• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

The Effects of Benchmarking Bias

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.
So why are we rehashing this which was already proven in 09?

Ahhh supposed to be fixed. Mihtly noble of you to test this for the community. Why isn't the guy who busted them imitially doing this?

Edit. That blogger already said its not better.

Yay! Intel smear campaign! :p
 
Last edited:
I hate to sound this way but the compiler in and of itself is one issue. A compiler that is generic is fine and a compiler that takes advantage of AMD is also fine. The entire problem is this:

The FTC’s administrative complaint charges that Intel carried out its anticompetitive campaign using threats and rewards aimed at the world’s largest computer manufacturers, including Dell, Hewlett-Packard, and IBM, to coerce them not to buy rival computer CPU chips. Intel also used this practice, known as exclusive or restrictive dealing, to prevent computer makers from marketing any machines with non-Intel computer chips.

In addition, allegedly, Intel secretly redesigned key software, known as a compiler, in a way that deliberately stunted the performance of competitors’ CPU chips. Intel told its customers and the public that software performed better on Intel CPUs than on competitors’ CPUs, but the company deceived them by failing to disclose that these differences were due largely or entirely to Intel’s compiler design.

Having succeeded in slowing adoption of competing CPU chips over the past decade until it could catch up to competitors like Advanced Micro Devices, Intel allegedly once again finds itself falling behind the competition – this time in the critical market for graphics processing units, commonly known as GPUs, as well as some other related markets. These products have lessened the need for CPUs, and therefore pose a threat to Intel’s monopoly power.

Intel has responded to this competitive challenge by embarking on a similar anticompetitive strategy, which aims to preserve its CPU monopoly by smothering potential competition from GPU chips such as those made by Nvidia, the FTC complaint charges. As part of this latest campaign, Intel misled and deceived potential competitors in order to protect its monopoly. The complaint alleges that there also is a dangerous probability that Intel’s unfair methods of competition could allow it to extend its monopoly into the GPU chip markets.

According to the FTC’s complaint, Intel’s anticompetitive tactics violate Section 5 of the FTC Act, which is broader than the antitrust laws and prohibits unfair methods of competition, and deceptive acts and practices in commerce. Critically, unlike an antitrust violation, a violation of Section 5 cannot be used to establish liability for plaintiffs to seek triple damages in private litigation against the same defendant. The complaint also alleges that Intel engaged in illegal monopolization, attempted monopolization and monopoly maintenance, also in violation of Section 5 of the FTC Act.

http://www.ftc.gov/opa/2009/12/intel.shtm

Now if the entire article is read then the entire crux of the issue becomes understandable. As to the compiler? It is advertised as:

Drive rapid development and winning performance with these suites of compilers and performance libraries

Intel® Professional Edition Compilers include advanced optimization features, multithreading capabilities, and support for Intel® processors and compatible processors. They also provide highly optimized performance libraries for creating multithreaded applications.

Now a good lawyer and a jury who actually has full comprehension may find no fault with this as it is good business. ie. why would Ford make a powerstroke disel turbo that would fit Chevrolet? Same thing there is nothing wrong with the compiler and I have no love for Intel either.

Is AMD 100% Intel compatible? No, absolutely not, they have not been since the 486! All Intel has to do is prove this and it should take care of this part of the case.

I think this is it.

EDIT:
Drive rapid development and winning performance with these suites of compilers and performance libraries

Intel® Professional Edition Compilers include advanced optimization features, multithreading capabilities, and support for Intel® processors and compatible processors. They also provide highly optimized performance libraries for creating multithreaded applications.


http://www.google.com/search?client...nced+optimization+features,&ie=UTF-8&oe=UTF-8 All about Intel compilers here.
 
Last edited:
I think this is it.

That's where you're wrong. We'll need a programmer to weigh in on this but if I remember correctly (searching in the dim recesses) the software that you run on your computer doesn't actually make specific system or register calls. That's handled by the OS and the BIOS/UEFI. When Intel states "compatible processors" they mean x86, x87, SSE, SSE2, AVX etc. NOT that it's architecturally the same or "compatible".
 
That's where you're wrong. We'll need a programmer to weigh in on this but if I remember correctly (searching in the dim recesses) the software that you run on your computer doesn't actually make specific system or register calls. That's handled by the OS and the BIOS/UEFI. When Intel states "compatible processors" they mean x86, x87, SSE, SSE2, AVX etc. NOT that it's architecturally the same or "compatible".

Well gee how do they get to be 100% compatible? Clone!

Like I said just go back a bit and run some older benches from the days when AMD ruled the roost. That will give you the answer. If it pans out that the AMDs do compete well with the Intel's then continue your quest.

EDIT: By the way do you know what a compiler is? Yes I went to collage for programming.
 
Listen, the whole point of this exercise is to take the same benchmarks that all of us use here and reviewers use repeatedly to determine which CPU's are "better" and see if there's any "bias" in the results. That's a very easy thing to do with a Via Nano CPU and Agner Fog's program (he's the one by the way that discovered the CPU bias way back in 2005).

And I have another question for you. IF as you've stated, "Intel's compiler is for Intel CPU's" then why does that same compiler, after being patched to remove the vendor ID check, code programs to fully utilize the accelerators in AMD CPU's? For that matter if "Intel's compiler is for Intel CPU's", why has Intel themselves stated that they've "fixed" the compiler so that it no longer uses the "vendor ID check"? Why would they have needed to if it was as you say it is?

Earthdog, this isn't about "smearing" Intel. This is about seeing if the programs we use to determine a CPU's performance level can be trusted or not. I would think that it is something we should all know. I'll post the results whether they show bias or not.
 
SSE2 SSE3...... :blah: are the same thing on whatever CPU they happen to be resident on, are they not?

Its all software code and operating systems.... when they ask for code, math and what not they are expecting an answer, not a possible 2 or 3 for the same thing.
 
SSE2 SSE3...... :blah: are the same thing on whatever CPU they happen to be resident on, are they not?

Its all software code and operating systems.... when they ask for code, math and what not they are expecting an answer, not a possible 2 or 3 for the same thing.

They serve the same function but they are not the same.
 
Listen, the whole point of this exercise is to take the same benchmarks that all of us use here and reviewers use repeatedly to determine which CPU's are "better" and see if there's any "bias" in the results. That's a very easy thing to do with a Via Nano CPU and Agner Fog's program (he's the one by the way that discovered the CPU bias way back in 2005).

And I have another question for you. IF as you've stated, "Intel's compiler is for Intel CPU's" then why does that same compiler, after being patched to remove the vendor ID check, code programs to fully utilize the accelerators in AMD CPU's? For that matter if "Intel's compiler is for Intel CPU's", why has Intel themselves stated that they've "fixed" the compiler so that it no longer uses the "vendor ID check"? Why would they have needed to if it was as you say it is?

Earthdog, this isn't about "smearing" Intel. This is about seeing if the programs we use to determine a CPU's performance level can be trusted or not. I would think that it is something we should all know. I'll post the results whether they show bias or not.

Hey it seems like a smear thing and I am interested in your tests. I could be wrong and I am glad to be because it keeps me honest.

As to your question? They did it to stop the winers who got big brother to step in and they are now serving other manufacturers with the best compiler out there.
 
So how is lets say SSE3 on an AMD difrent on an Intel?

How can you clock an i5 2500K to 5Ghz 24/7 on air and not a PhII x4?

They execute the same instructions but they are not the same and can not legally be the same unless Intel says so.
 
Hey it seems like a smear thing and I am interested in your tests. I could be wrong and I am glad to be because it keeps me honest.

As to your question? They did it to stop the winers who got big brother to step in and they are now serving other manufacturers with the best compiler out there.

First, there is not and will not be anything in my posts in this thread discussing the "merits" of one manufacturers CPU's over another. That is compeletly beside the point of this thread. I am examining the programs we all use to measure CPU performance to see if there is bias in the program.

Secondly, "Smear Campaign definition - 1. sustained slander: a concerted effort to damage somebody's reputation by spreading harmful rumors about him or her."

Nothing in my first post is either "slanderous" or a "harmful rumor". I only discuss past FACTS as a basis of HOW benchmark performance can be biased. It wouldn't matter if the author of the compiler was Intel, AMD, NVidia, Ford, Chevy or the President himself. We all deserve to know if benchmarks can accurately be relied upon to measure our CPU's performance.
 
How can you clock an i5 2500K to 5Ghz 24/7 on air and not a PhII x4?

They execute the same instructions but they are not the same and can not legally be the same unless Intel says so.

My answer to your question would be a 2500K runs cooler and has better temperature tolerances, it is there for better at dealing with the higher heat levels (or lack of) at the higher clocks on mainstream coolers, there is nothing to suggest a PhII x4 can't run at 5Ghz 24/7 if you could keep its temperatures under control.

But anyway, me answering that question is going off topic to overclockability and temperatures, i would rather keep it on topic.

They execute the same instructions but they are not the same and can not legally be the same unless Intel says so.

What do you mean by that?
 
Last edited:
First, there is not and will not be anything in my posts in this thread discussing the "merits" of one manufacturers CPU's over another. That is compeletly beside the point of this thread. I am examining the programs we all use to measure CPU performance to see if there is bias in the program.

Secondly, "Smear Campaign definition - 1. sustained slander: a concerted effort to damage somebody's reputation by spreading harmful rumors about him or her."

Nothing in my first post is either "slanderous" or a "harmful rumor". I only discuss past FACTS as a basis of HOW benchmark performance can be biased. It wouldn't matter if the author of the compiler was Intel, AMD, NVidia, Ford, Chevy or the President himself. We all deserve to know if benchmarks can accurately be relied upon to measure our CPU's performance.

Well it came out wrong on my end. But the fact is no matter anyones opinion it has be ruled on and that is it. So my view is just that, a view and I disagree with Intel's capitulation.

Why? Well it has been beaten like a dead horse and it is now revived.

Now on with the tests!
 
We'll just agree to disagree on that point and let it rest at that :p

OH MY GOD PCMark 7 is agonizingly slow on this CPU. I am sooo glad we've moved past single core 1.6GHz CPU's.
 
But it is architecture. They are different. Now my view is Intel had optimized code for their processors and a generic code for everyone else. They did not know if it would run well on other processors and that is that.

As Bubba said this is not about accusation of evil Intel and his agenda for world domination (where is Mike Myers when you need him?)
 
I'll keep you guys updated as I go through them but I'll post all the screenshots when I'm done with all of the results in spreadsheet or graph form in the first two posts so I'm not doing it multiple times. If anyone has any suggestions on programs to run let me know and I'll see if I can run them. I ran Aida as Via and AMD, as intel it refused to work and locked up repeatedly. Right now I'm runnind PCMark 7 but only the "computational" section of the benchmark. When I run anything that has a "storage" section it just sits there for a long long time. I might let it sit overnight and see if it really is an error or if it's just taking forever to finish.

Interesting, PCMark 7 computational score for Via = 500, for AMD = 402. Now for Intel vendor ID....
 
Last edited:
Back