On 6 Dec 2001 11:19:05 -0500, Warp <war### [at] tag povray org> wrote:
>Angelo 'kENpEX' Pesce <ken### [at] uniplan it> wrote:
>: Well... Actually my asm routines tell me something different... Mabye
>: you should read something like agner fog asm optimization. Many guys
>: think that asm is not useful at all for those things... Well, at last
>: on intel plattform you're really wrong...
>
> Tell me one thing: For which processor would you make the asm code?
>P-II code is optimized very differently from P4 code. Athlon is also another
>different story. So which are you going to optimize for?
> And what happens when in the future Intel and AMD make new processors which
>may require specific optimization (eg. because they are 64-bit)? Your asm
>code will be optimized for the old architecture and the new compiler will
>be unable to optimize it for the new processor. Are you going to write the
>asm code for the new processor?
Of course there will always be the plain c code for the processors
that are not supported
For now let's support the most widespread processors
The only real problem with intel 80x86 compatible processors is the
p4, that is really different (and it's really a piece of shit).
Post a reply to this message
|