Sign in with
Sign up | Sign in
Your question
Closed

AMD Won't Endorse SYSmark 2012 Benchmark

Last response: in News comments
Share
June 22, 2011 7:30:11 AM

AMD, nVidia, and VIA already left BAPCo. What does that say?
Score
21
June 22, 2011 7:40:08 AM

fleebAMD, nVidia, and VIA already left BAPCo. What does that say?

sysmark is still based on dual core testing, ie ... crap
Score
16
June 22, 2011 7:50:14 AM

fleebAMD, nVidia, and VIA already left BAPCo. What does that say?

Intel CPUs will be the world's champions from now on:) 
Score
-12
June 22, 2011 7:56:09 AM

Still, where is Bulldozer? i have expected it on 20.06 ^^. I went i7 2600k on month ago, could not wait anymore :( 
Score
-18
June 22, 2011 8:09:51 AM

We are committed to working with like minded companies that want to give consumers and business users an accurate, honest measure of what they can expect from their PCs and mobile devices,"

Yes thats all very noble but the point is if AMD are involved then Nvidia people will start shouting biased.
We already have the info needed to make an informed decision its out there on this and various other sites. Common sense says you look across a selection of sites to get a clearer more accurate picture. Some sites are after all leaning one way or the other.
Mactronix
Score
4
June 22, 2011 8:24:56 AM

I found some interesting read over here. Ok, It's old but I think it does cover the problems with SYSmark...

Quote:
In fact, Fog points out that even benchmarking programs are affected by this, up to a point where benchmark results can differ greatly depending on how a processor identifies itself. Ars found out that by changing the CPUID of a VIA Nano processor to AuthenticAMD you could increase performance in PCMark 2005's memory subsystem test by 10% - changing it to GenuineIntel yields a 47.4% performance improvement!
Score
18
June 22, 2011 8:50:01 AM

mkrijtI found some interesting read over here. Ok, It's old but I think it does cover the problems with SYSmark...

ermm...i believe that ^^
Score
1
June 22, 2011 9:41:44 AM

SYSmark should have been abandoned years ago by the companies that are leaving it now (its basically an Intel marketing tool anyway). What I hate is that tech/review sites still use it, even though they know how skewed it is towards Intel.

SYSmark = synthetic benchmark that falsely gives Intel the edge every time. Intel controls it and no input/suggestions are ever implemented from the other companies that finally said "enough".
Why no mention of Nvidia and VIA leaving at the same time as AMD? It was a mass exodus leaving only Intel. The deal is, these companies are finally fed up and are dropping this sham of a benchmark.

Score
15
June 22, 2011 9:43:38 AM

Yep, the big benchmarking companies have been in bed with the highest payer (usually Intel) for years now. Usually it takes the form of checking CPUID and using that to determine code path and optimizations. You then make the code for your preferred CPU more "efficient" and leave everyone else on a non-optimized path. This isn't only in benchmarking but also in software development. Lets say publisher A is making a game, and decides to "work with" Intel to "get the most" out of their code. Intel provides a SDK with a compiler they say will help optimize their code. The compiler does indeed optimize the code, but only for Intel branded systems, and further the compiler may intentionally compile the program so that it runs a very inefficient non-optimized path if it detects its running on a "non-Intel" system. After all Intel can claim they only support their own products and have no requirement to support competitors products.

This is why I take ~all~ "benchmarks" with a grain of salt. Different code paths can swing performance 10~40% depending.
Score
9
June 22, 2011 9:52:32 AM

crisan_tiberiuStill, where is Bulldozer? i have expected it on 20.06 ^^. I went i7 2600k on month ago, could not wait anymore


sometime between next month to late September area is all the info i got. Although this info been out for a little while.
Score
0
June 22, 2011 11:08:00 AM

fleebAMD, nVidia, and VIA already left BAPCo. What does that say?

Intel own Sysmark.
Score
6
June 22, 2011 11:19:35 AM

I just want competition and the best performing products that can be pushed out.
just for the record, I'm not a SYSmark fan as well..
Score
4
June 22, 2011 11:24:23 AM

yea this is pretty interesting although its common knoledge that those bechmarks are rigged .. it would be interesting to see how well some of the current gen amd processors really fair with un-biased benchmarks. not that i think they would surprise a lot but it would still be interesting .. i mean in the end i guess if you can play pretty much any game at 60fps or shrink and convert a dvd in 5-6 mins then you have a pretty fast system but both of those are either pretty gpu intensive or hard drive intensive better off with a raid or dual gpu set up. lol
Score
5
June 22, 2011 11:51:12 AM

In short, Bulldozer isn't competitive with Sandy Bridge on this benchmark (of course not, since it's design favors a different type of workload), so let's cry about it because they didn't change it for the Bulldozer type of workload where those two integers "cores" can show their advantage.
Score
-12
a b À AMD
June 22, 2011 11:56:57 AM

mactronixWe already have the info needed to make an informed decision its out there on this and various other sites. Common sense says you look across a selection of sites to get a clearer more accurate picture.
Correct.
Thankfully we're not limited to just the AMD or Intel marketing hype (or any other MFGR for that matter). We have many good independent review sites that test and report what they find.
Score
5
June 22, 2011 12:23:46 PM

this tells you how under-powered AMD cpus are, lol...intel for life
Score
-19
June 22, 2011 12:25:52 PM

This definitely indicates that SYSmark benchmark is a biased one, and I believe it overstates Intel’s CPUs performance.
Score
8
June 22, 2011 1:06:12 PM

To all the AMD haters, if AMD fails (folds) then expect chip prices to double.
Score
18
June 22, 2011 2:31:37 PM

Maybe it's jsut me but i think sysmark arnt teh only "intel-optimized" bench marking tools , i see this same pattern with futuremark and 3d mark for graphics , it always reads intel cpu as better regradless of the model (unless it is jsut a model taht is that much oplder than the amd model you test with. also whehn testing itnergarted graphic chips it will quite oddly favor , intel over invida and nivdia over ati NOW HOW CRAZY THAT !.

for this reason i gave up longa gout even bothering to read synthetic bench marks and i only look at the real world test , and in all hoensty i think tom's should DROP synthetic test
Score
5
June 22, 2011 2:41:39 PM

erhardmIntel CPUs will be the world's champions from now on

And all my friends said i was the best too!!! :D 
Score
-2
June 22, 2011 3:39:13 PM

Intel will compete with itself from now on and the irony will be if will lose against itself...
Score
3
June 22, 2011 3:41:38 PM

mkrijtI found some interesting read over here. Ok, It's old but I think it does cover the problems with SYSmark...

Quote:
Quote :
In fact, Fog points out that even benchmarking programs are affected by this, up to a point where benchmark results can differ greatly depending on how a processor identifies itself. Ars found out that by changing the CPUID of a VIA Nano processor to AuthenticAMD you could increase performance in PCMark 2005's memory subsystem test by 10% - changing it to GenuineIntel yields a 47.4% performance improvement!


where was this found and link please.
from the sounds of it the results are pre-programmed based on CPUid or would this be pre-programmed overclocking results?
point is the same result should be nearly identical every time (temp variance permitted) no matter what CPUid is used for the same processor.
this needs more investigating and i would almost like to see tom's do a benchmark with a single via, amd, intel cpu and change the CPUid's on each one of them for the results.
any chance of this Chris?
Score
4
June 22, 2011 3:42:04 PM

Shouldn't test performance with synthetic benchmarks anyways. Each cpu or Gpu has there own advantages, so if you are gonna buy 1000 gpu's for a super computer then they should test it on a smaller scale with similar programs. Even if you are running an intel which is superior in most aspects, there are still applications which Amd will smoke them in, same for Amd and Nvidia. Most people who buy these processors or gpu's in bulk "should" know that.
Score
0
June 22, 2011 3:58:57 PM

Quote:
Quote :
In fact, Fog points out that even benchmarking programs are affected by this, up to a point where benchmark results can differ greatly depending on how a processor identifies itself. Ars found out that by changing the CPUID of a VIA Nano processor to AuthenticAMD you could increase performance in PCMark 2005's memory subsystem test by 10% - changing it to GenuineIntel yields a 47.4% performance improvement!


Am I the only one who thinks that Intel should be fined in the tens of billions for this?

If I was king of the world, this kind of crap would not go unpunished. I would make Intel pay a ridiculous sum and split it in 3. State, GCC and affected competitors(divided in volume).

I bet you Intel can't wait to sell you celeron grade chips for the price of a SB i5.
Score
7
June 22, 2011 4:29:15 PM

@F-14
You can find an article here that uses Agner Fog's paper as a source:

http://www.osnews.com/story/22683/Intel_Forced_to_Remov...

You can find his blog detailing some of his findings with the Intel compiler:

http://www.agner.org/optimize/blog/read.php?i=49#49

The question is has Intel followed through with the requirement to remove the block to optimized code paths on non Intel processors. This is something I would love to see from this website too.
Score
2
June 22, 2011 4:29:30 PM

@F-14
You can find an article here that uses Agner Fog's paper as a source:

http://www.osnews.com/story/22683/Intel_Forced_to_Remov...

You can find his blog detailing some of his findings with the Intel compiler:

http://www.agner.org/optimize/blog/read.php?i=49#49

The question is has Intel followed through with the requirement to remove the block to optimized code paths on non Intel processors. This is something I would love to see from this website too.
Score
-3
June 22, 2011 6:08:43 PM

i guess AMD is not supporting it because they know Intel dominates most of there CPU's. :) 
Score
-10
June 22, 2011 6:40:41 PM

Its actually Tomshardware , anadtech ,firingsquads and so on fault for keeping on putting sysmark benchmarks up. If all these famouse sites stoped doing it since they know those benchmarks dont mean anything then it wouldnt be an issue.
Score
8
June 22, 2011 6:54:22 PM

No need for benchmarks.. real world performance is what counts, and for that you can use real applications and games.. Stuff that actual people use their computer for in the real world.
Score
5
June 22, 2011 7:02:11 PM

AMD was smart to do this. Why would you shoot yourself in the foot just to look good? AMD's quality over Intel speaks for itself. I've liked the way AMD's done business for a long time now, and Intel just seems sleazy, especially with benchmarks like SYSmark
Score
7
June 22, 2011 8:42:36 PM

master9716Its actually Tomshardware , anadtech ,firingsquads and so on fault for keeping on putting sysmark benchmarks up. If all these famouse sites stoped doing it since they know those benchmarks dont mean anything then it wouldnt be an issue.


Exactly. I always ignore the Sysmark portion of the benchmarks and look at the benchmarks for the applications types I plan to use with the hardware. I couldn't even tell you what scale Sysmark is using these days.
Score
0
June 23, 2011 12:37:09 AM

I'm more than ready to accept AMD's position. Anyone in the software engineering department knows that the majority of software out there work best with Intel processors, with a lot having to do with Intel-made compilers that these softwares use to compile their code.

Plus regular testing of hardware here at Tom's proves that too. Software like iTunes are obviously skewed towards Intel, it makes me wonder why it's still in the benchmark suite.
Score
-1
June 23, 2011 5:22:15 AM

I fully support AMD and will refuse to purchase any Intel product as the performance ratings using BAPCo are inaccurate and unreliable. I vote with my wallet.
Score
-1
!