• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Intel sued for stunting performance of competing GPUs

I refuse to buy Intel as this is well known fact that they have monopolised the CPU market. Hence why the EU and even the US are looking to sue.
They've been caught out for dodgy practices in the sales market whats to say they're not doing it in other ways aswell.

AMD took the opportunity and took a pay-off from Intel - understandably to further they're business as I'm sure they quickly realised that even if they sued Intel they wouldn't get much in return - just Intel being fined. Intel knew they've been upto dodgy stuff (more so than meets the naked eye) hence why they 'bribed' AMD.

64bit computing didnt become mainstream until AMD kicked it off with all they're Athlon64 models. Even Microsoft bent over for Intel and refused to launch WinXP64 until Intel had they're 'equivelant' chips out!

Maybe my outside influences are a factor in disliking Intel - Israel for example is a well known 'friend' of Intel with the israeli's going so far as to be boistrous about how the Core2 architecture belongs to Israel - because it was developed over there!!! wtf! -since when have you ever heard of a country claiming a companies technology and that company backing them up?. Intel assists Israel with investing heavily in they're country, in return giving them - the ONLY country in the world to have 50+ UN sanctions (and counting)- a heavy bias towards leniency from the US( who we all know favours big corporations), even though they are ironically carrying out the same acts they suffered under the Nazi regime (with the palestinians)

Besides all that - this is OLD news - I recall a good couple of months ago reading about how Intel only optimised they're compiler code for their own CPU's - this meant the competition would always run at a disadvtange.

You'll also notice this in the CPU instruction set arena - where most benchmarks are biased towards certain instruction sets which funnily enough Intel CPU's have but AMD CPU's dont.....
 
I was going to post the same thing roughly, I can't remember the specifics and can't be bothered to read it again. IT was essentially along the lines of having compiled code that would ignore using things like SSE and all its versions unless it detected a Intel CPU.

I'm sure it was a long time ago, not something they get away from and it really has nothing to do with GPU's. It doesn't matter if they crippled AMD to 10% of its gpu performance it would still beat Intel's current gpu's :p

They've been slapped on the wrist about it and threatened enough times with enough lawsuits over the years, aswell as enough people looking out for it that I doubt they'd do it again(obviously anyway :p ).
 
Still confused, the crux of the document seems to focus on CPU's, I might be wrong but surely you can't use any other manufacturers processor in an Intel board other than one from Intel?

"The agency also alleged that “Intel secretly redesigned key software, known as a compiler, in a way that deliberately stunted the performance of competitors’ CPU chips.”

What am I missing here?
The compiler is used to convert a programming language to the specific machine codes, now imagine your generic x86 compiler (supplied by Intel) actually deliberately cripples the performance of software running on non-Intel x86 CPUs.
 
Couple of things,

Is anything being done to help AMD?

And if a compiler issue gets sorted, would I get better performance from my pc on games?
 
How is this possible when Graphics performance on Intel boards is still much better than it currently is on AMD?

Maybe because the i7s are arguably 'better'. If they didn't mess with all this compiler stuff, i'd imagine the performance would be even better.
 
There's a few pretty glaring faults here, beyond the lack of a source for the information.

The OP says that intel have deliberately made processors that perform badly with competing nvidia and ati graphics cards. Given that intel don't make graphics cards, this is fairly retarded. There might have been a case if i3 included gpu but no pci-e controller, but as i3 will run nvidia or ati graphics cards yielding far better performance than onboard the argument makes no sense here either.

prohibiting Intel from making or distributing products that impair the performance — or apparent performance — of non-Intel CPUs

A phenom 2 will run very badly indeed on a X58 chipset. As such all intel motherboards are designed to perform well with intel's processors and badly with other processors. Who else thinks this is entirely reasonable?

I'd hope the compiler intel is distributing optimises code for intel chips. However I am assuming that amd have written a compiler that optimises code for amd chips. It's unreasonable to force intel to write a compiler optimised for their competitor if their competitor is unwilling to write one themselves.

If the entire market uses an intel supplied compiler, which intel haven't spent resources optimising for amd, then perhaps they can be sued. If Intel are forcing companies to use their compiler exclusively, then yes there probably is a case against them. I would like to know why the entire market is using the intel supplied one though, as I believe most code on my computer is compiled using gcc 4 and it's running fine.

Iirc, both c2d and amd chips use the "amd64" architecture flag if compiling on Gentoo, which suggests there's not much difference between the code that runs on each.
 
^ Good post JonJ678 - My feelings entirely

Edit..

I just read that the Intel x86 isn't open - x86 requires license from Intel; However, x86-64 require an additional license from AMD.

So surely that means that Intel can really do what they want with their x86 compliler.. more fool AMD for using it.
 
Last edited:
^ Good post JonJ678 - My feelings entirely

Edit..

I just read that the Intel x86 isn't open - x86 requires license from Intel; However, x86-64 require an additional license from AMD.

So surely that means that Intel can really do what they want with their x86 compliler.. more fool AMD for using it.

True, same could be said for x86-64 if AMD were to modify there software. If AMD and Intel were to battle it out with each other not renewing there licence AMD would cease to exsist and Intel would be back on 32-bit chips.

Messed up. Intel are not a good company this does not suprise me, maybe the AMD64s left Intel with a grudge to get one back at AMD.

AMD really need to step up there game and stop following Intel. I agree with standards being brought in so each other has the same starting block, from there we see who would be the better CPU manufacturer.
 
If i recall correctly, there are several parts to the claim.

I believe the case was that intel would threaten companies who would use AMD /competitors products, and give discounts to companies that use intel purely, thus monopolising those companies.

As per the compiler, I dont think it was the fact that it produced intel optimised code, after all there are AMD optimised compilers. I believe it's the fact that the compiler probabily did this without the programmer knowing, instead of producing a generic 32/64bit output ~ so when it came to benchmarking, intel ran the program faster and failed to disclose the fact that it was optimised for intel.

It's like saying Windows runs 10 times faster in an AMd system than it does in intel, failing to note the fact that the AMD system has been overclocked 10x the speed and is cooled with liquid nitrogen.

and as per the graphics: http://www.pcmag.com/article2/0,2817,2353938,00.asp something to do with blocking nvidia from using thier bus technology or something.
 
Certainly is something going on with the Intel compiler.

In the field i am most interested in number crunching the phenoms have always being signifcantly slower than core 2's.

Anyways in the latest version of the software i use (LLR) the AMD's have gotten a 33% speedup.
Speed increase is around 5% for the Intels.

Now is the programmer a coding god, well he is good but not that good.
That massive increase had to come from compiler optimisations.

This leads me to believe that now the 'cat is out of the bag' Intel have changed the behaviour of the compiler.
Seems it now optimises routines on both AMD and Intel chips if possible, before it optimesed a lot of routines only for Intel chips.

Intel will get a slap for this but they are trying to minimise the damage by playing fair now.
When it comes to court the current compilers will be squeaky clean.
 
I don't know of anyone that uses the Intel CPP compiler. Everyone uses either Microsoft's or GCC.

Total non-issue. It's like moaning that Apple iTunes doesn't support Creative Zen players... Why would they want to support third party products with their software?

And the 3DMark.exe thing.. this is nothing new. NVidia have been capturing the executable name in their drivers and applying bespoke optimisations to the rendering for as long as I can remember.
 
Everyone uses either Microsoft's or GCC.
I can name several compilers and these would not have been made and continually updated if they didn't have a large userbase generating them nice profits. You have simply named two popular ones, among many that should include intels.

I'm sure when people look towards making window programs they look towards microsofts' toolset, where as hardware manufacturers working with intel based systems and possibly other intel devices which normal compilers wouldnt be able to support, would want to use intels compiler and even continue to use it for all other, non intel related tasks for the sake of ease. Lets not forget that if a new instruction set is introduced to a new intel CPU, who is likely to have support for it in their compiler first?

Total non-issue. It's like moaning that Apple iTunes doesn't support Creative Zen players... Why would they want to support third party products with their software?
Apples and Oranges. :

This appears to be the case for intel's compiler.
The Intel compiler and several different Intel function libraries have suboptimal performance on AMD and VIA processors. The reason is that the compiler or library can make multiple versions of a piece of code, each optimized for a certain processor and instruction set, for example SSE2, SSE3, etc. The system includes a function that detects which type of CPU it is running on and chooses the optimal code path for that CPU. This is called a CPU dispatcher. However, the Intel CPU dispatcher does not only check which instruction set is supported by the CPU, it also checks the vendor ID string. If the vendor string is "GenuineIntel" then it uses the optimal code path. If the CPU is not from Intel then, in most cases, it will run the slowest possible version of the code, even if the CPU is fully compatible with a better version.

This vendor-specific CPU dispatching decreases the performance on non-Intel processors of software built with an Intel compiler or an Intel function library - possibly without the knowledge of the programmer. This has alledgedly led to misleading benchmarks[17]. A legal battle between AMD and Intel over this and other issues has been settled in November 2009[18]. In addition to this, the US Federal Trade Commission has filed an antitrust complaint against Intel.[19]
Source: wikipedia :3
 
Intel's a niché compiler. There is no getting away from that fact. People use the Intel compiler if they want the ultimate performance on Intel CPUs only.

Microsoft's and GCC are the most popular yes. That's why I stated them. Most desktop applications are compiled using these compilers. I suppose I should have also mentioned Borland's but that's fallen out of favour a bit these days.

It comes down to testing. Microsoft, GCC et al test their compilers to generate compatible instruction optimisations for a wide variety of processors and architectures.

Intel however only tests their compiler's optimisations on their own CPUs. It would be crazy for them to release a compiler that uses optimised code paths on CPUs/architectures they have not tested it on. This is the primary reason why Intel go one step further in their dispatcher routines by checking the Vendor ID string of the CPU.

You would be amazed of the subtle differences in behaviour between AMD64 and Intel EM64T. Microsoft, GCC et all are *still* optimising their compilers for this instruction set today. For instance a lot of the early x64 CPUs had quite significant bugs in their implementation and compilers must check for this and use a different, more conservative, code path.
 
I'm a little confused. Is this a whole lot of rage directed at Intel as they optimised software for their processors? This feels almost as ill informed as the EU getting upset about internet explorer in windows.
 
Back
Top Bottom