Mail Archives: pgcc/1999/05/25/21:20:32
At 12:28 AM 5/26/99 +0200, Marc Lehmann wrote:
>On Mon, May 24, 1999 at 05:56:16PM -0400, Adam Schrotenboer wrote:
>> Noticed that there is aparrently some interest in EPIC, which at this
>> time is not yet supported by GCC.
>
>I am sure gcc will support Merced when it will released ;)
>
>But first of all, if you want technical rather than strategic info I'd
>recommend to look at the GCC EPIC page:
>
>http://www.goof.com/pcg/epic/
>(I'm currently in the process of adding new links to that page)
>
>> My q is as to whether or not it is an open std, or is Intel going to
>> charge a head and torso for any other chipmakers to use it?
>
>EPIC is an architecture like RISC or CISC - you can patent aspects of it,
>but not the idea itself.
>-
>What Intel does is to sell the IA64, and, yes, they are fairly
>closed-minded about who gets which information.
Note that AMD K6-3 was not much faster than AMD K6-2 for most programs,
which was a big dissappointment actually. I expected more out of it.
PII-450 equals K6-3 at 475Mhz (applications using a lot of L2, and
considering speed of cache at K6-3 i expected more out of it).
IA-64 should be having a lot of instructions that cray processor has also
for bitboards, which should give this processor an interesting performance,
like the alpha 21264 does.
>> Will AMD be able to use it in an upconming K8 or whathever the next one
>Good question. Don't think so.
>
>> will be??? Or Cyrix (heard some talk of AMD buying Cyrix, though I don't
>
>Even better question, I don't think Cyrix will, they've got quite a few
>problems to solve with their current designs ;)
>
>At the moment it seems that the only other counter-player against the
>merced is the elbrus-2k, which might get delivered even earlier than the
>merced!
>
>--
> -----==- |
> ----==-- _ |
> ---==---(_)__ __ ____ __ Marc Lehmann +--
> --==---/ / _ \/ // /\ \/ / pcg AT goof DOT com |e|
> -=====/_/_//_/\_,_/ /_/\_\ XX11-RIPE --+
> The choice of a GNU generation |
> |
>
>
- Raw text -