Hello all,
I originally intended to sit back and watch and see who bites on this one ,
since it is actually not on topic for this newsgroup. However since it
appears that sgopus has completely missed the point of the article's target
audience and Ed's reason for posting it here, I have decided I should add my
5 cents worth(inflation), especially since I have also read a few posts
where someone advises someone to stay away from AMD and Via because, "they
are crap."
Also, since I have always used AMD with no more problems than people with
Intel processors, I think that this post and link is completely appropriate,
so people with AMD/VIA might finally see that the problem is most likely not
the processor alone, but the program that doesn't run as efficiently as it
should. I never trusted benchmarking software anyway. lol
So in response to sgopus:
Of course this is directed at programmers who use the illegal Intel compiler
which Intel was recently forced in an antitrust settlement with AMD to
change, BECAUSE it CRIPPLES AMD processors. . . . Duh!
A few paragraphs down the writer quotes Agner Fog who states, . . .
"Many software developers think that the compiler is compatible with AMD
processors, and in fact it is, but unbeknownst to the programmer it puts in
a biased CPU dispatcher that chooses an inferior code path whenever it is
running on a non-Intel processor," Fog writes, "If programmers knew this
fact they would probably use another compiler. Who wants to sell a piece of
software that doesn't work well on AMD processors?"
(
http://www.osnews.com/story/22683/In...from_Compiler_,
5th par.)
So the article IS NOT just for "AMD programmers", it is for all programmers,
because most programmers write their software with the expectation that it
will work on ANY computer regardless of the processor the computer was built
with.
If Intel had seen fit to include in their white paper and other
documentation that the compiler works better with Intel processors by
priority, they would not have gotten into legal trouble to begin with, and
programmers might have been willing to freely seek out a compiler that works
equally efficient with both AMD and Intel processors. Instead they purposely
included code that intentionally cripples (or rather degrades the
performance of ) AMD/Via processors.
So as far as reading the WHOLE article, sgopus, even if you did read it all
I expect that you only paid attention to those paragraphs that look like
they support your argument when a closer look reveals that you are only
supporting Ed's point that AMD users should read this, because once the
program with the illegal compiler is distributed, it then becomes the
problem of the end user. Just think how many programs have been distributed
that work GREAT on Intel computers but CRAPPY on AMD/VIA computer.
I for one believe the end users of AMD should sue Intel as well for all the
money they paid to have a tech diagnose their computer who advised them to
replace their motherboard or whole computer with one that uses an Intel
processor because Intel computers are 'better'; . . . this just because of a
small piece of code intended to make Intel look better than AMD. LOL It's so
brilliant that they already made their money before AMD and the FTC finally
discovered what was wrong. LOL
So, I consider this matter solved and have no intention of discussing it
further with anyone from this point on.
Good article ED, thank you for the link.
Happy New year to all....
--
Peace,
Jeffrey W. Bowen
Remove NO SPAM entries from email address to send personal email.
Please post replies of successes or failures so we all can benefit from each
others' experiences.
"sgopus" <(E-Mail Removed)> wrote in message
news:7A11C1D5-AAF9-4640-A20B-(E-Mail Removed)...
> when you cite an article you should take the time to read it completely, I
> quote
>
> "Here's something you probably don't know, but really should - especially
> if
> you're a programmer, and especially if you're using Intel's compiler. It's
> a
> fact that's not widely known, but Intel's compiler deliberately and
> knowingly
> cripples performance for non-Intel (AMD/VIA) processors."
>
> So, if your a programmer and you use the compiler on a NON Intel processor
> machine, it may choose a non optimized path for your compile. This
> doesn't
> say any and all software may run poorly on a NON Intel processor due to
> Intel. if the same software is compiled on a Intel processor machine it
> may
> compile faster, not necessarly execute faster, once compiled.
>
> "ED" wrote:
>
>> The way I see it, a lot of AMD users are running equipment that are
>> performing below their designed capabilities
>> because the software they are using is jacked up. Many are complaining
>> about
>> slow performance. How much of
>> that is caused by a tricked up compiler ?. Most software on the market is
>> compiled, apart from those written in machine
>> language so I don't see how you reason that this article only applies to
>> programmers.
>> The end user is the one who ends up being cheated.
>>
>> "sgopus" <(E-Mail Removed)> wrote in message
>> news:CD1C59A8-0777-44C5-93FF-(E-Mail Removed)...
>> > This applies to AMD programmers who use the Compiler function, most
>> > USERS
>> > who
>> > have AMD processers will not be affected by this.
>> >
>> > "ED" wrote:
>> >
>> >> http://www.osnews.com/story/22683/In..._from_Compiler
>> >>
>> >>
>> >> .
>> >>
>>
>>
>> .
>>
Similar Threads
