Author: Bob Durrett
Date: 15:14:28 02/17/04
Go up one level in this thread
On February 17, 2004 at 17:47:56, Slater Wold wrote: >About $50,000 and 2 years. It might be worth it! think marketing! Bob D. > > >On February 17, 2004 at 16:56:46, Bob Durrett wrote: > >> >>If one ignores the many anticipated objections to doing this research, then what >>would it take? >> >>I can see the following process: >> >>(1) A list of blocks of code which could be "hardware-ized" is made and the >>list ordered in order of decreasing expected benefit. >> >>(2) Then, the powers that be would decide which blocks to do that to. >> >>(3) Then it would be done. Testing and tuning would follow. >> >>(4) Then a match would be set up with Kasparov. >> >>(5) Etceteras. >> >>By "hardware-izing" I mean creating a chip which could input it's input data, >>process that data, and output that data all within the time of a few clock >>cycles of a modern microprocessor chip. >> >>QUESTIONS: >> >>Which block of Crafty code would be chosen first? How much benefit to expect? >> >>Same questions for second block of code. >> >>Etceteras for the rest of the code until nothing is left. >> >>Bob D.
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.