Search found 6611 matches

by Uri Blass
Sun Nov 10, 2019 7:28 am
Forum: Computer Chess Club: General Topics
Topic: The day when imagination and will was stronger than artificial intelligence. Man against chess machine. November 2019
Replies: 29
Views: 1319

Re: The day when imagination and will was stronger than artificial intelligence. Man against chess machine. November 201

I see bad engine moves. Would a little bit of more time really change them? Yes. Father wouldn't be able to win his games if there was increment, because he's abusing GUI bugs and lag to do what he does, if increment was set to compensate for GUI bugs and lag, he wouldn't be able to do it. And if h...
by Uri Blass
Sun Nov 10, 2019 7:23 am
Forum: Computer Chess Club: Tournaments and Matches
Topic: Dylan Sharp Vs. Harvey Williamson (G4)
Replies: 235
Views: 9089

Re: Dylan Sharp Vs. Harvey Williamson (G4)

You have no way to know if only zullil can see his PVs.
The only thing that you can know is that you cannot see his PVs.

You have no way to know that zullil does not share his PVs with other people assuming the relevant people do not share them with all the world.
by Uri Blass
Thu Nov 07, 2019 6:40 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Limiting nodes to reduce engine strength
Replies: 18
Views: 1401

Re: Limiting nodes to reduce engine strength

Not a fan of a depth limit, since they can occasionally be quite "coarse" in terms of time. I have a pretty decent NPS idea in the works. I have normalized on 2800 Elo == 6M nodes per second (on my MacBook). Elo scales down the NPS and introduces evaluation randomness. And it is actually working pr...
by Uri Blass
Wed Nov 06, 2019 1:14 pm
Forum: Computer Chess Club: General Topics
Topic: italien opening
Replies: 13
Views: 1178

Re: italien opening

I think that even in the last case they can generalize from their experience against engines and play correctly moves that they know to be good in similiar situations. Engines and humans play so completely different chess that even if you get a variation where you perform relatively well against en...
by Uri Blass
Mon Nov 04, 2019 10:10 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Limiting nodes to reduce engine strength
Replies: 18
Views: 1401

Re: Limiting nodes to reduce engine strength

Certainly, it is time for engines to produce more features for human players rather than increases in elo strength. The Deuterium method is to use an exponential function to calculate the rating based on nodes: rating = 297.12 x Ln(nodes) - 976.7 This does not seem correct. Rather, the elo performa...
by Uri Blass
Sun Nov 03, 2019 12:56 am
Forum: Computer Chess Club: General Topics
Topic: italien opening
Replies: 13
Views: 1178

Re: italien opening

I understand now that you do not claim that the italien is better for black but only that black can get an advantage against old books. But new books with those lines score better against themselves from the black side, what is the explanation for that? It could be argued that all my claims are nul...
by Uri Blass
Sat Nov 02, 2019 3:51 pm
Forum: Computer Chess Club: General Topics
Topic: italien opening
Replies: 13
Views: 1178

Re: italien opening

My claims about it has been about unassisted engine chess (blitz from 5 0 to 12 2 time control) and assisted chess at correspondence time controls (+1 year per game), based mainly in positions where Stockfish prefers white but it's black that has the advantage, and the critical positions contain ma...
by Uri Blass
Sat Nov 02, 2019 7:49 am
Forum: Computer Chess Club: General Topics
Topic: italien opening
Replies: 13
Views: 1178

italien opening

Ovyron claimed that the italien opening is good for black. GM larry kaufman claims that it is good for white his sample pages about his new repertoire has the following information: "So how much of an edge can you get as White with my repertoire? Here are the Elo advantages for White after my prefer...
by Uri Blass
Fri Nov 01, 2019 6:07 am
Forum: Computer Chess Club: General Topics
Topic: Tactics cannot be very important for chess
Replies: 90
Views: 3833

Re: Tactics cannot be very important for chess

The fact that contempt makes engines play better in some positions is nothing more than proof that engines are far too passive when the current score is drawn. Adding points to the existing score is a stupid way to solve it. The target of contempt is not to play better but to beat weaker engines. C...
by Uri Blass
Thu Oct 31, 2019 9:51 am
Forum: Computer Chess Club: General Topics
Topic: Tactics cannot be very important for chess
Replies: 90
Views: 3833

Re: Tactics cannot be very important for chess

If one has built a fast-searching engine, one jolly well SHOULD tune it to seek tactical complexity! :P I don't think so. Or, at least, that experiment has been tried. Tuning a chess engine to solve tactical test sets definitely lowers the strength in game play. That's not quite what I meant. I int...