Author: Antonio Dieguez
Date: 07:14:45 11/17/99
Go up one level in this thread
On November 16, 1999 at 14:27:37, Eugene Nalimov wrote:
>Unroll the outer loop, so that iDelta[iDir] become a compile-time constant in
>the inner loop. Also, you now have 2 calls to AddMove - replace it by one, write
>something like that:
>
> for (iTarg = x+17; ONBOARD(iTarg); iTarg += 17) {
> iTargContents = PIECE_AT(psPos, iTarg);
> if (IS_YOUR_COLOR(iTargContents))
> break;
> AddMove(psPos, x, iTarg);
> if (!IS_EMPTY(iTargContents))
> break;
> }
for (iTarg = x+17; ONBOARD(iTarg) &&
!IS_YOUR_COLOR(iTargContents=PIECE_AT(psPos, iTarg)); iTarg += 17) {
AddMove(psPos, x, iTarg);
if (!IS_EMPTY(iTargContents))
break;
}
just a bir shorter.
bye.
>I suspect that ONBOARD(), PIECE_AT(), IS_EMPTY(), etc. are macro. Right? Also,
>try to inline AddMove().
>
>In any case the move generation performance is not very important - but you know
>that yourself :-).
>
>Eugene
>
>On November 16, 1999 at 00:26:17, Scott Gasch wrote:
>
>>Hi,
>>
>>I have been rewriting my engine from the ground up and trying to "do it right"
>>this second go 'round. For instance, I pruned the move data struct from a huge
>>56 byte struct down to an int to increase performance ;)
>>
>>I have recently been doing Vincent Diepeveen's test of move generation speed to
>>clock how fast my generator is. In the prior version it generated at about 3
>>million nodes/sec whereas in this version (with the much smaller move
>>representation) it is up to about 6 million nodes/sec on my AMD K6-3 450.
>>
>>Vincent's test is to generate all successors of:
>>rnbqkbnr/ppp2ppp/8/3pp3/3PP3/8/PPP2PPP/RNBQKBNR b KQkq e6 0 0
>>
>>in a loop, 2000000 times and keep track of how long it takes... then compute
>>nodes/sec. Vincent gets 15 million in the same position with a slower
>>processor!
>>
>>I have profiled my code and determined that the bottleneck is in my generation
>>routines (as opposed to AddMove) So my question is this: I currently generate
>>moves for a specific piece something like this:
>>
>> //
>> // Figure out how the rook can move.
>> //
>> for (iDir = 0; iDir < 4; iDir++)
>> {
>> iTarg = x + iDelta[iDir];
>> while (ONBOARD(iTarg))
>> {
>> iTargContents = PIECE_AT(psPos, iTarg);
>>
>> //
>> // The rook can move if the square is empty.
>> //
>> if (IS_EMPTY(iTargContents))
>> {
>> AddMove(psPos, x, iTarg);
>> }
>>
>> //
>> // Or if there is an enemy piece there to capture.
>> //
>> else
>> {
>> if (COLOR(iTargContents) != COLOR(iRook))
>> {
>> AddMove(psPos, x, iTarg);
>> }
>> break;
>> }
>> iTarg += iDelta[iDir];
>> }
>> }
wow! thats looks exactly extracted of my program.
>>However I see in GNUChess they use some kind of precomputed table. I am not
>>sure I understand exactly how this works. Also, does anyone else have any other
>>tricks to speed up move generation? I know that in the long run this really
>>doesn't make TOO much of a difference but I want my new version to be lean and
>>mean...
>>
>>As always, I appreciate the help.
>>Scott
This page took 0.01 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.