Author: Vincent Diepeveen
Date: 06:35:12 07/18/02
Go up one level in this thread
On July 18, 2002 at 08:51:53, Enrique Irazoqui wrote: >On July 18, 2002 at 08:39:20, Vincent Diepeveen wrote: > >>On July 18, 2002 at 08:38:16, Enrique Irazoqui wrote: >> >>>I have been offered a P4-2.53 with 533 bus. Are there any tests showing how this >>>processor does in chess compared with the AMD? >>> >>>Any information will be welcome. >>> >>>Thanks, Enrique >> >>divide by 1.7 for chessprograms to have the real speed of it in AMD K7 Mhz. > >I don't think this works, since it would make the P4-2.53 slower than the AMD >1800+ (1533), while in fact the 1800+ is not faster than the P4-2.2 > >Enrique Yes it is slower. If you already 'guess' without testing how fast it is for you, why do you ask the question? Vaste majority of chess programs if optimized for intel and K7, the 1 Ghz for a K7 is equal to 1.7Ghz for p4 is true. Special Fritz versions, DIEP, crafty, and another zillion of chessprograms that were accurately tested they gave this result. Also specint when not using a buggy intel c++ compiler is giving this result everywhere. For a *current* p4. We don't know what the future brings of course for the P4. However i wouldn't call a new chip a P4 anymore though they will probably sell it as an 'improved p4'. The bigger L2 cache is of course a joke for computerchess when running single cpu. The P4 Xeons do not have 512KB but 256KB L2 cache. Right now the K7 0.18 micron 1.73Ghz is king. for smp the dual k7 is king. there the diff between intel and amd is more than 1.7, because of a better smp design from amd (it's simply taken over from alpha, so not real new and of course still pathetic compared to super's).
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.