Analysts and others are raising questions over Apple Computer’s claim that its new Power Mac G5 is the world’s fastest personal computer. Articles at News.com, ExtremeTech. Apple is defending their benchmarks. In the meantime, the first reference of how faster the G5 is against the G4s, is available.
Apple Benchmarketing Raises Questions
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSNews.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
“For one they arenot telling anyone the specs of the Dell models used, no RAM no bus speed nothing and I asked and I got ignored and when I asked again, I got a look”
http://www.veritest.com/
“They claimed that they are the only ones that have backwards compatibility with 32bit, that on the x86 side you need to run a 64bit OS, or run the OS and Apps in an emulation mode, an untruth because the AMD Opteron offers the same functionality as the Power4 that Apple is using.”
And yet it hasn’t shipped so they can only compare it with processors that have. There was nothing deceptive about what they were comparing to. Everybody knew that the G5 hadn’t shipped either.
“Also, the Xeon is dead Intel has abandoned it like Moto abandoned the PowerPC development
HUH?! Intel hasn’t abandoned Itanium. As a matter of fact they’re investing so much money in that chip it has become a huge drain on resources.
“What I want to see is a test against Itanium and AMD Opteron, to me those are the benchmarks that matter because that is what people are moving too”
That would be an interesting comparison. And yet an even more interesting comparison would be to see how the G5 compares against both chips.
“Xeons pretty much dead as is thePentium 4.”
You might want to tell intel that, because I don’t think they know just yet. They keep investing in the chip.
“And I know, its an Apple sponsored event and Apple is going to evangalize and lie and cheat and steal in order to push their product.”
Just because its an Apple sponsored event doesn’t mean they lied or cheated. Actually the way the benchmarked the platforms was extremely fair. They could have made it FAR more disproportionate by being misleading, but it appears that they took the high road. They should be commended not chastized.
“The main thing for me is the price. And realistically what am I going to do with a 64 bit Mac.”
The price is significantly less than comparitive x86 hardware that is currently available. Realiztically, you are going to do the same things with a 64 bit Mac as you are likely to do on a 64 bit PC.
“These machines are mainly focused on the highend graphics worktations and not the desktop users”
Considering the fact that this is a desktop computer with workstation performance, it looks like they should meet the needs of both markets.
“How many working class slobs do you know that are going to pump out the 3 grand for a G5”
They come in less expensive configurations too. Additionally, these working class individuals can now opt for a G4 tower which experienced massive price cuts with the introduction of the G5. (Yes they’re still for sale on Apple’s web site and in Apple stores and everywhere else you have been previously able to buy Apple hardware.
“I do not know many that can and that will.”
These same individuals probably wont buy Dual Xeons either I’d imagine. (Especially considering the fact that they are $1000 MORE than the dual G5.
Thankfully, there’s a Mac priced for nearly every market demographic.
“I must say tho, I am impressed with Apples offings and I have always been their harshest critic and if they can impress me then I think that they have a winner. I just wish they were more open on the Technical side.”
Apple is a lot more open than what you give them credit for.
Years ago, when Apple was using NuBus and IBM was using Microchannel for their respective card expansion options, those were examples of proprietary hardware. You could only plug Microchannel cards into those IBMs, and you could not use them with any other PC (unless they licensed Microchannel from IBM.) Eventually, both the IBM PC division and the Apple designers came to their senses, and they switched to Intel’s PCI design, which pretty much the rest of the home computer industry had already moved to.
Proprietary hardware is troublesome, because it restricts the availability of expansion and replacement parts. You are either locked into the original vendor, or to the handful of hardware makers who have specific hardware license agreements with the company who invented the hardware platform in question. Over the years, a lot of companies (including Apple) have attempted proprietary solutions for memory, video, expansion cards, etc. They seldom succeed, unless they manage to get the rest of the industry to adopt it as a standard.
Writing an OS that is specific to your company’s computer architecture (such as OS X for the Macintosh or Solaris for Sun servers) is not an example of “proprietery hardware.” It’s an example of operating system software integration, and if vendor lock-in (for the complete system, not for replacement parts) doesn’t scare you, it can be a very good thing.
My G3 tower has been upgraded with a third-party IDE hard drive, a third-party G4 CPU, a third-party PCI SCSI card, a third-party Firewire CD-R drive, and lots of third-party memory. All of these parts were industry-standard items which could have been installed in almost any x86 box sold in the last few years, too (except for the CPU, which could be used on any open-firmware motherboard, but then you can’t drop a P4 onto an Athlon board, either.) If Apple used proprietary hardware, as you claimed, none of this would have been possible. I would have had to purchace my CPU, HD, memory, SCSI card, and CD-R from Apple themselves.
really that is what this is…pointless. Benchmarks means nothing because they are too easy to manipulate by either side. Get the box, work on some real apps and gauge the value of the thing then.
Personally I think it was a fair test because an INDEPENDENT firm did the testing
Yes, just like when an independent firm called Mindcraft proved beyond a shadow of a doubt that IIS is faster than Apache…
*sigh*
Actually Apple did give the specs on the Machines, pages 15, 16 of the following:
http://www.veritest.com/clients/reports/apple/apple_performance.pdf
In addition, since they give the Model number of the Dell machines in question, the information is available currently on the Dell site.
Yeah, it’s hush, hush when they publish what they have done and give you enough information to draw your own informed conclusions.
If you used your computer to make your living creating media of any kind, you would see the Mac as the best, most cost effective tool for the job. Why do you think it is used 10 to 1 in the entertainment industry?
I make my living creating software, rather than ‘media’. That being said, I spend a large amount of my free time doing ‘creative’ things, and writing software that is definitely more geared towards entertainment than my work. That being said, the reason that Macs would ever be used ’10 to 1 in the entertainment industry’ today is more because it’s been that way for a very long time, and very few of the companies developing the software used in that industry are willing to develop for a market that they believe is 10% of their overall market (the same mentality that leads people not to develop games and other Windows software for the Mac or *nix). The most obvious place for this is music software, where some of the leading developers have spent years blaming Windows’ architecture for the reason they won’t develop their software for Windows, and only recently has a significant amount of serious music software come out for Windows. A few at the top are still claiming that Windows is at fault for them not putting their software on both platforms, and they’re slowly losing market share (very slowly, though). The other problem, of course, is that good hardware for recording music on the PC doesn’t work as well for generic use (especially games), so most musicians will have 2 computers anyway (if they use their computers for anything else), in which case they might as well at least look at the Mac, where they’ll have a little more software and similar hardware costs (because some of the music-specific hardware costs significantly more than even the highest-end Mac).
This is where the average Joe gets lost in comparing the two platforms. One one side you have a commodity based mind set (lowest price quantity over quality etc. etc.) and on the other you have the creative professional (Real world speed, ease of use, cost of ownership, return on investment etc.etc. etc.)
Everything you cited except for ‘quantity over quality’ depends on your use for the machine, and the tools with which you are familiar. I have many family members that use their computers for very little until something in the computer dies or they find out about some ‘cool new thing’ that people can do with computers which their computer can’t do (like CD burning). They usually don’t care after about 5 years of using the same computer that they could probably just add X part (CD writer) to do that ‘cool new thing’, so they buy a whole new computer. Those people might save money if they bought a Mac, or they might not. If the Mac cost them $10 more and they replaced it just as often, they’d be out $30 in 10 years (the 1st computer, the 2nd, and the new replacement), and they probably wouldn’t care. When they see $3000 on a top-of-the-line computer, they reconsider whether or not they even need a new computer (and hey, if I really wanted to, I could show them a $3000 x86 computer). Personally, if I can’t decide for myself exactly what goes into a computer I’m buying, then there’s a very good chance I’m not buying it, even if it means I have to build one myself (and I usually do build it myself, because I can spread the cost over a couple of months without paying interest on a credit plan, I have spent over $3000 on a new computer with a monitor/keyboard/mouse, but I certainly didn’t put it all down at once).
It goes without saying that you will rarely see one type hanging out with the other since they are polar opposites when it comes to their view on life and on the value of a computer as a tool.
I’d just have to say this isn’t the case. Most of the people I know view their computers as tools, regardless of what kind of computer they own. Then again, most of the people I know are fairly creative types. The only reason a Mac user might not hang out with x86 users so much is because they get tired of the rest of the people messing with them for using a Mac.
What makes anyone think a Windows drone is going to ever accept the fact that an Apple computer is better than anything they can buy in the Wintel camp? To them, the advantages are just as non-tangible as their creative ability.
As long as there are people out there that try to show the good side of Apple computers in this way, you’ll find plenty of people that oppose them. Creativity isn’t just Photoshop, and at the same time many people get along just fine with the Windows version of Photoshop.
As long as they can run Doom III faster than their neighbor they are happy campers. ;o)
Damn, you mean to say that new G5 my neighbor ordered won’t be able to run Doom 3? I think he’s going to be pissed.
I am amazed by Apple fans’ lack of skepticism. Too many swallow what Jobs says hook, line, and sinker. This is why these misleading benchmarks piss me off. For some reason the Apple fanboys actually believe them. Do you have any idea how many posts I’ve seen claiming that a 1GHz G4 is faster than a 3GHz P4?
Well, I for one am a skeptical Apple fan, and I simply say ignore Apple’s benchmarks and wait for 3rd party application benchmarks, yet many seem content to bicker about Apple’s benchmarks…
I think what’s actually in question here isn’t Apple’s reputation towards lying on benchmarks, which is already about as bad as it can get. So ignore it, and let’s wait for an answer to the question that really matters: how does a G5 Mac perform in comparison to a PC?
In addition to the gcc compiler they should’ve tested an optimized compiler for both the Dell and the G5. Afterall, whether I use a G5 or a Intel based processor I want the fastest compiler available for that platform.
That would’ve been a better benchmark and benchmarks suck anyway.
Or is this topic very very stale?
Eugenia, something new please this topic is definitely exhausted.
– Apple gets too much credit for their computer. We should congratulate IBM instead. It’s their CPU, after all.
– Any Mac zealot claiming that “they” always had the upper hand in the hardware is an idiot. ~80% of the technology in the G5 is borrowed from the PC industry.
– Any PC zealot claiming they always had the fastest hardware is an idiot. They should learn history. Apple was once the leader.
– I love how both sides are using “facts”, especially the Mac elitists. Yes, Mathematica, Photoshop and other software like this might be faster on the Mac… But these applications were built & optimised for the Mac! I also like the Q3A benchie… They’re using a dual G5 against a single P4. Fair, isn’t it?
– Macs are excellent in multimedia/graphics/maths applications, it’s not because it’s faster in THOSE applications that it’s definitely the fastest desktop computer around.
– I wonder why they are using the XEON in the benchies. It’s a server chip. Ok, I know that it’s used in many workstations… but why aren’t they using the Opteron then?
– Apple might have win this battle, but they didn’t won the war!
Okay, that explains the System Specs that Veritest ran, I stick by my first conclusion. We still dont know everything and even the veritest PDF is still incomplete.
{ Just because its an Apple sponsored event doesn’t mean they lied or cheated. Actually the way the benchmarked the platforms was extremely fair. They could have made it FAR more disproportionate by being misleading, but it appears that they took the high road. They should be commended not chastized. }
It was not fair, even the Veritest document is incomplete. The margins on the tests were so small that if people wanted to argue the fact, how much RAM was put into the G5. The test doesnt say, they only say for the Dell models. If I put 2 gb or Random Access Memory into the Dells and put 4 to 5 gb of Random Access Memory, of course it will outperform the other machines. Secondly GCC is so buggy and slow on the PowerPC that GCC 3.3 didnt fix many of the performance problems, it fixed most of the code build reliability. But, that is beside the point. My point is when your reps make it a point to either hide, or ignore, or not answer questions regarding your tests or your configurations. It makes people think you have something to hide. And it doesnt matter how fast the G5 is, when you have those type of conditions it makes you look bad. Its similar to ford selling you a car and not letting you test drive it first. Whether Apple was honorable or not, when you remain so tight lipped about somethings, especially something trivial and that if it was revealed would make you look even better. It just doesnt look very good.
“one they arenot telling anyone the specs of the Dell models used.”
Once again Doughnut thinks his own laziness and ignorance forms an argument.
It’s actually called flaming, arrogance, hypocrisy, stupidity, ignorance… Doughnut.
It was not fair, even the Veritest document is incomplete. The margins on the tests were so small that if people wanted to argue the fact, how much RAM was put into the G5. The test doesnt say, they only say for the Dell models. If I put 2 gb or Random Access Memory into the Dells and put 4 to 5 gb of Random Access Memory, of course it will outperform the other machines.
How RAM intensive is the SPEC benchmark? I would think that anything much beyond a gig shouldn’t really matter. The amount of RAM only increases performance up to the point where the OS no longer has to swap to disk for virtual memory. Therefore, I think this is irrelevant.
>>The margins on the tests were so small that if people wanted to
>>argue the fact, how much RAM was put into the G5. The test doesnt
>>say, they only say for the Dell models.
PLEASE READ THE PDF! It does indeed say. In fact, both Dells had 2GB of memory, whereas the G5 only had 1.5 GB.
>> If I put 2 gb or Random Access Memory into the Dells and put 4 to 5
>>gb of Random Access Memory, of course it will outperform the other
>>machines.
You are right. It was an unfair test. Since the PowerMac had 33% less memory, I think the results are clearly skewed. In order to get the actual results, please multiply the reported SPEC ratings by 1.33 for the G5. Thank you and have a nice day.
(Actually I doubt the RAM has that much an affect on the scores, but it just shows your ignorance and that you are apparantly having difficulties reading the PowerMac G5 system stats table in that PDF)
“Yes, Mathematica, Photoshop and other software like this might be faster on the Mac… But these applications were built & optimised for the Mac!”
Woah there buddy, it was only a couple months ago that all the PC fanboys were saying that Adobe didn’t optomize photoshop for the Mac anymore which explained the company’s PC preferred page.
They were probably right. Photoshop wasn’t as optomized for the Mac as much as it was for the PC. And yet, the G5 beat it ANYWAYS! You guys can’t have it both ways.
“I also like the Q3A benchie… They’re using a dual G5 against a single P4. Fair, isn’t it?”
No, they had comparisons in which a single processor 1.8 G5 competed against a P4 and it won in the at test too. They then compared the dual processor G5 to the dual processor Xeon and it won in that test too. (Just keep repeating to yourself, “the bechmarks were fair” “the bechmarks were fair” “the bechmarks were fair” and then maybe you’ll get it.
“Macs are excellent in multimedia/graphics/maths applications, it’s not because it’s faster in THOSE applications that it’s definitely the fastest desktop computer around.”
Exactly. Its because its faster (or as fast) in all applications where there is same or equivilent software between platforms which proves that the G5 is faster.
“I wonder why they are using the XEON in the benchies. It’s a server chip.”
Because the Xeon is the fastest available chip on the market today. People are well aware of the speed of Xeon after having used it and experiencing its speed.
“but why aren’t they using the Opteron then?”
Because AMD isn’t finished with it yet would be the most logical conclusion that I could come up with.
“Apple might have win this battle, but they didn’t won the war!”
I’m glad you recognise that the war is still one as many PC fanboys claimed the war ended a couple years ago when the G4 didn’t get the speed that it is capeable of.
The 970 980 etc have a LOT of room to ramp-up. Although only time will tell, it looks like they have their work cut out for them.
“It was not fair, even the Veritest document is incomplete. The margins on the tests were so small that if people wanted to argue the fact, how much RAM was put into the G5.”
I’ll see if I can find it, but i distinctly remember reading that the x86 machine got 2 Gigs of ram and the G5 got 1.5 Gigs. I’ll try to find where i found that.
“The test doesnt say, they only say for the Dell models. If I put 2 gb or Random Access Memory into the Dells and put 4 to 5 gb of Random Access Memory, of course it will outperform the other machines.”
its funny you say that considering that the Dell actually got 500 more megs of ram than the G5.
“Secondly GCC is so buggy and slow on the PowerPC that GCC 3.3 didnt fix many of the performance problems, it fixed most of the code build reliability. But, that is beside the point. My point is when your reps make it a point to either hide, or ignore, or not answer questions regarding your tests or your configurations. It makes people think you have something to hide.”
What are you talking about? Nothing his hidden OR ignored.
“And it doesnt matter how fast the G5 is, when you have those type of conditions it makes you look bad.”
but there were no such condictions that you describe. The comparison was very well documented as has been reiterated time and time again on this thread.
“Its similar to ford selling you a car and not letting you test drive it first.”
It might be if that’s what happened, but it didn’t
“Whether Apple was honorable or not, when you remain so tight lipped about somethings, especially something trivial and that if it was revealed would make you look even better. It just doesnt look very good.”
As has been mentioned on this thread numerous times… They weren;t tight lipped at all. The comparison was very well documented.
Woah there buddy, it was only a couple months ago that all the PC fanboys were saying that Adobe didn’t optomize photoshop for the Mac anymore which explained the company’s PC preferred page. They were probably right. Photoshop wasn’t as optomized for the Mac as much as it was for the PC. And yet, the G5 beat it ANYWAYS! You guys can’t have it both ways.
Maybe. I wasn’t really interested by the Mac before, so I didn’t followed that story, hehe.
No, they had comparisons in which a single processor 1.8 G5 competed against a P4 and it won in the at test too. They then compared the dual processor G5 to the dual processor Xeon and it won in that test too. (Just keep repeating to yourself, “the bechmarks were fair” “the bechmarks were fair” “the bechmarks were fair” and then maybe you’ll get it.
Well, my source is here: http://www.apple.com/powermac/graphics.html
Exactly. Its because its faster (or as fast) in all applications where there is same or equivilent software between platforms which proves that the G5 is faster.
You can’t claim that… and I can’t claim that the G5 is slower EITHER. Why? Because we can’t test it. Apple made a paper launch and the benchies came from an “independant” reviewer. I don’t doubt that the G5 is an awesome CPU, but I doubt that it’s as good as they claim.
Because the Xeon is the fastest available chip on the market today. People are well aware of the speed of Xeon after having used it and experiencing its speed.
Fastest? I’m not sure. I’ve seen many benchies where the Opteron crushed the Xeon…
Because AMD isn’t finished with it yet would be the most logical conclusion that I could come up with.
Not finished? Then why are they selling Opteron servers? I must admit that you can’t find them everywhere, but they still exist!
I’m glad you recognise that the war is still one as many PC fanboys claimed the war ended a couple years ago when the G4 didn’t get the speed that it is capeable of.
Well, Apple/Motorola were so quiet that we did really thought that the war was over, but it’s nice to see that they now came back.
The 970 980 etc have a LOT of room to ramp-up. Although only time will tell, it looks like they have their work cut out for them.
I hope so. Competition is always good.
for all the so called experts trying argue who manipulates and who doesn’t should read this article in The Register
http://www.theregister.co.uk/content/39/31416.html
seems to be a fair article and well written.
arstechnica did a summary of the keynote, here is an excerpt…
“The Adobe, Emagic and Mathematica benchmark bakeoffs were a bit more believable and relevant. I wouldn’t be surprised at all if the dual-G5 really did out-do the P4* on these apps, because they’re very bandwidth intensive and they give the G5’s excellent frontside bus and memory subsystem a chance to shine. What makes a media workstation is bandwidth, and not just raw CPU power. In fact, high-end Unix workstations from SGI and Sun have always been beefier on the bandwidth side than on the CPU horsepower side–this is one reason why they’re so expensive. The G5 tower looks like a real contender when it comes to moving data around, and that’s what counts for pro applications.”
Intel is running with its tail between its legs right now scatching their heads. It’s nice to see Apple ahead of the other 95% out there again.
Face it, Apple has a great looking system, first 64 bit personal workstation on the market. Mhh beat the wintel crowd to that one(64 bit poor loosers).
Duu I think people are forgetting that the 970(980)is based on the POWER4 chip that outperforms Intel. Remember it’s based on the POWER4, POWER4 PC Zealots!
So it’s just natural for a single and a dual to outpace Intel based Dell system. Sorry losers risc isn’t for kids!
Hmm according to this article it runs Linux compilers faster. Then the G5 a great LINUX MACHINE as the article says…..!
There has been a significant brouhaha over the recent benchmarks and marketing hype from Apple. Truth be told, I personally don’t care all that much if Apple, Intel or AMD is at the top of the performance heap. Granted, I am very excited about the G5 – it is long overdue. Deep down, I’m an OS and interface junkie. I work on a tricked out Dell C840 laptop during the day at work running Windows XP. It has a gig of ram, a 1600×1200 resolution screen, DVD/CDRW combo drive and a 1.6 GHZ P4-M processor. Does it feel fast? Not particularly. A day doesn’t go by that I wonder what the hell the thing is doing. It thrashes a lot and while it never ‘locks up’ in the traditional sense, I find myself rebooting it a few times a week. I have yet to find a media player that doesn’t stutter horribly under the slightest load (Winamp, Real Player, WMP and a handful of other also-rans.)
I’ll concede that Windows XP is the best version of Windows yet. Take it for what you will. Mac OS X, however, is the best OS I’ve ever used. I’ve dabbled with various Linux distros and will admit that they’ve come a long way. I still believe, however, that the proper place for Linux is in the server room and the pathological tinkerer’s desk.
I’m writing this on a G3/500 Pismo Powerbook ,which, incidentally, has been running wireless since I bought it in 2000 – Intel’s Centrino campaign makes me laugh. The G5 might not be the fastest box on the block and, even if it were, it is a tenuous position. There is always something faster lurking around the corner. Apple has finally delivered a promising new architecture, and, perhaps more importantly, has a clear roadmap.
If you run Windows and like it, more power to you. You just don’t know what you’re missing.
He’s exactly right – we can tell if something’s fasr enough for us. I still contend that this is not a Mac vs. PC thing anyway (although Jobs & Crew made it that in their demo). But, you have to see things from the Mac user’s point of view. When the G4 came out, it was blazing. And then Intel starting shooting ahead and there were only little incremental updates to the G4. It was like trying to wring water out of a rock. It was really depressing.
So, allow us Mac users to celebrate. Macs are now back in the game and that makes everything more interesting.
You are 100% correct. Mac fans should be celebrating (though too many Mac fans’ idea of celebrating is baiting PC users into an arguement – the reverse is also true). They finally have a great CPU again. My intention here was not to rain on anyone’s parade. It SHOULDN’T be a Mac vs. PC thing. Like you said, blame Jobs and crew for that one. The true differences are likely small enough that they will make very little difference in anyone’s buying decisions (OSX vs. WindowsXP vs. Linux should make a MUCH larger difference).
I just want CPU mag to run a test and let everyone know the differences. I’m a great fan of Intel, but I prefere OS X, so at home I run a mac.
Akrout: The G5 is actually a derivative of Power4. It’s basically the same processor which we took from the Power4 and we brought it to G5. We’ve added in the G5 beside the processor, the Velocity Engine, which optimized with the same kind of pipeline, with more enhancements we were able to put there. It connected to the fastest front-side bus, which we developed with Apple, which is able to get up to 1GHz. For this we used the most leading-edge process technology we have today. 130 millimeters, with the silicon-on-insulator, copper technology, eight levels. This is our leading edge technology from architecture, micro-architecture optimization and process technology.