TalkChess.com

I tested one download link and it works and leads to this (Nero 5.3 src): https://web.archive.org/web/20060624001712/http://www.mit.jyu.fi:80/~huikari/nerowb53.pas Thank you Günther. I didn't know that one could use Wayback like that. Regards. Roland Wow, funny source code! no negamax, but in evalu...

hgm wrote: ↑

Sat Oct 24, 2020 12:56 pm

So the images are still wrong: they state there are 41K inputs, while in fact there are 2 x 41K inputs.

Sorry, had a wonderful bike tour today

Becomes clearer now, did not study the transformer yet. Found FeatureTransformer::Transform and UpdateAccumulator. In particular void Transform(const Position& pos, OutputType* output) const { UpdateAccumulator(pos, WHITE); UpdateAccumulator(pos, BLACK); ... Found orient(perspective, s) and kpp_boar...

After some thinking, I believe the weights are not shared between two halves and the initial NNUE graph with its weight enumeration was more correct and I was too hasty to change it. The upper black half has black king placement times 5 black piece types on 64 squares, and the white half, white kin...

After some thinking, I believe the weights are not shared between two halves and the initial NNUE graph with its weight enumeration was more correct and I was too hasty to change it. The upper black half has black king placement times 5 black piece types on 64 squares, and the white half, white king...

What am I missing? I don't see any real difference between the images (except that the latter is missing the ReLu symbol in the deeper cells). I don't see any reason why the KPST for both players should be the same (i.e. why each weight should be used twice). The situation is not symmetric, as one ...

AndrewGrant wrote: ↑

Fri Oct 23, 2020 11:33 am

The confusion might be that this image on CPW is wrong

Thanks for pointing that out, so the labels of the weights were corrected, which are the same for both halfs.

I made an experimental version on the engine with this modification: if FActive = AColor then begin if LValue > LBeta then LBeta := LValue; if (LBeta > AAlpha) or (LBeta = AAlpha) and (ADepth > 2) then LStop := TRUE; end else begin if LValue < LBeta then LBeta := LValue; if (LBeta < AAlpha) or (LBe...

Fixed depth of 3 without iterative deepening and max depth 5 for captures should be quite fast. And at that depth the deficits missing deep cutoffs is not that important. I modified my engine to overwrite alpha with -MAX in search and QS. Main depth (without QS) 4 is about even, 5 takes 50% longer,...

Further the cut-off conditions seem too weak to me >= instead of >, or <= instead of < https://github.com/rchastain/moustique/blob/master/chessplayercore.pas#L417 If I understand correctly, you would suggest to try the following modification? if FActive = AColor then begin if LValue > LBeta then LB...