Search found 333 matches

by noobpwnftw
Sun Jun 30, 2019 7:21 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Neural Net Endgame technique
Replies: 7
Views: 2296

Re: Neural Net Endgame technique

There is nothing wrong with evaluating a mate in 1 and mate in 10 equally in terms of winning probabilities. Cosmetic "fixes" (as long as averaging and heuristics involved) without the ability to precisely discover the mating sequence is just as lame as it currently is. In other words, if the eval f...
by noobpwnftw
Sun Jun 30, 2019 5:22 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Linux flavors - opinions
Replies: 29
Views: 4364

Re: Linux flavors - opinions

I've always preferred RHEL/CentOS due to the reason that it has a pretty closed package chain compared to the others. You know what to expect, including many problems due to obsolete packages of course, but few times it gives you surprises. My experience with Ubuntu is things often times works on mi...
by noobpwnftw
Tue Jun 11, 2019 9:10 pm
Forum: Computer Chess Club: General Topics
Topic: OKE - Opening Knowledge Engines
Replies: 94
Views: 7999

Re: OKE - Opening Knowledge Engines

The depth value at the root will be approximately 22 + (distance to leaf). Indeed, but there is still a horizon problem, although the chances of getting something wrong diminishes when depth goes higher. The search breadth should be considerably wider than vanilla search from root position due to M...
by noobpwnftw
Tue Jun 11, 2019 9:00 pm
Forum: Computer Chess Club: General Topics
Topic: OKE - Opening Knowledge Engines
Replies: 94
Views: 7999

Re: OKE - Opening Knowledge Engines

The depth value at the root will be approximately 22 + (distance to leaf). Indeed, but there is still a horizon problem, although the chances of getting something wrong diminishes when depth goes higher. The search breadth should be considerably wider than vanilla search from root position due to M...
by noobpwnftw
Tue Jun 11, 2019 8:39 pm
Forum: Computer Chess Club: General Topics
Topic: OKE - Opening Knowledge Engines
Replies: 94
Views: 7999

Re: OKE - Opening Knowledge Engines

I have built a database using a slightly modified SF to adjust MultiPV depending on shallow search results. It now has about 2 billion unique positions, each analyzed with depth >=22 and MultiPV >=5, my experience is that eventually which engine or parameters you use becomes irrelevant - it converge...
by noobpwnftw
Thu Apr 11, 2019 6:39 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: 7-men Syzygy attempt
Replies: 588
Views: 125056

Re: 7-men Syzygy attempt

I am not quite clear what the purpose of this is.
I believe that is used to filter known worse moves at root given WDL information.
by noobpwnftw
Sun Feb 24, 2019 4:02 pm
Forum: Computer Chess Club: General Topics
Topic: Further development of Lc0: Lc1
Replies: 19
Views: 3207

Re: Further development of Lc0: Lc1

I believe there is a misconception that being totally "zero" would introduce no bias at all, for one, statistically sound priors may not be the true distribution of their outcome under perfect play, those will in turn get magnified in RL process, in that sense the final strength is limited one way o...
by noobpwnftw
Thu Jan 17, 2019 7:38 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Generate EGTB with graphics cards?
Replies: 29
Views: 4995

Re: Generate EGTB with graphics cards?

There! I meant Xiangqi tablebases, I have never wrote a generator for chess.
by noobpwnftw
Tue Jan 15, 2019 12:06 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Generate EGTB with graphics cards?
Replies: 29
Views: 4995

Re: Generate EGTB with graphics cards?

They are time taken for each separate step, DTM is a bit faster because it reuses WDL tables and DTC usually requires more iterations. Note that almost a half of the time are spent on compression because they will quickly become huge without it. You need also consider the time needed to decompress s...
by noobpwnftw
Tue Jan 15, 2019 1:21 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Generate EGTB with graphics cards?
Replies: 29
Views: 4995

Re: Generate EGTB with graphics cards?

KBNK on single thread:

Build DTC:
45ms
Build DTM:
30ms
WDL with compression:
22ms W:721B B:751B
DTC with compression:
16ms W:2748B B:3060B
DTM with compression:
24ms W:2746B B:3060B