Author: Amir Ban
Date: 14:57:07 12/15/99
Go up one level in this thread
On December 15, 1999 at 14:39:18, Greg Lindahl wrote: >I'm interested in coordinating an effort to build the next >world-champion chess program. FPGA technology has advanced since Deep >Blue's construction such that replicating its architecture is now >inexpensive. All you need is a cluster guy (me), a pile of FPGAs >(corporate sponsor), someone who knows how to route a board evaluation >chip, and someone interested in building a message-passing chess >program which uses the board evaluation chips as an accellerator. > The chips need to do more than evaluation. For this architecture to make sense, they need to handle move generation and perform subtree search. That's what Deep Blue did. If you restrict these tasks to your cluster processors, they won't be able to gain much from your evaluation chips. They'll need to transfer the position and a big bunch of parameters, which may take so much time as to make it not worthwhile. In addition, evaluation is only part of the time a chess program takes, and there are some well known techniques like piece-square-tables to reduce it even more. Designing a chip that will do evaluation, move generation and partial tree search sounds to me like a task not to be taken up lightly. Just being experienced with Verilog doesn't mean you have it covered. >In the tradition of Deep Blue, I'm fairly clueless about chess. > You can stay clueless as far as this project is concerned. However you'll need to become very familiar with the way computers play chess, because the main design problem here is to transfer this to your proposed architecture in a way that makes sense. There are several aspects that may not be obvious at first sight, like repetition detection, which need to be handled or else your machine will make beginner-level mistakes no matter how much hardware you throw at it. It will help in this project if you are a design genius. >Unlike Deep Blue, I plan on making this engine available for more than >just a few games, so that its behavior can be studied. > >If you are interested in this project, please drop me email. I'm >especially looking for a person/group seriously interested in >constructing the overall program. I can provide the cluster & support, >find the FPGA corporate sponsor, and find someone to route the chip. > It's an interesting initiative, but I think you don't realize yet what you need here. Translated to software terms, you are saying: I have a compiler, and I know someone who can give us a PC, now all I need is someone to write the program. In short, you are not really bringing much to this party. You should study this problem more seriously and then look for people who can assist you on your missing skills. Amir >-- greg
This page took 0.02 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.