Search found 40 matches

by Laszlo Gaspar
Sun Jun 19, 2016 11:03 am
Forum: Computer Chess Club: General Topics
Topic: Questions about Timea and Darmenios (probably @CCRL+CEGT)
Replies: 4
Views: 1446

Re: Questions about Timea and Darmenios (probably @CCRL+CEGT

Dear Guenther and Norbert, I am the author of Timea. I participated in the past with Timea in different internet tournaments as a private engine but as it was not on the level which I considered strong enough to make it public I didn't do it. On the other hand as it is leaked now I don't have any ba...
by Laszlo Gaspar
Tue Dec 03, 2013 8:07 pm
Forum: Computer Chess Club: General Topics
Topic: Komodo-TCEC
Replies: 37
Views: 15178

Re: Komodo-TCEC

What about simply Komodo Champion? (referring indirectly to Don who finally reached the top and became champion with his creature)
by Laszlo Gaspar
Tue Dec 03, 2013 6:38 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: PSQ tables depending on king sides, pawn patterns etc.
Replies: 15
Views: 4764

Re: PSQ tables depending on king sides, pawn patterns etc.

Hi Sergei, I did exactly the same! I constructed 6*6=36 different PSQT and other tables, each one is a case of the kings' relative position (queenside, center and can castle kingside, center and can castle queenside, center and can castle both sides, center and cannot castle, kingside - and the comb...
by Laszlo Gaspar
Sat Nov 23, 2013 9:49 am
Forum: Computer Chess Club: General Topics
Topic: Don Dailey, 1956 -2013
Replies: 150
Views: 86547

Re: Don Dailey, 1956 -2013

I'm very sad. One minute silence...
by Laszlo Gaspar
Sun Nov 17, 2013 6:51 pm
Forum: Computer Chess Club: General Topics
Topic: Don Dailey
Replies: 116
Views: 45328

Re: Don Dailey

I' m shocked! Don, thank you all the work you gave us and your posts were always pleasure to read! :cry:
by Laszlo Gaspar
Sat Oct 05, 2013 3:04 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: How much to reduce ?
Replies: 10
Views: 2643

Re: How much to reduce ?

Hi, I use a table for nullmove reduction (nulldepth = table[depth]) so I can experiment on each depth separatly. My table is like this: [-1, -1, -1, 0, 0, 1, ...] (-1 means calling QS). I have gain from the nullmove test even at depth==0 where R==0 and call QS. At depth 3 R==2 and at depth 4 R==3 th...
by Laszlo Gaspar
Sat Jul 06, 2013 5:40 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Measure of SMP scalability
Replies: 86
Views: 12708

Re: Measure of SMP scalability

I can accept that it should work that way and it might explain why my SMP is not that good as the SMP in Crafty. But the question is what algorytm is implemented in Komodo and this was my idea. It looks like it is not DTS and I agree with Edsel in his opinion.
by Laszlo Gaspar
Sat Jul 06, 2013 10:04 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Measure of SMP scalability
Replies: 86
Views: 12708

Re: Measure of SMP scalability

I meant different 'iteration' depths and synchronization between iterations...
by Laszlo Gaspar
Fri Jul 05, 2013 8:58 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Measure of SMP scalability
Replies: 86
Views: 12708

Re: Measure of SMP scalability

In my first tries in SMP programming I didn't synchronize correctly the threads, the search simply fell apart. The different threads were working on different depths. I feel something similar here, it seems to me that in this case the search doesn't stop when the FIRST thread finishes the last itera...
by Laszlo Gaspar
Sat Jun 01, 2013 6:58 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Possible history table improvement
Replies: 14
Views: 4934

Re: Possible history table improvement

Thank you, Harald, for the warning! I wrote in fact a Java code in which it is correct but it was intended just a pseudo code anyway.