.
Did you look at the apps they compared the G5s and the Athlon64s with?
/. calls this a trouncing?
Word- It's Microsoft, no shit it's going to be faster on windows, who would have guessed that?
Premiere - The video app that sucks so hard on mac that Adobe stoped making it. Try the same functions on FCP and watch it come out a few times faster.
Quake 3 - A game, 'cause you know macs are what everyone uses for gaming, and developers spend just as much time optimising their mac versions.
Photoshop - The only relavant and fair app they bothered to test, and the G5 is noticablly faster than any of the Athlon 64 systems, beaten only by the Opteron.
And
"The worst tyrannies were the ones where a governance required its own logic on every embedded node." - Vernor Vinge
I shouldn't have to *lower* my standards so I can be fair to Apple.
Want design better than Apple's? liebermann
You're fucking kidding me, right? Every one of their
laptops but the 17" looks like a Dell re-hash and
the 17" looks like their designer stole Apple's
notes. The design of the desktops is top-notch... if
you're the kind that goes for mullets and muscle cars...
To each their own, I guess...
Mod me down and I will become more powerful than you can possibly imagine...
I cannot believe this is taken seriously on Slashdot.
Let's look into this more closely: the PCWorld team tested only four applications, one being Microsoft Word, FFS, and another being Premiere, which is no more supported on the Mac, runs in Classic and is leagues behind Final Cut Pro in terms of performance, as anyone with a clue in Mac video processing will tell you. This alone qualifies this comparison as biased in my book.
Where is the After Effects test ? And where is the Mathematica test ? Did you only know that any G5 will trounce an Athlon 64 in these apps ?
Also, looking at the results, I can hardly call it "trouncing the Mac". Only one in the four apps make use of the 2GHz' second CPU (Photoshop), and dutifully the G5 beats the PC in this test, and the scores in the other tests (not counting the Premiere's joke of an application) are not even that far apart.
Lies, damn lies, statistics, advertisements and benchmarks.
I was a diehard PC user until I got my ibook. I use it as my main computer because of the ease of use, the user friendly programs that come in the box. I can do all my video editing and such on programs that are easy yet powerful compare to the PC.
-joe
Aside from benchmarking Word for Mac against Word for Windows of all things, what does this actually prove? That Macs don't run software as well as Windows does when it comes to software that has been available for Windows longer? I'd be more interested in a price comparison between the systems.
No software-RAID setup on the Mac? Why RAID on the other machines?
Seems kind of one-sided.
Someone remind me why I should even consider a mac?
Superior user interface, user experience, and a complete package where all parts and software were designed to work together?
BSD underpinnings
Well-done GUI
MS Office, for those lovely proprietary file formats.
Next time I'm in the market, I'll be shopping hard for a MAC.
Get thee glass eyes, and, like a scurvy politician, seem to see things thou dost not.--King Lear
Most people (the majority of computer users) do not need 3+ GHz for a computer. If all you are running is Microsoft Word, the difference between 2.5 and 3 GHz is not going to be all that noticeable. Unless you run extremely high-end games or are heavily into video rendering or something extremely processor intensive, you won't normally notice it at all. Also, it didn't say whether there would be a fast bus speed. Unless there is, there's not much point in having an extremely fast processor, is there?
You Maccies wish OS X had that kind of power... If it was released for x86, no one would give a rat's ass. The software everyone uses runs under Windows. OS X is no more a threat to Windows than Linux is.
If it is this kind of bashing (a machine way outperforming an Apple machine), what is wrong with it? Healthy competition.
the point is that the G5 clearly doesn't win in terms of performance. you're asking whether or not the Athalon would run OS X... well... will the G5 run Windows? They're obviously runnnig two different operating systems. The interesting point here is not choice of OS, but performance of the hardware.
I think its because apple sucks. But dont look at me; I'm biased.
I've argued with benchmarkers over and over about this, Premiere is a lousy benchmark, used only by people who want to stack the deck against Macs. Premiere is highly optimized for PCs, and highly unoptimized against Macs. Fortunately that benchmark will go away soon since there won't BE any further Mac Premiere versions.
If you want to do a proper test, you'd use a crossplatform product that runs equally well on both platforms and is highly optimized for dual processors, like Discreet Cleaner or Combustion.
There's only one benchmark I can think of that is more worthless than Premiere, the "MSWord scroll test." For some stupid reason, some benchmarkers think it's a useful test to see how fast the can scroll to the end of a long Word document with the arrow key. Unfortunately, Word has a delay loop built into the scroll function, it even changes the delay loop depending on the speed of the CPU. The results are totally useless.
Macs == More Women
Someone remind me why I should even consider a mac?
Do you like expensive stylish-looking furniture?
I'm a chainsmokin' alcoholic sociopath, so-ci-o-path
I didn't see a detailed description of the Opteron system, but I suspect that it had two processors, as it beat the Athlon 64 FX-51 in most tests, despite having slower memory and a lower clock speed.
Those benchmarks aren't very good. I mean ... search and replace in Word? Quake 3? ("Hello, PC World, the 90s are calling -- they want their software back.")
Until Tom's Hardware, Anandtech, or one of the good sites covers this topic, I don't think we'll see a fair and thorough evaluation.
No, but both of them run Linux...
I seem to rememeber Apple (I think) saying that Premiere for Mac has issues that significantly slow it down compaired to the PC version. Wasn't there actually a story on Adobe's site not long ago that said how much faster Premiere ran on some new P4 than on Apple's latest hardware? I remember some big issue about that...
Ever since I switched, I have had no RPM dependency issues, not had to wait 4 hours for Gentoo to compile something.
I was running SuSE 8.1, and could not run Thunderbird because my glibc was out of date. The solution was o upgrade to SuSE 8.2. Great; upgrade the whole OS to run a newsreader. It runs fine on OS X Jaguar. I dont HAVE to upgrade to Panther in order to run it.
The fact is that OSX is the best of both worlds. It still provides the UNIX cli, X Windows for compatibility, and most importantly, a consistent GUI designed by professionals, not some hodgepodge of widgets designed by some college kids.
So, in fact, you are wrong. There are many reasons to go with Apple. My favorite is the way you install programs. No rpm -ivh --nodeps --force, just drag the icon where you want it, and presto. if you have no life, and would rather do extraordinary work just to get simple stuff to work (DVDs, etc), then by all means stick with Linux.
Nothing in those benchmarks makes me think that the Athlon "trounced" the G5. Look closely. The machines that beat out the G5 in *SOME* areas (not all) are running twice the graphics memory and RAID.
You can get a MAC with those specs. If you're gonna run benchmarks, run them with at least partly equal hardware specs.
One thing does stand out however - the specs of the alienware machine and the ployware machine are not that much different according to that table. So why is one behaving so differently?
.
If you really read the article, you'll see that they tested them without RAIS and with 128Mb Ram
Call on God, but row away from the rocks.
Unlike other zealots, I don't "push" it on you.. you have to take the red pill willingly the world is not black and white, you can use both or hate both OSs, until you have actually USED it for while can you really objectively form an opinion on WHY you would want it -joe
both 64-bit contenders
Both the G5 and the AMD64 are great chips, but they really only represent the intrustion of 64 bit computing in the popular consciousness, not the actual beginning of 64 bit computing.
Compare their performance with the last Alpha chip, development of which was cut off years ago, and tell me again how the best is being brought to us.
Even as Intel picks the carcass of Alpha to revive the still-born Itanium series, the killed off Alpha chip line has performance that embarrasses HP into covering it up.
"Provided by the management for your protection."
Jesus FUCKING Christ, this is the most self-serving obnoxious comment to appear in Slashdot for a long time. "Stay on your side of the railroad tracks"? "The speed of my processor defines who I am"? No, ramblings like the one you just wrote define who you are, and what you are, I'm afraid, is an idiot.
Mother is the best bet and don't let Satan draw you too fast.
I do. Standard BTO on the G4's and later
Compare the single G5's results with the dual G5's: only Photoshop makes use of the second CPU.
So they're basically pitting ONE G5 against ONE Athlon64 in the other tests.
But this is not the only incoherence with this test. Using a Classic, unsupported application like Premiere instead of the native After Effects ? Testing a Mac's performance in Microsoft Office ? This is a joke.
LALL!!
if I wanted a "user experience" I'd sit down in front of a pink iMac. if I want to get work done, I use my beige PC thanks
If i'd really read the articles.. *shudder* /. right?
This is still
Macs are fastar!
No they're not, and here's proof.
But macs are faster at speck, whatever that is!
No, they cheated on the test to make the PC look worse than it is.
But, um...Macs are faster at Photoshop!
Wrong, here's another test to disprove Stevie's secret filter test. They're even slower at Photoshop.
But they say they're faster!
Here the AMD absolutely trounces the G5 and leaves it in the dust.
BUT THE AMD DOESN'T RUN OS-X, W00T IN YOUR FACE PC BOY!
Give me the source code, and I'll make it happen in a week.
The Opteron Wins in all tests except three and two of those are game tests. Were you reading the same article I did?
go back to sleep under your bridge troll -joe
Another case of Lies, damn lies, and benchmarks.
Computers/OS's/Programming Languages/editors are tools. Use the one that best suits your needs.
Seriously though, I can tell you that OS X on a dual Mac is very lacking in power, this compared to a Intel/AMD solution. I have a dual 1ghz Mac on my desk right now, and you know what? I was so pissed off with the performance I bought Yellow Dog Linux. Now this thing smokes, probably faster than my dual Athlon 2000MP machine.
Until Apple and developers start optimizing programs to actually use both CPUs, we'll continue to see Mac's get trounced in benchmarks.
This is a test. This is a test of the emergency sig system. This has been only a test.
Working toward that. A bit difficult in the current local market, so I'll just be using Windows "in the office" for a few more months at least.
Specialization is for insects. - R.A.H.
To quote the Apple G5 page linked in the article: "the world's fastest personal computer and the first with a 64-bit processor". Note that Apple wisely makes no claims about the workstation market, either. I'm sure Sun, HP and SGI would have some comments if they did.
See: http://www.digitalvideoediting.com/2003/06_jun/fea tures/cw_macg5_interview.htm. Funny stuff.
[excerpt:]
DMN: Now, you're saying it's the first 64-bit desktop machine. But isn't there an Opteron dual-processor machine? It shipped on June 4th. BOXX Technologies shipped it. It has an Opteron 244 in it.
Rubinstein: Uh...
Akrout: It's not a desktop.
DMN: That's a desktop unit.
Akrout: It depends on what you call a desktop, now.
---
S
you wish... and so do I. so do most OS X users for that matter. who wouldn't like to make a $300 computer feel like a Mac?
alas, it's Jobs' way or the highway...
I am still waiting for the Zork benchmark. Exactly how fast can I kill the Ice Dragon with fire?
This is a test. This is a test of the emergency sig system. This has been only a test.
it is?
so macosx is the thing that defines it's a 'desktop computer', as in 'fastest desktop computer'? man, i've been so dumb, i thought that the fact that i could buy it and place it on my desktop and use for office apps&etc meant that it was a 'desktop computer'(no matter what os).
yeah flamebait, but it doesn't change things, apple advertises it as the fastest desktop computer, and first 64bit desktop cpu too... which they aren't, no matter if they're the fastest things to run macosx on.
world was created 5 seconds before this post as it is.
There's also the very noticable lack of the Pentium 4 Extreme Edition... They claim they were not available at the time, though AnandTech's Athlon 64 article includes it though it was published more than two weeks earlier.
Yeah, life isn't always that easy. I was very lucky. Good luck!
Stick Men
What's all this talk of 64-bit? 32-bits should be enough for anybody!
www.brownsauce.org
This isn't bashing. This is letting the Apple FanBoys know that their brief period on top is now over for the conceivable future.
Blar.
They run nativ apps. Both processors can process 32bit code natively, and they do it here.
And AMD should profit MUCH more from 64 bit than g5:
G5 runs the same, only in 64 bit (more memory/cache bw required)
K8 gets twice as much registers.
HI O WISE PRINCE. WHT TOOK U SO DAM LONG?
And why should I even want OS X?
Where else can you find an OS that can run a great video editing package like Final Cut, can run Photoshop, etc., and can also run all of your favorite *nix apps, natively?
Right now, OS X beats Linux's best desktop. That won't be true forever - just one more reason for a big enough hard drive to dual boot - but it's true right now. And as for Windows XP, all it really has going for it at the moment is a gaggle of software.
This wasn't a very useful comparison.
All but one of the Athlon machine were using twice the video memory and a RAID storage setup.
On the comparison with the one machine that did have the same video RAM and non-RAID storage, the mac did better on photoshop, which is apple's target market anyway.
Are Quake III and Premier native Mac OS X apps? Or are they OS 9 apps?
2GB dual G5 benchmarks
So I guess someone should set up a benchmark using linux, so we know once in for all which is faster.
:D
Anybody got $5,000 for hardware so I can setup a benchmark?
Jaysyn
There is a war going on for your mind.
I've never been marked "Flamebait" before, it should be exciting...
I remember when the first G5 benchmarks came out, how there were just oodles of posts saying "Speed is great!", "Woo-hoo! Macs are faster than x86!"
Now that these benchmarks have come out, the common thread seems to be "Speed isn't so important", etc.
hmm...
Let's face it... Performence really matters most if the system you're using does what you want it to.
For example, the Xbox version of Halo runs flawlessly and almost perfectly smooth... On a 733MHz x86 processor with 64MB of RAM. The PC port chokes my Athlon XP2000 with 512MB of RAM and a Radeon 8500.
Regardless of whether you use a PC or a Mac... If you're buying a system because you think it's the fastest, you're just kidding yourself. Pretty soon it will be outclassed by something better because that's just the way technology progresses. There is no winners circle for people with the biggest baddest rigs and at one point in time that old 486 on eBay for $9.99 Buy It Now was really the shiznit too.
Buy what you can afford that does what you need it to. If it's not the fastest game in town, so what? As long as you're happy with it, that's all that matters.
---
DRM is like antifreeze, to the MPAA/RIAA it's sweet, to the consumers it's poison.
Isn't it kinda funny that pc world did the compairson. Last I checked they were pretty much in the pocket of Microsoft to begin with. Not that I care a whole lot either way but untill I see a third party's numbers I'm not going to take this to seriously....
OK, Premiere 6 has optimizations for X86-64, but not for the G5. Hell, Adobe is discontinuing the Mac version of Premiere, because they see no reason to spend the time making a CPU-efficient version for Macs when Apple is marketing Final Cut Pro. Then they used Word..... Word for OSX, which is probably the only word processor on earth that is slower than Open Office..... And Quake 3 is also highly biased towards x86....
Let's see, why not run FCP vs Premiere 6, each software that uses it's native CPU to the limit? That would skew the test because the software was written differently? Well, guess what, if the G5 allows software to be written in such a way that it can be as fast as FCP, doesn't it have an advantage?
That's one of the reasons why Sto.pp(Stockholm PostProduction) went with G5's: Running Final Cut Pro, After Effects, custom software compiled for it etc, it was faster than the Opteron alternatives
I'm so glad that someone FINALLY had the balls to call The Steve on his evil marketing ploy. How dare he try to claim that Apple was somehow better, when every Slashdotter knows it's not!
(He won't give us the source for Aqua, see. We installed Red Hat the other day and are therefore qualified to fix all the bugs.)
At any rate: I just wish Apple would go away and stop competing. Trying to improve their hardware and software - and then selling it with such outrageous distortions - is bad, and hurts the industry. It really hurt my ego to think that some Photoshop dork has a better computer than I do. We can' t have that. Everyone knows Apple sucks. Er, suX0rs.
Benchmarks, really! Everyone knows they're crap. ATI had the balls to claim theirs was faster, when everyone knows nVidia rules!!!1! (Or was it the other way around? I forget.) As soon as the benchmarks said nVidia was faster, I threw that piece of crap ATI out of my machine and got the newest GeForce.
*Sigh*
ZOMG I WOULD LOVE TO KNOW ABOUT YOUR FEELINGS ON MACINTOSH VERSUS WINDOWS, VI VERSUS EMACS, AND HOW YOU'RE NOT A DORK
Hmm...
So this is how we benchmark two different platforms these days?
For everyone's information, I should not have to point the following out, but here we go... the benchmarks were taken from the following apps -
Quake III, developed on, and for, x86 over 5 year period of programming research and enhancement. Later ported to OSX in a week by OmniGroup.
Word, developed on, and for, x86, by the developer who also wrote the operating system running on the PC's. Ported by MBU to OSX.
Photoshop, Adobe develops Photoshop in a very balanced way for the two platforms, and these are the results for this test -
Fastest 50MB image = 17 seconds, G5 = 18 seconds
Fastest 150 MB image = 47 seconds, G5 = 51 seconds
The final test was a Premiere rendering, where almost all the systems tested did the job in 3 or 4 seconds. The fastest was 3 seconds, the G5 did it in 4. This is Premiere which no longer exists as a current ongoing product for OSX.
Does anyone see just how biased and unscientific this all is?
Oh, and I didn't mention that most of th PC's had double the graphics memory, and had RAID as their primary storage.
This article is FUD.
-Nex
This sig has been deprecated.
Most programs do not use more than one processor.
Photoshop does on some filters I think. Office
runs the spellchecker in another thread, but
usualy you do not gain much with 2 cpu:s.
I was with you till you mentioned the Photoshop score. So a dual 2 GHz G5 is only 10% to 20% faster than a single 2.2GHz Athlon, while the Athlon is arround TWICE as fast as those two 2GHz G5s in all the other tests. They did their best with the apps that would be commonly used, and I still think the AMD chips win.
Blar.
Firstly, let's clear our minds and not even compare Intel against AMD and Apple (IBM). The Pentium 4 was not designed to compeate against the AMD64 or the G5, if you want to bring Intel into the picture, let's see some Itialium 2 IA64 benchmarks.
In the fight between Apple and AMD, sure Apple said that they had the first, fastest, 64bit processor on the market, and they did-- as far as personal computers go. I don't think Apple really streched the truth that much, when the claims were made they were quite accurate. Unlike other companies who strech the truth considerablly in order to create FUD about their competitors...
As far as AMD goes, everyone is hyped up about them, but I'm not so impressed. The first Athlon 64 that was released was bosted to smoke the P4. This was hardly the case, in fact, the P4 2.4 GHz (800MHz) held it's own against the new AMD chip, and the 3.2 beat it in alot of areas, especially games which is one of AMD's major targets. If AMD 64 is so great why is the first chip on the market slower than the P4 which it was compeating against?
I used to love AMD, then the Athlon came, then the Athlon XP, and now my faith in them is dead. (Any chip maker willing to name a chip after an OS created by Microsoft should be shot IMHO).
So lets think about the facts with AMD. With their new Athlon XP marketing stratigy they outright lied to customers. The Athlon XP 3200 hardley compeates with a P4 2.4 GHz (800) let alone the P4 3.2GHz (which it's name implies it is comparable to). We're talking a serious difference in performance here, a difference that can be mesured with a stopwatch in some cases.
So so far AMD has named a processor after Windows XP, and lied to customers about the speed of their product. Sounds a lot like another company I know, unfortinately for AMD, they have lest than 20% of the market share.
Intel on the other hand spends billions each year on research opposed to AMD's millions. Intel has been a strong supporter of [GNU/]Linux and Free/Open Source Software. Maybe Intel could do more, but they certainly do more than AMD.
Personally, I don't think the PC market was ready for 64 bit computing. AMD (and Apple) jumped the gun on it. 32 bit for home computers would have been fine for another 2 or even 5 years. Intel was taking the right approach with releaseing the Itialium, and Itialium 2, both are native 64 bit, and both have had running versions of Windows for months. Having 64bit IA64 workstations on the market would make it possible to slowly bring over 64bit apps. Sure you wouldn't have 32bit support, but do you really want 32 bit apps on your 64bit system? or do you want to force developers to move to 64bit or be left behind. AMD64 will actually slow down 64bit application development more than accelerate it. The outlook now for developers is:
Intel still supports 32 bit, they have an interesting 64 bit processor, but thats just workstations.
AMD has a 64 bit chip, but it's not compatible with Intel 64bit, and anyone serious about high-end workstations will probably go with Intel which has a much better track record for stability and performance than AMD...
AMD still supports 32 bit though, and is compatible with Intel 32 bit.... hmm.
Welp! guess I wont waste my time with 64 bit just yet.
AMD has always fed off using Intel's x86 arch. The AMD64 is not compatible with IA64, which causes a platform split between the two competitors for 64 bit apps. If Intel maintains it's marketshare (over 80% last time i checked) then it will be a hard sell for anyone serious about 64bit to go with AMD over Intel, especially if the IA64 out performs the Opteron (my suspicion is that it does.)
Am I bitter at AMD? Sure, but if you use your friend google and spend a few hours looking up the facts, you'll find that mostly everything I say here is accurate. (my memory isn't so hot)
Apple is Apple. I love Apple, they are the cool of cool, but as far as processor wars go, bringi
Too bad, so sad, and even if that's not good enough for you. Going to the OS is what I would expect when we are discussing hardware. We win, so all you apple tarts can shive it. a HUGE NYAH FUCKING NYAH in your head.
Are these guys serious? ... sure, why not include it.
4 apps. That's it?
* Microsoft Word? Give me a break! Sure, you don't have many alternatives in OSX, but this huge application still suck on OSX. Microsoft hasn't gotten it right, and they probably never will.
* Premiere is AFAIK a PC app nowdays. You don't buy a dual PowerMac G5 and run Premiere?!?! Final Cut Pro is the only option, and it is extremely fast on Apple's high end hardware!
* Photoshop seems to be an application that's very well optimised for all kinds of SIMD implementations, and it's an ok benchmark.
* Quake 3
Wouldn't it be possible to run Premiere vs FCP? They must have very similar featuresets, so it shouldn't be a problem to come up with a benchmark that you could run on both?
How delightful that PCWorld has chosen to make things nice and easy to read for any small children who might happen to accidentally read this article! It's a pity that actual facts and content had to be discarded to make this possible.
The instructions run by 64-bit processors are for the most part identical to those run by 32-bit processors.
You know, the only thing that really goes slow on my PC is XviD encoding from MPEG2...
Granted, ripping DVDs isn't as popular as making MP3s, but I am surprised how little coverage it gets as an aspect of Mac performence. Is this a PC-only phenomenon?
---
DRM is like antifreeze, to the MPAA/RIAA it's sweet, to the consumers it's poison.
It was paper launched, and probably isn't going to be available till November. I see no problem with only doing benchmarks on currently available systems.
Good for Athlon! I just got a G5 and this doesn't make me feel bad for it, it does what I want extrodinarily fast and very well. What people are forgetting is now Apple is at least back in the picture and has more Ghz on the way. I am a mac person, but I am not going to spout off the macs do everything better all the time no matter what, that is silly. They are great computers, and OSX is awesome. And I think Athlon should be proud to have a great processor out. The more competition pushes the better Apple will be in the long run. What a waste of time to bitch at each other because apple got beat up in one magazines test.The only bummer is they should have used identical VRam amounts, that was unbalanced. When the 256 radeon comes out for Mac I'll have one the next day. Hey, instead of being a zealot for either side lets just chill and be glad the processors are getting crazy fast and getting faster every few months, imagine what the next year holds for us. Can't we just get along?
It's no big deal - we all kinew this was coming. You can't stay fastest for long, and I was under the impression that the new "Extreme" P4 was faster anyway. Needless to say, the performance of the G5 will improve after the next OS update, and more so when developers (including Apple) start to leverage the potential of its 64 bit addressing, different Altivec design and massive FPU performance. If there's less than 10% performance between the fastet Pentium, Athlon and G5 designs, then they're all FAR MORE dependant on application optimisation anyway!
That was classic intercourse!
My latest Linux desktop has the ability to mimic almost exactly the "Luna" Interface of Windows XP.
Although, I personally prefer to use XFCE4, which has had the appearance that Apple took for MacOSX for several years running...
Look up XCFE and take a look at the screenshots dated from WAY before MacOSX. Sure, XFCE4 may not have all the eye-candy (thus CPU Chewing) features of the Aqua interface, but it sure is userfriendly and very capable.
It loads blazingly fast and operates exceptionally well.
If you ignore the other uses of a tool, does that make the tool less useful, or you less useful?
I wanted to go get the rest of the specs on the Polywell Polystation Two that seems to be the hands down winner. My only problem is i can't seem to find it for sale on their website. http://www.polywell.com/products_1.html
Point it out if i'm wrong please.
The G5's Q3 scores tipped me off.. they're half of what they should be:
According to Bare Feats... it should get closer to 300 fps
makes me wonder about the other tests as well...
Microsoft announces new emoticon product ratings, gives latest Windows and Office products XP
Their point was that it is still apples to oranges. Word doesnt show anything about the CPU architecture, it only shows how well Microsoft wrote it on each platform.
Plus, unless an app is recompiled, the Altivec functions will hurt the G5. I don't think there are cases where there is a similar occurence on Opterons...
The machines that beat out the G5 in *SOME* areas (not all) are running twice the graphics memory and RAID.
Did you read the article? Like... you know, the benchmarks at the end? The system that won the most benchmarks, the "Polywell Polystation Two", had "only" 128MB of graphics memory and no RAID, just like the Apple.
If you read the article you'd also see that they tested the Aurora system with 256mb vid mem/RAID *and* with 128mb vid mem and no RAID. The difference in results was very small.
Also, the extra video memory won't help in most of the benchmarks. The only benchmark it would possibly help in would be a gaming benchmark, but Quake3 is not going to show any improvement moving from 128MB-->256MB. That's a 3 or 4-year old game designed to run with about 32MB of video memory... it doesn't benefit from 128MB of video memory, much less 256MB.
Try to read the article more thoroughly before posting...
OtakuBooty.com: Smart, funny, sexy nerds.
Where can I get OS X for AMD-64 then? Oh thats right, it only runs on those slow G5's!
will the G5 run Windows?
I would think that if you hacked NT4.0 Workstation, it would run Windows quite well - just not a modern version.
Apple Power Mac G5 1.8-GHz PowerPC G5 128MB
Quake III frames per second: 141
Alienware Aurora 2.2-GHz Athlon 64 FX-51 128MB
Quake III frames per second: 257
Looks like it's Athlon for me--anything less than 200 FPS gives me a headache after a few minutes.
In Microsoft Word we timed a search-and-replace of one word in a 1437-page document, and the execution of the auto summarize function on a 210-page document.
OK, so the Mac takes 5 seconds more than the PC.
But why is this called "trouncing" ?
I mean : I hardly edit docs which are more than 100 pages long.
I Don't use a screen that could render 300fps in Q3, and Premiere seems just quick enough as it is.
Trolling using another account since 2005.
I was surprised that a single CPU Opteron could be in the same performance ballpark as a dual CPU G5. Does the Opteron do more per clock cycle than the G5? Are applications not taking advantage of the second processor? Is there some other performance bottleneck, such as the memory subsystem? I look forward to finding out....
Out of the apps listed, the only one that takes advantage of 2 cpu's is Photoshop, and that's only for some operations.
Premiere is especially noted for its threaded suckiness on the Mac.
Must be nice to live in a world where people can pick and choose increasingly scarce tech jobs according to which company does and does not allow you to use *NIX on production machines.
To hell with maturity, *NIX or bust!
Mac OS X and Windows XP working side by side to fight back the night.
In addition to dubious selection of apps for the test, it should be noted the G5 will be noticably faster for Panther due to the fact that it will be built with a compiler to optimize for the new processor.
Offtopic? No. Redundant? Yes.
[quote]"A 64-bit chip can run longer, more complex instructions than a 32-bit one"[/quote] As I know, the 64-bit PCs have abandoned the old x86 instruction set (CISC, non-uniform, complex), and use a RISC instruction set, wich means shorter, simpler, faster, and every instruction has the same length. This report just said the contrary.
I'm being serious here... I can't tell which OS you are bashing... because all of the OSes in question have literally thousands of apps. But, I see that you stipulate worthwhile apps.
On this, David Pogue of The New York Times said, "In time, iSync may even become yet another siren, singing out a question to Windows users who already look longingly at iTunes, iPhoto, iMovie and iDVD: Who cares if 30,000 programs are available for Windows, if the five you want most are available only on the Mac?".
Now we are at the dawn of a new juncture... Apple apps on Windows. Only hours from now, Windows users around the world will be running iTunes. And they will really really really like it....hmmm. But where will they have to go for another fix?
It seens that all benchmarks were performed with off the shelf 32 bit software, running on 64 bit hardware.
Call me again when they got GENTOO compiled at full 64 bit and full speed on both platforms, and we get the real numbers.
I do not recall a P-4 being not nearly equivalent to a G5 in these conditions. Them we will see if AMD can push the 24 year old 8086 family to competitive levels with their 64bit patches.
-><- no
How fast can either one transfer a 17 Meg file?
What's your point? That just means the G5 is a versatile chip. The person who submitted the article stated that the Opteron beat the G5 to market by 4 months..well, so what? There have been servers running 64-bit processors for a long, long time. Opteron is nothing new in that respect (its hybrid design is admirable, but a necessity, now that most 64-bit CPUs carry them). The G5 however, was marketed as a desktop solution, and was sold as one. The Opteron just was not. Thus, the time scale is completely negligible.
Also, the 970 may be a server chip now, but again you're missing the point: Opteron was never meant for the desktop.
--- What
Does anyone have a working Linux machine with all of their standard software (Gnome, KDE, etc.) compiled around the Athlon64 CPU?
My next upgrade will probably be one of these machines. On the other hand, I don't see that happening until drivers and all of the software works as well as it does on i686-class Athlons.
Is it worthwhile to build one of these machines just yet? I can't see any reason to do it at the moment unless everything was in good working order. Is it still too early? I'd love to have real 64-bit Linux on my desktop.
We're all geeks, so we love benchmarks.
It's too bad that no rigor is applied to 99% of the benchmarks that are applied.
Raw CPU benchmarks like SPEC end up being compiler tests rather than processor tests.
"Real-world" app tests like this one are better, but only if the apps used are representative of apps used by the person reading the benchmark. They are not a realistic measure of holistic system performance.
Adobe Premiere? Come on! Does anyone on the Mac use that at all anymore? Is it even OSX native? Since Final Cut 1.0 came out 2+ years ago, anyone who considered that dog Premiere deserves what they get. Isn't there a better editing package on the PC, or is the Mac just that much better for video editing?
Microsoft products should be excluded from benchmarks on Microsoft's OS. Of COURSE Microsoft optimizes performance of their apps on their OS more than their apps on other OSs. That test is pointless.
The Quake test would be valid, except as many people here have pointed out, it's a 32-bit app, so it's not using any of the 64-bit capabilities of these boxes. When we get a native, 64-bit version and can compare it to two boxes with the same ATI video card, then it will be a valid test.
- Vincit qui patitur.
I never knew I had it so good with WordPerfect for DOS on my 286. It kept up with my typing just fine...
---
DRM is like antifreeze, to the MPAA/RIAA it's sweet, to the consumers it's poison.
So don't try and shift the blame, son. Apple backed a loser (G5 is a baby cpu from IBM...they should have bucked up for something with more chest hair).
So there.
Blar.
So I can get a machine that's slightly faster, costs about the same (Alienware -- I'm guessing), that has a higher TCO, running an operating system that's got severe security issues and one nasty GUI.
Or I can get an Apple, which might cost a bit more, has a lower TCO, running on OS based on UNIX, with a consistent GUI and a command-line interface, from a company that issues security patches in an extremely timely manner, where I don't have to install drivers to get my digital camera to work?
I'll stick with an Apple. My time is worth too much to waste it monkeying around with Windows and trying to get shit to work.
blog |
Since they supposedly are not testing application performance but system performance run FCP to encode the video on the Mac then watch as it blows away Premiere on any of the other systems. It is insane to use a Classic app that has been dropped from the Mac for a benchmark.
The only OS X optimized app (photoshop) ran the best on the G5. It was also the only app to take advantage of the 2nd processor.
Why don't they use Mathmatica or other such well ported apps to test? It is because they want to skew the numbers.
Panther and new applications that are coming out that are G5 aware will even further skew things in Apples favor. Panther has a noticable performance improvement over Jaguar.
As always, as soon as Apple starts doing really good things as a company the PC journalist fall all over themselves to spread as much FUD as possible. It wont be long till they are singing about the doom of Apple again.
According to the benchmarks, the dual 2GHz G5 beats the single processor Athlon64 3200+ and the single processor Opteron 246 at Quake 3 Arena.
The SMP 2GHz G5 also beat out the UP Athlon64 3200+ at both photoshop renders (by a large margin).
The UP 1.8GHz G5 came in dead last at *every* benchmark.
In short, title of article is accurate.
Adobe Photoshop is an image editing and manipulation package, originally used for retouching photos (hence the name) but nowadays capable of vastly more. For years it's been the standard for 2D graphical work in design, graphics and DTP, and with QuarkXpress* was part of the reason for the success of the Mac in those fields (there were no PC versions originally). Practically any professional graphics you see nowadays has had Photoshop used on it somewhere. Other programs offer some of the functionality - Paint Shop Pro, and The GIMP which aims to be an open-source Photoshop equivalent - and does pretty well for a lot of things - but Photoshop is still the professional standard.
(* QuarkXPress is pretty much the standard page layout package, most newspapers use it. It too was for a long time only available for the Mac.)
Actually try doing those things on a windows machine and realizing that most of them are not as you think they are. That's ok, somebody needs to help Jobs make hs boat payments.
Blar.
What, did they take memory OUT of the G5s to purposely f*** them on that test? 128MB? Yes, they suck like that (and, incidently, don't sell with less than 256 on the lowend, 512 for the machines in question).
In the cases WHERE THEY ACTUALLY COMPARED TO A MACHINE WITH THE SAME AMOUNT OF MEMORY, the G5 was still smoking. Plus, they put the g5 up against machines with RAID? Who the hell is fooled by bs stats like this?
I love the opteron, but gimme a break, this test is shite.
Those jokes are so freakin old they don't even touch funny.
But anyway, is this really a fair comparison? A lot of the performance that they measured has to do with how the software is written and compiled.
yo! mac trolls, check the benchmark first! the dual G5 only won the Photoshop tests against single cpu Athlon Fx's. It never won a single benchmark overall - the best scores in Photoshop went to the dual Opteron box!
and remember, the dual Opteron uses ECC memory - slower and not at 400MHz yet(*) (they probably had 333MHz in it).
(*) they said 'the PCs had 400MHz memory' in the Athlon64 vs. P4 bench- by the PCWorld definition, that doesn't include the Opteron ehich is clearly a workstation system. also, people who tried 400MHz non-ECC memory with their Opteron boxes said it doesn't work at all.
yeah...that's what I thought.
Blar.
has AMD had a decent processor that was fast and stable... I've never used one that was either of those. AMD has always sucked as far as I've seen. On the other hand, my PowerMac Dual Ghz smokes any AMD I've used (and these are recent ones too!).
So what am I buying and who do I believe?
Apple.
Can't wait to see the white background mac commercials with a bunch of whiners stating that their mac is faster than anything else then going through 25 excuses why other systems crush it.
Did you even read the article?
The dual G5 lost 4 of 8 tests and won 3 of 8 against the single Athlon64 3200+. The SINGLE G5 lost all tests to all other processors
Yeah, sure. The G5's won exactly 0 of the tests. The dual G5 managed to beat the single Athlon64 FX+ on the photoshop benchmarks. That was their only victory, but they still got stomped by an Opteron in those. And that wasn't a dual Opteron.
Yes, you will see a price-performace ratio change. The Dual G5 tower, which was the only thing close to competitive, costs about 3.5 grand. So, it's twice the cost and way slower. It's even 200 dollars more than an Alienware AthlonFX box! Try just buying a Athlon64 in a standard box and its a clear victory on every front, just as with the expensive one, just by a heck of a lot more.
First off, the artical is wrong. It states that apple made the comment that it was the first & fastest DESKTOP 64bit processor. The Operon is not a DESKTOP processor. And yes, it was the fastest when it was released, thats a fact. Is it anymore, prolly not - Thats technology though! And for all of you who were just waiting for some faster processor so you could bite at the heels of the Mac people - let me clear something up for you. PREMERE was NEVER made for MAC OS X. If they benchmarked this, they would have had to open up CLASSIC mode. If you want a comparison (windows people) Open up Virtual PC (yes there is a Windows version) startup Windows 98, open Premere and benchmark it. You will see a fair benchmark. Becides, my G5 will win overall because MacOS X is a hell of a lot more stable & secure!
and it can still dust most of the new 64bit chips
also you can have one today cheap
just check out ebay
...that RAID and additional RAM improves performance.
Oh wait, wasn't that what we were testing?
Make a $300 computer feel like a Mac.
You amuse me. All the comment appears to have meant is "you don't like it, fine, but I do, so I don't really care".
Really, I don't care either which is the top CPU, I run a 1.4 Athlon at home and have no real feeling that I need to upgrade... so why should a G5 owner really worry if it's a little slower than some other CPUs?
Good God! OS X is the WORST feature on a Mac. Terribly, non-intuitive mickey mouse homosexual GUI. If I bought a Mac, it would run *nix, but then again, I don't make a habbit of pissing my money away when it can be had for much cheaper.
Yes, Apple makes awesome hardware, but their OS is garbage and their prices are too high.
jim frost
jimf@frostbytes.com
In related news:
Trying to decide which applications to use for the benchmarks in his upcoming article Tom Mainelli scheduled a meeting with a mac genius at his local Apple Store. He asked the genius one question "what are the slowest, most loathed OS X applications which also are available for Windows XP?" The mac genius cut him off right as he was about to finnish his questions saying "Adobe Premier and Microsoft Office... Why?" Mailelli answered, "Oh, just wondering"
Unfortunately regardless of the benchmark results, You can still only run OS X on a mac. I guess I will just have to settle for the G5 since Microsoft can't seem to offer anything that is nice to look at/usable. Go Apple!
What I think would be a cool benchmark is to install Debian on both a G5 and an Opteron machine, and run apps from Debian on both. You'd be using the same OS, the same compiler, etc and it might show the differences due to the hardware.
:-)
If you really wanted to go crazy, you could install the Debian base system and then start recompiling stuff with more optimized compilers for each platform. (Assuming they are available for Linux.) At least you'd be running off of the exact same codebase for your benchmarks.
If anyone wants to give me a G5 and an Operton, I volunteer for the task
Funny, when Apple-benchmarks showed G5 to be impressive performer, macheads were screaminh "Woohoo! G5 is the fastest CPU in the world and the new Powermac G5 is the fastest computer in the world!". Now that some third-party benchmarks dispute that, they whine that "this is a joke!" or something.
Lesbian Nazi Hookers Abducted by UFOs and Forced Into Weight Loss Programs - -all next week on Town Talk.
I'm not taking any chances: give me one of each.
org.slashdot.post.SignatureNotFoundException: ewg
It's not 32 bits compatible, so even if it was out before the G5 doesn't interfere with the fact that the G5 was the first "desktop" 64 bits processor. See, if a processor cannot run any existing app on a given system it can't target it. 32 bits apps are what you get in the desktop world right now, if you can't run them, you are not a desktop thingy, period. The Opteron was and is still marketed as a server/workstation processor. The G5 can obviously act as a workstation processor since it have lots of bandwith and the architecture to act like one but, since it's 32 bits compatible, it can also be used on the desktop and run existing desktop apps. In a few years from now most apps will be 64 bits capable, in even more years all apps will be 64 bits capable, only then will the Opteron be usable in a desktop fashion.
As much as you don't want to acknowledge any benchmark provided by any Mac sites because they are zealot benchmarks that can't be trusted, you cannot acknowledge any benchmark by a PC magazine because they are zealot benchmarks that can't be trusted...
ooohhhh, I forgot, they don't do that in the PC world! riiiight.
I wonder how these machines would perform if the tasks were performed simultaneously. I often have time consuming tasks running in the background. I suspect that besides OS X's awesome process manager, the dual processors would really help in this case. It's the least they can do if they won't run apps that actually use both processors.
t'nera semordnilap
The one-size-fits-all approach to these things are pointless. I really do not care about how fast Quake runs because I don't play it. I couldn't care less about Premier either. I use my PC to browse the web, SSH into remote servers to code, and compile code. Except for the last one, any difference between processsors would be minute. Performing a few tests and then claiming one processor is better than another is just silly.
EvilCON - Made Famous by
SGI had 64-bit WORKSTATIONS for YEARS before Apple started marketing their solution as the "First" 64-bit Workstation.
IBM had 64-bit Workstations for Years before Apple started marketing their solution as the "First" 64-bit Workstation.
Sun has 64-bit Workstations for Years before Apple started marketing their solution as the "First" 64-bit Workstation.
Again, what is your point? Apple's marketing department is just as full of crud as everyone else.
BTW, The Intel Xeon was "never" meant as anything aside from being a Server chip, but that didn't stop people from making workstations out of them. UNIX, in general, was never meant as a Desktop Operating System, it was always geared up as being a Multi-User Server based system... But, that didn't stop anyone from making UNIX workstations, even Apple.
I say, if it works in areas that a product wasn't initially intended for, then by all means use it... The main reason behind the Opteron being 'geared' more for Server use is some architectural design differences that equate in higher fabrication and thus higher retail prices... Most 'desktop' users aren't willing to spend that much for a 'dekstop' system.
Again... I am still failing to see your point... (Did you even have one?)
If you ignore the other uses of a tool, does that make the tool less useful, or you less useful?
which one should I get? Apple or AMD? I want to access email, surf the web and do a few spreadsheets.
For crying out loud. Grandma doesn't care how many teraglops the whozit can do. Neither does the very VAST majority of users. But I suppose I'll be hearing that "I just have to have one of those 64 PC's. Why? Because they're the fastest!"
Sheep I tell you. Sheep.
as for athlon, pentium etc. Free/Net/Open BSD Well done GUI (winxp and linux arent really as bad as mac fanboys/zealots claim) MS office Next time i am on the market, i will be shopping hard for something not a MAC( especially since it wont burn a hole in my pocket)
They must have had a hard time getting the Opteron to beat the G5 since they apparently chose rigged benchmarks.
to begin with the mac had less graphics memory and a less advanced grphics card but still turned in a fine job on quake.
There were no tests of large memory moves, a particular strength of the apple G5.
for example, the photshop tests used files of only 50Mb. That's pretty small of a pro-user.
Now in detail: there are only two tests where the opteron significanly outperforms the G5: Word Document tests and an Adobe Quicktime render, the rest are close enough to call a tie: no one machine is best.
Notice that the "quicktime" tests were not done using apple software but actually adobe premier, and adobe has long ago announced that the apple market is not something they are optimizing Premier for.
Now look at the only benchmark where there is a notable difference: WORD. These are all going to be integer operations, which we acknowledge is going to slightly less fast on a g5 then a x86. These are all going to take place on a platform where the APP and the OS share code. Finally word is not an application likely to be highly optimized for speed, espeically on a mac. Moreover, its not going to be using any dual processing capability.
If you wanted to turn the tables I'd suggest picking say PDF rendering, let the Opteron use adobe's product and let apple use their own.
They also did not do any tests that reveal the fact that g5 dual processing scales well.
Finally if you stripped all the security features out of the mac OS, removed the quartz graphics, and changed it to a bare bones primary color (fisher price) interface rather than aqua, you could probably get a faster machine too. but then you would also have a windows experience too.
Some drink at the fountain of knowledge. Others just gargle.
I just sent this letter in to PC World. I think it pretty much covers all the mistakes they made in the cross-platform benchmarks.
>--
I have been a long time reader of PC World, and have much respect for your magazine. However, I am yet to see a more abject review than the "64-Bit Takes Off" what was presented in your November 2003 Edition.
Let's start with the choice of Microsoft Word. Undoubtedly a widely used piece of software, and Microsoft incredibly allowed Office v.X for the Mac to receive a number of features that the Windows version is yet to receive. There is, however, one thing that Microsoft will not allow Office for the Mac to achieve; and that is performance parity. To add to this, much of the codebase of Office v.X is left over from the good ol' days of MacOS 9 - reflected in the fact that Office is still a Carbon app. So, although Office on the Mac is extremely widely used, it's of dubious use as a means of comparing performance between processors. Unless, of course, all you do is Office and it's not presently running fast enough for you.
Next. Premiere. This is what stunned me. There is a reason that Premiere doesn't work very well on the Mac. This is because absolutely nobody who does video editing on a Mac uses it. Period. Final Cut Pro wipes to floor with it; not only in functionality, but performance also. Of all the ways you chose to benchmark the G5s, this surprised me the most.
In the Quake test, the Mac was hamstrung by the fact that it only had a 128MB video card in it. I also may be wrong in making the assertion, but doesn't the 256MB ATI 9800 Pro run at a faster clock rate than its 128MB cousin? This would account for quite a performance differential. Despite the fact that Macs aren't really known for games, no other computer with a 128MB graphics card beat it.
The next test was Photoshop. This is the one app you benchmarked in which some 64-bit optimisations have taken place for the Mac, and is also an app that many people use on the Apple platform. In this test, the G5 beat everything on offer from the x86 world by quite a handy margin.
What makes this even more impressive is that the G5 system you benchmarked is running on a stop-gap operating system release from Apple. OS X 10.3, codename Panther, has been specifically designed to take advantage of the G5's 64-bit CPU structure; it's out in barely a week.
I would certainly be interested to see a re-run of the tests, if you think that this feedback is valid. Cross-platform benchmarks are notorious for being difficult to standardise; I do, however, believe that if done properly they can be both useful and interesting.
-- james
Fuck the laptop, did you see the PCs they were selling? Good god, I didn't know you could do that with x86 hardware...
Jaysyn
There is a war going on for your mind.
This is about as foolish as comparing megahertz, and PC folks still don't get it: Mac SPECmarks and PC SPECmarks are completely different, a Mac SPECmark is significantly bigger, with more SPECs per mark. You're comparing Apples to... well... I dunno, the PC is just fruity.
Just look at the title
Athlon 64 vs. Apple G5 Systems: Not Even Close (chart)
I wonder if this was impartial (:
Let's face it.. Mac users and PC users are fundamentally different kinds of users, just as Mac's and PC's are fundamentally different kinds of Computers. Apple stopped being in the hard-core geek world back when they stopped making the Apple 2GS. Since then, Apple has been in the foo-foo world of grapic arts and tabulating scores on second grader's papers. Apple's software suite is limited to these sort of applications, too. I just think it is funny that Mac, having been a "touchy-feely anti-geek" sorta platform where everything works in a single click (mainly because there is only one button on the mouse) and mr. bluebird and happy help bubble live a quiet, happy life on a quiet, happy appliance of a computer, starts stomping around trying to look all powerful and geekish because their new 64 bit processor is so amazingly fast-- Who cares? Am I going to buy a Mac, ever, for any reason, ever? (even if it is fast) No. Mac is for touch-feely grapic artists and school teachers. Mac is an appliance. It is not a computer. 'Nuff said.
That was the graphics memory column. Doh! Both machines had 1GB of RAM.
From the article:
"Most of the PCs used dual, RAID-striped hard drives; the Apple systems did not."
And Microsoft Word? I wouldn't count on that to be optimized for OS X.
I've also heard that ATI drivers are not optimized for platforms other than wintel, it's like, a known thing. The upcoming XF86 4.3.x drivers (not the current ones, the drivers planned for xmas time) are going to be optimized. No word on OS X/64bit though.
Try this test using Intel chips for x86 comparisons... It will be closer I would bet. AMD 64-bit stuff is optimized heavily to increase execution of 32-bit apps.
or so i think.
SIGERR: laziness exceeds quota
You know, Apple new said workstation, only personal computer. It would be nice if reviewers and news people on the internet were blievable.
We're speed testing 64-bit chips using (old) 32-bit applications
Premiere runs in Classic Mode. It's not even running in OS X. This should not have been used.
One of the appplications tested was Microsoft Word. I won't even comment on that one.
"All machines were tested with 1GB of RAM and the ATI Radeon 9800 Pro graphics card; the Mac version of the graphics card has a maximum of 128MB of RAM, while the high end for PCs is 256MB."
"Most of the PCs used dual, RAID-striped hard drives; the Apple systems did not."
Most of these tests (including, possible, the Quake test) are restricted not by processor speed but by a memory bottleneck. The G5 (and most likely the AMD) processors were data starved. This is espicially true in the Microsoft Word and Premiere test. This is more of a test to see how well the applications are writtem to maximize cache loading and memory transfer. The Photoshop test has the same problem.(if using the default memeory and scratch disk values).
Further, their tests of Quake didn't amtch those Apple performed. I doubt that the Quake applicaiton was compiled on the G5. It most certainly wasn't optimized for the G5 (Apple did this and the results beat those of the AMD machines...using the same type of graphics cards) See http://www.apple.com/powermac/graphics.html
The bottom line: This test is more of a memory and disk access test. There is not a single tet here that stressed the speed of the processor. Not one. What we need are applications tested on compiled on each machine and stress CPU (rather than memory) performance.
Come on guys, start practicing, you're going to be typing this a lot.
Speed doesn't matter
Speed doesn't matter
Speed doesn't matter
Speed doesn't matter
Speed doesn't matter
Speed doesn't matter
I agree. Combustion benchmarks are what I want to see.
The new Macs aren't great values either, as the top-of-the-line G5 ($3549 as configured) costs about $200 more than the similarly configured Alienware Aurora. (Prices do not include a monitor or speakers.)
Ah, the well known high quality brand Alienware, of course Apple doesn't have a fighting chance against a brandname like that. Alienware's years of service in the Fortune 500, the constant stream of quality awards in consumer magazines. Apple's crazy to ask 200 dollars more, the G5 should be half the price to even be considered.
And the speedtest, well, we all know everybody is still stuck on the Mac with Premiere as their main application running in Classic. Why didn't they test Netscape Navigator 3.0 as well? I mean, that's todays most popular webbrowser!
good thing that i don't use Photoshop the whole day... only half of it. i would have never noticed that 20% of my time! can you say altivec? it's also a good thing that i like many people buy their computers for the applications they are going to use, not the shiny cases or 3% savings. i'm feeling downright taken advantage of! but i feel better when i look at my shiny g5!
I've not heard of a 64-bit version of Photoshop
for an OS which will run on the Opteron.
If the applications tested on the apple were
compiled 64-bit, while those on the Opteron
were 32-bit apps, that could make a significant
difference to measured performance.
No... you'll never know who's faster. it's who marketing department who's fastest.
And the response from the benchmarkers "How dare you! We use Macs everyday for... yeah, we use them every, DAMN, DAY!!!
||| I still can't believe Parkay's not butter.
If you had bothered to actually read the article you would see they are comparing stock machine to stock machine. Apple doesn't ship with RAID, nice troll though.
quote from article
"However, while Athlon owners must await shipment of Microsoft's 64-bit Windows XP (or choose a 64-bit Linux OS) to use the chip's 64-bit capabilities, Mac G5 owners have a 64-bit-capable desktop out of the box."
see you in 2006.
They only gave the macs 128mb of ram....everyone knows that the g5 does loads better with a bunch of it.
Okay, any given app can give one platform or another an edge. But let's look at some hard facts about the 4 apps they chose to run, shall we?
:-)
1) Premiere 6. This program is used by thr minority of video editors on the mac platform for GOOD reason... it's TROUNCED by Final Cut Pro 4's performance.
2) Photoshop. I'll give this one a fair shake as the Mac and PC are equally mediocre. Thankfully, there are plugins for the G5 on this, so if the PC wins, I'll give it the nod fair and square... but wins by less than a 10% margin.
3) Word. Written by Microsoft. Do I need to elaborate? I didn't think so. Does anyone really believe this is a good performer on the Mac's side? Anyone? Bueller? Bueller?
4) Quake 3. A poor indicator for modern gaming performance... especially since a sustained FPS of over 50 frames a second is not noticable to a human eye. Even then, the PC is winning by roughly 20%. I'll give them the nod, but I would have liked to have seen something running on those Radeon 9800's that uses per-pixel shading.
Overall, it looks like the PC is slightly faster and I'll give it the nod. I don't think it will last, however. Prepare for the 3.0 Ghz G5 by next June. I think IBM will be able to play well as industry leader. I'm glad that the Mac is running neck and neck these days, as my OS has really always been the deciding factor in efficiency for me... speed just helps me with my 3D modelling.
"Politicians find new names for institutions which under old names have become odious to the people."
weeooeeeooeeooeeooee
only the g5 can claim to be used in the second fastest super computer in the world. although i still think i would prefer an amd chip. price is a factor, as well as modifications, and choice of operating systems.
You want to see which one blows up the best???
The POINT is that Apple never marketed the G5 as the fastest workstation. All Apple marketed the G5 as was the a) first 64-bit desktop (and if your definition of desktop differs from "a pre-built box from a well-known company that an ordinary human might buy", that's your problem, not Apple's), and b) the fastest desktop around at the time.
Saying, "Ooh! Ooh! New computers have come out! There are benchmarks against computers Apple wasn't talking about! The G5's not the fastest! Apple LIED!" is just plain dumb. Of course faster computers will come out! Apple isn't dumb enough to think or claim that their first-generation G5s will always be the fastest, and anyone who thinks they were claiming that is dumb.
And does anyone else see the possible conflict of interest with PC World running these benchmarks? Now, note that I'm an Apple fan. However, I won't completely believe any benchmarks that are done by anyone with an interest in seeing either side win. And it would probably be best if both computers were running something neutral, like a Linux or a BSD. Does anyone really believe these benchmarks are any more fair and unbiased than Apple's own???
Dan Aris
Fun. Free. Online. RPG. BattleMaster.
Further the PC World tests run here are much different than the tests run against the G5. This is no explaination of this.
Sure... Apple fans love the results from Apple's benchmarks and hate others. Apple detractors hate Apple's benchmarks and love ones that show slower scores. Neither one is a reflection of reality. Benchmarks all suck. Kind of like statistics, you need to know what the biases are of the person doing the reporting.
That being said, I find that these benchmarks are lacking. The scores for Permier and Photoshop are close, the Word test is silly and I've always thought measuring FPS on a game is really silly. (Mainly because it measures the video card most of all, not the rest of the machine.)
to quote the article
.....
"Fifteen years passed between the first 32-bit chip and a full 32-bit Windows. The 64-bit switch may be faster: Both chips and OS are here."
It then goes to show the a list of dates for the 32bit era, with the first Intel chip being made in 1985, so I assume they are saying that windows 2000 was the first 32bit desktop OS?
But, if they are including Windows 3.1 as a desktop OS then surely the should include NT3, which although looking like 3.1 was actaully a full 32bit desktop OS. Which would make their statement blatently false?
Or have I missed something
try to make ends meet, you're a slave to money, then you die
http://www.discreet.com/products/combustion/
http://www.discreet.com/products/combustion3/
Well, there's Windows, Linux, and the BSDs. Windows has Cygwin, as well as AT&T's commercial Unix environment. Linux/BSD have WINE, to run the Windows apps.
I admit, that might not exactly be "native", but it's no less "native" than running Unix apps under OSX.
Slashdot gets worse every day... Pipedot: News for nerds, without the corporate slant
Um. "trounced?"
Perhaps the submitter of this story should try "sometimes beats the G5, and when it does, it's by a small margin." Only one benchmark had the G5 being "trounced," IMHO.
Did he read the article? I did. Looks like the PC trolls are upset that the G5 can handle it's own and need to fuel the constant war of words. I'll bite. But for god's sake, get rid of such a grossly innacurate headline. It's not even close to reality and the poster has to know it.
"Politicians find new names for institutions which under old names have become odious to the people."
I am seriously considering getting a G5 just to run Matlab simulations. Where's the tests of something that may stress the hardware a bit, like Mathematica or Matlab?
These benchmarks are a bad joke. My pentium II or Athlon box runs Word pretty fast.
..don't panic
At best the P4 Emergency Edition is in a dead heat with the A64FX with the FX being the gaming / productivity king and the EE being the media production king.
Dissolve... Resolve... Evolve...
-----
---- "Excuse me. Where's the children's gun section?"
...Is that once again we're painfully shown how hard it is to benchmark cross-platforms. And how desperately most magazines fail in doing so every single time they try it.
I'm a Mac user, but I'm sorry to have to tell you: MacWorld doesn't have a fucking clue about benchmarking, they never had and aparently aren't learning anything.
There are about 8 different versions of Quake3 1.32 for Mac OS X for example, the latest one being a LOT faster than the first one and, judging from the sub 300FPS scores Macworld has repeatedly produced now, I must assume they use the first version (test5). Head over to www.quake3world.com/forums and check the Mac discussion for real benchmarks, you'll be surprised to find the G5 being much faster.
Then look at scores of G5s using 512MB RAM, vs Using 2GB of RAM. The difference is up to 10%!
The list would go on infinitely. Unless you want to include bias into a review, you really shouldn't benchmark cross-platforms without having a clue.
And I'm sorry to say that I think neither PCWORLD nor Macworld have one.
When will we ever learn to stop using terms like "whopping." What is "whopping" today is "wimpy" in a few years. It's been that way since computers were invented. Think about a new 386 computer with a "whopping" 8 MB of memory, or a whopping 700 MB hard drive.
Read the sentence again -- the one about a "whopping terabyte (1000GB) of physical memory" -- and you see that the sentence has the very same meaning if the term "whopping" is removed.
Quake 3 requires a console command to enable SMP support, which is off by default. Premiere only uses a single CPU. Word almost certainly also uses only a single CPU.
Every benchmark test is basically useless. It really comes down to what machine can help you get more work done or play cool games on. If you are into games, get an Alienware with XP. If you are do video editing or photo manipulation, Apple is the way to go. Linux pretty much can handle the rest. I would just like to see Linux preinstalled at the stores. Me? I'm running Debian on my desktop at work. And I'll be purchasing a dual G5 in the spring for home because I'm really wanting to get more involved in video editing and my iBook just doesn't cut it.
VirtualPC doesn't work on the G5, even though it's quite happy on the G4.
Your Mac got beat fanboy. Just accept it. That is the way it always has been, and always will be. Macs will be slower and more expensive, and wankers like you will make up excuses for why they really would be faster if only...
First off, you want to compare the same application to the extent you can, or is this a faulty methodology? By your logic, maybe they should've used Premiere Pro (which is far more stable and more powerful than its predecessor. Maybe you would like to compare ClarisWorks or OpenOffice instead?
Well, let's do a little examination of real benchmarks and use some deduction and estimation to get some better facts than you present. The first thing is this barefeats.com article comparing the dual G4 1GHz to various flavors of G5, including the dual G5. The Dual G5 scores 836 in the After Effects versus the Dual G4's score of 355. About 2.5x the performance. Great. Let's put that in our back pocket for now.
If one looks at the Mediaworkstation.com benchmarks for After Effects where they pit a dual G4 1.0GHz versus a dual Athlon MP1800+, the dual Athlons are nearly double the performance of the dual G4s in many benchmarks.
Now, a dual Athlon MP1800+ is limited not only by its lower bus speeds and non-integrated memory controller, but by its relative lack of internal registers as compared to the Opteron/Athlon64. You'd be lucky to get the equivalent performance of a P4 3.2GHz HT on the dual Athlon MP1800+ box. Yet the single-processor Athlon64/Opterons seems to be much faster than that P4 box according to PC World.
My point is that, if you want to compare your "gut feeling" based on extension of real benchmarks (versus your complete lack of evidence and questionable comparison methodology), I have the feeling that the dual G5 at best will approximately match the best single-processor Athlon64/Opteron configuration. That's a far cry from your claimed "trouncing" of the Athlon64 by the dual G5.
If you want to argue for the dual G5, why don't you simply state something such as the fact that FCP4 is, for now, the only editing program this side of a $30k Avid or $150k Discreet that will run on an average consumer-level PC? FCP is one of the de-facto standards in the editing world now. Regardless of the power of a computer, it could be a far more compelling reason for a certain niche of users to go Mac than Wintel.
One of the things I find really amusing is that everyone is using AMD originated technology except intel. Transmeta Efficeon, Apple's G5, AMD Athlon 64/Opteron, Nvidia Athlon XP chipsets and the Microsoft Xbox all use hypertransport.
That's got to make you think that AMD isn't just the technology follower like they used to be.
Exactly. That's really all this puffed-chest speed comparing comes down to. Glad I could amuse.
> If you compare the prices for the AthlonFX to the dual G5,
> you will see a similar price/ performance ratio
Yeah - if you're blind and look the other way you will.
How on earth do you figure that? A single Athlon 64 machine is way cheaper than any G5 and beats it in all the tests (single processor - 'cos obviously comparing a dual G4 to a single Athlon 64 is a useless comparison). A G4 may look cool, but nobody is ever going to argue that it gives a good price/performance ratio (well the might if they're a Slashdotter, but they'll lose).
Sorry, couldn't resist.
Okay, so far, all tests are fairly inconclusive about which system is really, overall, THE BEST. Nobody's done a truly fair test, but maybe that's partially because there isn't one that can be done...
/. whose main reason for not switching to Mac is because it's not fast enough? And I would express the same doubts about the opposite camp.
/Insignificant rant.
What this article says to me, though, is that it's okay to stick with AMD. They're doing fine. Which is good, since I can't afford to spend over about $750 on a computer. So I won't be getting any of this stuff for quite some time anyway.
If I DID have the money, it would still tell me to stick with AMD, because it does better with the stuff I use the computer for. That should be the basis of a buying decision.
Most of you people obviously aren't concerned with such silly details. You're concerned with keeping the war alive, the war against/for Macs! Honestly, is there any devout non-Mac-user on
But it's ridiculous to HAVE camps. Just get the computer that's best for you, let the other guy get the one that's best for him. When you argue, those of you that compare certain zealots to paid employees look a bit like you're waiting for your next paychecks, too.
End the silliness.
For quake3, the performance difference between a 128MB and 256MB card is non-existant. Quake3 does not come close to using 128MB, much less 256MB. In fact, the 256MB card is more apt to perform slower because usually cheaper memory is used in the 256MB version. The G5 got beat in Quake3 because macs suck at games.
Wow, you are obviously a brave person to call someone a fanboy, hiding behind that anonymous coward catch-all there.
While the grandparent post is definitely someone who at least admires Macs this does not make them an automatic fanboy. Apple was blasted for posting benchmark results that were in their favor, now that someone posts that the G5 got "trounced" we are not allowed to examine the test methods? Simply examining the test methods and pointing out possible problems does not make a person a fanboy.
Besides, no matter what kind of fan the grandparent poster, hype7, is at least he had the courage to post using a real account instead of hiding behind the ubiquitous anonymous coward...
Sapere aude!
doesnt seem too kosher, why almost all the PCs in the test have the benefit of RAID 0 volumes and the Apple should not... atcually I think its very odd I cant find any specs on the HDs in this article. the GFX cards are mentioned, the RAM is mentioned, but HDs are probably the most significant area where performance can lag. You dont know if any of the PC test machines are configured with ATA or SCSI or even Serial ATA like the G5. Maybe they made all the PCs have RAID 0 to even the playing field with Apples SATA Drives, but im not really sure that logic flies, cause a striped RAID array can dramatically improve a computers efficiency. And it still doesnt fix the point where we dont know what types of drives these are. Thats pretty shoddy testing or shoddy reporting. Maybe selective reportingsince talking about the Hard Drive Latencies isnt nearly as exciting as Quake benchmarks or muckracking. Its interesting to note that you need SoftRAID to do the same thing on a Mac, maybe they didnt feel like coming up with the change to buy a copy. Either way thats seems liek an awfully glaring hole in their analysis, and if the analysis is indeed valid, bad form not to provide these specs. And yes for all you Mac hard-asses, i can concede that premiere is horrible on the mac and a piece of trash in general (go FCP and AVID, id recommment vegas video on pc rather than premiere), word is a totally useless comparison no matter what (gee when im typing 60 wpm i really am taxing that computers capabilities). And if you buy an apple branded UNIX workstation to play Quake, u have serious inadequacy issues.
Btw there seems tobe a consensus that the G5's architecture really starts to shine when loaded with 2GB RAM or higher, Im nto sure exactly why that is, but it would be interesting to see all the test machines have their memory bumped up and then make a new comaprison. If i look i may be able to cite a benchmark test where actually increasing the G5's ram to 2GB (apples recommended amount btw) resulted in near 200% performance increase in certain test scenarios. AND I STILL DONT KNOW WHY PC/MAC COMPARISONS ARE BEING MADE. They are fundamentally so different that making a truly scientifically unbiased comparision is nearly impossible.
Why would anyone benchmark a system using and OS and software that wouldn't be used in the 'real world'?
How accurate could those benchmarks be? They simply wouldn't equate to anything that anyone could relate to.
I can imagine it now.. "Wow, this new Apple GXX Based computer is Super Fast on the Amalgamated Systems Wiffle-Waffle Test Suite! It totally beats the pants off the latest AMD Akronis 64-X2 based system!"
Three weeks later...
"Why is my Apple GXX Machine so much slower when rendering scenes then that AMD Akronis 64-X2 Based system? It was so much faster on the artificial test suite..."
Note, you can just as easily switch the imagined Apple and AMD based systems for the other...
There is simply no way to trust any kind of artificial test, like you are suggesting. The only thing to trust would be more real world tests.
If you ignore the other uses of a tool, does that make the tool less useful, or you less useful?
"There is, however, one thing that Microsoft will not allow Office for the Mac to achieve; and that is performance parity. To add to this, much of the codebase of Office v.X is left over from the good ol' days of MacOS 9 - reflected in the fact that Office is still a Carbon app. So, although Office on the Mac is extremely widely used, it's of dubious use as a means of comparing performance between processors. "
Comparing the performance of Office is a non-dubious means of comparing the end-user experience. As the test shows, if you are using Office on the Mac on your G5, you just won't be getting the same performance you could be on the x86 side. If that's not to your liking, you have to swap to a different office application.
It's not really apples to oranges, it's more how the user will feel about it. As you've pointed out, the user won't feel as happy with the performance of the G5 because MS has been dragging its heels with Office. That's not a fault of the benchmark...
--
Internet Explorer (n): Another bug -- that is, a feature that can't be turned off -- in Windows.
Oh come on now. Apple is only going to please it's own installed base withthe G5 machines unless they can break into the new super-computer-cluster market currently owned by lintel.
So what most inmportant are plain SPECMarks and the linpack matrix benchmarks done in double precision
floating point on a 64-bit linux install for the AMD-64 chips.
In any case it looks like Intel has really dropped the ball with it's Itanium white elephants.
Martin
Take a closer look at the numbers. The G5 got destroyed in the photoshop benchmarks. Read the part at the bottom, "LOWER is better."
Dirk Meyer, chief architect of the alpha, is chief architect for the Hammer (Opteron, Athlon64 series).
[RIAA] says its concern is artists. That's true, in just the sense that a cattle rancher is concerned about its cattle.
Ah, but being a zealot for one side or the other is so much fun! Nobody here really cares whether the benchmarks mean anything or not; The numbers are like "top trumps" scores used to fight your opponent.
:)
For the record, I switched to a Mac because it looks pretty and has a nice developer tool suite installed on it, not quite as good as MS Visual Studio but better than anything I've seen on Linux (while still remaining a big fan of Linux and a heavy Win2k user/developer.)
I agree that we need to see benchmarks of the systems running native 64 bit code. This should happen on both the Athlon64 and G5 systems however. I suspect you'll find the Athlon64 also benefits greatly from its 64 bit mode.
The previous G5 comparison benchmark, done at NASA vs. the P4 architecture, described performance per clock cycle. That's a weakness of the the P4's architecture due to design choices by Intel. This is not the case with the Athlon64 systems however. Athlon64 has a to gain when it goes fully into 64 bit mode. There are twice the number of registers, and SSE2 instructions have been added. This gives the Athlon64 similar vector capabilities to those found in the G5.
Note also that the Athlon64 systems use Hypertransport, which is to my understanding what gives the G5 its memory bandwidth. (Feel free to correct me.) While only the FX and Opteron models have dual channel memory interfaces, it puts these systems closer to a performance parity with the G5. The integrated memory controller may even give it an edge.
In short, both are formidable entries into 64 bit space for mainstream systems. I wouldn't assume that G5 will maintain a performance lead over the Athlon64 until better benchmarks have been run.
GPL: Free as in will
(Disclaimer: Article Not Read Thoroughly)
More to the point, did they bother to apply the AltiVec patch to Quake3?
If not, then did they similarly cripple Q3 on x86 by disabling all SIMD-like support?
What's the frequency, Kenneth?
>>Quake III, developed on, and for, x86 over 5 year period of programming research and enhancement. Later ported to OSX in a week by OmniGroup.
This benchmark may indeed be a steaming pile of crap. But.
It was a point of pride for John Carmack that Quake III was developed in parallel for Windows, Linux, and Power Mac platforms, was continuously tested on all three during development, and was released for all three nearly silmultaneously. All three ports share 90%+ of the same source code. It's not really honest to suggest that the Mac version hasn't been around for as long or hasn't received the same attention as the PC versions.
It has endured as a standard benchmark in the PC world because it has been shown to be a fairly sensitive indicator of system bottlenecks. Changes to, say, RAM speed, FSB speed, CPU speed, and video card speed all show up in Quake III framerates, which isn't true for many other games.
I don't know about OSX specific ports - perhaps it's true that it was ported to OSX in a week. That could indicate that Quake III was very well-put-together in the first place, making the port relatively simple, rather than that it was shoddily ported. The popularity of the underlying engine for use in other games would lend some support to this idea, as well.
Fair go! Apple's claims were about personal computers, not servers or workstations. The Opteron is not intended to be a personal computer processor; that's what the Athlon 64 is for.
I wonder why these benchmarks contain measurements for an Opteron but not for an Itanium 2 processor? Maybe it's because they're using 32 bit programs for the Athlon64/AthlonXP/Opteron/P4 measurements, which run like a dog on Itanium. Wait a moment, why is an article on 64-bit processors using 32-bit code for its performance measurements? Hmmm...
ahh, makes more sense. But I do think that a REAL combustion test would be a lot of fun to conduct.
Actually, you missed the great little blurb at the bottom of the benchmarks--"All machines were tested with 1GB of RAM and the ATI Radeon 9800 Pro graphics card; the Mac version of the graphics card has a maximum of 128MB of RAM, while the high end for PCs is 256MB. Most of the PCs used dual, RAID-striped hard drives; the Apple systems did not. We retested the Alienware Aurora with the 128MB Radeon 9800 Pro card and without RAID for more-direct comparison with the G5 systems."
So the magazine did try to compare as closely as possible between the systems... Further, the point of the article was that single processor AMD and Intel systems were handily beating the G5's AND cost significantly less, even with better video cards and RAID arrays...
That's pretty rich.
The only rigged tests are the Apple ones. G4 is 275% faster! No wait, it's slower, but the G5 is faster when we use these artificially low SPEC scores from our paid consultants instead of the official spec.org ones! Oooh, and the secret photoshop bakeoff that nobody can duplicate!
PC users don't claim the Apple tests were rigged because they lost (like you're doing), they are actually correct.
It got beaten by one computer, the polywell polystation2. Which seemed to outperform almost every other computer used in the test in all tests except the quake test and the search/replace.
I'm vey curious as to whether or not the RAID arrays had anything to do with the benchmarks on the other scores.
"Because on the desktop side of the story Linux is to OSX as SCO Unix is to... any other Unix out there."
Not true.. OSX was really nice when it was introduced, but I think the modern Linux desktops (kde && gnome) have actually started to eclipse it.
I was a Mac user from 1989 to around '99, and for a long time, I couldn't stand anything except for the classic MacOS. But then I discovered that people can actually use other things in an efficient manner.
One of the few multitasking benchmarks I've seen recently is at ExtremeTech. The Pentium 4 did much better than the Athlons, presumably due to HyperThreading.
> Where else can you find an OS that can run a great video editing package,
Windows!
> can run Photoshop,
Windows!
> etc., and can also run all of your favorite *nix apps, natively?
Windows! (with cygwin)
and Windows has about 9999999999999999999999 times more apps and games than MacOSX.
Even Linux on x86 is good, since it can run Windows apps natively via wine.
Any news about Apple porting McOS X to post-x86 64b uProcessors (int3l / AMD)???
You have been a great audience!
Cheers!!!
Summary: - Premiere is about the slowest video editing program on the Mac. Carbon, singlethreaded, not really MP aware. - Word on the mac is far slower than Word on the pc. Again, see above, plus to make it funnier Microsoft based Word X on Word 2001, right down to leaving in the 31-character filename limit. Not exactly an optimized program. - and Q3... Their numbers on the dual G5 are low. I have one; and they're either using the wrong version of Q3 for X (i.e. generic vs. altivec, or the 1.31 revision instead of 1.32b), or they're flat out lying. Interesting that their tests don't see much difference between the 256MB and 128MB cards on the PC side, as I remember Q3 being very graphics-card bound. Here it seems to mean that +128MB=+3PFS.
for the price of 1 Apple dual g5.
I dont have $3000 to spend on a computer.
I can get an athlon 64 system for less than
a thousand dollars.
Apple doesn't make the g5 processor, IBM does.
AMD doesn't make an Athlon 64 computer, everyone else does.
> Well-done GUI
ahhahahahahhahaha hahhahahahah hahaha
hahahahhahah ahahhahahaha
ahghahahahhashdshhsd hasdhashhash
hahshsahas hsahshahsahsahsahhsa
OSX is a GUI made by homosexuals for homosexuals.
I have been a long time reader of PC World, and have much respect for your magazine. However, I am yet to see a more abject review than the "64-Bit Takes Off" what was presented in your November 2003 Edition.
Dude, you're a hardcore Mac user. You MAY read PC World regularly, but I doubt they'd believe it. *I* don't believe it.
Let's start with the choice of Microsoft Word. Undoubtedly a widely used piece of software, and Microsoft incredibly allowed Office v.X for the Mac to receive a number of features that the Windows version is yet to receive. There is, however, one thing that Microsoft will not allow Office for the Mac to achieve; and that is performance parity.
Let's start with the choice of Photoshop Apple regularly uses. The PC guys bitch that it's tweaked for the Mac, blah blah blah, and the unwavering Mac zealots scream that Apple chooses Photoshop for nearly every single test it does because that's what the common man uses. Same applies here. Everybody uses different software, I try not to use either of these. But they have to use SOMETHING. You demand that the Mac platform be given all the advantages possible in software choice and configuration while simultaneously arguing that the PC's advantages are unfair. Get a life.
Next. Premiere. This is what stunned me. There is a reason that Premiere doesn't work very well on the Mac. This is because absolutely nobody who does video editing on a Mac uses it. Period. Final Cut Pro wipes to floor with it; not only in functionality, but performance also. Of all the ways you chose to benchmark the G5s, this surprised me the most.
When Apple releases Final Cut Pro for Windows, I'm sure they'll use it in their benchmarks, notwithstanding Apple's mandatory speed slowdowns added to the Windows version...
In the Quake test, the Mac was hamstrung by the fact that it only had a 128MB video card in it. I also may be wrong in making the assertion, but doesn't the 256MB ATI 9800 Pro run at a faster clock rate than its 128MB cousin? This would account for quite a performance differential. Despite the fact that Macs aren't really known for games, no other computer with a 128MB graphics card beat it.
As Jon Rubenstein always says, it's not about processor speed, it's about "total system performance". When Apples start shipping with the same hardware that PCs do, they'll use that.
The next test was Photoshop. This is the one app you benchmarked in which some 64-bit optimisations have taken place for the Mac, and is also an app that many people use on the Apple platform. In this test, the G5 beat everything on offer from the x86 world by quite a handy margin.
So the one test you have no problem with is the one test Apple won. Interesting. Don't you see a bias in your thinking?
What makes this even more impressive is that the G5 system you benchmarked is running on a stop-gap operating system release from Apple. OS X 10.3, codename Panther, has been specifically designed to take advantage of the G5's 64-bit CPU structure; it's out in barely a week.
Yeah, and Longhorn is due out in a year or two. Perhaps we should wait for THAT to be perfectly fair. NO! The test uses what is out RIGHT NOW.
I would certainly be interested to see a re-run of the tests, if you think that this feedback is valid. Cross-platform benchmarks are notorious for being difficult to standardise; I do, however, believe that if done properly they can be both useful and interesting.
Translated: I would certainly be interested to see tests in which Apple win every event hands-down. I will welcome these and only these as just and righteous. Any tests in which Apple loses are obviously flawed somehow. If Apple loses the next set of tests I will claim that cross-platform benchmarks are notorious for being difficult to standardise and in your case are obviously incorrect, as the Mac did not win. I do, however, believe that if done properly (you didn't, since the
My memory is fading on this issue....but was there even a Premiere 6 version made for MacOS X? If there wasn't...then what you see there is a Premiere 6 test run from MacOS 9 Classic. And that would be slow from the get go, and of course make the G5 look so poor, which if you have used FinalCut Pro on the G5 even a little bit, you'd see the big improvements made over that running on a G4.
I'd like to see them run tests with equal version applications...something new too. How long has Premiere 6 since been sunset? Cripes.
First why are they called bench marks? Second From everything I have read or seen on the G5, G4, P4, Opteron, Athlon 64 is that everyone has found different numbers and no one believes them one one way or another. the supposed apple bench mark lie was later supported by NASA tests yet after that other tests came out that said it was still a lie... So when it comes to this test. I will have to say I will put it in the the same pile as the rest.
"Despite the fact that Macs aren't really known for games, no other computer with a 128MB graphics card beat it."
What?
Can you read a chart? The very first AMD system has a 128MB 9800Pro.
G5-1.8GHz: 147/141
G5-DUAL-2GHz: 294/207
FX-51: 335/257
"Photoshop... the G5 beat everything on offer from the x86 world by quite a handy margin."
The SINGLE Opteron-2GHz system beat that DUAL-2GHz G5.
SPECfp base2000
2Ghz G5 - 840
Opteron 146 (2Ghz) - 1291
SPECint base2000
2Ghz G5 - 800
Opteron 146 (2Ghz) - 1170
SPECfp rate2000
Dual 2Ghz G5 - 15.7
RackSaver RSN-1164/op (1.8 GHz Opteron) - 22.5
SPECint rate2000
Dual 2Ghz G5 - 17.2
RackSaver RSN-1164/op (1.8 GHz Opteron) - 24.0
These numbers seem to back up the PCWorld tests.
oh? and way back when macs were benchmarked as being so much faster than PCs, the benchmarks only compared a mac with scsi hard drives against a pc with ide drives (this is way back), was that one sided?
face it, both systems are too different to benchmark.. any software you compare is going to be equally about benchmarking the software's optimization, maybe even the compiler's optimization, rather than the cpu.. Both platforms cannot run the exact same code, so it brings in too many outside variables into the mix to really say which chip is faster..
its like comparing apples to oranges, pardon the pun
To do a benchmark correctly, one would install Linux on all the systems. Then run Gimp, Open Office, Video Gimp, and a few other Linux apps.
This would give you more correct results than Window against OS X.
Does Linux run on G5's?
The above is not worth reading.
I have an excel spreadsheet with an optimization problem for determining employee schedules I run.
I ran it on a 90 mhz Pentium with 32 MB of RAM at work and it took 58 seconds.
I ran the exact same problem on my new dual 2 Ghz G5 with 1 GB of RAM, and it took 79 seconds.
Thus the original Pentium is faster than the fastest Mac! By a lot! Boy, am I mad I wasted all this money on a Mac!
I'm not kidding about the times. I'll send you the spreadsheet, if anyone wants to replicate these results.
I know there is no such thing as a direct-comparison cross-platform benchmark, but I figure benchmarks should do one of two things:
1. try to do a spec test, to gauge the "raw" performance of the machine that well-programed software could use.
2. compare real-world applications to see how the machines perform in some actual usage scenarios.
This looks like an attempt at 2, but it's not.
If anyone thinks Word runs unacceptably slowly on the G5, they must be doing some crazy word processing. Get Nissus Writer or something powerful.
And as for Adobe Premier, try comparing it on PC versus performing the same actions on the same movie in Final Cut Pro on the Mac.
If you're measuring a FPS with PC's with a 256MB video card, throw in the Mac with Radeon 6800.
I'm sure that if I had acess to the Athlon 64 machines and took my time, I could come up with a set of "benchmarks" that shows the G5's much faster. - Phat Tony.
Can anyone tell me how to set my sig on Slashdot?
AMD and Intel are looking over their shoulders. In reality with OS X thrown in they should be looking forward at the leader, Apple. The fact that these kinds of tests, skewed or not, are being run and looked at says a lot for the G5. How far have we come in a year!
My karma is getting better everyday.
Processor comparisons are only meaningful if some attempt is made to provide parity in other subsystms, such as mass storage, and graphics cards.
When you read the article, you'll see that all of the PCs were running with RAID arrays for mass storage. The G5 was not, even though RAID can be had for a Mac OS X system.
The G5 had a video card with half the video memory of the PCs.
In what sense can this be seen as a comparable test?
"We hamstrung a G5 by giving it a slower disk subsystem, and a video card with less memory, and it didn't perform as well as these workstations." Duh.
What we are talking about here are two completely different software and hardware platforms. No objective comparison can be made. Let's just stop this pointless bitching. Both sides are to blame.
On a side note:
I am not a Mac user. I love OSX however. If OSX was available for x86 I would be first in line.
er... according to the article, the Athlon system came in at $200 cheaper. When you're dropping $3000+ on a computer, an extra $200 isn't way cheaper.
T Money
World Domination with a plastic spoon since 1984
How nice. I, personally, like my area very much (Upstate NY, just south of the Adirondacks). However, the nearest Apple store, until just recently, was in Syracuse, nearly an hour's drive away (and that was in the back corner of a CompUSA, or something). Now there's a Best Buy about 20 minutes away. The closest "real" Apple store is in Crossgates mall...in Albany, 2 hours away.
If you have multiple Apple stores near you, good for you. You probably either live in Souther California or somewhere between Boston and DC. But most of the country doesn't have Apple stores all over like that. But that doesn't mean we should move.
Dan Aris
Fun. Free. Online. RPG. BattleMaster.
No... not evar! ALl mac users are teh ghey!
Flame on!
Gabriel Ricard
"claims to be the ... only 64-bit processor for the desktop and workstation market"
Whatever happened to SGI machines with MIPS RISC chips? AFAIK, the O2 had 64bit processors. They may not be as fast as a G5, but they were still 64 bit.
The truly funny thing about all this is all the Macultists are hopping mad and jumping all over this.
Despite the obvious fact that we will not know the "truth" until we have 64bit versions of these programs, even then, if AMD (or, god forbid, Intel) proved to be faster, they'd still find SOMETHING wrong with the tests. Even if they tests truly were fair, they wouldn't believe them.
It's not like this is going to change anyone's minds: if they like PCs, they'll use those, if they like Macs, they'll use those.
Wouldn't you expect a person that buys an Alienware box to shop at the Men's Warehouse?
If some no-name company rushes an Opteron box out the door a few days before Apple's announcement, can you really gloat about it? I mean, fuck dude, get a life! While Apple is out there mass-producing killer machines, these jokers are niggling about whether it's the first one or not. This is not a patent application, loser--it's marketing. Get it through your thick head.
Look, Objective-C will always be slower because of OO late binding issues.
Wouldn't it be much more fair to test the CPU and I/O systems with SPEC and TPC-C?
Oracle is available for OSX. I'd like to see some TPC scores for this platform.
If you knew anything about Quake you would realize it won't even use 32MB of memory on the video card. Guess what, Opteron and Athlon 64 is much faster than the G5. It has to be with a faster bus and onboard memory controller. Stop trying to deny that.
ignorance is bliss. googlefiberatx.com
It is generally understood that on Solaris UltraSparc, the 32-bit Oracle version will outperform the 64-bit version for databases that do not need large SGAs (shared memory/db cache).
There are LOTS of cases where a 32-bit app is better on a 64-bit cpu.
When Apple releases well documented (and fair IMHO) benchmarks everybody screams that benchmarks don't mean anything. When a magazine called "PC"World runs benchmarks - people think it is the word of the Lord. Common people - use your brains. Both AMD and Apple have great processors. Both top-of-the-line systems cost about the same. That means that there is real competition in the marketplace - which is ALWAYS a good thing.
...And when they came for me, there was no one left to speak out for me." - Martin Niemoeller (1892-1984)
He isn't saying that the RAM is what's increasing the performance, it's that as far as he knows, the 256mb card has a higher clockspeed. That would certainly affect its performance.
There is no gravity...the earth just sucks.
I was reading your post wondering how you may know Tom Mainelli or whether *you* were Tom Mainelli (it'd be pretty coincidental, you must admit) when I realized what you were getting at ;)
There are not enough exclaimation points in the world nor bold-enough text to truly describe how tickled I was by your post. Your wit and clever prose rivals that of the great Ms. Parker.
"The Windows guys don't do it, and the Mac guys mock them as unintelligent, unimaginative copycat second-class citizens."
Well you know what they say, if the shoe fits...
Or, more accurately DEC arrived well before the party.
http://en.wikipedia.org/wiki/DEC_Alpha
64 Bit workstation machine, introduced 1992, even ran Windows NT.
First 64 Bit desktop indeed!
The G5 got beat in Quake3 because macs suck at games.
You mean because Q3 is x86 naitive code that was PORTED to run on mac.
And how is that different from any other popular game? Hence why the G5 and ALL macs suck at games.
The funny thing about the PC World review is Apple actually beat out Athlon and Opteron on the only test that counts, Photoshop. Everyone knows that Premiere on the Mac is a dead App not being updated in over 2 years and poorly written at that. PC World having to use Premiere as a benchmark is laughable and desperate. They should have used Final Cut Pro (even Express) against the latest version of Premiere performing similar tasks. That would have been a fair test of real word Apps. I don't know who would have won this test had it been done right, but at least it would not have left a stink. Shame on you PC World for giving into childish tactics!
One user activation key for everything in the VMS suite.
You just need the media.
That PC World test (as they admitted) simply used the testing of MacWorld for the Quake 3 scores of the G5s. MacWorld didn't do their homework and used a lesser build of Quake 3, and now PC World just repeats the same mistake! Doesn't anyone in journalism do things without being sloppy or cutting corners these days?! BareFeats is a one-man operation (with some assistance) and BareFeats knew to use build 1.32b instead of build 1.32. BareFeats shows the dual 2.0 G5 beating all those other systems, with the G5 scores of 339 and 342 FPS at 1024x768.e Feats' P4 scores are in line with PC World's as well!
http://www.barefeats.com/g5b.html
Bar
Our new herbal MHZ enhancer will make her moan in appreciation! You'll feel like a real man, instead of a little boy! Maybe she'll even stop using Windows behind your back!
Well that is just plain wrong. The article says the Radeon 9800 Pro was used for both and just the memory size has changed. This would mean core clock speeds are the same for both. As far memory speeds, they are most likely the same, but if anything, the 256MB version would use slower memory to cut costs.
And ZDnet.com, Computer Shopper, and blah blah blah? I remember I saw something alike from PCMag. Am I right that they belong to CNet?
There is a spark in every single flame bait point.
When Apple releases Final Cut Pro for Windows, I'm sure they'll use it in their benchmarks, notwithstanding Apple's mandatory speed slowdowns added to the Windows version...
*cough* QT for Windows is purposely hindered BY windows *cough*
As for the rest of what you said. Uhm, repetitve is the word I'm looking for...
Apple never said it was the first 64bit workstation seeing as SGI and Sun have had 64bit workstations for what 10 years now. Before you open your mouth get your facts. Also Final Cut pro smokes premier for speed across the board. Its a known fact that Premier is NOT well coded for the Mac.
These are benchmarks, 99% of the time they will be unfair/biast. And I've said the same in previous tests where the results were the other way around. There's no point in making a serious issue about it. If you're concerned that much with gaming performance get a gaming console, not a PC.
Their tests are largely I/O-bound and video card related, too. It's a system comparison, not a processor comparison. If you have different I/O or video card, you'll get different results.
So who has the faster processor? Who knows. I suggest you buy the system you like the most.
I think the /. community should put more effort to stop the FUD on Slashdot by modding-up the responses to ignorant and stupid posts, such as the MidnightBrewer's. As it is now, people will only read FUD, something that degrades the overall value of /. news...
Cheers!
You know the second the combustion results would be posted you'd get:
"APPLES DON'T COMBUST AS WELL BECAUSE THE COMBUSTION PROCESS IS X86 OPTIMIZED!"
or
"PC DON'T COMBUST AS WELL BECAUSE THIS STUPID APPLE ZEALOT DID THE COMBUSTION!"
If I was a PC zealot I would still be disgusted at PC World. There tests do nothing but illustrate the bias in the industry. Once a test is published like this, it destroys the credibility of the PC base. Comparing a dead-end aging App like Premiere on the Mac which was ported from an old obsolete Premiere on Windows. It gives the impression that the author was hard pressed to find a slower App on the G5.
How hard would it have been to compare the much more popular Final Cut Pro against the latest version of Premiere on a task by task basis? Not hard at all. What a shame!
I had a chance to try a 3.2 GHz P4, an Opteron, a dual G5, a 3.06 GHz dual Xeon, and an Athlon 64, using a suite of productivity and multimedia apps. Here's the verdict:
The P4: Very very fast.
Opteron: Super fast.
Dual G5: Really really fast.
Athlon 64: Totally fast.
Dual Xeon: Nice 'n fast.
Telling results! Unfortunately since I have put so much effort into accurate, impartial analysis of the test results, and participating in all the arguments with disbelievers and naysayers, I have not had a chance to get any work done for months. But who wants to use CPUs for productive tasks anyway, when it is so much fun to sit back and watch them "trounce" each other!
The tone of the letter you wrote was great; it is nice to see a well behaved advocate. However, you read the benchmarks wrong.
The only place the G5 did better than any of the x86 CPUs was when they used two G5s. All of the x86 configurations were single-CPU machines. The single G5 configuration had worse numbers than any of the x86 counterparts. So your comment:
was wrong, as was your comment about video memory.
Caveat: I'm not trying to validate the benchmarks, but I get sick of lazy moderators.
Eclispse? Really, I can't stop laughing... wait, give me a minute to breathe.
Linux desktops still don't begin to hold a candle to Mac OS X for cross application integration and consistency. The nightmare that won't die that is the world of X windows applications is so completely inconsistent in GUI design, that switching between applications is like moving between different operating systems.
Nice try. Eclipse. Yeah right. Linux desktops will catch up with where Mac OS X Jaguar was a year ago by next total eclipse of the sun, which is due, when in a few decades? But by then of course, Apple will have released a half dozen major upgrades to the functionality of Mac OS X.
Amen. These tests are irrelevent as performed.
FCP doesn't run on Windows, and hence, not on Linux with WINE either.
Remember, your parent said "a _great_ video editing package" not some Windows also-ran.
The next test was Photoshop. This is the one app you benchmarked in which some 64-bit optimisations have taken place for the Mac, and is also an app that many people use on the Apple platform. In this test, the G5 beat everything on offer from the x86 world by quite a handy margin. The Polywell Polystation with Two 2-GHz Opteron Model 246s won that one in both the 50MB and 150MB. It beat the Mac by 1 second in 50MB and 4 seconds for the 150MB. And final cut pro isn't available for Windows and Mac OS X won't run on an Athlon so how would you test that? Also, this is testing the highest end pcs for both the Apple G5 and the Athlon 64/Opteron, so why shouldn't they use the best video card available? They aren't running the 64 bit version of windows so why didn't you mention that in your email?
If you need that much CPU power, you presumably have some compute-intensive app to run with it.
Or you could just benchmark how long it takes Java to redraw a window.
So then, who really cares if the athlon64 is X% faster at this task or Y% slower at that task, when compared to the latest apple. The point remains that X and Y are not 250. If you sat down at one and ran photoshop you would not go to the other and think "wow this is really slow". You'd have to sit there with your stop-watch to see if one or the other was faster.
No one switches to Apple because they are faster than AMD/intel. Similarly noone switches the other direction because of speed. You switch (or continue using whatever you're using) because you enjoy the environment, or because you are impressed by the general look and feel, or because all your friends use a (foo) and you want to use a (foo) too.
The cool thing about both the athlon64 is that it's noticably faster than the duron under my desk, and I can run the same software. The cool thing about the G5 is that it's noticably faster than the G4 under my buddy's desk, and he can run the same software he already uses. They are both really nice advances, and to compare the two is stupid. Even if the athlon64 was 85% faster than the G5 at every task my buddy still wouldn't buy one. My evidence: My Xeon 2.4 IS 85% faster than the 500Mhz G4 he has now, but he still uses the mac.
Aside from that, compare $3500 machines is great for the "my mac penis is bigger than your amd penis" crowd, but for those of us who want to spend $1000 for a computer, it's still athlons, celerons, and iBooks. In the grand scheme of things a comparison of consumer-priced systems is far, far more interesting, as the volume is much, much higher. When 64-bit athlons and G5s make their way into inexpensive desktops and laptops, then lets all get worked up about it.
"Despite the fact that Macs aren't really known for games, no other computer with a 128MB graphics card beat it."
That is so not true! The alienware aurora with 128mb vram beat the G5 in QIII fps. That would be the first athlon system listed in the table.
That said, I don't know that I agree with this idea that the athlon killed the G5 in every category. It looks to me more like for each category there was an athlon system that could beat the G5, but no single athlon system had a speed advantage in all the categories. In other words, mixed results.
Second: Yes, systems using the Opteron were for sale 2-3 months before people started taking delivery of Power Mac G5's. But... you do, of course, realize that AMD positioned the Opteron as a server chip, yes? Let's see... a 64-bit server chip. Hmmm. Uh, no, I can't really say I consider that a "first." You'd have to ignore a lot of IBM POWER chips, Sun SPARCs, HP PA-RISCs, and MIPS things over the last 10+ years.
But oh, you say! BOXX took those server chips and put them into specialty rendering workstations, some of which can sit on a desk! Isn't that a first? Hmmm... seems to me that at least 7 years ago I was working on an SGI Indy, and even with that goshawful-heavy huge monitor on it, the desk didn't collapse.
So, yes. AMD's Opteron -- intended for an entirely different sort of system than the PowerPC 970 used in the PowerMac G5 -- started becoming available a few months before people started getting their PowerMac G5's. This is every bit as relevant, and apt, a comparison as pointing out that sometimes one automaker will introduce their large new SUV at a different date than another automaker's compact coupe.
The Athlon64, on the other hand, is targeted at desktops, and looks like it'll outperform the G5. However, from the time the G5 launched to the time the Athlon64 launched, it sounds to me like Apple actually did have "the fastest, most powerful personal computer," and that was enough time for them to convince over 100,000 people to order them.
(As the The Register points out, Opteron sales seem to be off to a fairly modest start . 5,000 or so units in a quarter? That's what, about 1/20th the rate of PowerMac G5 sales?
Self-fullfilling prophecy.
People think Macs suck at games.
Nobody makes games for the Mac.
Any games on the Mac are therefore not native, or not optimized.
These games suck relative to their native ports.
GOTO: start
For a review of the potential of a new platform, we'd like to see what's possible, not what the past has shown us. If the G5 could be the greatest gaming platform in existence, maybe the world could start to shift that direction.
Doubtful. Then again, so is Linux on the desktop.
The G5 is available TODAY and you can go out and buy a G5 Mac if you want right now. Where is this "Athlon64" cpu? Vapourware. Gotta love you PC-heads with your constant hopes and dreams about beating Apple even though you know it is impossible. Keep dreaming though, penguinheads, and maybe one day we'll let you use our "grown up" computer: Mac OS powered G5! Still the fastest, most advanced computer money can buy.
If your definition of 64-bit is a 32-bit operating system around a 64-bit chip, then the G5 is a 64-bit platform. Mac OS X 10.2.7 (and the upcoming 10.3) is not a 64-bit operating system. This is particularly frustrating because Apple's marketing machine has very carefully crafted their message to make a reasonable person believe the operating system is 64-bit, especially if you download and read Power Mac G5 Tech Overview (PDF). Apple says about the G5 version of Mac OS X that it runs all of your software -- and runs it faster -- with a version of Mac OS X Jaguar specially tuned for the PowerPC G5 processor, providing a seamless transition to 64-bit power. That's only the beginning of the smoke and mirrors. The 64-bit power only gives users two things: the operating system can address up to 8GB of RAM, though user programs are still limited to 4GB, and some of the G5 numerical hardware is available with a special version of GCC (3.3). That is very far from what I thought. In fact, we returned the G5 we got last week for a full refund (didn't have to pay the 10% open box fee either), after about 2 hours on the phone. Buyer beware.
Also consider the source of the article. PC World are blind, rabid Mac-haters. Show me some third party benchmarks and I'll give a shit. Well, no, still probably not even then. Point being that these stupid holy wars mean nothing, absolutely nothing in the real world. Let's play out a scenario:
Geek: My Mac is better than your Winblows machine!
Hot chick: Um, right.
How about another? Ok.
Geek: Linux rulez all! BSD is dead!
Boss: I don't care. We bought 5000 Windows licenses. We use Windows.
Geek: I refuse to use Windows, it stifles my Linux fag superiority complex.
Boss: Ok, you're fired.
This is about how much these things matter. Get over it.
"...the benchmarks should speak for themselves."
I think 5 pages of comments in under four hours say otherwise. Honestly, when have benchmarks ever spoken for themselves?
I wish that my inferiority complex were as good as yours.
-RenderHead
Someone should point out that the Mac was running on 2 CPUS not 1 like the Athlon 64 and Athlon FX
The Mac has two CPUs, but I'm not sure that both were actually being used in any of the tests. As far as I can tell, none of the tests used was multi-threaded, so will only use one CPU. A dead giveaway is when the dual 2 GHz system is just a little faster than the single 1.8 GHz.
Now if they were to run all tests simultaneously... that's a result I'd like to see.
I write this note just after reading the article, some posts I completely agree with, and making some google searches:
- the comparison is at least strange because the test is based on Premiere (not so much a success, and especially not now) and MsWord (really not a perf. master on Mac), so IMHO only Photoshop stays as a comparison (why not comparing MP3 encoding or easy to run stuffs like that?). The FPS in Quake is comparable but really not that much interesting as depending on such configuration mainly on the graphic card.
- the PCs used and cited at the beginning (Alienware, Falcon, Voodoo, ABS) are all mono Athlon, most of them $3500, except the ABS at $1900, are all mono Athlon 2.2Ghz, comparable to the Apple 1.8Ghz mono G5, $2400 by Apple (just do the math, Athlon speed per second/2.2*1.8 are really nearly the sames as the G5).
- The "winner PC" is a strange "Polystation Two" which is not really on Polywell catalog, but close to a Polystation 2020 (dual Opteron), priced around $3200 (one Opteron) or $4100 (dual Opteron). Whatching at the results, so close to the dual G5, I really believe a dual 2Ghz has been used for the test (>$4500)
This leads to real doubts on the serious of this comparison, but if we just clear the record and base the comparison on the Photoshop indicators/ the machines/ the prices/ the manufacturers we first obtain an interesting conclusion:
"The Athlon 64 (and Opteron) is really close to a 64 bits G5 and the machines based on those processors and adapted architectures are comparable in real-life use."
We also obtain some interesting information in term of PC/Mac comparison as the 3 first (Alienware, Falcon, Voodoo) are quite comparable to Mac mono-G5 1.8Ghz (clean, well built, powerfull), even if they are not made by a "big name" with worldwide distribution. Well, they are also more expensive ($2400 for the G5, $3500 for the Pcs, even the 1.8/2.2Ghz doesn't make the maths). The comparison is different with the Polystation (not nice looking, close to a server, clearly not the engineering of the Mac) for which the price difference is even bigger ($4000-$4500 for the PC, $3000 for the dual G5 2Ghz).
My conclusion for this odd "audience seeker" article would be:
In the top-desktop offer, real bad news for Intel in term of speed, very good news for Apple in term of price.
ClaudeBBG
As I mentioned in another post, one should also keep in mind that none of the tests used was multi-threaded (AFAIK -could be wrong). Thus the "dual G5" was really running as only a "single G5". The giveaway is to compare the dual 2 GHz results with the single 1.8 GHz results.
Now let's run all those tests simultaneously and see who really wins!
the G5's tested had 1/2 the ram and "standard" HDD instead of RAID.
That's not a comparison, that's apples and oranges.
-elmar-
All right, I get it; the Mac benchmarks were done by Macworld. But the point still stands that the comparisons were done by a primarily Windows-oriented magazine, and several people have called the methods into question (eg, almost none of the apps even *used* the second processor, and the PC TestWorld suite, or whatever, doesn't run on the Mac, etc). From what I can see, this cannot be called an unbiased comparison, any more than any paid for or done by Apple, Dell, M$, or Sun can.
Dan Aris
Fun. Free. Online. RPG. BattleMaster.
Not that anyone is counting, but where I work I couldn't care less about certain benchmarks or assertions. All I care about is render times. We've tested both Xeon's and P4's and G5's to determine which was going to be the best bang for the buck... and truth be told, the dual G5 was considerably faster for the price than either the P4's or the Xeon's in After Effects and Combustion. It was significantly faster in some Photoshop tests and equal/lesser on others and pretty much on even ground with the P4's in Lightwave renders. None of the above tested apps are 64-bit as of yet. The "G5 optimized" plug-in for Photoshop is barely a hack for certain features, it certainly does not make Photoshop 64-bit, only certain plugs and functions. For us to purchase an equivalent Xeon box it was going to run us $1,200 more than the G5, so we've opted for the G5. From my understanding Panther will increase the G5's overall performance from 10-40% depending on the task. Even though Panther is not 64-bit throughout it does have tweaked math libraries and a new kernel which gives it a push. Being that the G5 is essentially running 95% 32-bit code one should be a little more impressed with the benchmarks. When they release a full 64-bit OS I'm sure we'll see the beast unleashed. Sheesh. They are just tools after all folks. Get over it.
When will we quite caring which one is faster? OS X is what I want. I don't care what hardware I have to use. Everything currently on the market is fast enough for me and OS X is the nicest OS available so I want whatever I need to use OS X. It's that simple. Sure I can go with AMD and get a chip that may be faster than a G5 but what am I going to run on it? Windows XP 64? No thanks. Linux? That would work, but wait... OS X has a lot of the features that I like in Linux. It has features that I can't get in Linux. It costs more, but to me the extra price is worth it.
How is it that one careless match can start a forest fire, but it takes a whole box to start a campfire?
...for the ones that tests a particular program's performance.
If I never use Word, Premiere, Quake III, or whatever is being benchmarked, then what good does the benchmark tell me? There are too many programs that use processors in different ways for such benchmarks to be generally useful. If I'm going to be running programs A, B, and C, then the only benchmark that's useful to me is one that separately tests programs A, B, and C on different processors, and gives me the individual results for each. If processor #1 is better on programs A and B, but worse on program C, then saying that it's "overall" faster is still meaningless, especially if I don't use A and B nearly as much as I use C!
I'm beginning to think that benchmarks need to be done on a software-by-software basis. Game performance is irrelevant if I'm only using Final Cut; word processor performance is irrelevant if I'm only using Matlab. Benchmarkers should stick to individual program performance, varying only the hardware used.
And so you all don't think I'm biased in favor of Apple, I don't own any Apple products, and I never have, and I probably never will. My machines at home are a P3 and an Athlon XP running Red Hat and Win2k, respectively.
"Destroy science and religion. Science would re-emerge exactly the same; but not religion." - Penn Jillette, paraphrased
You mean because Q3 is x86 naitive code that was PORTED to run on mac.
Uhh, no. Thanks for spouting though. Windows/Mac/Linux were all developed at the same time. The Mac version actually came out first. But please, don't let FACTS stand in your WAY.
Is anyone else sick of these online ignorant spastic ravings from such imbecilic internet intransigents?
uh, 1st off, quicktime, premier, word, and quakeIII all dont have 64-bit support. premier (as far as i know) and word both dont have dual proc support. Computer Magazines suck at everything besides telling what prebuilt computer incompetant "tech" shoppers should buy Windows XP 64 is still in beta and the G5 is only running a tweaked, special version of Jaguar. Panther offers more support for 64-bit. OS X overall is not a true 64-bit OS. Who in there right mind would run a dual-G5 rig specifically for Word and Quake III? you get a Mac for graphics and video design, not typing up documents (however important that may be, not the primary goal) Quake III is OLD. it doesnt even touch the Radeon 9800's potential is a pretty useless benchmark. Who would buy a $4000+ machine (incl. monitor) to play quake 3? now i realize they had to make do with what is currently available, but this is the shittiest benchmark you could ever do to compare a Mac and a PC. why even do the article? why not wait a few months and do something that actually indicates performance? did i mention Computer Magazines are blind? this was a completely pointless article, there are way too many variables. also, dont tell me that they tested it using software that most people would commonly use. bull. wait 4-5 months when all the software has 64-bit support and then run the benchmark.
why are people comparing these two?
The p4 isn't capable of SMP configurations. So comparing them means only talking about single processor machines.
Thing is, people who are hungry for performance aren't going to go for the single processor g5, they're going to buy the dual. People who aren't so concearned about the most horsepower they can get, but want a new bus and 64 bits for apps that come out 5 years from now (mac users keep them that long) aren't going to get that from the p4.
So really, the p4 benchmarks seem of dubious utility against the g5. Anybody considering a g5 mac probably has needs that a p4 equipped machine won't satisfy.
The relavent comaprison are Xeons and 64 bit SMP processors from intel and AMD. The P4 doesn't fit here.
"Let him go, Ralph. He knows what he's doing." --Otto Mann (simpsons)
would be to put two video/graphics power users in the same room one on a AMD/PC and one on a G5 and monitor their vital signs and brain waves. Give them a creative task and a timeframe for completion. Something real world like that. It could even turn into a Reality TV show!
My karma is getting better everyday.
Was sure I replied to this before...
anyway, didn't mean you amused, I agree with you as you can tell. Was the post I replied to that was amusing.
Still, if it makes you feel better you are highly amusing.
Sorry but this is incorrect. The G5 only uses Hypertransport as a interconnect to PCI-X. The G5 System controller has a built-in 128 bit 400 Mhz Memory Controller the supports DDR. HT is not involved in the memory path at all.
Here are some more details:
http://www.apple.com/powermac/architecture.html
You need two G5's to come close to any of the other chips including pentium 4. So that means that the other chips are twice as fast as the G5.
Fast Federal Court and I.T.C. updates
The G5 got beat in Quake3 because macs suck at games.
I'm one of those "Mac guys" and if someone came to me and said they wanted a computer for playing games I'd tell them NOT to buy a Mac. They're fine for casual games, but if that's your main purpose, a Mac just isn't the right tool for the job.
I use my Mac for "computer stuff" and got a PS2 for playing games. I don't see any need for games on the Mac, beyond playing on chess servers or Yahoo games and whatnot.
I really don't understand why people want to use PCs as primary game machines, myself. A console and games for a few hundred seems to beat keeping up with the latest CPUs and video hardware ($$$) to play the latest PC games. OK, I know why people do it, I'm aware of the advantages - I just don't relate.
Every time my windows box has to go down for it's weekly security patch, that's time wasted. Everytime I have to go into my routers, block new ports, turn ports off in windows and then run tests to ensure I haven't been infected, that's time wasted. Every time I have to install drivers because I plugged a fucking mouse or USB hub into my computer is time wasted.
It doesn't matter how small amount of tiem it is, it's more time wasted and it adds up.
T Money
World Domination with a plastic spoon since 1984
G5 System controller has a built-in 128 bit 400 Mhz Memory Controller the supports DDR
Correction, the memory bus is clocked at 200Mhz, but it does indeed support DDR. This gives it the effective bandwidth of a 400Mhz bus.
Windows has Cygwin,, as well as AT&T's commercial Unix environment. Linux/BSD have WINE, to run the Windows apps. I admit, that might not exactly be "native", but it's no less "native" than running Unix apps under OSX.
If you think that running X on Cygwin is comparable to running X on OS X, or that running Office on Wine is comparable to running Office v.X (which is a Carbon and therefore native app) on OS X, you haven't spent much time comparing. OS X is a BSD with a Mach kernel. Cygwin is a POSIX environment running on a non-POSIX operating system. Wine is a very, very, very good but nowhere near complete reimplementation of the Windows APIs for Linux/x86. When you run FCP, Office v.X, Quicken, etc. on the Mac, you are running COTS apps native, not with a reverse-engineered API that almost but not exactly replicates the characteristics of the OS the software was intended for. When you run XEphem, or Gimp, or ssh on OS X (which by the way comes with ssh and sshd, with emacs, with tsch, with gcc, with apache and samba, and to which you can easily add bash, lynx, wget, etc.), you are running on libraries that were written as part of the operating system, not add-ons. I can't call either Cygwin or Wine "as" native as running UNIX in OS X - it's not something that Windows was intended to do or that LInux was intended to do in the way OS X was intended to run the things it does. Key thing is you can expect the average OS X box to have certain UNIX features, which you can't expect on the average Windows box, and there are COTS programs written with the Mac in mind that perform functions you can't find in COTS software for Linux or the other BSDs. It's the best of both worlds argument.
And the hardware is pretty slick on its own. You can with a lot of thought and effort configure a dual-boot Debian/XP Alienware box that I would admit is probably superior in many ways (speed, software availability, overall functionality) to anything Apple sells - though I'd be surprised if you found it sitting on a shelf configured that way for you by the manufacturer. And you'll have to dual-boot to get the benefits of the *nix and the consumer OS, which you won't have to do on the Apple.
I'd say that the main arguments in favor of using FreeBSD and Linux for a primary geek desktop are philosophical/ideological, vocational, or aesthetic: either you do it because you have a dedication to the ideal of free software, or because you enjoy working to improve the software all the time, or because you like the incredible amount of control free operating systems give you over every detail of your work environment. I can respect all those reasons, but they don't add up to superiority in any absolute sense over OS X. There are things OS X can do that they can't do as easily - not yet, anyway.
(Admittedly, I think Office v.X is overrated in comparison to Office for Windows, which, like it or not, is the better product; but Office for Mac is a far different experience than emulation via Wine.)
What are we supposed to wait 3 years for XP and OSX along with Premire and Photoshop all to go 64bit then benchmark?
The benchmark is correct. These are the apps that people are going to be running NOW and for the next few years so they are the ones which matter.
If you wanna get rich, you know that payback is a bitch
-Lucas
The entire article is AMD this, Athlon that with as side of Intel... I'm not even sure why they included the G5s in the benchmarks.
All of the mac systems only have 128 MB of ram. That's not even enough to comfortably run my iBook, let alone a dual processor G5. Plus 64-bit applications are bigger and use more ram (every int is 64 bits).
Are they even benchmarking 64-bit applications? Premiere, Photoshop, Word and Quake III are benchmarked... But nowhere does it say it's the 64-bit version of these. Also, Most of these products are completely different between Mac and PC versions making the benchmark essentially pointless.
Did you really think that 'PCWorld' would publish a benchmark that gave any favor to a mac? That's what I thought.
Mod PCWorld +1 Flamebait
Today in the news, an unnamed x86/Windows PC enthusiast said something disparaging about Apple.
"We don't know how to take this. This sort of thing has just never happened before, Ted.", said Steve Jobs.
In other news, an x86 enthusiast called mac users "dumb. I don't know why they don't just use Windows. I don't like it when people don't do what I want them to."
"Intel counters that its 32-bit processors are plenty powerful to meet user needs"
a sp
http://www.eweek.com/article2/0,3959,642927,00.
Give Apple credit for true innovation.
I think they wrote the perfect article to get Ad revenue. But i digress.
The testing scheme they used seems more then a bit suspect. If they had done ANY research, they would know Premier is hobbled, cludgy, slow, and unuseable on a mac. Word on windows loads itself in to the system to get performance increases on x86. Photoshop is hindered by disk access as much as CPU. When they take away the raid, the scores seem a little more real. (you CAN do raid in the G5 box, albeit software, its still faster). And Quake3, not so much hindered by video ram, as with a more then likely slower clocked video card, not up to date Quake version, etc. Also notable, is unless im mistaken, NONE of the apps, except photoshop, would be useing the dual processors. There is NO link to the test methods (app version numbers, weather the G5 plugin on PS was used, what version of word...). Macworld may have done the Mac tests, but they only did it to what PCworld asked, or i dont think they would have agreed to do a premier test. Benchmarks are benchmarks, but these seemd to be a little more then bent. Good for ad $$ though....
"Stuff... In my home!? NEVER!" - Zim on Invader Zim
"I want the toilet seat!" - Little Dog on Two Stupid Dogs
The truth is over here, and my G5 workstation is way over there...
Manipulate the moderator system! Mod someone as "overrated" today.
Let's deconstruct the differences between what is 'commonly' referred to as a 'Desktop' and a 'Workstation'.
Desktop - Used in the home or by students to perform basic computing tasks, such as email, basic home office tasks, perhaps some gaming as well as amatuer hobby level tasks; like photo-editing, simple video editing and music editing/swapping.
Workstation - Typically used in the office or workplace of professionals, normally able to run all the same software as a 'Desktop' version of the platform, but typically geared towards more professional pursuits, such as Computer Aided Design and Computer Aided Machining, Scientific Calculations, Professional Quality Video, Photo and Music Editing.
Now, let's look at what the Apple 'Desktop' Computer is marketed for...
Well, first... They call it the "World's Fastest Personal Computer" both 'Desktop' and 'Workstation' fall under that umbrella... I could stop there, but I won't...
Under the heading of Architecture...
"The Power Mac G5's new system architecture delivers phenomenal throughput for the most intensive image-editing, rendering and scientific computing tasks - " This statement refers to what is commonly considered 'Workstation' work...
You know what Fanboy? I am going to stop right there.
Apple's Marketing department fed the world a lie about being the first 64-bit Personal Computer. They weren't. They could have been more truthful and said, "The first consumer priced 64-bit PC from Apple" while slightly misleading, it would be very truthful...
Go ahead and read Apple's web-site. I did.
If you ignore the other uses of a tool, does that make the tool less useful, or you less useful?
You are missing the point.
The point is that there is a huge variance in the test results ranging from virtually none to a large difference in favor of the Athalon64. When you look at which apps this is occuring in you see it is in case where one could expect strong differences between the PC and mac versions of the software: Premiere and Word.
Given this its reasonable to assume that you are measuring the software in those cases. And had you chosen mac optimized titles you would flip this around. Indeed there is some logic that says you should keep the tasks the same (rendering pdf) and choose the optimal software for each platform if you want to have a software neutral test of the hardware.
in any case the best I can conclude is that on non-optimzed systems (on either machine) the hardware performs about equally well, and the software makes the bigger difference than the hardware.
certainly the athlon64 is not trouncing anything.
that said I think the G5 has a lot more room for growth since intel (and amd) have announced today they are having problems getting further speed scaling. Meanwhile the g5 is loping along at 2 Ghhz. See you next year.
rant
/rant
I've always found the labeling scheme behind "Desktop," "Workstation," and "Server" machines to be nothing more than a way for companies like HP and Dell to charge people and corporations hundreds, if not thousands more than by implying vastly superior quality when they might only get a few incremental upgrades such as extra RAM, a bump in speed for the CPU, and possibly a faster video card.
good post. the stupid mac lover responses to this are exactly as i expected.
Have any applications besides Adobe's been recompiled for G5 yet? The G4 to G5 jump is more of a change than the x86-32 to x86-64 jump and you need to recompile apps to handle G5 in order to see G5 gains.
Same thing happened years ago for MIPS, SPARC and Alpha although Alpha was more of code structure "all the worlds a 32 bit quantity" issue.
I VERY sincerly doubt MS apps have been recoded for G5 performance!
...not mentioning the fact that Opteron beat the G5 to market by over 4 months...
*sigh* If we are going to get into a pissing match as to whose 64-bit CPU was introduced first, then it still goes to Apple/IBM/Moto. Anyone remember the PPC 620 which was released in 1995? Also, if AMD introduced their Opteron 4 months ago, why did they have to introduce their "desktop" CPU only a month ago (remember, Apple's claim was the first 64-bit "desktop")? And why can't I buy one at Dell? Geeze, get real. This article is merely for the benefit of their readers to make them feel good about being behind the curve with respect to innovation.
It matters a lot to me! Now that x86 kicks ass again, I can finally face myself in the mirror. My dog licks my balls again.
There seem to be lots of complaints about the fairness of the benchmarks.
If you want to compare the performance of CPUs with different ISAs, then you need to have the benchmark source code to compile it to the target ISA. This also brings the performance of the compiler into question but there's just no other way to do it if you want to compare CPUs with different ISAs.
One benchmark that people in the industry use to measure relative CPU performance across different ISAs is the SPEC benchmark. Just about every single computer maker from Dell to HP to Sun have submitted scores. Apple has not. This is in an of itself very telling. What is Apple afraid of? I'm hoping IBM releases a computer based on the G5 so that we can get some idea of the SPEC performance. Apple seems to believe it has something to lose by submitting a SPEC score.
Don't like SPEC? Please suggest some other CPU intensive applications to benchmark that have source code and publish your results.
They're also systems from Alienware and Falcon NW. These are two companies with serious support, serious reputations, and also serious prices. Try putting together the systems from newegg and you'll find a price difference of 30-40%. That comes out to be a lot more than $200 difference.
Just to clue you in on why they didn't use a more recent game:
If you pick too new a game, the performance bottleneck will more likely occur at the video card. If you pick an older game, the performance bottleneck will more likely show up at the CPU level.
So if you're benchmarking a CPU, it makes no sense to run halo at 1600x1200 with AA and AF turned on.
WinNT was a 32-bit os (even on 64-bit chips-alpha), but it was used for 'servers' and 'workstations' and was pretty expensive, and I am assuming by both apple's use and the articles use they mean "generally consumer available" when they say desktop.
Windows 3.1 was not even an Operating System, any more than X-Windows on *nix is an Operating System. The operating system was MSDOS (most likely version 6.22) (or could be DRDOS, with incorrect error messages). This also includes windows 95 (and possibly 98). (See the Caldera-Microsoft lawsuit for some of the details, or at least those published)
Windows IA-64 is the first 64-bit windows, and as another poster posted, it lacks a heck of a lot of features from x86 windows. So I suspect that x86-64 will be highly problematical in terms of support for a long time.
Is that why quake 3 was released simultaneously for Mac and PC? Properly ported code is simply code. I'm assuming they use a mac compiler...
*sigh* I seem to be attracting trolls today...
Apple markets the G5 as a desktop, to the general public (not admins, not geeks, ordinary people). The previous 64-bit systems have been marketed as workstations or servers. Regardless of your semantic quibblings about what exactly constitutes a desktop vs. a workstation, this is what they were sold as, to the general public. Sure, I could buy a whopping expensive server or workstation from whoever I wanted, put it on my desk, and call it a desktop. And it wouldn't be any less true. But that's not the point. It's not my terminology we're dealing with here, or yours.
Apple doesn't care what you call the G5, or what anyone else does. They call it a desktop. Perhaps they made this decision so they could market it as the "first 64-bit desktop" (which I consider synonymous with Personal Computer, as they were using the term). It was clear the message Apple was trying to send. Whether it was a purely marketing-based decision or not, when arguing with Apple's words, you at least have to accept their terminology. You may not agree that that's what a Personal Computer is, but I, and many others, think it's a reasonable definition, so to talk about Apple's statement, you simply have to accept it. Otherwise the argument doesn't even make sense.
Dan Aris
Fun. Free. Online. RPG. BattleMaster.
Reality distortion filter *on*
While no sensible person would argue deployment numbers and market share, anyone who thinks Apple is 'struggling' as a business is struggling with reality. Apple is one of the few profitable companies out there, with a 'devout'(dare I say rabid/zealous user community), rapidly growing laptop market share and a strong foothold in many facets of the computing/entertainment industry. Regardless of top speed shootouts and partisan OS arguments, Apple is no Be or Amiga. They are not going anywhere...If you want to talk about a struggling company, then I'd talk about AMD or anyone else going head to head with Intel or MS.
.1% of the car market, and even though you could buy a Vette for half as much as a Porsche Turbo (and more importantly it might be faster even) people will always buy Porsches and their limited market share will always mean that a porsche has the freedom to innovate regardless of bottom line. Or something like that...
Apple will never 'lose' to Intel or MS because they aren't really in competition with them. They may eternally remain less ubiquitous, but the old tired logic of automobiles rings true -- regardless of validity -- Porsche will never sell more than
We apologise for the fault in this post. Those responsible have been sacked. -- Signed RICHARD M. NIXON
the mac tests are performed with 128 mb RAM, wich is the bare minimum to run osx. I'd like to see the benchmarks with similar hardware setup (aside from the cpu)
No offense meant, but you're wrong about the video card bit. The 256MB Radeon 9800 Pro runs at the same clock speed (yes, core AND memory) as the 128MB one, and unless you have heavy AA and AF on you won't even notice the extra memory. Benchmarks have shown that the 256MB one offers something like a 1% performance increase unless textures are huge or AA+AF are heavy. Perhaps you are thinking of the Radeon 9800XT.
The 1.8GHz Mac was also listed at $2999, thats almost $500 more that reality.
Go figure.
Now it's your turn to be sarcastic. =) I just checked my system prefs, and you can tap on the pad for a click, but not a control click.
Thing is, I know I saw someone doing this on a powerbook, so they must have had some third party utility to do it.
It's not offtopic, dumbass. It's orthogonal.
I don't know a whole lot about different programming techniques on different operating systems, but I do think that programs can be optimized for s certain OS.
All of the programs in the benchmark are natively Windows with the possible exception of Quake III (is that native to Windows or Linux?). Why don't they throw into the mix programs written specifically for Mac that have been ported to PC's? Could the results possible come out differenty?
If nothing bad (*cough*) had ever happened to DEC, how would our desktops be now? We wouldn't be stuck with THESE wimpy processors, that's for sure.
The AMD system smoked the G5 because windows has better software than apple. The 4x difference clearly indicates this. Just another reason why people are far better off using windows than a mac.
I was NOT making qualitative comprisons, only noting functionality.
There are many many reasons why I do not use OS X, and they are none of the above.
Slashdot gets worse every day... Pipedot: News for nerds, without the corporate slant
Apple markets the G5 as a desktop, to the general public (not admins, not geeks, ordinary people). The previous 64-bit systems have been marketed as workstations or servers.
Ie, Apple produced a 64-bit computer that was marketed towards the general public more so than previous 64-bit computers. This is not the same as producing the first 64-bit computer - that suggests a technical feat, independant of marketing.
Apple doesn't care what you call the G5, or what anyone else does. They call it a desktop.
Ie, when Apple define "desktop" to include Macs, but to not include other 64-bit computers, then Apple produced the first 64-bit desktop. What a surprise. Just like when they defined desktop/personal-computer to include Macs, but not other computers that were faster than Macs, they were producing the fastest desktop/PC.
Is going by Apple's definitions really fair or accurate? AMD could pull the same trick - I bet that the marketing for 64-bit Windows based machines will at some point become far greater than that for Macs, so we could discount Macs on that ground - or possibly on cost too. Or perhaps just define desktop not to include Macs for no reason at all - after all, AMD/etc can define it however they like, just like Apple do.
You may not agree that that's what a Personal Computer is, but I, and many others, think it's a reasonable definition, so to talk about Apple's statement, you simply have to accept it.
So you'll accept people saying that other computers were, or will be, in fact the first 64-bit Personal Computer?
'If it was released for x86, no one would give a rat's ass'...
Why do so many x86ers say they want it ported then (I don't know a single mac user who wants x86 hardware)? I always thought people generally liked the OS but didn't like the 'limited' hardware options...or were interested in trying it without making the large investment in equipment.
We apologise for the fault in this post. Those responsible have been sacked. -- Signed RICHARD M. NIXON
Whatever, who is David Pogue and why should I care what he wrote? When I had my Mac the only one of those programs I ever used was iTunes, and since switching to a Windows XP laptop I like Winamp a lot better anyway. iSync a killer app? Surely you are joking...
I call BS. A library is a library, there's no difference between the C library shipped with OS X and the one that comes with Cygwin, it's just an interface between user apps and the kernel. In fact, I was easily able to compile Kerberos and build a kerberized ssh on cygwin, a feat which took me much longer on OS X because I had to figure out it's weirdo "Framework" system of library management.
Notice that the Mac tests included a dual Opteron 246 in order to beat the Dual G5 in the photoshop benchmarks.
The dual Polycomm dual Opteron is around $4200 in the closest configuration that I could build on their site.. but that was with dual Opteron 244s because they don't offer the 246s in the BTO site yet.
The opteron that beat the Mac in Photoshop isn't even available yet.
Hmn...
Not to mention that there are fairly significant changes between the G4 and the G5 that haven't been optimized for yet.
Athlon is a great chip, but it's been around longer than the PPC 970. There is better compiler support for it and the software vendors have had more time to tweak for it. Not to mention that IBM has diverged from Moto in some very real ways that affect performance on G4 optimized code.
They are both plenty fast, but we are just beginning to see what the G5 can do while the Opteron is a bit farter down the track in that regard.
ffakr.
I'm not feeling witty so bite me
The G5 v. Opteron tests were hand timed. I'd say there's a lot of room for improvement in testing methodology there.
I love you, lets make babies.
Man you are all zealots!
You criticize the Apple zealots for defending the platform of choice.. The Microsoft zealots defend an OS that's full of holes. You question the legitimacy of PcWorld and MacWorld benchmarks.. And for what meaningful purpose? So you can say "PC Sucks" or "Mac Sucks"? Does it really matter? The G4 was sorely in need of an update and it got one in a big way. AMD once again kicked ass with both the Opteron and Athlon64. I don't care who was first to market.. I am a Mac guy because I can't stand the kludge that is Windows. I like the Mac interface.. I am not a game-obsessed kid who needs a GeForce5456456456 with a terrabyte of on-board memory so I can play the latest version of My computer is a tool and it works great. I have a dual G5 at home and a P4 1.8 at work. Both do what they need to do. 600+ posts of mostly banter. Shame on you all. Go do something productive!
Check pricewatch you twat. And FYI, Opteron was available before the G5 by months.
I mostly play first person shooters and strategy games (both real time and turn based). FPS are lousy on a console because a mouse and keyboard are much better for quickly turning and aiming while moving. And strategy games are practically non-existant on consoles.
As for using the Mac for gaming, it also depends on what you want to do. If you want to play the same games your friends are playing, forget it. But if you want to play great games that were popular on the PC 6 months to a year earlier, then it's not a big deal, most of them are ported.
You're about the most sensible person in this whole stupid discussion. Including me.
Dan Aris
Fun. Free. Online. RPG. BattleMaster.
I can understand the reasoning behind testing Photoshop and Word, but the problem is they both represent old codebases for a different architecture (old Mac Toolbox).
There's been a lot of Carbonization work, and they run well from the user's perspective -- but if you watch top, they also waste a lot of CPU time even when idle. This makes them poor candidates for benchmarking the system itself. It would be nice if there was way to benchmark apps written for Mac OS X against counterparts written for Windows. Also, Panther is singificantly faster than Jaguar in many cases.
Premiere seems a particularly odd choice. Don't hold me to this, but I was under the impression that it has long been much slower than it should be on the Mac. Maybe that was After Effects. Also would have been nice if they used the newest version of Photoshop, which undoubtable is more tuned to Mac OS X.
- Scott
Scott Stevenson
Tree House Ideas
It looks like the G5 holds its own over the opteron and the 3200+ as well as against the P4.
LOL... The G5 (single cpu) got stomped even by the P4, and absolutely clobbered by the athlons.
With the dual cpu configuration THEN you get something roughly equivalent to its competitors... but at a much higher price!
1MB L2 in the Athlon hopefully means Apple will have to add some more to the G5. The availability of better (256MB) video cards for PCs also needs to be addressed, or Apple's ability to make outrageous marketing claims that get people fired up and creating a buzz for their products will be hampered. Hopefully the next rev will see some of the G5's shortcomings addressed. That, coupled with Panther, should make things interesting when this shootout is repeated in a few months.
Either way, guys, it's all good. These are both great chips and mean cool things for their users. Don't lose the forest for the trees =) err... wait, maybe it's me missing the point: this is /. afterall, so I guess yelling at each other is par the course... forget everything I said, gentlemen. Carry on!
Just out of curiosity, /. if you never read it?
How and why are you posting on
Oh those were the days, last month, when we were on top of the world. I remember hearing them rejoice at their secret MUG meetings -- "We're number one! We're numbe...Damn!...We're number two! We're number two!"
I do find the 'fanboy' phenomenon interesting -- from both sides. I do not have a brand spanking new Mac or a brand spanking new PC but I find it interesting how both sides identify with the 'winners' in theses shootouts. I guess it is similar to sports fanatics -- I don't understand what makes 'fans' identify so hard with 'teams' of players who could give a rats ass about the fans -- whenever I hear anyone chanting "we're number one" I always wonder what that 'we' is supposed to mean. They are not part of the team. I did not have a hand in designing the G5 or the Athlon. What would I be celebrating?
I would like to take a chance at diffusing the mac zealot/PCs 'Rox'er experience -- I have never met a Mac user who in any situation incited a pro-mac rally. Quite the opposite, I find most mac users tend to be quiet about their choice of OS if for no other reason than the ignorance of others makes it rather annoying ("can you run Word?", "But there are no appllications for the Mac.", "can you play such-and-such game?")
I have used Macs for nearly 20 years -- as well as PC's running Windows(the last was Win 98) and Linux (ppc variations and RH). I could care less what others use.
I only become zealotous when I read others insane rantings (the tired mac users are gay thread, the 'zealot' thread) When I respond, I only want to provide a measured response. I don't care if my platform is the fastest -- we know that is always a short lived position. I do like to share the fact that I have always been productive using macs and that is the most important metric I can think of.
We apologise for the fault in this post. Those responsible have been sacked. -- Signed RICHARD M. NIXON
PCWorld is taking their readers for a ride...and maximizing ad impressions. Sort of like the old days when small time sites with banner ads would create scripts to maximize ad impressions and click-throughs to milk the ad agencies of ad fees...
Think about it:
They create a benchmark test
The benchmark test they create ensures controversy by throwing into the mix AMD, Intel, Apple, Microsoft, OSX
They use similar systems, but tweak some systems with graphics cards with more memory and probably faster clock speeds
They use similar systems, but some have raid, some do not
One system has XP Professional, others have XP Home, are they all using the same filesystem? XP Professional has a different filesystem not available on XP Home iirc
They use similar systems, but the software fails to use the second processor on some tests
They use the same graphics program (Quicktime), but it may use a different codec on the different systems
They almost get it right, but not exactly
They create controversy between the warring camps/fans, creating a buzz
The warring camps/fans wait with bated breath for the next fair benchmark, whether from PCWorld, or elsewhere, for validation of their opinions
PCWorld laughs all the way to the banner advertising bank
PCWorld plans next benchmark, with almost, but not quite equal hardware/software preparations
If PCWorld doesn't have the resources to ensure a fair test, then no one does, and there is no such thing as a fair test. But then, a fair test wouldn't get the masses riled up, and wouldn't have the masses waiting for the next one, and wouldn't create the controversy necessary for the buzz to maximize page views...
By reading the posts here, I've come to a conclusion that I've reached on other topics, such as Sun and SCO's future, MS, Adobe, and Macromedia's future (and later Oracle's as well) in the face of the GPL, and AMD vs. Intel vs. Apple. The conclusion is basically that some users that don't have a strong preference for one brand or technology (and don't waste money on overpriced sports team branded clothing/logos, don't get starstruck by movie stars, etc.) will spend money on technology as value investments. Those that do have strong, can't beat my technology no matter what, preferences, will always look at benchmarks through a prejudicial eye to justify their decision in their technology choice.
Is comparing a Pentium IV 2.8 at $225-$250 to an AMD XP 2800 at $157 a fair comparison? Or would a fair comparison be a Pentium IV 2.8 at $225-$250 vs an AMD XP 3000 at about the same price?
Is ease of use a huge issue when power users are the main subjects of Apple vs Windows vs Linux tests, especially when the graphics abilities of Apple are invariably used?
Is ability to run Photoshop a make or break situation for Linux vs Apple when Photoshop actually can be run on Linux?
Will a dual G5 box be such a great choice for running Photoshop in a production environment when you can put together a cluster of a dozen to dozen and a half Linux/OpenMosix boxes for the same price running Photoshop (and everything else in the production environment)? Even when (did it already) the shared memory patch enables huge performance gains on Photoshop and other apps?
From my own point of view, the bottom line comes down to: what is the best performance/value I can get from the same dollar on each platform? Are the apps I need available on the platform? Yes? Is it the best value? Yes? Then that's what I go with.
ISPs are good examples of this. I know one ISP reseller personally. He sells dsl services. He's a member of linux user groups and participates on the linux mailing lists. And has contributed code to various foss pr
Because it's actually based on Server 2003. Server 2003 dropped 16-bit support, and doesn't include Windows Media or the Luna interface (by default, but you can get those parts back).
That's probably what they're talking about.
Fuck Beta. Fuck Dice
You have Windows software? Great -- Athlon64 is probably going to be the best bang for buck you can get. You have Mac software? G5 for you, baby. Use the system that's right for what you need to do.
As for me -- as a Unix sysadmin, I have several options: Mac, PC running Linux, Sun are the three main ones. I've opted for a Mac -- recently upgraded from a 500 MHz iBook to a 1.25 GHz PowerBook (*drool*). Frankly, the G4 is fast enough for me; I don't need the speed of the G5 (I'd like it, but I don't need it, and I can't afford it anyway.)
In short -- any system you buy these days is "fast enough" for the average Joe, and if you aren't the average Joe -- if you need speed and lots of it -- you're not going to be looking at PC World to figure out what to buy. More likely is that you'll get evaluation systems from a number of vendors, and test them yourself with whatever you'll be using them for. This whole thing is utterly irrelevant, if people would just stop and think for a bit.
That can prove my 2MHz Z80 is faster than the G5... This test is flawed.
The best thing I can come up with for a fair test is doing a verity of optimized operations in FORTRAN and then in C on both platforms. Try everything from doing YUV transforms to physics simulations (Motion of a alpha particle... or something like that). Do some DCT and then try sorting bits and doing ridiculous amounts of memory accesses.
BUT, Can the AMD do this??
http://www.ohlssonvox.com
Are we joking here? They tested 4 applications with 2 tests per application. How about some tests in scientific computing? Those are the people who are really going to love 64-bit. Their test notes were ridiculously vague, and make independent verification almost impossible. "We exported to the QuickTime format"? What the Hell does that mean? What video codec? What audio codec? Was it just video encoding? Did the two labs even use the same codec? Were there filters applied?
I'll wait for someone like ArsTechnica to do a review of the two systems rather than accept the conclusions of people who think that media frameworks are synonymous with video codecs.
Interesting (and funny) look at the review As the Apple Turns
The proper arrangement of the words (based on the porting argument) is: "Mac games suck."
It is not the Mac that sucks--it is the game software that sucks. As has been proven over and over, When developers actually care about the performance of an application on the Mac (rather than merely its existence) they can tune it to be as fast or faster than its x86 brethren.
That being said, no one buys a Dual G5 strictly for games--while plenty of gamers will rush for the Alienware Aurora no doubt. I am USING a Dual G5 right now and I have played my share of before-mentioned "sucky Mac Games" and they are plenty fast for me....280 vs 350 FPS isn't really gonna make a difference...my eyes can't tell the difference, and no one's reflexes are that good.
Remember....before you port the code to the Mac, it needs to be finished for the PC....so the development time for the Mac version was likely significantly strained in order to meet the launch date.
Remember the converse: Improperly ported code is simply crap.
Oh yeah!
.
These hard working scientist know how to measure benchmarks: a tolerance of 1 second! And even if that is not enough, they won't use longer lasting benches - instead they use files so small that the difference between those platforms is completly within the tolerance.
And if they can't use larger bench files - why didn't they publish the numbers with a tolerance of 1/10 of a second? Does this mean they have no clue how to exactly measure the needed time of the photoshop actions?
Interesting also that they thought to be so "clever" not to use After Effects - that might have ended in a debacle for them we have seen not so long ago...
The most important thing those benchmarks show is, that these testers are to lazy, and the journalists don't have a clue what those numbers mean...
By the machine you enjoy. Be happy. Live longer. Simple ;-)
-psy
A. Adobe Premeire!? Adobe freaking I haven't been developed for the Mac in a few years Premeire!???
Moderators! Change this topic title.
Word -- "auto summarize" --
Moderators! Toss it all out.
Furthermore, I have seen independent numbers on the web giving higher numbers for Quake III.
As if. Don't believe everything you read about the new G5 in PC Magazine.
Excellent observations. I hope you get a good response to this from PC Magazine as things really are skewed. As for your reference to the video card:
The Radeon 9800 Pro vs. the Radeon 9600 Pro. The 9800 Pro has nearly double the throughput for its memory and faster clock speeds as well for the processor. At least this is what I've read before. I cannot now put my finger on the original article that I read, but the information can be found through google by searching for "Radeon 9800 Pro" and "Radeon 9600 Pro" in a comparison context.
If anyone can find a very good comparative article, please post it as I would like to reexamine the differences myself.
The bottom line appears that the G5 will really still outshine or be right above par with the AMD 64 chips given the above observations. I would like to see gentoo compiled from scratch on both machines and do some real number crunching, app running and comparable tests. Lets also see filesystem performance tests and various other data transfer tests (network, firewire, usb 2, pci, etc) as well.
Just go to Apples front page.... it says "the world's fastest personal computer" dan aris, you're a moron...
Because it can be dangerous to your karma to post 'bad' things about the Mac. Kind of like saying something bad against the PA.
that is a damn ignorant opinion. Mac OS is solid. Hardly "crap". You obviously have a pro athlon bias, for whatever reason. Certainly you also have some awful longevity and reliability malfunction to run Windows XP and Linux on Athlons.
I've worked in the industry for four years and I've seen many athlons, NONE of which have been in the data center because of their gross unreliability. Who cares if an intel box is $50 more expensive???!?! Mission critical systems deserve an extra $50!
Back to your original statement, however, Apple does not make crap. Their operating system is well designed, well implemented, built on a solid foundation, and leaps and bounds ahead of windows and linux in the UI department. Calling it crap only reduces your comment to a bias-driven bucket of suck.
The List of Grievances with Slashdot.
hahaha , as far i read the article , i was wondering ohhh those stupid apple zealots would post replys like :
OHH those benchmarks are tricked blah blah
ohhh old aplications ohh ohh ohh
ohh ohh aplications are too old
bla bla bla bla
i have only to say TAKE THAT INTEL & APPLE FUCKING ASSHOLE ZEALOTS
What point is there to this kind of benchmark when the entire platform is so different?
I mean, who on earth really finds them relevant? Is anyone going to be choosing one over the other because of how fast they can search-and-replace in Word? Is anyone actually still running Premier on a Mac? Is any serious gamer even going to consider a Mac? Do graphic designers really use Gaussian blur that fucking much?
We all know that both chips are pretty fucking fast. Is that really in question here?
Don't become a regular here -- you will become retarded.
Need I say more? It's just that, for some strange reason, Apple's marketing gets WAY more scrutiny than anyone else's. Not sure why.
Dan Aris
Fun. Free. Online. RPG. BattleMaster.
Before it's release Quake 3 was demo'd on a Mac.
Quake 3 is written in C.
It was not ported from assembly x86 to assembly PPC.
There will be little hand written assembly if any, unless it directly relates to a video card.
is it me?? or did they change the entire page??
i hate microsoft.
I love macs, use them, hate windows and all that, but isn't the point they are trying to make that they can have the fastest COMPUTER? Their chips may be slower, but if they can speed up the rest of the computer with components that are not available for the mac (such as the video card), isn't that fair? I think the point is that if you want to buy a top 500 supercomputer, you can't exactly get it from a website and credit card deal...but of those computers that you can just buy off of a website with a credit card, macs are not the overall fastest? It doesn't matter so much to me, since i'll never be able to afford any of these machines, but i think people in general are comparing the chips, not the computers.
Ie, Apple produced a 64-bit computer that was marketed towards the general public more so than previous 64-bit computers. This is not the same as producing the first 64-bit computer - that suggests a technical feat, independant of marketing.
Why would Apple's marketing campaign suggest anything other then marketing to you? Every technical feat that comes with a press release has marketing involved. We should all know this by now.
No need to be sorry, thank you for the correction. As I'd noted, this was an area I was a bit fuzzy on. In any case, it means that the Athlon64 and G5 share yet another common trait, the integrated DDR memory controller.
Really, it's silly how many people are balking at the possibility that AMD's 64 bit chip could be faster than a G5. I suppose that's the Jobs effect at work.
GPL: Free as in will
It seems to me if AMD wants to trully beat Apple in the battle for the 64 bit desktop they are going to need a real SIMD soltution. Why doesn't AMD just license Altivec? MMX, MMX2, SSE, SSE2 and 3DNow! are never going to measure up to Altivec.
These days, everyone is faking their registers, the real number is huge, even the RISC chips are doing it, just like x86 have since the Pentium Pro.
So, two choices. Come clean, like the ia64 chips, with 128, and hope you guessed right about the number, or pick a good number to pretend to have. You don't really need 32, if the chip pretends for you, but you sure need more than 8. Thus the 16 that AMD went with. Internally, I would guess there are many more.
We shall see.......
Plato seems wrong to me today
Have you actually looked at the webpage for the G5? If that is something being marketed to "ordinary people", then I've severely underestimated the general public's grasp on computers. That page is very much geared toward professionals. It is splattered with benchmarks showing floating point performance and photoshop performance. Not to mention their price point, which is well beyond what "ordinary people" will be paying for a computer to check email and browse the web with.
I won't argue that their terminology is wrong b/c the line between desktop and workstation has at the very least been a gray area in recent history. But if they want to use that gray area to include their machine, which they market to professionals and geeks, in the realm of desktops, then I think it's reasonable that that same expansion encompass other computers that are marketed to the same segment of users. If the Opteron is indeed faster, then they were not the fastest in their class when they came to market, regardless of what they want to call that class.
Of course, it's all typical marketing fodder. The G5 is a sweet machine. Who cares if it's fastest in its class or not?
- b
How is it FUD to use benchmarks that reflect the real software people run on their machines? You're saying Apple users don't run Quake, Word, etc.? That's not true.
Is anybody going to decide between a G5 and Athlon based on Word?
Quake, maybe. But I think the point is the benchmarks come off as testing system performance, but they're really benchmarking app efficiency. Maybe some lower-level tests would work better.
- Scott
Scott Stevenson
Tree House Ideas
But thats not what it says, it doesn't mention "consumer", it's an assumption .... NT4, even NT3 was as much a Windows desktop OS as 2000.
....
It may not have been promoted as such, but it was. NT4 even came with DirectX
The article claims that it took 15 years for a 32 bit Windows desktop to be made - it didn't. NT would have that title.
Windows 98/95 definately aren't full 32 bit operating systems, but they get pretty close, especially in the ME incarnation - I belive you can strip all the 16 bit libraries and programs out and still have a functioning OS.
And I seem to remember OS2 being promoted at one time as a consumer desktop OS, although obviousley it's not Windows.
try to make ends meet, you're a slave to money, then you die
wow. an excellent use of language, and i learned something (never heard of that word before). but the word is actually spelled anschluss...
Needless to say, I also find PC World's selection of benchmark applications uninteresting to say the least. I would expect any serious comparison of high-performance workstation CPU's to use accordingly demanding applications which place an emphasis mainly on the CPU and not on the graphics subsystem.. HMMer and Mathematica come to mind...
i think the Opteron may just be faster. know what?
Don't care.
why?
because the last time i had to race my computer was.......NEVER!
the day i'm doing anything, especially if it's work, the outpaces a dual-2ghz G5, is the day i quit that job cause i'm working way to hard.
add in the fact that this was a comparison between a server chip and a desktop chip(Apple said fastest PERSONAL computer), the fact that the tests were not extensive or even handed(no comparable benchmark software was used cross-platform), and that no matter how much faster a PC may be than my Mac it still has Windows on it, and i double don't care.
an Opteron with Linux on it is a fine machine in my opinion, but i still wouldn't give up a G5 for it. i like being able to run PS 7, Office, Quake Arena, and BitchX, Emacs, and X11, not to mention OS9 apps all on one box without rebooting, and still have an Apache webserver or network file server running. OSX is what Linux wishes it was, and will be someday, but not yet.
when are people going to learn that the Mac is a total experience, and that is what seperates the Macs from the boys.
--from CyberdogOSX
You're smoking something or you didn't read the article very carefully.
Both systems had Radeon 9800Pros. Nowhere did it mention 9600Pros.. The only diff is that the Mac Radeon had 128MB of graphics RAM (you can't get the 256MB version apparently) while the PC version had 256MB.
It wouldn't matter for these tests anyways. Quake3 doesn't even use 64MB of video RAM, nevermind 256.
Thanks for pointing that out. You're correct. Notwithstanding his other points are very valid and I'd sure like to see some straight comparison tests, which I'm sure we'll eventually see.
drag and drop? that's too much work.
apt-get update ; apt-get -u upgrade
Upgrade an OS? apt-get dist-upgrade
How do you upgrade OS X? How much do you pay to upgrade it?
What's the big deal with installing Linux apps? It's easy.
college kids? Where did you get that? Most OSS developers are experienced professionals with 10+ years experience.
OS X is great, but your points were mostly incorrect, except for this:
would rather do extraordinary work just to get simple stuff to work (DVDs, etc), then by all means stick with Linux.
I don't use Linux for this sort of thing, though I have before. But you're right that multimedia takes a small bit of configuration to get working. It's not that hard, though.
#!/
that didn't make any sense at all.
you're telling me your girlfriends name is Amiga? Or that people with Amigas have girlfriend?
If Amiga is your girlfriend, yes I'd like to fuck her box and shoot a huge load of cum deep inside her and impregnate the bitch. You can deal with what comes later.
Amazingly, given all the raves about how bad the test are, the non Mac boxes are ordered as expected given those tests. So, if you are using any of the these programs, you can determine exactly where the G5s fit in. Sorry Mac fanboys, G5's are slower when running these programs. Deal with it.
I did some tests on a cpu heavy render.
my G5 DP 2Ghz : 16min 7sec.
a 2.6 DP Xeon : 15min 48sec.
I guess a 3.06 Xeon should do 15min 20sec.
and a 1.25 DP G4 in about 32 mins.
The G5 is slower than the Xeons, but not that much. That said, Combustion is in no way optimised for the G5.
AfterEffects 6 benchmarks shows that the G5 dominates on CPU bound tests. I don't think anyone did AE tests using an hardware RAID 0 on the G5 so it's wh it loses in disk bound tests.
http://www.macworld.com/2003/09/reviews/macworldla bfirstg5testresults/
As you can see the numbers match up with the PC World chart for the Quake III & Photoshop tests. But you'll also notice that MacWorld says that "All systems had 512MB of RAM" So, in actuality the G5s in the tests had half the RAM that PCWorld said they did. I'm not sure how much a difference that would have made but I'm sure it would have helped the G5 numbers somewhat.
The machines in the article bore no resemblance to the sorts of machines people would normally buy. Who on earth would pay that much for an Athlon 64 system, when you can get one for half that price? The Mac however, can't be had for half that price.