BareFeats is continuing their benchmark articles for the fastest workstation/desktop machines around.
Apple’s G5 Power Mac versus Pentium 4, Dual Xeon, and Dual Athlon
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSNews.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
Looks like the G5 is holding its own against similar machines. Its a big improvement from the older barefeats tests where the G4 was only half as fast as the competition.
damn apple for not having a g5 ibook out yet if it was out i could get a desent notebook but i am glad they finaly have a good cpu praise IBM
They never seem to publish thorough benchmarks. Even if it is showing my beloved Mac ahead of the game, or keeping up with others, I would completely disregard any benchmarks from this source. Sorry, Bare Feats.
the machines are great, what lacks behind is still the OS.
These are comparisons of rather generic cross-platform applications.
So sad to think that people will use Cinebench or AfterEffects to compare systems. You could look at comparable numbers, or even numbers that marginally favor the Xeons; when in reality you would buy the Mac to run Final Cut Pro HD or Motion to do real-time effects that just aren’t possible on the other systems.
Conclusions: Dual G5s are substantially faster when running optimized code. Apple makes better software.
Apple does make better software. I bet that if there was an OS X release for Intel/AMD machines that was able to run every software that XP runs, I bet the difference in performance would be staggering. An intel/amd based computer running OS X would beat a G5 running OS X no matter what.
the g5 is a 64 bit proccessor so its not fair to comapare it 2 anything intel makes because there 64 bit chip sucked and as for amd vs the IMB G5 i dont even want to think about comparing because i like both companys
How well can you compare benchmarks when the test machines aren’t even running the same operating system? Here’s what I’d like: and unbiased benchmark and comparison of a dual 2.5GHz G5 system, a dual 3.6GHz 64 bit Xeon system, and a dual 2.4GHz Opteron system. All should be using the same memory (or at least the same amount and speed, PC3200 is what should be used for all the above sysems), preferably the same video cards if applicable to the tests, and be running the same operating system, whether it’s FreeBSD, Linux, NetBSD, etc.
The benchmarks used should definitely not be tailored to a specific platform (as Photoshop is, for example). Applications should be compiled specifically for that platform.
Really, is it too much to ask for? If these places have the resources to get their hands on systems like these, can’t they at least be bothered to set up a controlled benchmark environment?
http://www.thejemreport.com/modules.php?op=modload&name=News&file=a…
This is what a good benchmark should be.
“Looks like the G5 is holding its own against similar machines. Its a big improvement from the older barefeats tests where the G4 was only half as fast as the competition.”
Half as fast, and a quarter as hot
.
You are right that these benchmarks are not really unbiased, but the method you mentioned is biased as well.
Why? Because if you choose, let’s say, Linux, you will use an os that is optimised to run on x86 processors (simply because most of the developers own them: how and why do you optimize for a G5 if you will never own one?), so it will likely be faster.
The same is true for the benchmarking programs you use. Of course you can use an open source benchmark that you compile yourself with gcc on all your test machines. But again, gcc is optimised for x86 as well and PPC code will be slower in comparison.
It could even be that a closed source benchmark on different operating systems is more unbiased than your approach:
If you consider that a company releases it’s software to x86 and PPC and expects sales which are reason enough to optimise for both systems, then you will get the same software on different architectures which is optimised as much as possible.
In a nutshell: It’s not true that running the same OS and the same benchmarks on different system leads to more comparable benchmarks.
I noticed that the G5 lost in one test, the Cinebench test. I was reading on Cinebench’s website, that there is a G5 optimized version of the benchmark program – The testing site doesnt show any mention of that. On the program’s website, they claim a 20% boost in performance, by my calcluations, that would blow away all the computers, even in the first test.
Sorry if I have read wrong, and they used this optimized version.
If they wanna do benchmarks, it’s best to compare them once the nexus rendering engine comes out. Its optimised for mac and pc so these comparisons should be fair
The problem with that is, someone would eventually say. “It’s totally unfair, these test were optimized for <insert processor who wins test here> more!”
/me throws two pennies in the bucket & walks away
Because if you choose, let’s say, Linux, you will use an os that is optimised to run on x86 processors (simply because most of the developers own them: how and why do you optimize for a G5 if you will never own one?), so it will likely be faster.
Probably, but at least it’s more fair than having one system run Windows and the other OS X, as long as it’s the hardware that is being compared.
And you’re right that a closed source benchmark could be less biased, but I was thinking more along the lines of “real world” tests, perhaps OGG Vorbis encoding, AES encrypting/decrypting speeds, rendering, compile time, etc.
It’s improbable to have a perfectly unbiased benchmark, but I guess that what I was trying to get at was that at least it could be done much better than this.
I guess fair is fair… Photoshop & After Effects are REAL WORLD examples of what people really, honest-to-god use… Yes, they are different code bases, different OSes… so what?
I am happy to come home to my G4 after a day woikin’ on a PC and XP. IMHO
Jb
Photoshop & After Effects are REAL WORLD examples of what people really, honest-to-god use… Yes, they are different code bases, different OSes… so what?
What about those of use who wouldn’t run Photoshop and AfterEffects on that hardware? We would have no way of judging the performance of the hardware for other duties if only applications that are optimized for one architecture and operating system are used.
It’s no secret that Photoshop runs best on Apple computers, but what I’d like to know is whether the hardware is truly faster, or if it’s only the optimizations that make it seem that way. In fact, if I got a Mac, I most likely wouldn’t be running any of the Mac-optimized applications; I’d like to see whether it would be worth it or not in terms of performance.
I guess the benchmarks are real world enough… the thing that gets me is the PC hardware is old as! A 3GHz P4 is *not* state of the art any more, neither is a 2.1GHz Athlon. I’m not totally up with Opteron speed measurements, but I’d take a bet that there’s something faster than a 2GHz part.
Everyone knows that the G5 does more per clock than the P4. Big deal; that’s a big factor in why they don’t clock as fast. The thing that actually matters is how the equivalent chips clock against one another; which doesn’t mean the ones with the same number of MHz. Same number of dollars would be a much more useful figure to compare on.
The G5 did get beaten in one benchmark; when they found a Xeon chip made in the last two years that made 3GHz, it actually competed with the G5. I’d be interested to know how an Athlon64 would go; but it seems putting a 3800+ in the mix might have proven embarrassing, and we couldn’t have had that now could we…
How well can you compare benchmarks when the test machines aren’t even running the same operating system?
If you are just testing the CPU that is correct, but if you are buying a Mac you are most likely buying it for the entire experience – and that includes OS X.
There is nothing wrong with benchmarking a system but you have to know that is what you are benchmrking.
My own opinion is there is no clear leader unless you are doing very specific tasks. They are all in the same ballpark but each CPU has it’s own strengths, Opteron has low latency memory access and strong overall, P4 strong integer performance and best compilers, G5 strong on Floating point but relatively weak gcc compiler (IBM compiler is said to be much better).
http://www.barefeats.com/pentium4.html
a typical mac site
a site dedicated to benchmarking yet they tell us little about what is in the machines.
but the biggest joke is comparing a $3k dual 2.5ghz mac to:
dual 3.0ghz xeons that can be had for over $1k less
dual xeon 2.4ghz that can be had for $1.5k less
dual athlon 2.1ghz that is so old you could buy for $2k less
a single pentium 4 3ghz that would cost over $2k less
a dual opteron that is clocked at the same speed as the older dual g5.
so where is the dual xeon 3.6ghz for this test….put apples best up against intels best.
so where is the single cpu intel pentium 4 3.6ghz for this test?
so where is the dual opteron 250 running at 2.4ghz for this test?….put that mac up against amds best too.
so where is the single cpu AMD Athlon 64 FX-53 in this test?
so where is the single cpu amd athlon 64 3800 in this test?
so where are the g5 laptops for mobile power use?
fact is you can both intels and amds fastest offerings in laptops.
fact is you can buy both amd and intels best for less money.
what a joke benchmarking against cpus that came out years ago.
both intel and amd have much faster offerings now out. they have faster memory, they have faster system busses, they both have pci express. the pcs have faster consumer and workstation 3d video cards available for them too.
and all of that can be had for less money!
a comparison of diverse machines with different oses is perfectly valid.
computers are tools. they are bought to solve problems. the single biggest factor in almost all cases is cost of the tool. so when benchmarking, what needs to be compared is what it costs to get the tool and how long it takes the tool to do its work.
“What’s interesting is the fact that the dual processor G5/2.5GHz blows away dual processor Xeon 2.4GHz. In other words, here’s two dual processor machines running at about the same clock speed. Yet, the G5 wins. That gives Mac fanatics something to cheer about.”
sounds like a typical, ignorant mac user that is hung up on the mhz myth. it isn’t about the mhz, its about the cost and what you get for that cost. fact is you could buy two dual 2.4ghz xeons for the cost of the dual mac. its two years older for crying out loud.
no comment either concerning a lower score with the g5 featured here- http://www.barefeats.com/g5c.html and yes, they are running the g5 optimized test.
bare feats is a farce.
How well can you compare benchmarks when the test machines aren’t even running the same operating system? Here’s what I’d like: […] and be running the same operating system, whether it’s FreeBSD, Linux, NetBSD, etc.
The benchmarks used should definitely not be tailored to a specific platform (as Photoshop is, for example). Applications should be compiled specifically for that platform.
What the hell are you whining and complaining about? These are application benchmarks, you tard. If you want synthetic benchmarks head over to spec.org and you can find all of them you want. However, these benchmarks are aimed at people actually considering using these applications on operating systems that can actually run them (i.e. NOT Linux, FreeBSD, etc.)
I have no problem with the software selection. But the hardware selection for the benchmark makes no sense.
They are not comparing equal price points nor equal performance levels so it seems a little mac biased- the very best Apple has to offer compared to a xeon and opteron that have been out since late last year.
A benchmark comparing a Dual Nocona 3.6Ghz workstation and a dual Opteron 2.4Ghz to the lastest G5 would not only have been more interesting but also useful.
I noticed that the G5 lost in one test, the Cinebench test. I was reading on Cinebench’s website, that there is a G5 optimized version of the benchmark program – The testing site doesnt show any mention of that. On the program’s website, they claim a 20% boost in performance, by my calcluations, that would blow away all the computers, even in the first test.
Barefeats is running the Cinebench beta for OS X, which has been optimized for the G5… but futher optimizations are coming, so the G5 score will probably increase by another 10 or 20 per cent.
And for folks looking to see top-quality PC hardware in the Barefeats bakeoffs, just hold your horses. They’ll be testing a dual Opteron at 2.5 Ghz in a few weeks. (Barefeats has to borrow computers for testing… So sometimes you have to wait for the latest and greatest).
FYI: Barefeats isn’t a huge testing lab; it’s a hobbyist site, but the tests they run seem to be quite thorough and unbiased. They’ve trashed Apple as often as they’ve praised it.
I would like to see a few additional Real World tests, just for interest’s sake, of the best against the best. Photoshop on both platforms is a good choice (and I think AMD’s chips do better than the G5, pound for pound… ), but I’d also like to see renedreing and compression tests with Final Cut Pro vs Premiere, After Effects against Motion compositing, and so on.
All the Apple fanboys believe that Apple software running on Apple hardware is what makes our beloved platform King; I’m interested in finding out if it’s so. 😉
The x86 hardware is on average about 2 years old. A 3Ghz P4 is used in $600 mid range home PCs.
How about running 64 bit Xeons and Athlons/Opterons using 64 bit windows? Unfortunately the G5 might be embarrassed by genuine competition.
That’s a very well reasoned and thought out argument you’ve got there. Nice work.
You could, of course, fairly compare the G5 to, oh I don’t know, an Athlon64… and there are 64 bit Xeons out now aren’t there?
Frankly a dual G5 doesn’t really compete in the same market as a single P4 – there aren’t that many multithreaded programs to take advantage of the second processor, and the thing costs a small fortune by comparison. A benchmark saying that a $3000 Mac beats $700 Pentium chips doesn’t really prove anything.
The x86 hardware is on average about 2 years old. A 3Ghz P4 is used in $600 mid range home PCs.
How about running 64 bit Xeons and Athlons/Opterons using 64 bit windows? Unfortunately the G5 might be embarrassed by genuine competition.
Exactly! Apple is a sissy computer for pansies. Everyone knows it. The one-button mouse sucks!
Look at this link, from six months ago, pitting a dual 2.0 GHz Powermac against dual 3.2 Ghz Xeons in a top-of-the-line HP workstation running Linux. Yeah, OS X will be going 64-bit in a few months with Tiger, and the current PowerMac is 25 per cent faster than this one, but look what happens when you use serious scientific software like HMMer and Bibble that really tests the whole system…
It ain’t pretty, mac fanboys. There’s no way Apple can catch up. It’s over. You lost. Go back to Momma.
http://www.popularmechanics.com/technology/computers/2004/4/desktop…
Everyone seems to be missing the point of benchmarking. When I go out and shop for a machine, I look for something that will run the apps that I use the fastest. I’m not interested in a completely “unbiased” CPU benchmark, because in real life, I will be using applications that WILL BE BIASED, based on what I am doing w/ the machine.
For example, if I’m looking for the machine that runs Photoshop the fastest, I could care less about which CPU compiles C code faster while running Linux. It seems at the moment that a Dual G5 2.5 loaded w/ gobs of RAM and running OS X would be the best way to go for that. I already know that the cheap PC guys will say that I can get PC hardware for cheaper, but the fact is that they aren’t getting paid to run Photoshop and other graphics apps, and I am. Something that people often forget is that time=money. I don’t care if I’m paying $1000 or even $5,000 more for the hardware because I’ll make that back with less than half a day’s work. And since I will be more efficient, I’ll be able to go on to the next job quicker.
On the flipside, If I needed a machine for 3D modeling and animation, I’d be buying the latest and baddest AMD system with a pro video card, because I know that will get the job done the fastest and easiest. But again, in this situation, an UNBIASED test does me absolutely no good.
Get it??
After Effects is a great comparision of cross-plateform testing. Most post house use the G5 just for After Effects, because the numbers are evedent that the G5 does a great job with it. I want to see lightwave 3d and Maya unlimited tested, maybe even Combustion. Can’t remember if I saw a lightwave test, but it would be interesting. Makes me want to get a G5 2.5 now.
what is x86 i see it all the time on os news and know nothing about it? The macs do have a couple advantages to the athlon 64s i actully have one of them and that is that osX actully is a 64 bit os unlike xp which to my knowledge is only a 32 bit os
Wait, I hope you’re being sarcastic.. because otherwise you’d just be severely retarded:
“Not being able to run SPEC tests, we turned to BLAST and HMMer, which are DNA and genome-sequence matching tests, as well as to Bibble, a batch image-processing application. The problem is that these tests do not run on Windows XP. In frustration, after running the SPEC tests on the HP xw6000 workstation, we installed Linux on the HP, which allowed us to run the new tests. And we were surprised. The G5 was 59.5 percent faster than the HP at processing 85 high-resolution color photographs totaling 684.6MB of data. In the HMMer tests (61.3MB of data), Apple was 67 percent faster than the PC and under BLAST (32.8MB), Apple was 85.9 percent faster.”
So.. you’re not serious right?
Yeah, I was being sarcastic. 🙂
You’ve been misinformed. OSX as it currently stands, version 10.3.5, is NOT a 64-bit OS. It contains libraries that allow for 64-bit computing, but as a whole it’s not (yet) a 64-bit OS.
As some people have already pointed out, they havce tested the fastest G5 against much older hardware from Intel (presently 3.6GHz) & AMD (presently 2.4GHz). If one were to scale the numbers for 2GHz Opteron (and it does scale almost linearly with the clock, unlike it’s Intel counterpart) in all of these benchmarks to 2.4GHz, it beats the G5 in nearly everything. So where’s the alleged ‘feat’ here, barefeats?!
Sorry… I was trying to be funny, by writing like one of the usual dismissive PC guys around here who take pots shots at Apple, adding heat, but no light to the debate. Like Anonymous (IP: .tpgi.com.au) who said that the Barefeats site only uses old PC hardware because the PowerMacs would be embarrassed by hot PC hardware.
I’m all for debate. And many smart people here can alter my point-of-view on occasion. Anonymous didn’t even try. 🙂
But I also wanted to make certain that some PC advocates would actually read the posted link, which shows a Dual 2.0 Ghz Mac toasting a Dual 3.2 Xeon when serious number crunching is involved with several highly-regarded science apps. And the HP was way more expensive than the PowerMac.
Even offtangent (IP: wayne.edu) doesn’t quite get the Barefeats tests. He’s right; a dual Opteron running at 2.5 Ghz is probably going to best the PowerMac in many tests. Maybe all. Most Macintosh people are probably OK with that.
Here’s the point, for the Macintosh community.
Over the last year, Apple has gone from selling elegant, but anemic hardware in August, 2003 to selling some serious kick-ass hardware that can nail Intel’s Pentium or Xeon hide to the wall, and go toe-to-toe with the best from AMD. And it’s price competitive.
And it only gets better. Tiger, a true 64-bit OS, is just around the corner. So even an old G5 is about to get much faster. And IBM seems to be fixing the problems at Fishkill, so faster Macs will probably be announced at MacWorld in Jan, 2005. Maybe even a G5 PowerBook. And when you pair Apple hardware and software, the computing experience is among the best in the world. Both fast and powerful.
I know; I know; I should be working for Apple’s PR department. It’s taken seven years, but Apple is finally firing on all cylinders under Jobs’ leadership. The company will not only survive, it should even regain some lost marketshare.
We don’t want to change the world, or make everyone throw their XP or Linux ‘puters off their balconies. We just want Apple to have a proud and successful future, and that’s looking as if it just might happen.
And that will be good for everyone, given Apple’s history of innovation.
Let me get this straight. A hobbiest site managed by people in their spare time put together a near-real world benchmark with the available tools at their disposal. The people in this forum complain about every aspect of it. Here’s an idea, get off your duffs and do it yourself if your so damn good at it and have so much free time. You want raw numbers, look at the FLOPS/$ ratio of the Xserve superclusters compared to their Itanium and x86 counterparts. Look at their raw FLOPS/node performance as well. The G5’s are fast machines when code is written to take advantage of the hardware. It’s the same connundrum that a software company faces when utilizing MMX extensions on x86, utilizing the EPIC architecture of the Itanium or the massive parallel vector architecture of the NEC and Cray processors. So shut the f-up already. We don’t have enough chees to go with your whine.
I wish my processor ran at that it only runs at 2.01 GHz
Wait until these applications on the G5 get compiled with XCode 1.5. I’ve seen considerable (>25%) speed improvements with my simulations compared to XCode 1.2.
Amen to gankaku
i want a g5 ibook but from what i have read it wont play the game i want it to play so i could only use it for web browsing and school work so i might as well go with another athlon64 for a laptop even though the osx is sweet
Apple is laughing all the way to the bank thanks to the AMD nerds and PC apologists.
Why?
Because you have effectively changed the debate from “Apple’s are slow,” to “Apple isn’t the fastest.”
Apple couldn’t of been happier about the uproar last year when they got tens of millions of dollars of free advertising when the web trolls were debating if Apple was #1 or #2. Talk about repositioning.
It’s beautiful. It’s why study after study show Mac users have higher IQ’s. The PC guys don’t realize when they are being had.
All said and done, the moral of the story is this: comparing two different machines with different CPUs is a bitch.
“The Dual Athlon 2600+ (2.1GHz) system had 1GB of DDR memory and ran Windows XP Professional.
The Dual 2Ghz Opteron system had 2GB of PC3200 DDR (cas 2) memory and ran Windows XP Professional. (We hope to test this once a 64 bit version of Windows has is available and fully supports our test software.)
The Opteron system was provided for testing by XiComputer.
The Dual 2.4GHz Xeon system had 2GB of PC2100 DDR (cas 2) memory and ran Windows XP Professional. (Hyper-threading was enabled.)
The 3.0 GHz Pentium 4 system had 2GB of 400MHz PC3200 DDR memory and ran Windows XP Professional. (Hyper-threading was enabled.)”
Where are the 3D games test results ?
Comparing a G5 with a dual FX-53 2.4 Ghz,Opteron 250 2.4 Ghz ( Did you know this is socket 940 to ?) would have been more objective in my humble opinion.On what decision was the test based if i may ask? The avaible hardware at hand? Can’t be the prize . A dual FX-53 with mid-end graphics card , simple case ,watercooling, wouldn’t cost more then a G5.
“…I don’t care if I’m paying $1000 or even $5,000 more for the hardware because I’ll make that back with less than half a day’s work…”
Damn, pee drinking pay is that high? So what is that, something like over $200,000 or $1,000,000 a year? Do you get health coverage with that?
In any case I agree with your point.
so if “apple” useres have higher iqs what about the people that use both and prefer there wintel pc? are they just crazy or what?
“so if “apple” useres have higher iqs what about the people that use both and prefer there wintel pc? are they just crazy or what?”
I wouldn’t say higher IQs but a lot of Mac users are in the creative field…
I haven’t yet met a person who works with both platforms prefer PC, all that I know prefer the Mac. They do get enticed by some initial costs and switch to pC but soon after start to complain.
Raw speed isn’t everything… UI is! I still love to go home to my G4/Dual 867 it just feels better.
BTW someone mentioned that Photoshop always runs better on Mac, well wasn’t there a little to-do a little while ago when Apple stated that their software runs best on Wintel?
all, of course, IMHO
Well, now you have. I use both at work, for compatability reasons. I think the Mac looks better (especially with native widgets for Safari), but that’s about it. It’s slower than the PC desktops we use, and of course none of the applications I want to use are available for it. Even Eclipse seems to run slower on the Mac. No rhyme or reason for it, but that’s just how it is. It’s slower and doesn’t run the applications I need, so I prefer Windows.
Intel’s whole advertising campaign was centered around megahertz which now seems to be irrelevant.
What are they going to have now, a CISC revival?
Well, it is another spitting contest. My computer spits faster and further than your computer. In industry time is money (real money) while at home time us usually imagined money. If my computer takes 3 hours to render a movie and yours takes 2 hours to do the same job I am not forever depressed. I will just go eat a meal while it crunches. Whoopee.
Let’s face it. Benchmark tests using real applications are actually the best method of determining relative speeds. The speed of a computer is tied to its’ internal architecture, the efficiency of the compiler to optimize code, and the person who wrote the code.
Many end users are finally figuring out that MHz is only part of the picture. If all you want to do is surf the web, email, and word process then even a relatively slow ?rocessor is adequate.
How about all of us growing up by not trying to prove that we can spit further and faster than the other person. rofl
My favorite benchmark is the one that my dad does on his systems, which, because they’re not identically spec’d is his yard stick to see how much computing power your $ buys. As all of his home computers have cost about $2000, it’s actually not a bad benchmark and gives an idea of how computers have evolved over time.
His test: have the computer compute PI until it crashes. He then notes how long it took and how many digits it got to.
That’s the test I’d like to see: Take what are supposed to be two comperable systems, give them equal ram and drive space, use programs optimised to take advantage of each plaform’s strengths and see how much PI you get.
Hey, the Opteron (so also the Athlon64) beats the G5 on most benches. Fine
The “Mac fanatics” spend years ranting about the “Mhz myth”, but then make comments like this:
“What’s interesting is the fact that the dual processor G5/2.5GHz blows away dual processor Xeon 2.4GHz. In other words, here’s two dual processor machines running at about the same clock speed. Yet, the G5 wins. That gives Mac fanatics something to cheer about.”
Are they stupid, or just unashamedly hypocritical ?
ok i am not a mac fanboi but i do use macs and my pc i love my pc to death (maybe its because i built it) but macs really equal to the pc in most ways get over it or the pc is equal to the mac in most ways idont care how you look at it the hardware is equal the oses are not
Hey drsmithy, think about what you just said for a moment and see if you can figure out why it was the most insipid thing yet in this forum:
The “Mac fanatics” spend years ranting about the “Mhz myth”, but then make comments like this:
“What’s interesting is the fact that the dual processor G5/2.5GHz blows away dual processor Xeon 2.4GHz. In other words, here’s two dual processor machines running at about the same clock speed. Yet, the G5 wins. That gives Mac fanatics something to cheer about.”
Are they stupid, or just unashamedly hypocritical ?
LOL….
pc users for years have talked about how much faster they are than macs.
the ill informed just mentioned mhz.
mac users tried to fight back by talking about other components of the whole system. but the macs have soundly been beaten in nearly all forms of benchmarking for many many years.
savvy pc users of course know that it is more than just the mhz of the cpu alone.
but now we have a site dedicated to macs running benchmarks of a brand new, very expensive, high end workstation against 2 and 3 year old pcs and making the claim it is faster yet it has nearly the same mhz per cpu!
what the authors fail to mention in the case of the mac versus the pc is the following when compared to 2 and 3 year old pcs:
agp bus speed
pci bus speed
amount of ram
amount of cache on cpu
video card model and amount of memory
hard drive type and interface
hard drive rpm
hard drive cache amount
etc etc
so whats funny and ironic, is that the mac now has better specs on both mhz for the cpu and all those other things as well when compared to TWO and THREE year old pcs.
however if the author was to review against currently shipping and less expensive pcs with components like:
top end amd athlon 64 and opteron cpus
amds hypertransport bus
intels raw clock in the 3.6ghz pentium 4 and xeon
larger caches on cpu for some of the models
8x agp and 16x pci express video cards
high end consumer video cards available for pc but not mac
workstation video cards available for pcs but not at all for macs
10,000 rpm sata 150 drives that ship from the maker (they can be added to a mac after buying slower models that have to ship with the mac to start with)
16mb cache sata 150 drives for pcs (they can again be added to a mac after buying slower models that have to ship with the mac to start with)
scsi hard drives are not even an option from apple on their workstations any longer
faster memory speeds on pcs
dual channel memory on pcs
faster cd and dvd burners in pcs
etc etc
the irony is the mac has to compare on cpu mhz alone to win the benchmarks and to do that the author used much older technology that is now so outdated its hard to find except in used pcs on ebay.
compare the total system as anyone with any knowledge always has, and currently shipping machines that cost the same will lead to the pcs handily beating macs. more interesting however is that well built single cpu pcs that cost substantially less will likewise win all but a few benchmarks. since the release of the dual 2.0ghz g5 this has been shown to be true over and over from many different third party reviewers.
but it is nice to know that apple can at least trot out one model as doing reasonably well, even if it is very expensive. now if they could give more bang for the buck on their consumer and portable machines.
What is it with benchmarks and the particular brand of complaining found in this, and other, forums?
The Bare Feats benchmarks can be considered suspect in that the configurations and testing methodology are not fully documented for the results (although the actual tests are), there is no preamble that documents what they’re trying to achieve, and they draw conclusions without analysis.
However, as benchmarks are a comparative tool, there is no reason to consider the results themselves suspect. Those making criticisms about the ‘old’ CPUs being tested conveniently ignored this paragraph:
PC fanatics can cheer about the fact that there are PCs running at over 3GHz. In fact, we should have full test results on a Dual 3.06 Xeon and Dual 2.5GHz Opteron within two weeks.
Now, if in two weeks the results aren’t there, then continue with your attack =)
an objective reviewer would have waited instead of posting half done tests.
secondly-
“PC fanatics can cheer about the fact that there are PCs running at over 3GHz. In fact, we should have full test results on a Dual 3.06 Xeon and Dual 2.5GHz Opteron within two weeks.”
3.06 xeons are over a year old too.
2.5ghz opterons are not made. the fastest run at 2.4ghz.
so the question is to be begged, who are these guys and what do they and what is their agenda?
current xeons run at up to 3.6ghz, have a faster 800mhz bus, have faster memory, and have pci express video.
Couldn’t agree more.
come on
its hard to get your hand on a 3.6 xeon
its not possible to get your hands on a 2.5 opteron
it’s equally not easy to get a 2.5 G5
Result? *They* got 3ghz xeon, 2ghz opteron and 2.5 ghz G5
the g5 win by a good margin in most tests.
Memory in the pcs was all cas2. afaik apple uses cas3, it should kills benchmarks.
most of the arguments in this thread are :
– pc are better because i prefer them
– macs are better because i prefer them
who cares ? Just take a look at the bench. the G5 performs fairly well, and benches always gives only a mixed result. It probably runs in the range of the 3.6 xeon (very new btw), and the 2.5 opteron. Probably a bit slower, a bench would tell that anyway.
So, couldn’t anyone find that out by himself instead of yelling how much xxx is better or how flawed is a benchmark ?
There are way more flawed benches, by the way.
P4 sucks because its not a 64 bit processor. G5 is nice because it was built buy a couple of great companys. the amd 64fx is a great chip. comparing is only useful if you need to know which runs certin things faster because my g5 might be quick with photoshop but my friends p4 might be faster at loading MS word. a benchmark is only useful if you know what you want it for. Trolls love benchmarks because it makes the other seem like an easy tarte or make it easy to say it was rigged but thats probly not true its just that some systems and software are better set for certin work loads and some things are optimized for certin things i really dont see why people keep saying oh the won a bench mark haha you suck deal with it the mac has some good stuff and so do pcs
Both Mac’s and PC’s run great these days(3004-2004), do you think any company these days would have slow machines. Surely not Apple nor Alianware, etc. Both risc and x86 have leap-frogged each other throughout the years, it will happen again. So what is the fuss?? Give me both, I will use them without complaints and I do!
sorry to disappoint but you have made several mistakes and you keep repeating them. ive seen some odd things you’ve written in several threads.
“P4 sucks because its not a 64 bit processor.”
that is untrue. both pentium 4 and xeon have models that run 64 bit oses and code.
http://www.anandtech.com/news/shownews.aspx?i=22700
“what is x86 i see it all the time on os news and know nothing about it? The macs do have a couple advantages to the athlon 64s i actully have one of them and that is that osX actully is a 64 bit os unlike xp which to my knowledge is only a 32 bit os”
another false statement. os x is not a 64 bit os. windows xp however is fully 64 bit on itanium and is in prerelease (free) beta status for amd 64 and intel 64 bit extended.
for someone that doesn’t even know what x86 is, you sure do toss some info around like you are very sure of yourself. relax a bit. write less and read more might help.
and f-
“its hard to get your hand on a 3.6 xeon”
no as a matter of fact you can order one and have it in a weeks time via delivery from dell and others. the dual 2.5ghz g5 however still shows 4-6 weeks delivery time on apples web store.
and to show you what a dual 3.6ghz can do, check out these numbers:
http://www.digitalvideoediting.com/articles/viewarticle.jsp?id=2816…
Haha, for those of you still beating each other up over this tired topic, lemme try to explain again why these “biased” or “unbiased” or whatever benchmarks you wanna call them are completely IRRELAVENT in the real world in this day and age. I do a lot of freelance work here in L.A., and the fact is that there are Macs and PCs everywhere you look in my industry. Here’s an example of why this is:
You just graduated from USC or Art Institute or whatever, and you’re out looking to find some video work. Well, these days you better get really familiar (if you went to those schools you probably already are) with Final Cut Pro, and fast, if you wanna work at Discovey, ESPN, Disney, etc.. Oh wait, FCP only runs on a Mac! How are those benchmarks gonna help you now?
Another situation — similar from the first but in a slightly different area of Hollywood. You’re fresh out of GNOMON <http://www.gnomon3d.com/title.html> and you’re submitting demo reels to different studios hoping to find work. In this case, your work was probably done in Maya or some other high end 3D software, and you’d be absolutely crazy to run that on a Mac because everybody knows there isn’t any pro video card support for OS X and that stuff runs 50 times better on PC hardware anyway. AGAIN, what do those benchmarks do for you now?
If all you do is play games, use a pirated copy of MS Word, check your hotmail account, and troll these forums hoping to pick a fight, then you don’t need those benchmarks either. What you need is to get a life.
your right you rock
according to that artical i couldnt go out and buy a 64 bit p4 because im not an OEM so again i say that the P4 is not 64 becaus A. only some of them are B. Intel still trys to get people to belive the MHz Myth and C. because i can just plan not belive