Author: Robert Hyatt
Date: 17:12:07 01/22/00
Go up one level in this thread
On January 22, 2000 at 19:59:54, Tom Kerrigan wrote: >On January 22, 2000 at 19:33:51, Eugene Nalimov wrote: > >>Let's assume you have a fastest race car in a world, for example a jet car that >>can drive with a supersonic speed. Then you are replacing its engine by a >>VW-beetle (old one, or new one - doesn't matter) engine. Resulting car will be >>absolutely incompetive not only with "proper" version of itself, but with any >>"normal" car, including VF-beetle itself. > >All evidence indicates that DB would run just fine on a PC. Evidence from Hsu >himself. Like the 40k instructions per node (presumably for full evaluation). >And the matches that Hsu played between commercial programs and 100k NPS DB. > >-Tom You overlook the hardware point. The equivalent of 40K instructions. But when done in the hardware design of DB. This does not mean that 40K instructions on an X86 would replicate this evaluation... It also means that when you design hardware, you think in a different box than you do when writing a program. The parameters are different for making decisions. I would design something one way if I knew I would be able to parallelize the pieces in hardware, I would write it another way if I was worried about performance on a serial Turing machine type gadget..
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.