Scene: Fallujah, dusk. A squad of marines is on patrol for insurgents. A bombed out office building stands before them, barely standing and pock marked. The squad officer stops.
Officer: All right men, we're gonna check out that building. I want Johnson and Tex to take the point, Smith, Freddie and Joe provide cover, the rest of you follow me. He starts offs, but Johnson says "\/\/+Ph!?!? i d0N'+ \/\/@N+ +0 g0 d0\/\/N +h@+ $+r33+. \/\/hy d0 i h@v3 +0 g0 d0\/\/N +h@+ $+r33+?"
Before he can reply, Joe pipes in "y3@h, i d0N'+ \/\/@Nn@ c@MP, i n3v3r g3+ @Ny/The squad leader is dumbfounded by this, beginning to formulate a thourough bitching out when the Tking starts.
And that is why the US never again tried to use videogames for recruiting.
For the third time, I'm still waiting for evidence of "SPEC is subject to all kinds of problems.", why Apple doesn't use your "benchmark" to advertise their products, these supposed Intel optimizations, evidence of why spec isn't accountable, and why I am "blinded by marketing".
Uh, how exactly do you get "GCC, is said to generate code that less well optimised than Intel's" from "GCC, is said to generate code that less well optimised for x86"?
I repeat, since you again seem to have poor reading comprehension skills, how is spec testing the compiler? There's no specmark for the efficiency of the compiler, nor does it matter, because the goal of spec is to provide a benchmark for the CPU under the best conditions. You arguement about standardizing compilers is equivalent to having a horse (G5) race a F1 (Opteron), under the "standard condition" of a dirt track. It's ludicrous and retarded.
Hmm, does this appear to be vector processing done by a compiler? If I "have no idea how compilers, CPU architectures, system architectures are designed", then how come I'm the one saying that hand coding is a bad idea? Even a child of 5 could understand that "out of order instruction" means that it's unlikely your "tightly written asm" is going to be executed the way you want it, and add "superscalar", and it becomes completely out of reach for humans to keep track of all the depenencies.
I also need to add two new "claims" of yours to my list: Why I "have no idea how compilers, CPU architectures, system architectures are designed", and "people are still smarter than compilers".
And if "its not worth my time to argue with you.", then quit. You're only providing more and more evidence that the G5 is crap, that stupid fanatics will use any excuse or attack to desperately try to show it isn't, and that I can refute any delusional idea that you can come up with.
It's not a unrealistic as you make it seem. If the locations were chosen randomly, then sure, you'd be right, but I would think that they had the GPS in the cabs for some time, collecting data as to the most popular areas, and then choosing 40 of them.
Because I don't have benchmarks about what an Intel Apple would use, and I guessed that it would be comparable in speed to an Opteron, given the year or so it'll be till we see an x86 powerMac.
I'm still waiting for evidence of "SPEC is subject to all kinds of problems.", why Apple doesn't use your "benchmark" to advertise their products, these supposed Intel optimizations, evidence of why spec isn't accountable, and why I am "blinded by marketing".
How is spec testing the compiler? I don't see any spec compiler benchmarks. The only relevance of the compiler to spec is the code it produces to test the processor, analogous to testing a car (CPU) using a certain octane gas (compiler). As for me knowing how compilers work, I certainly know that they're better equipped to produce optimal code for a modern CPU than you.
Throughput is how fast a CPU can commit instructions, though you wouldn't know about modern CPUs with superscalar processing and out of order execution by your claim that hand coding asm is faster. "Goodness"? I didn't realize we were discussing CPU aesthetics, not performance. BTW, bus speed and cache are dependent on the CPU.
Go read the register link, "GCC, is said to generate code that less well optimised for x86". You're an idiot. The controversary was about Veritest using GCC instead of Intel's compiler to produce favorable results to Apple, but again, you wouldn't know that because you say that Apple did the tests, and don't tell me to "google", it's your job to provide proof for your own claims.
What's a better CPU architecture? You certainly wouldn't be a valid judge, not knowing about modern CPU design. Instead of some conspiricy that x86 owns the market by simple being entrenched, isn't it simpler that x86 has had enough development pumped into it to make it fastest?
I also remind you, since the RDF appears to have made you miss them, the 5 claims you made that neglected to provie proof for.
My archeology teacher always said "Archaeology is the search for fact. Not truth.
If it's truth you're interested
in, Doctor Tyree's Philosophy
class is right down the hall."
Also, Linpack is used instead of Lapack presumably because Lapack is geared towards true supercomputers with massive bandwith like Blue Gene and Crays, not clusters.
My stats are not misleading, the 2 way spec measures throughput, while the uniprocessor measures raw speed. System architecture doesn't account for the vast majority of the G5's crappy throughput. If you just want raw speed, the G5 is 71% as fast as the Opteron, or the Opteron is 41% faster - a "29% difference" is a worthless number without direction.
Spec tests the processor, not the compiler. The point of the benchmark is to use test the system under optimal conditions, and Apple used GCC, which is bad at optimizing for x86. Re: hand coding - go learn something about the complexity of modern processors and you'll find that hand coding has gone the way of lookup tables.
How is spec not accountable? People bitched and moaned about Apple's spec fraud, why wouldn't the same happen if it was Dell? You want spec test conditions? Top500 seems to be missing that. You want verification? I refer you to the Apple fiasco. Top500 seems to be missing that as well.
You points about the G5's supposed effeciency and use in consoles have no bearing on it's performance in benchmarks.
I'm still waiting for evidence of "SPEC is subject to all kinds of problems.", why Apple doesn't use your "benchmark" to advertise their products, these supposed Intel optimizations, evidence of why spec isn't accountable, and why I am "blinded by marketing".
Show some evidence for "SPEC is subject to all kinds of problems." The only recent spec controversay is Apple's dishonest G5 benchmarks, your paranoid anti-Intel conspiricy theory aside, and the fact that you bring up GCC shows you know nothing, since only 4/14 specfp programs are C, the rest are fortran. Also, have a look at comparison of Intel GCC and PGI compilers. Hand coding?!?! Are you insane? Do you have any idea of the complexity of a modern processor's scheduling? Go talk to a comp arch professor, and get bitchslapped for retardedness. If GCC sucks for the G5, then use something better, nobody's forcing you to use it.
Ah, I see what the problem is. Steve Jobs Reality Distortion Field has given you selective reading comprehension. "LINPACK has been largely superceded by LAPACK
which has been designed to run efficiently on shared-memory, vector
supercomputers." Note that it was designed to run efficiently on shared-memory, vector supercomputers, therefore providing a better benchmark of said supercomputers superness. The reason Linpack is used is because clusters would completely choke on this test. All of this however, completely ignore the fact that you're trying to use a test of a supercomputer to decide what a processors worth.
Linpack is a one class of algorithms, SPEC is 14 fp and 12 int classes, selected to provide a *wide* variety in algorithms.
Find a CPU, not supercomputer, benchmark that shows the G5 is faster, otherwise it's just your word that your Andersonesque calculations are worth more than the accountable SPEC.
But all this really doesn't matter because of one question, if the top500 results prove the G5 is so much better, why didn't Apple use them?
If you'd read the link I provided, you'd see that Lapack was created because of changes in supercomputer architecture, providing a more accurate benchmark.
Linpack is type of test: linear algebra, and again, if you'd bothered to read the links, or even my post, you'd notice that 26 different types of test > 1 type.
Finally, you're using a very muddled secondhand source based on 1 test to desperately hang onto your delusion that the G5 is anywhere near the Opteron.
Continue reading the following until you realize that the G5 is crap. crap crap crap. And keep your pie hole shut before you remove all doubt about being a fool as well.
SPECint_rate2000
2200 Opteron 68.1 64.2
2200 PowerPC 970 21.5 20.2
SPECfp_rate2000
2200 Opteron 69.1 63.9
2200 PowerPC 970 20 19.2
If you don't like the facts I use, instead of absuing your mod points (ie modding something overrated that hasn't been modded and is at 1), go refute the facts.
It's not a comparison of speed, it supposed to generally indicate how well designed the processor is. It's like comparing a 911 with a 300hp engine to a Mustang with a 600hp engine, both may get sub 4 0-60 times, but clearly those German engineers made a better designed automobile, and could make it even faster easier than the Mustang hackers.
Rpeak is hardly a valid benchmark, since it can only be achieved with instructions doing basically nothing, and the erratic differences between Rpeak and Rmax on the top500 make it even more useless.
The G5 has similar architecture to the POWER4+, so I guess its SPECfp2000 would be around 1400 +/-100.
And finally, these benchmarks don't matter if you have the program that you intend the machine to run, say Photoshop, and it's faster on the Mac than x86.
This still leaves the results I gave that the G5 is as third as slow as an Opteron in both int and fp rates valid.
Slashdot Mirror
Scene: Fallujah, dusk. A squad of marines is on patrol for insurgents. A bombed out office building stands before them, barely standing and pock marked. The squad officer stops. /The squad leader is dumbfounded by this, beginning to formulate a thourough bitching out when the Tking starts.
Officer: All right men, we're gonna check out that building. I want Johnson and Tex to take the point, Smith, Freddie and Joe provide cover, the rest of you follow me.
He starts offs, but Johnson says "\/\/+Ph!?!? i d0N'+ \/\/@N+ +0 g0 d0\/\/N +h@+ $+r33+. \/\/hy d0 i h@v3 +0 g0 d0\/\/N +h@+ $+r33+?"
Before he can reply, Joe pipes in "y3@h, i d0N'+ \/\/@Nn@ c@MP, i n3v3r g3+ @Ny
And that is why the US never again tried to use videogames for recruiting.
Hehe, how ironic that a Locke is espousing a Hobbesian view.
Yes, but is the Royal Navy full of cannibals?
For the third time, I'm still waiting for evidence of "SPEC is subject to all kinds of problems.", why Apple doesn't use your "benchmark" to advertise their products, these supposed Intel optimizations, evidence of why spec isn't accountable, and why I am "blinded by marketing".
Uh, how exactly do you get "GCC, is said to generate code that less well optimised than Intel's" from "GCC, is said to generate code that less well optimised for x86"?
I repeat, since you again seem to have poor reading comprehension skills, how is spec testing the compiler? There's no specmark for the efficiency of the compiler, nor does it matter, because the goal of spec is to provide a benchmark for the CPU under the best conditions. You arguement about standardizing compilers is equivalent to having a horse (G5) race a F1 (Opteron), under the "standard condition" of a dirt track. It's ludicrous and retarded.
Hmm, does this appear to be vector processing done by a compiler? If I "have no idea how compilers, CPU architectures, system architectures are designed", then how come I'm the one saying that hand coding is a bad idea? Even a child of 5 could understand that "out of order instruction" means that it's unlikely your "tightly written asm" is going to be executed the way you want it, and add "superscalar", and it becomes completely out of reach for humans to keep track of all the depenencies.
I also need to add two new "claims" of yours to my list: Why I "have no idea how compilers, CPU architectures, system architectures are designed", and "people are still smarter than compilers".
And if "its not worth my time to argue with you.", then quit. You're only providing more and more evidence that the G5 is crap, that stupid fanatics will use any excuse or attack to desperately try to show it isn't, and that I can refute any delusional idea that you can come up with.
Tried that already. I lost and got 5 years for burglary.
There's always real-life Risk, though it seems that after Germany stopped playing it kinda died.
I'll bet he'll be so shocked he'll have a heart attack!
Almost made it?!? What, you mean you didn't keep trying until you succeeded. Quitter.
It's not a unrealistic as you make it seem. If the locations were chosen randomly, then sure, you'd be right, but I would think that they had the GPS in the cabs for some time, collecting data as to the most popular areas, and then choosing 40 of them.
Because I don't have benchmarks about what an Intel Apple would use, and I guessed that it would be comparable in speed to an Opteron, given the year or so it'll be till we see an x86 powerMac.
I'm still waiting for evidence of "SPEC is subject to all kinds of problems.", why Apple doesn't use your "benchmark" to advertise their products, these supposed Intel optimizations, evidence of why spec isn't accountable, and why I am "blinded by marketing".
How is spec testing the compiler? I don't see any spec compiler benchmarks. The only relevance of the compiler to spec is the code it produces to test the processor, analogous to testing a car (CPU) using a certain octane gas (compiler). As for me knowing how compilers work, I certainly know that they're better equipped to produce optimal code for a modern CPU than you.
Throughput is how fast a CPU can commit instructions, though you wouldn't know about modern CPUs with superscalar processing and out of order execution by your claim that hand coding asm is faster. "Goodness"? I didn't realize we were discussing CPU aesthetics, not performance. BTW, bus speed and cache are dependent on the CPU.
Go read the register link, "GCC, is said to generate code that less well optimised for x86". You're an idiot. The controversary was about Veritest using GCC instead of Intel's compiler to produce favorable results to Apple, but again, you wouldn't know that because you say that Apple did the tests, and don't tell me to "google", it's your job to provide proof for your own claims.
What's a better CPU architecture? You certainly wouldn't be a valid judge, not knowing about modern CPU design. Instead of some conspiricy that x86 owns the market by simple being entrenched, isn't it simpler that x86 has had enough development pumped into it to make it fastest?
I also remind you, since the RDF appears to have made you miss them, the 5 claims you made that neglected to provie proof for.
>Printing a book of this length one off will cost you more that the $19 street price.
I seem recall paying $0 to print 161 pages of Dune double sided, and the University was none the wiser about picking up the tab.
My archeology teacher always said "Archaeology is the search for fact. Not truth. If it's truth you're interested in, Doctor Tyree's Philosophy class is right down the hall."
Bart: Thank you, thank you. [walks up to Russian man] So, you're from Russia huh?
Russian: Da.
Bart: Ya drunk yet?
Russian: [depressed] Da.
Well put. It's good to see someone that understands the complexity of high performance computing commucation.
Also, Linpack is used instead of Lapack presumably because Lapack is geared towards true supercomputers with massive bandwith like Blue Gene and Crays, not clusters.
They let you make pretty tables
To answer your question and clear up the incorrect info posted:
#systems, %, sum Rmax, sum Rpeak, #processors
Intel+AMD = 358 71.6 895150 1480709 266242
Power+PC = 77 15.4 616413 897521 251502
All = 500 100 % 1686933 2632133 580336
%Rmax, %Rpeak, %processors
Intel+AMD = 53.1% 56.3% 45.9%
Power+PC = 36.5% 34.1% 43.3%
On a side note, how can I format this better, I've tried some html tags but slashdot ignored them.
My stats are not misleading, the 2 way spec measures throughput, while the uniprocessor measures raw speed. System architecture doesn't account for the vast majority of the G5's crappy throughput. If you just want raw speed, the G5 is 71% as fast as the Opteron, or the Opteron is 41% faster - a "29% difference" is a worthless number without direction.
Spec tests the processor, not the compiler. The point of the benchmark is to use test the system under optimal conditions, and Apple used GCC, which is bad at optimizing for x86. Re: hand coding - go learn something about the complexity of modern processors and you'll find that hand coding has gone the way of lookup tables.
How is spec not accountable? People bitched and moaned about Apple's spec fraud, why wouldn't the same happen if it was Dell? You want spec test conditions? Top500 seems to be missing that. You want verification? I refer you to the Apple fiasco. Top500 seems to be missing that as well.
You points about the G5's supposed effeciency and use in consoles have no bearing on it's performance in benchmarks.
I'm still waiting for evidence of "SPEC is subject to all kinds of problems.", why Apple doesn't use your "benchmark" to advertise their products, these supposed Intel optimizations, evidence of why spec isn't accountable, and why I am "blinded by marketing".
Show some evidence for "SPEC is subject to all kinds of problems." The only recent spec controversay is Apple's dishonest G5 benchmarks, your paranoid anti-Intel conspiricy theory aside, and the fact that you bring up GCC shows you know nothing, since only 4/14 specfp programs are C, the rest are fortran. Also, have a look at comparison of Intel GCC and PGI compilers. Hand coding?!?! Are you insane? Do you have any idea of the complexity of a modern processor's scheduling? Go talk to a comp arch professor, and get bitchslapped for retardedness. If GCC sucks for the G5, then use something better, nobody's forcing you to use it.
Ah, I see what the problem is. Steve Jobs Reality Distortion Field has given you selective reading comprehension. "LINPACK has been largely superceded by LAPACK which has been designed to run efficiently on shared-memory, vector supercomputers." Note that it was designed to run efficiently on shared-memory, vector supercomputers, therefore providing a better benchmark of said supercomputers superness. The reason Linpack is used is because clusters would completely choke on this test. All of this however, completely ignore the fact that you're trying to use a test of a supercomputer to decide what a processors worth.
Linpack is a one class of algorithms, SPEC is 14 fp and 12 int classes, selected to provide a *wide* variety in algorithms.
Find a CPU, not supercomputer, benchmark that shows the G5 is faster, otherwise it's just your word that your Andersonesque calculations are worth more than the accountable SPEC.
But all this really doesn't matter because of one question, if the top500 results prove the G5 is so much better, why didn't Apple use them?
The basic SPEC methodology is to provide the benchmarker with a standardized suite of source code based upon existing applications that has already been ported to a wide variety of platforms by its membership. The benchmarker then takes this source code, compiles it for the system in question and then can tune the system for the best results. The use of already accepted and ported source code greatly reduces the problem of making apples-to-oranges comparisons.
If you'd read the link I provided, you'd see that Lapack was created because of changes in supercomputer architecture, providing a more accurate benchmark.
Linpack is type of test: linear algebra, and again, if you'd bothered to read the links, or even my post, you'd notice that 26 different types of test > 1 type.
Finally, you're using a very muddled secondhand source based on 1 test to desperately hang onto your delusion that the G5 is anywhere near the Opteron.
Continue reading the following until you realize that the G5 is crap. crap crap crap. And keep your pie hole shut before you remove all doubt about being a fool as well.
SPECint_rate2000
2200 Opteron 68.1 64.2
2200 PowerPC 970 21.5 20.2
SPECfp_rate2000
2200 Opteron 69.1 63.9
2200 PowerPC 970 20 19.2
If you don't like the facts I use, instead of absuing your mod points (ie modding something overrated that hasn't been modded and is at 1), go refute the facts.
Coward.
Re GTA, I meant if the car was $5k. I should've used grand larceny as it covers both.
Hehe, I wish I had enough time and an xbox for San Andreas to be affecting my brain.
It's not a comparison of speed, it supposed to generally indicate how well designed the processor is. It's like comparing a 911 with a 300hp engine to a Mustang with a 600hp engine, both may get sub 4 0-60 times, but clearly those German engineers made a better designed automobile, and could make it even faster easier than the Mustang hackers.
Five things: Linpack was obsolete in the 80s, so Lapack was created.
A one program benchmark is essentially worthless, so spec tries to remedy this with 12 integer programs and 14 fp programs.
Rpeak is hardly a valid benchmark, since it can only be achieved with instructions doing basically nothing, and the erratic differences between Rpeak and Rmax on the top500 make it even more useless.
The G5 has similar architecture to the POWER4+, so I guess its SPECfp2000 would be around 1400 +/-100.
And finally, these benchmarks don't matter if you have the program that you intend the machine to run, say Photoshop, and it's faster on the Mac than x86.
This still leaves the results I gave that the G5 is as third as slow as an Opteron in both int and fp rates valid.
Yeah, sorry about that. I've been drinking and sorta forgot the date, so if you could forget about it that'd make my life much easier. -John Titor