[mdlug] [Fwd: Re: [opensuse] 64 bit vrs 32 bit advantages speedetc.]

[Aaron Kulkis]

Adam Tauno Williams wrote:
>>>  From a neutral perspective Intel wants to sell there chips.  BUT they 
>>> know how to make their CPU's run faster. 
>> True.  But releasing a compiler with a deliberately
>> constructed sabotage (not using ANY vector operations
>> if a non-Intel chip is discovered, even though there
>> is a manufacturer-independant way of detecting the
>> presence of SSE, SSE2 and SSE3 capabilities) is
>> highly remeniscent of Microsoft doing a deliberate
>> search for DR-DOS (which was actually MORE stable
>> than MS-DOS) and issuing warning messages to
>> scare the consumer.
> 
> Ah, the rush to conspiracy.

If it wasn't a deliberate plan, then why in the
hell was it put in the compiler????


 >  Perhaps (a) it is a bug,

Checking for the string "GenuineIntel" is not
necessary to determine whether a CPU has any
SSE/SSE2/SSE3 capability.

And such a test does not end up in the code
by any sort of accident, unless you believe
that Intel employs and infinite number of
monkeys to generate its compilers and other
software.

 > or (b) a
> disagreement over the spec, or (c) they've encountered SSE/SSE2/SSE3
> implementations that don't work as reported so they err on the side of
> paranoia [not necessarily a bad thing given the complexity of a modern
> CPU and not knowing what the application will be used for (say, maybe,
> airliner navigation)].  I don't think anyone who has been involved in
> the development of complex applications will have any problem believing
> in (c);  everyone else will accuse them of foul play.  And if it is (a)
> or (b) it may take a v-e-r-y long time to resolve; understandably so.
> 

Bogus excuse making.

If an alternate supplier's chips don't work
PROPERLY, then I would hardly fault Intel for
that...but blindly assuming that ALL other
manufacturers chips cannot properly execute
SSE/SSE2/SSE3 instructions is, like I said,
likely a criminal act.

Based on the outcome of the DR-DOS/Windows sandbagging
suit, if AMD decided to pursue the issue, Intel would
not have come out with a favorable judgement.

Remember, at the time, Microsoft was VERY popular
with the American public (and therefore the potential
juror pool), and MS folded because they knew that
the suit was unwinnable because they had committed
an act of chicanery for the purpose of making a
detrimental impact on a competitor's product.

The same principle applies here.