Author: Robert Hyatt
Date: 08:51:10 01/01/03
Go up one level in this thread
On January 01, 2003 at 11:00:36, Vincent Diepeveen wrote: >On January 01, 2003 at 05:23:57, Jeremiah Penery wrote: > >>On January 01, 2003 at 01:58:01, Vincent Diepeveen wrote: >> >>>Itanium2 blows the 21264 away for me. It's just that crafty cannot >>>profit too much from the next major advantage McKinley offers offers >>>over the 21264 for IPC: >>> >>>Look to this table: >>> >>>McKinley 6 instructions a cycle (bundles) >>>Alpha 4 instructions a cycle >> >>If you're going to post such simplistic tables, without much understanding of >>the data contained in them, here is a 'table' for you: >> >>McKinley - strictly in-order execution >>Alpha - out-of-order execution >> >>Make of it what you will. > >Read other postings from Bob. He's just talking about that if you know >how the vector machines (like the 1Ghz Cray) work that you can program for >it very well. > >Yet crafty doesn't profit from exactly that with the McKinley, which is >weird IMHO. Potentially it's 50% faster for him. He is the one who claims >he knows how to write for it. As usual, I have _no idea_ what you are talking about. Mckinly produces good numbers for crafty (NPS). Just not as good as a 21264. The 21264 is a better processor at the moment, and it should be since it has been 10 years in the making. Mckinley is only a couple of years old going back to the itanium 1. I _do_ understand the difference between a super-scalar architecture and a VLIW architecture. VLIW has been around for 30+ years...
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.