TalkChess.com

I will be there, I informed Bart Goldhoorn too. The coming days I will try to add a few bugs, at least that is what usually happens when I start hacking the engine a few days before the tournament.

Now I sound like Henk!

in my regular search I never promote to bishop because that is clearly a waste of time. Is this really worth it? I always consider all legal moves in search and never even considered doing anything like that. Besides there are other ways to avoid wasting time on bishop underpromotions rather than c...

If you want perft to be a representative measure of the speed of an engine it should indeed focus more on capture-only generation. Perthaps it would be useful to define a function perft(N, M) which does N ply of full-width search followed by M ply of captures only. Most people use perft only for te...

I suppose that I go overboard looking for better performance. And after all is said and done there may not be much gained. And probably I make things harder on myself than need be. And there is something to say for smaller code when it comes to performance. I am curious about one thing though. Why ...

In my old engine (the one I'm still playing with) I also have a switch statement in my domove() with 15 branchless cases (move types), but I find it difficult to maintain. Each time I change or add something I have to modify most of these cases. In a new version of the engine I'm working on I just h...

I guess this is the most telling. Traversing all the squares of a slider in capture only mode on an empty board using two moves and a jump. wbcns: movsx ebx,[bns+esi+ebx] mov edx,[brd+ebx*4] jmp [wbcf+edx*4] It is impossible for any C compiler to beat that! This is typically what humans do when the...

The good old days! The first chess program I wrote during 1977/1978 was in assembler too. I remember the move generator being quite fast 15 kns, but the program overall did 700 ns on the Heatkit H8 (2MHz. Intel 8080 with 16k static memory ). In the beginning I used 2 cassette drives, one to load the...

Nightmare (10 threads) takes 19 ply and 14 seconds to solve this position with ~270 million nodes. It uses aspiration window values: 50, 150, 450, 900, and Infinite. These are rule of thumb values, I never optimized them. info depth 15 score cp 79 time 180 nodes 1900795 nps 10559972 tbhits 0 pv h5h6...

I expected LeelaRB to win the tournament, but somehow it didn't live up to expectations. Indeed it did not do that well. In the absence of Arasan and Texel I thought it could fight for the win, but at the end of the day the £100 GTX 1050 was probably not enough. A £700 - £800 RTX 2080 would have be...

Henk, this probably has to do with the Mc.Mahon system. Nightmare was very lucky this time, two of its opponents were so friendly to let their clock run out of time in a drawn endgame, so it should probably have had 1 point less. Both Texel and Arasan and were missing and that also increases ones wi...