Author: Robert Hyatt
Date: 15:09:03 12/21/99
Go up one level in this thread
On December 21, 1999 at 16:49:18, Greg Lindahl wrote: >On December 21, 1999 at 09:35:04, Tom Kerrigan wrote: > >>Greg, you keep repeating that FPGAs have memory and that this solves the memory >>problem. > >I'm just brainstorming. You seem to think that when I put out an idea, that I am >claiming that I know the answer, or that I have a degree in FPGA design. I >don't. > >I pointed out that there are 2 obvious objections to the "it can't be done >without memory" claim. The first is that FPGA boards come with SRAM and that >works fine with algorithms such as FFTs, and the second is that an eval-only >chip doesn't need the huge memory that a full-up chess chip uses. > And there is another bad assumption. I use one 64kb array _just_ to figure out whether one side has an outside passed pawn or not. one 64kb array for detecting a left-most distant passer and another 64kb for a rightmost distant passer. and two more for potential distant passers via majorities. That is 256kb before we even get started. I would probably need 1-2 megs of memory to store the patterns I used in my eval. that was _megs_ not _kilobytes_... >Since I don't have any design in mind at all, just a couple of SWAGs (which >stands for "stupid wild-ass guess"), I could care less if memory is or isn't a >killer problem. I'd just like to think about a fast chess engine. > >>This shows that you have not done one single little bit of logic design in your >>entire life. And if you did, you probably got an F in it. > >Can you please stop with this kind of language? I'm getting tired of your abuse, >name-calling, and straw-man attacks. > >>I think it's sad that you're clinging to this silly fact to save your entire >>project here. > >And here I thought I was having fun brainstorming. Dang. > >-- g
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.