Slashdot Mirror
Apple Hardware VP Defends Benchmarks
He said Veritest used gcc for both platforms, instead of Intel's compiler, simply because the benchmarks measure two things at the same time: compiler, and hardware. To test the hardware alone, you must normalize the compiler out of the equation -- using the same version and similar settings -- and, if anything, Joswiak said, gcc has been available on the Intel platform for a lot longer and is more optimized for Intel than for PowerPC.
He conceded readily that the Dell numbers would be higher with the Intel compiler, but that the Apple numbers could be higher with a different compiler too.
Joswiak added that in the Intel modifications for the tests, they chose the option that provided higher scores for the Intel machine, not lower. The scores were higher under Linux than under Windows, and in the rate test, the scores were higher with hyperthreading disabled than enabled. He also said they would be happy to do the tests on Windows and with hyperthreading enabled, if people wanted it, as it would only make the G5 look better.
In the G5 modifications, they were made because shipping systems will have those options available. For example, memory read bypass was turned on, for even though it is not on by default in the tested prototypes, it will be on by default for the shipping systems. Software-based prefetching was turned off and a high-performance malloc was used because those options will be available on the shipping systems (Joswiak did not know whether this malloc, which is faster but less memory efficient, will be the default in the shipping systems).
As to not using SSE2, Joswiak said they enabled the correct flags for it, as documented on the gcc web site, so that SSE2 was enabled (the Veritest report lists the options used for each test, which appears to include the appropriate flags).
Really?
If you want OSX, you'll need to get the PPC.
If you want Windows, you'll get the x86.
If you want Linux, you can pick up 10 and build yourself a cluster for the price of one of these new machines.
I have been pwned because my
At least everything that they did seemed to be amply documented.
I found that to be refresing especially in light of all the recent benchmark tests that have not been so forthright with all their methods and procedures.
If everyone benchmarked with open source compilers, there would be none of the shady benchmark-specific optimizations you'd expect to see in proprietary compilers. Everything would be above the table.
And that's not to mention the benefits for OSS compilers. Imagine the kind of resources and funding processor companies would dump into open source compiler projects if they were going to be the basis for their benchmark scores instead of their closed source proprietary compilers.
"The worst tyrannies were the ones where a governance required its own logic on every embedded node." - Vernor Vinge
Why on earth would you want to separate the software from the hardware? This isn't a IBM vs Intel comparison. This is an Apple vs Dell comparison. Apple is selling a platform, not a bunch of PCB boards. I sure as heck won't use GCC to compile SAS or Oracle just before I put up a mission-critical database server...
Unless mankind redesigns itself
I hadn't looked through the detailed report before - one interesting thing was that they physically removed one of the processors for at least one test (SPEC CPU 2000). I seem to remember some people claiming some of the spec tests were unfair when run on a DP system... well there you go.
It really seems like they tried to do a pretty even evaluation. And again, if the benchmarks were so off then why was the performance on the G5 apps so good? And that was without G5 tuning most likely.
"There is more worth loving than we have strength to love." - Brian Jay Stanley
that all software vendors have to be honest now, or just Apple?
Further I notice he didn't mention the problem of not doing comparisions to AMD.
While I can understand his reasoning, the fact is that most software on the PC runs under VC or Intel's compiler. It doesn't run under gcc. The benchmark might be a fair Linux/OSX comparison but implies something about Windows/OSX that is incorrect.
I'd also like to see the tests done under Mathematica and Photoshop discussed more. Apple's had a history with photoshop so there is prima facie reasons to distrust it. But the Mathematica test, which seemed the most exciting to me, is what I'd really like to see.
Realistically though the tools for Apple, including graphics drivers, are all very beta. So we should see improvements with time. And realistically benchmarks are typically kind of deceiving as an indicator of real world performance.
So any word on these other questions?
PS - I love OSX and would love to make a Mac my primary machine. If only Project Builder was up to the task so I could abandon Visual Studio. But I am excited about the G5, but I think Apple's "questionable" tactics have brought a lot of unfavorable press that more honesty would have avoided. Personally I think being within 10% - 15% of the top end PC would have been fine.
It seems to me that this rep from Apple, unless I am very naive, is being very candid and honest with us. It seems that, by showing us the complete specs on the benchmarking, they are doing what they claim to be doing. Thinking differently, and giving us (for 3 grand) an honestly faster machine. I appreciate the prompt frank response from Apple on this controversy. I am typing this on a PC, simply because I could build it myself for less money than I could buy a nice Apple. Ah, the life of a poor student...
Karma: Can there be a void?
.. -. - . .-. .-. --- -...
I have to say, this puts things in an interesting light.
Does a company, in trying to be fair as it seems in this case, get penalized for choosing the best optimization and not testing with the worst optimizations(as per their views)?
In looking at other sites like Tom's Hardware and Anantech, I think the answer is simple: Show all of the results, both the good and the bad. That way, it removes the spectre of doubt in peoples' minds that fairness wasn't present during testing.
Personally, I don't have the funds to get a G5 based system. It just isn't in the budget. But then again, the only reason I would buy a G5 system over an x86(Opteron or P4) would be to run Mac OSX. :)
I'm guessing that tests will be conducted by various groups over the next few days to either validate or invalidate the tests. Sounds alike like that whole MS/cost analysis/web server speed fiasco all over again.
Despite the tests, for Mac users who wish to stick with Mac OS X, the G5s are as fast as they come.
Winged Power Photography
Fuck those $750 PCs, I'm getting me a $3000 Mac.
Q: You know what a $750 dual 3gHz Xeon PC is called?
A: "Stolen Goods"
Really, the point for Mac users is not so much whether or not the G5 wipes the floor with the Pentium, but whether or not the long period of performance stagnation is coming to an end, and whether or not top-end Mac performance will once again be reasonably comparable to top-end PC performance. And it looks like the answer to both questions is YES! (FINALLY!!!)
It really seems like they tried to do a pretty even evaluation. And again, if the benchmarks were so off then why was the performance on the G5 apps so good? And that was without G5 tuning most likely.
Oh, yeah. Steve probably said "hey, vendors, come on over and do a little demo. Yeah, it'll be a duel, but don't worry about recompiling for the G5 (which is supposed to be trivial). We'll just see what happens."
Look -- they spent every last minute they could optimizing the builds they used for the demo - don't doubt it for a minute. On the other hand, every last minute probably wasn't all that long, and the demos did kick ass.
But let's call an Apple an Apple. This was a DEMO. Smoke and mirrors were involved. But I drank the cool-aid; I believe it's faster. Dunno how much, but I don't really care. Mostly I'm just happy it kicks the crap outta the systems they're shipping now.
The Apple guy is both correct and wrong.
Correct in the sense that he wasn't necessarily being unfair. I don't think Apple was raelly cooking the books here. OTOH benchamarking is quite difficult.
No, it would not be fair to compare intel compilers to gcc compilers. But what about, say, another non-hardware tied compiler? Look at it this way - 3dmark scores on graphics cards. Theoretically it should give a good impression on thier relative hardware - but we all know that it doesn't necessarily. It may do something bad on one system, great on another, one system may cheat and have special code to work better with that particular test.
Same here. Ideally you would find many benchmarks, not just gcc, but both with all optimizations on, with all off, both with the best compilers, worst compilers, and middle of the road. You also need memory intensive, processor intensive, grpahics intensive, floating point, integer, and many others to get the full idea and compare it to what you need to do with the computer. For many of the crowd that worries over this stuff overclocking can become an issue also.
This is why benchmarking is as much art as science. I care about all those numbers - I have code compiled specifically for my athlon-mp's, some generic, and some optimized for p4's for the consumer tasks. On our computation cluster we use specialised compilers. I care how it runs on all of it for real world use. But no hardware manufacturer does those extensive of a tests - they pick the best of the ones they can claim "fair" on usually.
And lastly, in the end, who cares? Unless you are regularly running 4 hour jobs from a console it is irrelevent. It is more important that you are productive with the interface and that is personal choice. Few consumer tasks (and even programming tasks) require that power - and the stuff that does is generally handled by specialised hardware. Then if they have the fastests today they won't tomorrow.
------- Sorry about the spelling, I suffer from two problems. Dyslexia makes it difficult to spell well, lazy makes it
Is there good reason to believe that the same compiler will produce relatively as well-performing of code for one chip it supports as it does for another? I don't think so.
In this case, performance will in part be a function of how mature and optimized the generation of code for the advantages of that particular chip is.
Because there is no guaruntee at all of fairness by using gcc for both processors, except of course if we had the expert opinion of someone intimately familiar with gcc's code generation for both processors, using gcc for both processors would seem to be little more than a marketing tactic to give the appearance of fairness and credibility.
It seems to me that a better test is to take the best compiler widely available for each chip, and then run your tests with the produced code. Now, this isn't necessarily real world application testing, but that isn't what we are necessarily looking for here.
How well the processor performs with code generated by the best generally available compiler, is, apart from extraordinary measures, the best prediction we have of how generally the processors will compare for any given well-written, production quality code.
.....can be trusted 100%. Only Apple would exaggerate for marketing purposes.
I have this theory. A 2Ghz twin G5 system is really fast. And if you have some money to spend and you want a really fast system and you'd like to run OSX then you could do worse than buy one.
Now wash your hands.
All this talk of gcc removing a variable is naive at best, misinformation if the speaker is knowledgable on the subject. Gcc is not a constant, the quality of it's code optimization varies from platform to platform. To be more specific, gcc is used by Apple to build MacOS X and Apple has been improving gcc PPC code generation. Apple provides gcc to Mac developers. Apple is also IBM's partner in the development of the PPC970. Gcc is the developer optimized compiler for the chip in many ways and is more comparable to Intel's compiler in this respect.
If everyone benchmarked with open source compilers, there would be none of the shady benchmark-specific optimizations you'd expect to see in proprietary compilers. Everything would be above the table.
No. Benchmarks would become less realistic. There is nothing wrong with proprietary compilers. If they use proprietary techniques not available to gcc, so what. The only consideration is whether the compiler is available to other developers. The Intel compiler is available under Windows and Linux so it would be completely fair to try it and gcc and pick the faster of the two.
You know, I always thought that this would be a good idea for Slashdot. I mean, you guys must have some pretty interesting contacts by now, use some of them to do a "news" article or two on your own. I'd still keep the old Slashdot question/answer interview around because they are interesting and good for the people who don't have time to do a traditional interview.
Sapere aude!
The "status quo" crowd that jumped all over Apple this morning for the "fake" benchmarks and "dishonest" wording will still find lots of reasons and ways to disparage the fruit company, simply because Apple isn't doing what they want - building the best, fastest, and most cutting-edge computers for $400.00.
Forget about Serial ATA - (Apple is the first top-tier manufacturer to make this interface stardard across their high-end machines.)
Forget about the new motherboard featuring HyperTransport, PCI-X, and the IBM-fabbed 1GHz northbridge chip. Oh, and 802.11G, USB 2.0, FireWire 400 and 800, and Bluetooth, too.
Forget about the imagination and creativity that goes into making a project like this go from concept to reality in eighteen months.
Why support a company like that? Bunch of dirty liars - there's no way a 2GHz chip could be faster than my Intel/AMD/whatever86!
Maybe it's not ultimately faster (although Greg's comments seem to indicate that the playing field was pretty equal). I don't buy "fast". I buy well-integrated tools that help me get work done, and in turn, bill clients. So I (still) use a Mac.
Jeez - to hear people around here, you'd think that innovation, style, performance, and the courage to move forward agressively and definitively with new technologies doesn't come at a price.
What other comapny would develop all these technologies to hardware and software maturity as part of a new hardware platform, then bring it all to market with system software already written (by the same vendor, I might add) to take advantage of new hardware features?
Those things DO come at a price. The price begins at $1999.00 for the 1.6GHz G5, or $799.00 for an eMac.
As long as there are people who just want to get work done on their computers without hiring an IT department or worrying about who is responsible for which component of the system, Apple will still be around.
I bill around eight hours a day with my Macintosh - the $400.00 price premium over PC hardware at the time I bought my G4/800 simply isn't an issue - over the lifetime of the machine, I'll probably bill at least two hundred times that amount for work made possible by its existence.
That $400.00 up-front cost means that I don't have to spend my time - my extremely expensive and finite time - having to deal with at least two vendors just to get a system with competitive hardware, a competitive OS, and support for them both. If your time isn't valuable, by all means cheap out and build your oft-touted (and perfectly capable) PC from parts you buy at Frys. $400.00 means nothing to professionals - it's cheap support insurance.
I hope Apple sells a TON of these machines - because they're practically the only personal computer company willing to take the initiative and responsibility for supporting hardware and operating system on equal terms.
Perhaps if Apple stressed the cost of ownership point to more people, they'd have higher sales. Our small business has nearly thirty Macs. I'm the lone IT person, spending an entire hour a week on supporting a bunch of artists and their Macs. What similarly-sized Windows-based business can make that claim?
While I appreciate the opinions and arguments of the author, I am dismayed by the constant references back to the BeBox. Yes the BeBox and Amiga (similar case) were excellent platforms when they were introduced. Yes, each had quite revolutionary ideas. Unfortunately, neither caught on. Both were targeted at geeks - a niche market with questionable budgets, based on the "$1k is TOO much to pay for a box" posts.
Apple does not invent everything, although historically they tried! Apple is clever at evaluating technologies, combining them into their existing product, and making the results available to the mass audience. I agree that many of the individual accomplishments are unremarkable, packaging all these into a new release is impressive. And doing this three times (10.0, 10.1, and 10.2) in about three years is amazing for any product. Doing this with an operating system is unparalled*.
Take a step back and evaluate Apple's announcement in the broader industry context. You may not be amazed, but I think you'll be impressed.
.
* I suggested that Apple's OS release schedule was unparalled based on the number of features being introduced. This isn't an achievement driven entirely by the programmars in Cupertino. With the Mach / BSD underpinnings, open source software can be easily ported over to the new machine (see: fink, apt-get, etc.). In many cases (e.g. Safari, ProjectBuilder, etc.), Apple is applying a flashly UI on top of standard source. The result is a relatively small number of programmers producing a large number of features. Compare this to Microsoft which uses entirely custom code and where you need a large number of programmers to get a small number of features. Compare this to the Linux model where you need a large number of programmers (as most are part-time), to get a moderate number of features. Solaris and other commercial Unixes also have this advantage, but neither has been quite as driven; I don't understand why this is so.
If this analysis is correct, OS X should have an impressive feature and Microsoft will need to change their OS development model. Linux / BSD, if they can avoid fragmentation, will continue to provide much of the R&D that the other OSes rely on.
Given one hour to live, the student replied: "I'd spend it with professor FP who can make an hour seem like a lifetime."
While surprising and most certainly refreshing to see that apple is serious about their claims, serious enough to publicly rebuke the claims almost certainly first brought to the big light by /.'s earlier article, this may only be leading into a circle of prooving and disprooving.
I believe it would be best for apple to answer with a full fury of tests to truly show the full range of operating prowness of the G5's vs the P4's, etc.... at least initialy, and to from there LEAVE IT ALONE. Cause no matter how many tests they do, no matter how much proof ... there will always be people out there ready to bring flames over nothing.... For instance this guys claims that FP isn't all that important, and that int tests are basicly all that matters for the majority of users.
He and others will stick too their guns even if they have only a couple benchmarks to cite as being supirior (kinda like the G4's and their altivec/photoshop optimizations of yester-year).
Apple needs to make sure that they have a clean image of being flatly open on their claims, and then to move on without being bogged down in an obvious quagmire of platform evangalism. The truth is, their strongest advantage remains the OS and not their hardware's direct horse-power. Of course the G5 along with all the goodies they come with are incredibly great, but this isn't apple's mainstay... it's simply another selling point.
If they become entrapped in having to proove themselves through benchmarking every new release, it won't be long before their entire image would have to live up to being ahead at all costs.... and guess what... they ARE going to fall behind again.... and then they'll leap ahead again.... and then they'll fall behind... etc.... And every down cycle will be worse, since the specs will be much more associated with their image.
keep your strong point in innovation apple, and if youve got the great hardware... great.... but don't get stuck in the mind-less mhz/spec race that has stagnated computer innovation for the love of ego's.
just my 2 cents.... I develop ASP, and love win 2k adv srv, ill never use anything but unix/linux for my networking gear, and OSX keeps me damn happy when i want to do anything not mind-numbing. Cisco IOS is arcane but makes me feel good. I am biased towards all platforms.
--Idiots, Every single one of YOU, A flaming mass of conglomerated morons, hey wait a second, isnt that how RAID works?
2xCPUs would cost around $1400
Motherboard $300
'Cos everyone knows all you need is a motherboard and processors. Didn't you work in IT at a company I used to work for? You're the one who took the RAM out of my computer and said you'd be "right back", aren't you?
Excercise for you:
Add the cost of Bluetooth, PCI-X, 802.11G, Gigabit ethernet, SATA hard drives and controllers, DVD-R drive, power supply, all the other hardware stuff I've forgotten, plus iTunes, iDVD, iMovie, and the ten or so other bundled applications on the G5s, a Unix-based operating system with superior usability, and one year of free warranty and support for ALL of that stuff.
How much does your dual Xeon cost now?
So many people are slamming Apple for posting biased benchmarks. Yet, I found it very interesting that Apple posted one benchmark which showed the G5 being the slowest machine: the SPECin_base2000, single processor mark. For someone posting completely untrue and biased benchmarks, showing a last place finish shows that not everything was biased in favor of the G5.
Is the PowerMac G5 the "world's fastest personal computer"? Probably not, but it may be the first 64-bit personal computer to ship to the masses (ie. bought in a store like CompUSA). I wonder if AMD will move up the Opteron release now or if Intel will drop the price on their Itanium. If so, then people who want 64-bit x86-compat CPU's should thank Apple for bringing them their CPU's faster. =)
Anthony
I seems that some people just like to bitch.
./'s favorite vendor.
I know I do, and that's why I'm writing this.
I can't figure out why so many people post to threads like this and
bash Apple, while saying that they would never buy a machine from them
anyway. What's the point in that? Would the industry be better off if Apple
didn't exist? Would you finally be happy if everyone went out of business
except for Dell, only selling boxes pre-loaded with Linux, for $299?
If that was true, Lindows should be
And those that say that they could build a machine themselves for way
less than a Mac, if Apple had a build it yourself, parts in a bag option for
$500 less, then people would still bitch that for that price, it should come
fully assembled.
Although yes, I am a "Mac guy" (but I've got Windows, Linux, Solaris, IRIX,
NeXT and a few other boxes on my home network), regardless of my
prejudice for the platform, you have to acknowledge what a beautiful
$3000 machine the G5 is. Clean inside and out, plenty, plenty fast for
the years that you'll have it in service, arguably a better OS than any
Linux variant and absolutely better planned out and cleanly feature
rich (and economical) than any Windows release. I was doing some
admin work on Win 2000 server today, what a disorganized, steaming
plie that thing is. Some say it's superior, I think it might be the absolutely
worst collection of software ever crammed into one box. Pheeeewwww!
But I digress. I have come here to praise the Power Mac G5....
One of my favorite things about the G5 (and I know that non-Mac users
think than Apple just makes pretty boxes), is indeed, the pretty box.
J. Ive did such a restrained design. So clean and minimal.
There's a guy with rare discipline and insight.
The new design language, aluminum and circular hole accents, also
seen in the iSight and hints of it in the line of new aluminum PowerBooks,
in my opinion is the best we've seen in the 2nd Jobs era at Apple.
I liked the clean white, crystal and chrome designs of the G4 iMac and the
iPod but this new design language is going to make for some other very
exciting products. The new display line will be beautiful, wrapped in a
thin sheath of aluminum. Will a future iPod have the look of a large-ish
Zippo lighter? What would an all-aluminum G5 iMac look like?
I'm just glad that Apple's still here, still thinking different, and still making
insanely great products.
Dell? HPQ? Gateway? Lindows? Sony? (Well, Sony's trying).
The parts bin at Frys? That little shop in the strip mall that sells cases and
motherboards? For the most part, all of that is commodity crap. Even if
you throw on your free homemade Linux on it, it's half-assed at best,
even after hours of effort.
Apple is the only computer company left that's doing anything that really matters.
Like it or not.
This seems reasonable of Apple now. There are many applications compiled on Windows that don't use Intel's optimizing compiler. Indeed that's the norm, since most Windows developers use Microsoft's compilers that ship with Visual Studio and other x86 environments like Linux are dominated by gcc development. You have to buy Intel's compiler separately and add it to your development environment in most circumstances and it ain't cheap despite the obvious benefit getting better x86 optimized apps released has for Intel. The biggest difference AFAIK is Intel's good work in optimizing for their SIMD style instructions like SSE2, where their compiler does a much better job at parallelizing multiple serial operations into a single SSE op. The difference this makes to some code when comparing Intel's compiler to Microsoft's compiler on the same CPU can be dramatic, even 2X or more on specific benchmarks.
All in all I think this was a fair test of these CPUs, it was a level playing field. OTOH we know Intel can do much better with their compiler, but only some developers use their compiler. It would be interesting to see just how much of a benefit Apple could squeeze out of non gcc compilers, probably not as much as Intel, perhaps not anything, it depends on the work they or IBM et.al. have done on their compilers. You just know if it was to Apples advantage they'd have compiled with their best compiler and dont teh comparrison with those numbers vs Intel's so this situation has been contrived to an extent.
With Intel charging what they do for their compiler developers can be reluctant to pay extra for it, I expect almost everyone (on Windows) would use it if it were free. I know I would, but I can get by without it. I don't really have much sympathy for Intel here, they make billions of chipe, make significant performance claims based on their own compiler, yet charge for it to the point where many developers simply stick with Microsoft's compiler that they've already spend a fair bit on. Now Intel is upset that Apple used gcc, well more people might use Intel's compiler if it were easier to aquire, and clearly it would benefit Intel. If they want to run there business where everything is a profit center and they don't have to be smart enough to evaluate obvious but intangible benefits that's their business, but this is part of the price you pay for charging an arm and a leg for your compiler when you should be in the hardware business and giving your compiler away to help your customer gain the benefit of faster code from the applications they purchase. In the meantime specbench numbers for Intel are simply bogus for many applications.
a 64bit desktop computer for general use.
Certainly at $3000 the dual 2Ghz is pricey, but look at what you can do with it. This computer can work with video, audio and bitmaps NOW and it doesn't take Joe average weeks to figure out.
Only an idiot would use a shotgun to kill a fly, or a semi-trailer to bring home the groceries, but both have their place and purpose. I'm sick of idiots claiming you can create a Linux cluster to get the same power at the same price, but then not mentioning the applications they will run and more significantly their price.
Reality is MacOSX works, it works well on a G4 and even better on a G5. I'll bet no-one in your neighbourhood will buy a NEW Opteron workstation, but a few will buy a G5 Apple.
The less you know about computers and computing, the more appealing Apple's Products become.
There is something for everyone, BSD Unix for geeks, and a great interface for the rest. If only they cost a tad less!
Several other posters have noted that GCC/970 is really not the same compiler as GCC/Xeon. Sure there may be a bit of code in common between the versions, but the job of a compiler is to produce object code... and by definition, the object code for 970 is different from that for Xeon.
What matters to a purchaser is "How much performance can *I* get out of this machine". If I am performing CPU-intensive scientific calculation that require the fastest CPU I can find (at least for a given number of kilodollars), I'll almost certainly spring a few hundred extra for the compiler that produces the fasted object code on that platform (if needed, there's nothing ruling out GCC automatically because it's free).
It happens that for a Xeon or P4 (or Opteron, for that matter), the compiler that produces the fastest object code is ICC. Intel has done an amazingly good job with their compiler.
Now, sure, I *could* get a similarly optimized 970 compiler for comparison.... if one existed, that is. It looks like right now, GCC is the best you can get on a 970. It doesn't do a buyer any good to know that IN PRINCIPLE a more optimized compiler could be written.
All that said, the 970 looks like a very respectable chip. And Apple is selling their new machines at a very competitive price; and Macs have extremely friendly and stable OSs. All that means that it is probably well worth buying a PowerMac even if it will crunch big computations a few percent slower than a more expensive Xeon. But still... the "GCC is the common element stuff is pretty darn bogus."
Buy Text Processing in Python
Let's face it: in their own way, Apple is being quite fair. Everybody in the free software community uses gcc, and publishing SPEC scores on x86 gcc is valid and useful.
However, IBM probably has C compilers for the POWER architecture that produce far more optimized code than gcc. Why hasn't Apple licensed and ported this technology?
Apple needn't resell such a C compiler, but critical system binaries (i.e. the kernel) could be recompiled for much better performance. Granted also that IBM is unlikely to support Objective C anytime soon, so such a compiler is only marginally useful.
However, Apple positively wastes these POWER chips without a vendor-optimized C compiler.
I'm becomming really dissapointed with what seems to be the majority of posts on /. (and especially in this thread). I was anticipating a lot of /.ers going on and on about how sweet the tech specs are on the PowerMac G5 hardware. It shouldn't matter what religion you are (M$, Sun, *NIX, Mac, IdogAppleToSoundSmart...), the hardware freeking rocks! Just like my attitude towards BeOS - I don't necessarily care whether the thing will gain 82% market share, its just cool shit.
The weakest point of Mac systems for many years has been slow bus speeds. Nobody's challenging the bus speeds and they're much, much faster. If you had a bus this fast on the G4 systems, they would dramatically improve their real-world performance.
RAM capacity is also not under challenge. So, for 23999 I can get a system that would permit up to 8GB of ram on the system.
Just those two unchallenged figures make this much more than just another boring speed bump hardware upgrade.
If they're providing the actual compiler flags they used and the flags used disprove one of the doubter's claims (no SSE2 use) then maybe Apple is *not* just making stuff up?
It is quite difficult to produce better code than gcc, and my tests on powerpc (granted, those were a few years back using xlc on RS6000 with AIX 4) showed that xlc produced code of about the same quality -- sometimes worse, sometimes better.
The gcc "Haifa" scheduler was donated by IBM Haifa, by the way, so I think it's not surprising that gcc produces good code on powerpc.
On Intel it's quite the same, except that gcc does not vectorize code. From what I have seen, however, icc's vectorizer is not very useful either. I recently tested ogg-vorbis (which is a plain C floating point intensive benchmark) with icc 7 and gcc 3.3 and the gcc version was actually faster than the icc version (on my Athlon XP, target CPU pentium3) despite icc having vectorized several loops.
So all this "vendor-optimized C compiler" stuff is really besides the point. No C compiler will ever be able to match the quality of hand optimized assembler code, and the most important code (ffmpeg MPEG-2 decoder and MPEG-4 codec) has already been hand-optimized. You might be able to squeeze anoter 5 percent out of your code by using a vendor C compiler with insane optimizer settings, but what good is that if the end user is only going to use gcc anyway. I know I am, so I find the numbers for gcc actually more useful for comparison purposes than some vendor C compiler comparison.
Also, we don't want to encourage vendors to produce super vendor optimizing compilers, we want them to optimize gcc (so that everyone benefits, not just their users). So the more benchmarks are done using gcc, the better!
Forget the debate over if restricitng GCC is fair, let's think about what we are trying to measure!
What it comes down to is the speed of the system, not the chip. The 1000+hp dragster is a useless vehicle to me because I don't need to go in a straight line at over 200 mph. What I do need is to be able to accelerate in traffic, handle corners, etc.
To me, a computer is a system. So I don't really care if the G5 is cranking out power I will never need. I feel like the G4 and the P3 are plenty powerful chips if the OS is built to be efficient and the supporting components to the system are configured correctly.
These days, my biggest reason to upgrade to a new computer is desire for faster system components. I wish my P3 had firewire (might go buy a card), I wish my PowerBook G4 (400mhz mind you) had BlueTooth (not 1600 extra mhz).
Benchmark the user experience, give a review that is more like Automobile or Road and Track. Tell us the zero to sixty and then move on the how the G5 handles in the turns. Tell us if the wind noise is less than a Xeon. Tell me if it has power windows. But don't spend 90% of your marketing materials telling me about the engine, that is only 10% of my buying criteria.
You just gotta TRY OS X!
I only came here to do two things; kick some ass, and drink some beer...looks like we're almost out of beer.
I've read through most of the streams on this topic over at Ars, here and Geek â"most of it consisting of shrill whining over Intelâ(TM)s compiler not being used. Iâ(TM)ve looked at all the views, read the study -and from what Iâ(TM)ve CAREFULLY read, the testing methodology was fair.
3 1.html
c s/power_ps3q02- khalid.htm
...declarative boobs.
... also noted that Intel's chips perform disproportionately well on SPEC's tests because Intel has optimized its compiler for such tests.â
I feel those shouting the loudest -scanned, rather than read the report.
Typical of the 'outrage':
http://news.com.com/2100-1042_3-10206
"It wasn't really a fair test," said Gartner analyst Martin Reynolds, who said that the Dell machines are capable of producing scores 30 percent to 40 percent higher than those produced under Apple's methodology [using the Intel compiler]...â
-Well DUH! I guess we then should have expected Intel to trot out a hand-coded/Spec optimized version of their compiler for the G5 too! Idiocy.
âoe...In response, an Apple representative said it wanted to compare hardware performance, so it made sense to use the same compiler on the Mac and the Dell. The SPEC benchmark tests measure the performance of the hardware and the compiler. âoe
-Lets get real here! NOT normalizing the compilers on each systems is nutso â"even to me (which isnâ(TM)t saying much.)
âoe...Joswiak said that the Power Mac settings were representative of how the final machines will ship, even though a few settings did differ from the way current prototypes are configured. As for the Intel-based PCs, he noted that some of the settings that have been criticized were chosen because they actually improved the performance.â
DUH again! Even I saw that! Hey people, the methodology rationale was even explained in the study! Read it!
The bottom line of most of the PC-lumpenproletariat out there seem to be: âoehey dude, ya needed to use the Intel compiler because the Specmarks are, like, soo-ooo much higher on the Dell. As for the G5, use whatever â"it sucks anyway.â
Another big moan out there was that the Dellâ(TM)s Hyperthreading was turned off. They donâ(TM)t seem to realize is (according to DELL) is that, re SpecMark, this was to Dellâ(TM)s advantage!!!!
http://www.dell.com/us/en/esg/topi
(...As was the fact that the Dellâ(TM)s were packing 512 MB more RAM than the G5! (Which NOONE seemes to have noticed. btw.)
âoe...Peter Glaskowsky, editor-in-chief of Microprocessor Report,
Damn right they do! To put it mildly.
Iâ(TM)ve read scads of articles on the various, ahem, Spec-specific âoptimizationsâ(TM) theyâ(TM)ve built into their compiler. Great too if youâ(TM)re comparing one Intel product against another â"but other than that, itâ(TM)s just marketing fluff, IHMO.
"...Jobs on Monday also showed demonstrations in which the new Power Macs outperformed the Dell by greater than 2-to-1 ratios on several programs...Reynolds says he has no reason to contest those claims. âoe...the application benchmarks look quite credible," Reynolds said.
Those usage tests may also be more important than synthetic benchmarks, he said. "The SPEC benchmarks aren't that relevant anymore. People now are looking for things like multimedia (performance) and content management."
Agreed. I also think there is just too much marketing driven Spec-chicanery going on out there for them to be considered meaningful benchmarks -if they ever were.
Anyway, the telling of the tale will be on actual boxed applications.
And although I may be surprised, I would place big bets (right now) that the G5 system -especially running Altivec-aware, 64bit recompiled applications, -will run (multiple) circles around the best MP PC versions that are out there (right now.)
But Intel and AMDâ(TM)s caldrons are busy bubbling â"and the landscape may change radically by October (doesnâ(TM)t Intel typically intro their new stuff in September and March?).
Even so, it will be interesting to see the price point any new uber-systems come in at. Right now, (unless you, brrrr...., 'roll your own'), theyâ(TM)re priced in the workstation stratosphere.