Intel to terminate AMD x86 license?!

[Apokalipse]

Intel's management would have to be idiots to want more competitors.

Of course, but that's missing the point.
It's illegal, immoral, and bad for the consumer when a company goes out of its way to crush a competitor.
the way the market is supposed to work, is through making competitive products and marketing them well.

If they don't like a competitor, then stiff ****. the markets in democratic countries, by design, are made to have competitors competing by making products/services and marketing them. It wasn't designed for, and is in fact designed against, companies crushing competition.
Unfortunately no system is perfect. However, AMD does have the option of fighting against Intel's practices via the legal system. which is what they're doing.
Hopefully other companies will benefit from the fight aswell.

How is it Intel's fault that the average consumer is uninformed or picked a brand they were familiar with?

didn't I just say it was besides the point?
Of course, marketing isn't illegal, unless they deliberately deceive customers.

 

[Puddle Jumper]

Of course, but that's missing the point.
It's illegal, immoral, and bad for the consumer when a company goes out of its way to crush a competitor.

Is it completely fine when a company breaks a contract they agreed to?

this is more than just favoring or providing optimisations for their own chips. They deliberately made inexcusably slower code paths which execute when a non-intel CPU is detected.

I still don't see what is surprising abut that, its a Intel Compiler. If you have a problem with it there are other options.

 

[Da Viksta]

well i personally think its not on to deliberately slow down non intel chips...thats called anti-trust and anti-competitive behaviour.

 

[Apokalipse]

Is it completely fine when a company breaks a contract they agreed to?

they haven't broken it. Also, I think it depends entirely on the circumstances.

I still don't see what is surprising abut that, its a Intel Compiler.

Do you understand the distinction between providing optimisations for your own chips, and deliberately slowing down competitors chips?
It is illegal to stifle competition in that way for a reason. It's just one of the things in AMD's antitrust case.

If you have a problem with it there are other options.

Not when programs come precompiled by others.

well i personally think its not on to deliberately slow down non intel chips...thats called anti-trust and anti-competitive behaviour.

If you had a look at some examples of the assembly code generated, you wouldn't say that.
Take a look here, for example:
Slashdot | AMD Alleges Intel Compilers Create Slower AMD Code

 

[Da Viksta]

If you had a look at some examples of the assembly code generated, you wouldn't say that.

why would you say that?

 

[Puddle Jumper]

they haven't broken it. Also, I think it depends entirely on the circumstances.

Are those circumstances if the company at fault is AMD or not?

Do you understand the distinction between providing optimisations for your own chips, and deliberately slowing down competitors chips?

Yes I understand the difference but the end result is the same, stuff compiled with Intel compilers runs faster on Intel cpu's, which again isn't surprising.

Not when programs come precompiled by others..

Then blame those others. They had the option to use any compiler they wanted and they picked intels. If you have a problem with that don't buy or use their software.

 

[Apokalipse]

why would you say that?

AMD Alleges Intel Compilers Create Slower AMD Code

I noticed this problem back in January of 2004, with Intel C++ 8.0, and went through heck over nine months with Intel's customer support to get it fixed until I eventually had to abandon their compiler.

On any non-Intel processors, it specifically included an alternate code path for "memcpy" that actually used "rep movsb" to copy one byte at a time, instead of (for example) "rep movsd" to copy a doubleword at a time (or MMX instructions to copy quadwords). This was probably the most brain-dead memcpy I'd ever seen, and was around 4X slower than even a typical naive assembly memcpy:

push ecx
shr ecx, 2
rep movsd
pop ecx
and ecx, 3
rep movsb

They responded with completely ridiculous answers, such as:

"Our 8.0 memcpy was indeed optimized for a Pentium(r)4 Processor,when we reworked this routine we used the simplest, most robust, and straightforward implementation for older processors so that we didn't need the extra code to check for alignment, length, overlap, and other conditions."

BS. I went and added the following line to the beginning of my source code:

extern "C" int __intel_cpu_indicator;

then I added:

__intel_cpu_indicator = -512;

to the "main" function.

This forced Intel C++ to use the "Pentium 4" memcpy regardless of which processor in in the machine. It turns out that their special "Pentium 4" memcpy which I tested thoroughly in all kinds of situations, and it worked perfectly fine on an AMD Athlon and a Pentium III. I pointed this out to them.

I received the following response:

"The fast mempcy is over 2000 lines of hand coded assembly, with lots of special cases where different code paths are chosen based on relative alignment of the source and destination. ... If the performance of memcpy/memset only are improved for Pentium III will that satisfy you?"

I answered "No," saying that I needed support for AMD processors as well. I also gave them a copy of my own memcpy routine that was 50% faster than theirs--and just used MMX. They closed the support issue and did nothing to resolve it.

I switched back to Visual C++.

AMD Alleges Intel Compilers Create Slower AMD Code

Are those circumstances if the company at fault is AMD or not?

The circumstances is what the company is forced to do to even exist in the market, when their competitor is out to crush them.

Yes I understand the difference but the end result is the same

no, the end result isn't the same. Instead of getting the performance they should, they get deliberately crippled performance.

Then blame those others. They had the option to use any compiler they wanted and they picked intels. If you have a problem with that don't buy or use their software.

You don't know what compilers they used.
Also, they probably don't know that Intel cripples non-Intel CPU's. It's not like they tell people "we cripple our competitors CPU's"

You can't shift the blame on the people using it. the blame rests on the people who designed it to behave like it does.

That's like blaming a racing driver/team if a competitor swapped their fuel for one that gives 20% lower power, or something - while that competitor used the proper fuel.

 

[Puddle Jumper]

You don't know what compilers they used.
Also, they probably don't know that Intel cripples non-Intel CPU's.

You can't shift the blame on the people using it. the blame rests on the people who designed it to behave like it does.

I'm pretty sure all of the people with CS and CE degrees know a little abut compilers.

You can't shift the blame on the people using it. the blame rests on the people who designed it to behave like it does.

According to that a knife maker is responsible for a stabbing because they made the knife sharp.

 

[Apokalipse]

I'm pretty sure all of the people with CS and CE degrees know a little abut compilers.

Intel doesn't tell people they cripple their competitors CPU's.
Even if they did know, it still doesn't make it right. The competitors CPU's are still not performing like they should, because Intel crippled them.
There is no excuse for writing your compiler to deliberately slow down your competitors chips.

According to that a knife maker is responsible for a stabbing because they made the knife sharp.

your analogy isn't valid. The behavior of the knife depends on the person who uses it.
the compiler behavior depends on the person who wrote it, not the person using it.

 

[zmatt]

Boohoo for AMD, for over 20 years they have been able to not only thrive in Intel's crap but have beat them in several cases. Why should we feel bad for them now? If AMD had invented the microprocessor they would be doing the same thing. The only reason they are in this position now is because they dropped the ball on the AM2 transition and let Intel get the lead. You can't do that against an opponent like Intel. AMD is paying the price now for their lack of initiative, that's how the market works. The weak die and the strong survive.

Apok, if you feel so strongly about it then I suggest you try and be proactive instead of arguing with us. I'm sure there are groups out there of people who agree with you who want to get the playing field leveled, I encourage you to join them.

I have little respect for someone who sits and complains but I have the greatest respect for someone who fights for what they believe in.