TalkChess.com

Is it possible to make the games available?

The link at http://www.ssdf.bosjo.net/ssdf.htm#games does not work.

I don't get that. 'Very bad' here means below alpha minus a rather small margin? Because that is what it automatically boils down to: if only captures stand a chance to raise the score above alpha, you will only search captures, and prune all non-captures. No, there is an eval test, and the margin ...

chrisw wrote: ↑

Thu Jul 09, 2020 7:39 am

There’s a great deal of power in “averaging”.

See the Central Limit Theorem for some mathematical justification of this.

Last is when I review other code from other engines, I see razoring and futility pruning performed when entering a node to search (before generating and looping moves) and I will also see futility pruning pre-move-make (within the move loop). Razoring as the term is commonly used is a pre-search pr...

fjpv wrote: ↑

Wed Jul 01, 2020 10:02 am

Trying to download the Arasan 22.1 Windows installer and I receive a message 404 Error - Page Not Found.

Please try again - should be fixed now.

--Jon

Arasan 22.1 is available from https://www.arasanchess.org . Changes in Arasan 22.1: 1) Remove potential passer scoring. 2) Tighten validation of hash/killer moves 3) Fix Resign threshold option (CECP mode) 4) Simplify lmp threshold computation. No longer different for PV. 5) Modify formula for lmp c...

But the 6 dtz isn't read by the computer.

Not clear what program/interface you are using?

Several TCEC games have been featured on https://www.chesspublishing.com recently. Formerly they used to pretty much ignore computer games, but now the GMs writing their columns are paying attention.

--Jon

Fixed-depth testing is not really an established method for eval tuning.

The standard way to tune eval parameters is the "Texel" tuning method, more properly termed logistic regression. Lots of discussion of that in this forum, if you search.

--Jon

Obviously my experience is absurdly out of date, but I am curious what things are like now: not just whether one can get high-performance Java (one always could if one wrote everything in terms of primitives and arrays) but whether one can write at a high level of abstract and still get good perfor...