Search found 2086 matches

by Michel
Fri Jan 03, 2020 9:08 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: RMO - Randomized Move Order - yet another Lazy SMP derivate
Replies: 114
Views: 24358

Re: RMO - Randomized Move Order - yet another Lazy SMP derivate

Perhaps now is a good time for me to plug my test methodology? I have invented a "game" for my personal study. It is the simplest possible game I could imagine, so that I can focus on the pure task of finding a better search. 1. A move is represented by a 5-bit number, between 0 and 31. 2. Players ...
by Michel
Fri Jan 03, 2020 8:50 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: RMO - Randomized Move Order - yet another Lazy SMP derivate
Replies: 114
Views: 24358

Re: RMO - Randomized Move Order - yet another Lazy SMP derivate

Sigh... Everyone claims "it is a mystery that Lazy works so well". I do not see the mystery. Occam's razor says that we should first look for the simplest possible explanation. And the simplest possible explanation is that Lazy has almost zero overhead. The overhead is so small, that (combined with ...
by Michel
Fri Jan 03, 2020 3:12 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: RMO - Randomized Move Order - yet another Lazy SMP derivate
Replies: 114
Views: 24358

Re: RMO - Randomized Move Order - yet another Lazy SMP derivate

_If_ Lazy would be really worse than YBW in TTD (I appreciate Lucas' high quality data, but there should also be data on the YBW side for comparison) then its strength must come from widening the search. However it is not clear to me what mechanism in Lazy would be responsible for widening the searc...
by Michel
Thu Jan 02, 2020 6:17 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: RMO - Randomized Move Order - yet another Lazy SMP derivate
Replies: 114
Views: 24358

Re: RMO - Randomized Move Order - yet another Lazy SMP derivate

I found this thread http://www.talkchess.com/forum3/viewtopic.php?f=2&t=61131 where it is claimed that Lazy is worse in TTD. But it is not clear to me if the analysis is valid since some of the other threads may be searching deeper than the main thread and this would skew the reported results - I t...
by Michel
Wed Jan 01, 2020 10:52 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: RMO - Randomized Move Order - yet another Lazy SMP derivate
Replies: 114
Views: 24358

Re: RMO - Randomized Move Order - yet another Lazy SMP derivate

I found this thread http://www.talkchess.com/forum3/viewtopic.php?f=2&t=61131 where it is claimed that Lazy is worse in TTD. But it is not clear to me if the analysis is valid since some of the other threads may be searching deeper than the main thread and this would skew the reported results - I th...
by Michel
Wed Jan 01, 2020 9:54 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: RMO - Randomized Move Order - yet another Lazy SMP derivate
Replies: 114
Views: 24358

Re: RMO - Randomized Move Order - yet another Lazy SMP derivate

I agree with that thought, in that I would not just blindly commit something because its an elo gainer. But its not hard to construct a reasonable argument for why LazySMP might out perform other methods. You can argue that YBW is just a cheap way of making a minor effective speedup, and that LazyS...
by Michel
Tue Dec 31, 2019 7:09 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: RMO - Randomized Move Order - yet another Lazy SMP derivate
Replies: 114
Views: 24358

Re: RMO - Randomized Move Order - yet another Lazy SMP derivate

If this gets better game results despite worse time to depth, then it must be searching moves that were incorrectly pruned normally. And it seems that this incorrect pruning has been happening often enough to have a measurable impact. So it seems like it would be worth researching what moves get se...
by Michel
Tue Dec 31, 2019 2:23 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: RMO - Randomized Move Order - yet another Lazy SMP derivate
Replies: 114
Views: 24358

Re: RMO - Randomized Move Order - yet another Lazy SMP derivate

This basically means that your whole world of pencil and paper is an illusion. Perhaps a big part of reason is the chaotic nature of modern search (with all the path dependt effects like history ordering and its effect on lmr etc.). That and the chaotic nature of thread scheduling, becoming chaos^2...
by Michel
Tue Dec 31, 2019 9:26 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: RMO - Randomized Move Order - yet another Lazy SMP derivate
Replies: 114
Views: 24358

Re: RMO - Randomized Move Order - yet another Lazy SMP derivate

I know, chaos sounds a like a bullshit smoke screen for what we don't understand. But sometimes it's the right way to think about a problem. Look at meteorology. They don't try to prove whether it will rain tomorrow by using pencil and paper anymore. They run huge simulations and just accept the Or...
by Michel
Wed Dec 18, 2019 4:13 pm
Forum: Computer Chess Club: General Topics
Topic: If there were one engine I could bring back and covert
Replies: 40
Views: 6448

Re: If there were one engine I could bring back and covert

Psion Chess by Richard Lang (https://www.retrogames.cz/play_430-DOS.php) I played a lot of games against it as a boy, and over the course of one summer holiday proceeded from losing at level one to winning at next-to-last level. I remember the engine as very stable, level-headed opponent. Also, it ...