Search found 2668 matches

by mcostalba
Thu Aug 24, 2017 2:29 pm
Forum: Computer Chess Club: General Topics
Topic: Natural TB
Replies: 330
Views: 78173

Re: Natural TB (take 2)

You treat them differently: TB loss returns immediately, TB win is ignored. <cut> That does not seem to make a lot of sense. Yes, I treat them differently, and not only when entering in TB, but always :-) The reason for this is to avoid odd sacrifices and TB scores to be exposed to user. If I don't...
by mcostalba
Thu Aug 24, 2017 2:07 pm
Forum: Computer Chess Club: General Topics
Topic: Natural TB
Replies: 330
Views: 78173

Re: Natural TB (take 2)

But Michel is right. The "clean" solution to the problem has been pointed out many times in this thread. The simple idea of the solution is that you don't take the TB cutoff if the TB win or loss value is inside the (alpha, beta) window. Yes, to use TB scores in analogy with upper/lower bounds of T...
by mcostalba
Thu Aug 24, 2017 12:31 pm
Forum: Computer Chess Club: General Topics
Topic: Natural TB
Replies: 330
Views: 78173

Re: Natural TB (take 2)

Can you please better explain your point? I was clearly not talking about TB draws, so I suggest you carefully re-read what I wrote. Ok, in this position after the winning side capture Rxa5 black is losing and TB return loss. [D]4k3/8/8/p7/R7/8/8/4K3 w - - 0 1 Instead in this position [D] 4k3/8/1p6...
by mcostalba
Thu Aug 24, 2017 11:53 am
Forum: Computer Chess Club: General Topics
Topic: Natural TB
Replies: 330
Views: 78173

Re: Natural TB (take 2)

I'm not sure it is helpful to treat TB wins and TB losses differently in the middle of the tree (unless you are talking about win/loss relative to the root position). It should not matter whether a TB position is entered by a capture by the winning side (in which case the TB probe returns a TB loss...
by mcostalba
Tue Aug 22, 2017 5:30 pm
Forum: Computer Chess Club: General Topics
Topic: Natural TB
Replies: 330
Views: 78173

Re: Natural TB (take 2)

I will properly address your post later. I'd like just to quickly present my point of view. TB in normal games are useful for 2 reasons: to show the engine the winning path and to prune useless subtrees. My view is that the latter is what gives the biggest advantage, the first one is not so importan...
by mcostalba
Tue Aug 22, 2017 11:56 am
Forum: Computer Chess Club: General Topics
Topic: Natural TB
Replies: 330
Views: 78173

Re: Natural TB (take 2)

That doesn't seem different from hash cutoffs truncating the PV. This is normally solved by not not doing cutoffs in PV nodes. There are 2 critical differences. 1. A TT entry is not an insurmountable wall, it is depth dependent, so at next iteration, with increased depth, I can overcome any TT prob...
by mcostalba
Tue Aug 22, 2017 10:38 am
Forum: Computer Chess Club: General Topics
Topic: Natural TB
Replies: 330
Views: 78173

Re: Natural TB (take 2)

As has been pointed out many times by many people already this particular problem can be fully solved within the context A/B search by regarding TB scores win/loss scores as bounds (like TT upperbound and lowerbound scores). The point is if you return from a TB probe immediately (no matter with wha...
by mcostalba
Tue Aug 22, 2017 7:48 am
Forum: Computer Chess Club: General Topics
Topic: Natural TB
Replies: 330
Views: 78173

Natural TB (take 2)

After one year I return at my attempt with Tablebases in Stockfish. My taget is still the same, to make TB to look more natural, without strange scores, hiding mates and unnatural sacrifices...but keeping same ELO level of current implementation. This year I am trying a much more conservative approa...
by mcostalba
Sat Jun 03, 2017 1:28 pm
Forum: Computer Chess Club: General Topics
Topic: Scientific American article on Computer Chess
Replies: 52
Views: 12969

Re: Scientific American article on Computer Chess

I sort of agree. Is there any computer application which you'd say is really AI? Today in 2017 there is broad consensus that deep learning mimics what can be defined as AI definitely more than a traditional software (chess engine or other). The main difference is that in a deep neural network there...
by mcostalba
Sat Jun 03, 2017 5:15 am
Forum: Computer Chess Club: General Topics
Topic: Scientific American article on Computer Chess
Replies: 52
Views: 12969

Re: Scientific American article on Computer Chess

The term AI associated to a chess engine is totally misleading. There is no AI in a traditional chess engine more than can be in a CAD or in a word processor. This article is a blast from the 90's, but we are in 2017 now. This is just a gross pitch to IBM Watson that is just basically a text search ...