We don't, but it has been raised before that it is more "sabotage" than "optimization", and as far as I know has been resolved before. The FTC, on the other hand, thinks otherwise. They seem to be pretty confident they can do this even without AMD on their side. That leads me to think they might just have something big, an ace up their sleeve so to speak.
Are you talking about the FTC institution/entity, or the government-employed people in it? Whether they crap on Intel right now or not, the FTC is going to keep on existing, so perhaps you are talking about the people involved who wish to keep their particular jobs in the FTC somehow by taking a crap on Intel?I wouldn't equate their desire to do something with themselves (which is the only thing that keeps them employed) as being tantamount to them having confidence on this particular matter.
Yeah, Intel really is going to be screwed with this. I mean seriously, the FTC pretty much gets to pick judge, jury, and venue. What does Intel get? Sure they can send the best lawyers in the world. But in the end, the FTC is bound to decide that something is wrong and fine them for it.Both, neither the government institution nor the people employed by it risk anything by pursuing this action against Intel.
But it does give them something to fill their calendars with, and is expected to drag on to 2012 so from a government employment perspective its full of win.
That is but one of the many differences between being sued by another business versus the government. When it is another business usually there is an over-riding profit motivation, and time-to-profit drives the timeline. And risk of counter-suit drives down lawsuits based on false or rash merits.
None of that applies here. But then again it is with this in mind that one expects executive-level decision makers to strive to do what they can to stay off the radar of a government agency like the FTC.
For Intel there is absolultely no upside to this, it is all downhill and the only unknown is just the pace of the decline from here out.
While the end result is the same, I don't think that's exactly what the FTC is complaining about. For software to perform better on Intel CPUs versus AMD CPUs, these two could happen:
1.) Intel made a compiler (or modified it so) that, quite naturally, produces code that is optimized for their processors. Whether it will also fit like a glove for competing processors is not a concern.
2.) Intel modified the compiler not for optimization concerns for ther processors, but rather to deliberately slow non-Intel processors.
You are saying it is case #1, and your reaction is valid for it.
The FTC, however, is clearly claiming #2.
My personal opinion is that the reality is most likely #1. I will, however, be interested in seeing the developments in this case against Intel.
Okay, but how is that any different from
Code:if(CPUID = Geuniune Intel) { Execute [b][u]accelerated[/u][/b] code path } else { [b][u]normal[/u][/b] code path
In their case they're claiming Intel forces AMD and other processors to use a slower code path.
How do we know that Intel isn't forcing it's own processor to use a faster code path rather than slowing down the competition?
If the code is done correctly to detect genuine Intel processors then it shouldn't work on any other processor, right?
If this goes through, I can see Intel releasing a junk(their junk might still end up being faster than others on the market but lets say it's still junk) compiler in the future only to optimize it for their own processor at a later update, or they may choose to only update their compiler code for the Intel line while leaving AMD's stagnant and choosing not to develop them since apparently neither AMD nor Microsoft can do so either.
If Intel releases an SSE4.2 compiler, then decide to add an SSE5 code path at a later date for Intel processors only, how is that anti-competitive?
What if they decide not to allow their compiler to work on AMD and say there are other compilers out there for people to use?
We're not talking 10% difference in performance here. More like 200%.
Fox, V8, assuming it is "door #2" what exactly makes it illegal or anti-competitive (in the unlawful sense, all business efforts are supposed to be anti-competition to begin with) for Intel to make their own compilers favor their own hardware and put their competitor's hardware at a disadvantage?
Could you imagine if SUN expected its customers to rely on IBM's compiler team for a compiler that could/would generate the best binaries for use on Sparc platforms?
It sounds absurd because it is absurd, so I am wondering how/why/when/where the world of competition got so twisted around backwards with Intel that they are to be demonized for trying to make their products look best when using their software to compile the code?
AMD and Intel cross license SSE based instruction sets. If their compilers artificially ignore these standards, they are competing unfairly and thus subject to monopoly regulation. I guess if they claimed the compilers weren't compatible with said instruction set, or were specifically only for Intel processors, they could get away with it, but they're marketed them as x86 compatible compilers and are expected to work to that standard, not some subset there of.
And? If it is an issue, software developers would use a different compiler. Intels compiler has no baring on my processor choice.The difference here is that the Intel compiler would have worked just fine for AMD, but they had to go and look up vendor id to prevent AMD CPUs from using the optimized compiler.
And? What does their cross license have to do with the compiler? Its a pretty big leap to say "They share a license therefore their compiler should use amd SSE"
Could the compiler use it? yes. Does it? No. Does that mean that AMD's license has been violated? Absolutely not. Just because intel's compiler doesn't use their OP_Codes, doesn't mean that someone else can't.
As I said before in this thread. Intel isn't the only compiler manufacture out there, they aren't even the most popular one (Though, they are thought of as the one that generates the fastest code).
The compiler market is different from the processor market. Saying that a compiler makes the processor market anti-competitive is ludicrous.
Well, at least we could pay off the national debt with the money they take. (What does the FTC do with their fines?)
Does it matter? As long as the compiler generates SSE code, it is an SSE compiler. Even if it disables that code on other SSE enabled CPUs it is still and SSE compiler. This has nothing to do with licensing.Do they advertise it as SSE compatible or Intel only SSE compatible. In general when you cross license something, it means you are agreeing to adhere to that standard. Not just when it's continent to you.
Then what were you trying to say? If it didn't violate their license, I don't even see why YOU would bring the license up. Cross licensing doesn't mean that all the sudden compilers that support SSE code have to support (or can't disable) for every SSE licensed company or processor.Did I say it violated AMD's license? Strawman? I was arguing it was a deceptive practice. The fact that they reversed this practice generally affirms its uncompetitiveness.
Nobody. We are talking about legality, not popularity or fairness. Intel making and marketing a compiler that generates code that only works on Intel CPUs is not illegal.And? Who ever said popularity and fairness must walk hand in hand?
HA HA HA, nice one. You know, there are ARM compilers that *GASP* are tied to ARM processors. They don't make code that will run on the x86, isn't that despicable! ARM must be anti-competitive. So is IBM, Sony, Samsung, Motorola, VIA, and anyone else that makes a microprocessor.Yet they were tying their compilers to their processors to gain a distinct advantage. Not related to the processor market? Ludicrous?
It's under chapter 5. No money is involved. It's a ultimatum, clean up your act or face breakup.
Does it matter? As long as the compiler generates SSE code, it is an SSE compiler. Even if it disables that code on other SSE enabled CPUs it is still and SSE compiler. This has nothing to do with licensing.
And for the record, they DID adhere, very strictly, to the SSE standard. They just didn't implement it on every processor. Implementation has nothing to do with adherence to a standard.
Then what were you trying to say? If it didn't violate their license, I don't even see why YOU would bring the license up. Cross licensing doesn't mean that all the sudden compilers that support SSE code have to support (or can't disable) for every SSE licensed company or processor.
Nobody. We are talking about legality, not popularity or fairness. Intel making and marketing a compiler that generates code that only works on Intel CPUs is not illegal.
HA HA HA, nice one. You know, there are ARM compilers that *GASP* are tied to ARM processors. They don't make code that will run on the x86, isn't that despicable! ARM must be anti-competitive. So is IBM, Sony, Samsung, Motorola, VIA, and anyone else that makes a microprocessor.
The code SSE code, as others have demonstrated, worked just fine on AMD cpus. It wasn't intel only, that WOULD be a violation of the SSE license. It just used CPUID to route the a non-intel CPU away from the SSE code. The SSE code was valid, worked with an AMD cpu, didn't exploit any undocumented, unknown SSE functionality. It generated SSE code which people using AMD CPUs could see and, with some editing, execute.I would say if it only generates SSE code for Intel processors it's an x86 compiler with Intel only SSE code. As I said in my original statement, if they advertised it as so, it probably wouldn't have been a problem.
Duh, Just because the standard says how to determine if a processor is SSE capable, doesn't mean that a software model has to follow that standard to generate SSE code.Duh part of the SSE standards are how to determine what instructions are available on the processor. I guess adherence is only a standard when it's convenience.
Unless I missed something, their software was compatible. SSE code didn't suddenly stop working on AMD CPUs even if intel's compiler didn't allow it to run on them. Underhanded, yes, adhering to the standard? yes. AMD and intel also agreed on how CPUID would work, Intel adhered to that standard and used it to filter out how the code would behave.My point was if you license Intel technology, you pay for the benefits that come with that technology, both software and hardware compatibility.
Understood, you take the approach that this is illegal because intel is intel.What may be legal to some, is not legal to others. Monopoly's must adhere to a higher standard.
No, it is completely relevant. What you are arguing is just as absurd as the statement about IBM et al.WTH total poison in the well and PS your analogy doesn't really hold water. Does ARM ignore instructions based on architecture? Are you claiming unfair practices? If so please elaborate, othewise you're just muddying the waters with irrelevances. You can bet that all the above companies, especially IBM, are smart enough to uphold whatever standards/patents/contracts they enter into. Licensing is more than just an annoyance to them, it's their cash cow.
The code SSE code, as others have demonstrated, worked just fine on AMD cpus. It wasn't intel only, that WOULD be a violation of the SSE license. It just used CPUID to route the a non-intel CPU away from the SSE code.
The SSE code was valid, worked with an AMD cpu, didn't exploit any undocumented, unknown SSE functionality. It generated SSE code which people using AMD CPUs could see and, with some editing, execute.
Think of it this way. x264, a popular encoder, will use certain SSE instructions depending on the CPU detected (Even if the other SSE instructions are present). x264 still claims to use SSE code, even tough it disables certain SSE instructions based on CPU, should we start complaining and saying that they are false advertising for not following the SSE standard? No, that's ridiculous.
It is the same thing, saying that Intel should be held liable because they would criple code for the competitors processor is ridiculous, the only thing more ridiculous is saying that the reason it is wrong is because they aren't obeying their SSE license (Or whatever you are arguing, you keep saying that it is both an issue and not an issue. Which is it?).
You might as well say that their compilers can't say they support x86 instructions because some compiler flags would produce code that would prevent the code from running on a non-intel CPU.
Duh, Just because the standard says how to determine if a processor is SSE capable, doesn't mean that a software model has to follow that standard to generate SSE code.
Unless I missed something, their software was compatible. SSE code didn't suddenly stop working on AMD CPUs even if intel's compiler didn't allow it to run on them. Underhanded, yes, adhering to the standard? yes. AMD and intel also agreed on how CPUID would work, Intel adhered to that standard and used it to filter out how the code would behave.
Understood, you take the approach that this is illegal because intel is a monopoly (intel).
No, it is completely relevant. What you are arguing is just as absurd as the statement about IBM et al.
anand just put up an article on this law suit, here's something interesting, FTC demand:
"To stop selling things below cost. The FTC is defining this as being the average variable cost plus a contribution to Intels fixed sunk costs in an appropriate multiple of that average variable cost.
first how do FTC actually know what some tech is worth, I mean a i7 920 is worth 200 or 250 or 300? how does anyone know? I find that very dubious and impossible to enforce. even more considering performance level of i7 there's absolutely no competition at that level, intel can even set the price 350 and people might still buy it.
another good one,
"For Intel to stop badmouthing competing products unless they have solid scientific evidence." I think badmouthing is a industry tradition, like when intel calls amd x3s defetive products, but how is this something bad? I think it's just a marketing trick. certainly nothing to sue over.
one thing I do agree on,
"Intel cannot require OEMs to purchase only Intel CPUs and GPUs" intel's been doing this too much and it certainly slows down competitions not with superior technology but with market share clout. I think this type of tactics allow Intel to hold down to market share without invest in making their products better which is bad for progress.
They seem to be pretty confident they can do this even without AMD on their side. That leads me to think they might just have something big, an ace up their sleeve so to speak.
It's the government. I wouldn't bet on them being either competent or thorough. The only thing I would count on them being is: expensive.
I agree. It looks like lawyers who grew up with computers are finally taking over from the older generation. The FTC compaint is surprisingly up-to-date and well versed in technology, compared to what we've seen anywhere in the past. I wouldn't be surprised if some of them are seasoned AnandTech readers.After how botched the Microsoft monopoly case was, this suit looks to at least have a better understanding of the issues at hand.