[RC5] New Intel compilers...
bwilson at distributed.net
Mon Dec 9 13:23:21 EST 2002
Yes, as noted elsewhere in this thread, there is a difference between
optimize (improve) and optimal (perfect, or best possible). A
compiler optimization for Intel may improve performance on the AMD,
yet not improve it as much as is possible.
It only makes sense that two chips written to run the same instruction
set and operating under the same global constraints of die size,
thermal limits, speed of light, and the rules of numbers and math will
generally benefit from the same optimization techniques. Trying to
design a chip that was vastly different would be prohibitively costly
and serve no real purpose.
Bruce Wilson <bwilson at distributed.net>
PGP KeyID: 5430B995, http://www.toomuchblue.com/
"I want to move to Theory. Everything works in Theory."
--John Cash, id Software
| -----Original Message-----
| I would not say that! The differences dominate. Especially in
| RC5, which
| requires a very high
| intructionen throughput , special opimizations must be
| appied, that do
| not exist for the PIII.
| E.g, the PIII likes an 3-1-1 uop intruction pattern, the Athlon
| something completely different.
To unsubscribe, send 'unsubscribe rc5' to majordomo at lists.distributed.net
rc5-digest subscribers replace rc5 with rc5-digest
More information about the rc5