|
|
Posted by Gene E. Bloch on 04/20/07 20:15
Maybe Intel processors are just incredibly difficult to code for.
Or they just planned incredibly far ahead...
Note the use of "incredibly" in both hypotheses :-)
"Smarty" <nobody@nobody.com> wrote in
news:5eydna5A3vY_x7zbnZ2dnUVZ_qWvnZ2d@adelphia.com:
> I don't think I am missing the point. Your point just doesn't make
> sense. Why would Apple pay them for the ***past 11 years*** to
> optimize code on Intel processors???
>
> Smarty
>
>
>
>
> "Jim" <no@spam.plz> wrote in message
> news:no-6BD56B.13365114042007@west.100proofnews.com...
>> "Smarty" wrote:
>>
>>> Jim,
>>>
>>> At risk of seeming a bit disagreeable, I would think that
>>> porting an OS in
>>> 1994/5 to Intel chips of that vintage has little or no relevance
>>> to optimizing X86 code 12 years hence. What I am saying is that
>>> the numerous profound changes to hyperthreading, multicore
>>> platforms, and the instruction
>>> sets of SSE, SSE2, MMX, and all the other changes to the Intel
>>> family in the
>>> ensuing 10 years time has rendered their optimizing skills as
>>> being of little or no modern value.
>>
>> You've missed the point. The engineers at Apple now were the
>> engineers at NeXTSTEP then. They've been optimizing code on the
>> Intel line since the 90's when they released 3.1.
>>
>> The test will be in releases, no?
>>
>> --
>> Edo ergo sum
>
>
>
--
Gene E. Bloch (Gino) ... letters617blochg3251
(replace the numbers by "at" and "dotcom")
Navigation:
[Reply to this message]
|