Search found 10 matches

by kaseldop
Thu Jun 27, 2019 10:52 am
Forum: Main Lobby / General Discussions
Topic: We don’t need many NPS for chess
Replies: 35
Views: 6156

Note also that lc0 has played more than 220 million games to reach this strength. Stockfish, in a certain sense, plays from a priori principles. So while stockfish crunches numbers in the game itself, lc0 has crunched them previously to learn the game. All the chess programs play millions of games ...
by kaseldop
Mon Feb 18, 2019 11:42 am
Forum: Main Lobby / General Discussions
Topic: La Regence
Replies: 17
Views: 3055

Hi, I think the company went bust pretty soon after release, so no more models.
Soon after someone connected with that company produced a lovely looking wooden machine called the L'Empereur.
Richard Lang was program writer, would like one of them in my collection.
by kaseldop
Wed Jun 06, 2018 8:59 pm
Forum: Main Lobby / General Discussions
Topic: different super connie versions?
Replies: 8
Views: 2223

The Constellation did have a official usa rating of 2018 from 1984.
2 or 3 models played in the US Open
Even though it is way to high it was the early days of humans facing strong programs and that makes a big difference, how much experience in playing computers.
by kaseldop
Thu Jan 11, 2018 11:14 am
Forum: Main Lobby / General Discussions
Topic: 2018 Spacious-Mind Swiss Chess Computer League - Division 2
Replies: 112
Views: 11034

Following with great interest. Well done.
by kaseldop
Thu Jan 11, 2018 11:12 am
Forum: Main Lobby / General Discussions
Topic: How do yiu determine ELO by seconds?
Replies: 7
Views: 1157

It was often supposed that a doubling of speed gave program a 80elo increase. But I seem to remember Selective Search magazine brought this down to 50-60 as programs got stronger. Also the increases applied to computer verses computer, slightly less against humans. There was a computer that was ente...
by kaseldop
Fri Jan 05, 2018 10:31 am
Forum: Main Lobby / General Discussions
Topic: 2018 Spacious-Mind Swiss Chess Computer League - Division 2
Replies: 112
Views: 11034

Hi, these Tournaments are brilliant to follow. More interesting than one held between the stockfish's, Houdini 's etc when can't understand most of it and too many draws.
Your work is much appreciated.
MARK
by kaseldop
Fri Jan 05, 2018 10:23 am
Forum: Tournament Results & Games
Topic: Millenium Chess Genius Pro-Exclusive Genius
Replies: 25
Views: 8798

Thanks for carrying out these match's. You will usually have a more one sided score when the same programs play a match when the only difference is speed of processor and RAM size, compared with different programs which generally get close to rating expectation. The reason is that the slower Genius ...
by kaseldop
Fri Jan 05, 2018 10:06 am
Forum: Tournament Results & Games
Topic: CGPro v Tasc R30v2.5
Replies: 34
Views: 9576

Always enjoy following match's like this. Well done and thanks

MARK (KASELDOP)
by kaseldop
Fri Mar 24, 2017 5:21 pm
Forum: Main Lobby / General Discussions
Topic: Hiarcs Forum 10th Birthday Party Competition!
Replies: 40
Views: 8989

10 Years Competition

Hi
My predictions are :-
1 Hiarcs
2 The Baron
3 Nemeton
4 Arminius
5 The King
6 Sparticus
7 Embla

Last round moves by Hiarcs = 53

Not confident but let's make a start.
by kaseldop
Fri Dec 18, 2015 11:59 pm
Forum: Main Lobby / General Discussions
Topic: Millenium new release of chess computer
Replies: 676
Views: 171930

Hash Tables

Regarding the earlier discussion whether larger Hash Tables ALWAYS produces better results. I remember in the 'Selective 'Search' magazine, (which I still have from 1992 until when it finished 2012?) where program owners who had the very latest PC's approx 1998, were struggling to find the right Has...