I thought the PC crowd was critical but as dedicated as Apple users are I am a bit surprised about this thread. It seems a loyal Apple user who currently owns a G4 thinks Apple is misleading them about the actual performance of the G5. I am not going to rant and rave as I think it best for you all to make up your own mind. I am quite happy sitting here with my P4 2.4 ghz desktop replacement laptop in Windows Bliss. [www.haxial.com] there is also another thread at [AMD Zone] the threads keep building [The Register]
21 Comments
Comments are closed.
well, i dont watch too much TV, but i have never seen a computer ad on tv…but everywhere i go i see apple ads on billboards! everywhere the same white macs or an ad for their music dll site…
face it, when the best the PC community can come up with is “dude, yer gettin’ a dell” and nothing matters more than price point…well, then why not buy a yugo with mentality. after all, a car is only a car and NOTHING more. my guess is you’ve never thought of that angle. nope. apple doesn’t even need to advertise on tv to sell their product, and they only do so very sparingly, and only to announce a newly released product. hmmm, it’s obvious who is grasping in desperation. face the music pc geek. you’re a sellout to a price tag. enjoy your ramen noodles.
“i like to get my hands dirty”
you said a mouthfull with that comment. buy a pc and you have no choice but to become an IT expert. i’ve worked on both platforms for over 5 years and trust me…something as simple as loading a font on a pc is nothing more than an exercise in bullshit. hell…a pc can’t even dislplay a discernable font name in it’s font directories. i want helvetica and yet i get “helv_5578BB_o7650” what the fuck? idiots.
scroll wheels? is that the BEST you’ve got? go back to your windows-i-wanna-be-mac interface and shut the hell up. i can buy a third party mouse with a friggin’ scroll wheel for my mac if i’m that attached to that clunky bullshit feature. if that is what you are basing your buying decision on, then hell…YOU DESERVE A PC. cheap shit.
“As for the $129 they want for Panther…I think that’s absurd. Notice the complete silence in the audience when he announced it. It’s a minor point upgrade, and it should be free (10.2 to 10.3). No way should it he expect people to pay another $129 when they have already shelled out $129 for Jaguar, another minor point upgrade (10.1 to 10.2).”
Wrong. 10.2.1, 10.2.2, 10.2.3, 10.2.4, 10.2.5, and 10.2.6 were minor point upgrades, none of which Apple charged a cent for. 10.1, 10.2, and 10.3 are major point upgrades. I, for one, will gladly pay for 10.3 (albeit with my .edu discount!)
“This casts doubt in my mind on any contention that Appe has corrupted the tests — particularly when the “authoritative” site has “AMD” in its name. Please.”
I would, when like for like AMD chips have been spanking both intel and apple chips for countless years, and when the hardcore users and fans of AMD chips know more about tinkering with the hardware than the average apple user does about a two button mouse.
“We have the ergonomic elegance of OS X and you guys have a virus infested, cluged work environment known as Windows.”
you guys got scroll wheels yet? ;-)
how about more than one mouse button by default. ;-)
nice of you to point out we have a work environment though: mac’s look nice, windows work. :-D But whatever floats your boat, mac’s never have done for me, i like to get my hands dirty, and have more than adobe products and outdated games to load on to my system.
What about the real world tests? The G5 smoked the DILL or did Apple just make that up as well ? As for panther well that is about 2-3 years ahead of wndoze. You know why M$ call it longhorn ? It will make all the little dicks think they have a longhorn.
Uhhh, anyone who believes what was written at the Haxial site ought to chant to themselves that not everything on the Internet is as it seems. Both their Windows and Mac software they write is some of the worst crap around and they just seem to share the same code (so you have twice the amount of crappy apps regardless of platform). The guy lacks any sort of credibility either based on the snippets he responsed to from some of the chatter as a result of his op-ed. I don’t agree with some of vitrole that was posted as feedback at the MacNN site but there are trolls always on either side that write such feedback. This guy has gotten more than his 15 minutes worth of fame and continuing to redirect attention his way does not help anyone.
As for benchmarks, anyone who is not gullible should realize that such tweaking and stretching of the truth is normal across the industry. Sorry, I don’t believe everything I read and hear. Also, Mr. Jobs’ famed RDF wore off on me a long time ago and I always take what he says with some wide latitude and skepticism. The bakeoffs they did I also take with a grain of salt. I’d prefer to draw an opinion on these new systems once they’ve been released and how they really stack up under real world usage with a variety of applications.
Zealots exist on all platforms (hardware and software) and this age old argument of fastest, biggest, best seriously is mundane and non-productive. I say the more choices, the better because it provides a more competitive environment where ultimately, consumers win. And as for such users, get the tool that best fits your needs.
Take everything with a grain of salt. Neither Apple, Dell, Intel or Microsoft will lie about what they do, they simply won’t tell you the whole truth.
That’s Life Folks…
I won’t repeat the optimizations, etc. of Intel and Apple. The only thing that matters to most people is real-world performance and here it appears that Apple is way out in front (notwithstanding the elegance, stability, and power of OS X) on many important apps such as Mathematica and Photoshop. As a scientist I am more interested in floating point and again Apple is way out front. Currently integer performance is comparable.
IBM will scale this rapidly and it supports up to 16 processors. I will use MPI to do serious number crunching for bioinformatics, etc (and 64 bits is important here).
I just LOVE this line (from :
“The G5 seems more of a workstation platform, and the benchmarks shown certainly were not of the desktop variety, nor where the pricing which ranged from $2,000 to $3,000. No one pays $2,000 to $3,000 for a desktop computer. They pay sub $1,000. The eMac is a desktop class computer. That is in the $1,000 range. For a workstation that is another story. So what we are truly talking about is 64 bit workstations. The PowerMac G5.”
That’s a pretty good indictment of PC desktop performance right there. I’ll pay $3,000 for a desktop Mac. Not a PC, sure as heck. They’re cheaper for a good reason.
(Yeah, that was a troll. Couldn’t resist. Hey, everybody’s doing it! Hmm…Using that argument, maybe I should get a Dell after all…)
won
The genuinely funny thing is that this has quickly degenerated into a “mine is bigger” argument that misses the main point: Apple’s G5 dual processor is as fast as the fastest dual processor Windows machines. It may be a little faster or a little slower overall – and appears to be a lot faster in floating point operations (which are important to Apple’s graphics and music customers). Yet last Friday, everyone was talking about the megahertz gap, with Apple being hopelessly behind. That’s not true anymore. The single processor machines are a little expensive, but no more than Apple’s past offerings. The dual processor one is currently a bargain. I’ve priced the closest equivalent at Dell and it did come out to a grand more. I assume that price will drop faster than Apple’s.
Just who consecrates the “amdzone” site as official? Why should I trust that any site named “AMD” will be less biased than Apple?
As for the Haxial site, it speaks of a Dell with a 3.2 ghz P4 when Intel itself only just announced the existence of that chip yesterday (Monday). Dell might list it as an option, but don’t be surprised if it won’t be available for some time yet. (Not to mention that the difference between 3.06 and 3.2 clock speed will be negligible.)
The Haxial site also downplays the importance of floating point speed, even as it also complains that Apple didn’t test the P4’s SSE2 floating point instruction set. Why make an issue of this if it’s not important in the first place? Now it’s true that if all you do on your computer is e-mail and word processing and the like you will never require high floating point performance. But a big part of Apple’s marketing strategy is the “digital hub” concept in which the Mac is used for music, photos, and videos. And THESE functions (as well as sophisticated games) make great use of floating point math.
But the MAIN point here is this: these benchmarks are comparing a 32-bit P4 chip running 32-bit software to a 64-bit G5 chip running 32-bit software. So it’s not really a fair test of these two architectures (as others have pointed out). The G5’s full capacity is not being used in these tests — not by any measure.
Besides, even at that, Apple has acknowledged that its tests show the G5 to be somewhat slower than the P4 in integer performance (when both are using 32-bit software). So what’s the point of “rigging” tests if you still show that your competitor wins?
This casts doubt in my mind on any contention that Appe has corrupted the tests — particularly when the “authoritative” site has “AMD” in its name. Please.
In addition, an anonymous user on another web site, under the moniker, “Mr Know-it-all,” adds the following observations:
“I haven’t read the full original report, but the Haxial site’s interpretation of some of the excerpts looked questionable to me. The excerpt purporting to show SSE was disabled, for example, in fact says nothing of the sort – it simply appears to say that for machines in which SSE2 was not implemented, 387 instructions would be used instead. Which would not apply to P4 or Xeon boxes.
On the flip side, Apple chose to use GCC rather than the Intel compiler because it’s a poorly-kept secret in the industry that Intel has specifically designed that compiler to grossly inflate its SPEC ratings.
It may or may not be true that the tester allowed GCC to take advantage of G5-specific features but not Intel-specific ones; as I said I haven’t read the original, but I don’t see the evidence to support that in Haxial’s report. But no one in their right mind would use SPEC ratings using Intel’s compiler on Intel boxes to compare Intel chips to anyone elses. This test is clearly more “fair” than any of the other reported ratings.”
I am a dedicated Apple user, but it’s important to remember that they are just a company and will do what they need to do to sell computers and stay in business. I still recall the pre-clone era where they produced expensive, garbage equipment like the PowerMac 5200, and were only challenged when companies like PowerComputing continually produced better mac-compatible machines for less money.
It’s seems to me to be a fairly common practice to mention those benchmarks that favor your machine and if you need to to some tweaking to make that happen, then you will do it. I don’t discredit Apple for their benchmarks…I just think it’s a serious thing to call the PowerMac G5 “the world’ fastest personal computer” when raw speed would indicate otherwise. I’ll know for sure when I get PhotoShop running on one and can flip between PS and Illustrator and InDesign and my other apps instantaneously and maintain a decent workflow.
As for the $129 they want for Panther…I think that’s absurd. Notice the complete silence in the audience when he announced it. It’s a minor point upgrade, and it should be free (10.2 to 10.3). No way should it he expect people to pay another $129 when they have already shelled out $129 for Jaguar, another minor point upgrade (10.1 to 10.2).
I would definitely like to see know about the system modifications anyone makes before running a spec test. Granted, I care much less about spec tests than “real” world tests. But I think it’s good that the documentation is available. Like others, I wish it were available for everyone’s tests.
And yes, as the previous poster noted, Apple fans are particularly hard on the company as well. Want to touch a real sore spot? Just say 15 inch powerbook.
And finally, the three threads mentioned (or first three anyway if you keep adding) are all really the same info. The haxial one is probably the most complete. He doesn’t need to posture as a mac lover though. No need. He should just state the facts.
Take care,
g
Idiots and Apple Haters Abound –
The first item I noticed after the Jobs intro was a slavering mad mention in the San Jose Mercury accusing Apple of unfairly profiting by having the gall to CHARGE MONEY for the upgrade from 10.2 to 10.3 (Panther). Pick the one downside item in the presentation to bitch about – typical press reaction to another apple leapfrog over the competition. I guess in their narrow view, Apple shouldn’t charge for 100 new features and blazing speed …
That said… it is only cheating if you are breaking established rules and are doing so deceptively. Apple isn’t trying to put anything over… they’re simply doing what anyone else does: accentuate the strengths of their product. No bluff was really made, as Apple is totally up-front about the tests, if you look. Also, ultimately, consumers and prosumers don’t need to worry about benchmarks… just look at the side-by-side Photoshop and Logic/Cubase tests. That real-world enough for ya’?
What people fail to understand is that x86 companies have been tweaking their benchmarks for ages. They have even gone so far that totally abuse or “sponsor” (i.e. buy) sites, publications, analysts and benchmarks. You ‘ve heard AMD accusing Intel. You’ve heard Sun’s response to Dell’s FUD. You remember M$ petshop fiasco. You are bored to death reading about NVidia and ATI…
Mac users are usually only devoted insofar as Apple’s products deserve devotion. We have very high standard for Apple Computer, Inc, and are the most vehemently critical and dissappointed when Apple fails to deliver. Remember, this isn’t devotion to a cult or something… it’s devotion to a superior product. We’ll call a bluff by Apple and give them a harder time than anyone else will.
Apple tells plainly what it did in its bench marks, something Intel and the Window’s crowd never does. The truth is that Dell and the Window’s Intel world do the same, they tweak their systems to make them stand out better than they actually do. At least, as said before, Apple admits they tweak the tests…. don’t hold your breath waiting for Intel to do the same!
dD
What the nerds always seem to forget is the HUMAN speed element in the long standing debate over which platform is the fastest:
Bottom line:
We have the ergonomic elegance of OS X and you guys have a virus infested, cluged work environment known as Windows.
Nough said.