New engine releases & news 2021

[Edsel Apostol]

New Invictus version here:

https://github.com/ed-apostol/InvictusC ... s/tag/r382

I hope to see a plus 100 Elo gain but 50 would be great

Gauntlet run for the Gambit Rating List.

http://rebel13.nl/a/grl.htm

Refresh every 30 seconds.

Thanks Ed for including Invictus in your tournaments. Looking good so far.

[Rebel]

New Invictus version here:

https://github.com/ed-apostol/InvictusC ... s/tag/r382

I hope to see a plus 100 Elo gain but 50 would be great

Gauntlet run for the Gambit Rating List.

http://rebel13.nl/a/grl.htm

Refresh every 30 seconds.

Thanks Ed for including Invictus in your tournaments. Looking good so far.

You are most welcome. Unfortunately there was a disconnect, so only 577 games.

Code: Select all

Finished game 573 (Amoeba_3.3 vs Invictus_r382): 1-0 {Black disconnects}

On the other hand...

Code: Select all

Estimated elo gain for Invictus_r382
Elo pool : 3009
Invictus r340 : 2979.0
Invictus_r382 : 3102.9
Difference : 123.9

[!]

[Edsel Apostol]

New Invictus version here:

https://github.com/ed-apostol/InvictusC ... s/tag/r382

I hope to see a plus 100 Elo gain but 50 would be great

Gauntlet run for the Gambit Rating List.

http://rebel13.nl/a/grl.htm

Refresh every 30 seconds.

Thanks Ed for including Invictus in your tournaments. Looking good so far.

You are most welcome. Unfortunately there was a disconnect, so only 577 games.

Code: Select all

Finished game 573 (Amoeba_3.3 vs Invictus_r382): 1-0 {Black disconnects}

On the other hand...

Code: Select all

Estimated elo gain for Invictus_r382
Elo pool : 3009
Invictus r340 : 2979.0
Invictus_r382 : 3102.9
Difference : 123.9

[!]

120+ Elo is great if it holds. Can you send me the Invictus.log in the engine folder?

[Rebel]

http://rebel13.nl/dump/invictus.7z

log + pgn

[Archimedes]

Dear Norm,

Is it possible to improve Fire time management so it not uses too much of its time. I am playing with it right now and most of the time it always gets in Zeitnot. Can you introduce Move Overhead and Slow Mover option from Stockfish in next Fire? I want to be able to tweak the time management...

If you compile Fire for yourself, you can change the integer variable move_overhead_ in chrono.h. With a value of 50 time forfeits almost gone (tested on Android).

[hiblom]

DoctorB release 1.2.0

* tapered evaluation
* better move ordering
* slightly improved SEE score handling
* proper implementation of "stop" and "quit" commands
* various other improvements

expected CCRL Blitz rating of about 1380

[Gabor Szots]

chai 2.5.5 by Jakob Hackstein, Germany
https://github.com/jakhac/chai

Approximate strength: somewhat above 2000 Elos.

[Graham Banks]

Thanks a lot Edsel!

One question: did you manage to fix exception error bug that caused the engine to deliver no move and resign the game ?

All the bugs I'm aware of are fixed by now. Unless something new comes up.

Emailed you the debug file for a time loss.

[Edsel Apostol]

Thanks a lot Edsel!

One question: did you manage to fix exception error bug that caused the engine to deliver no move and resign the game ?

All the bugs I'm aware of are fixed by now. Unless something new comes up.

Emailed you the debug file for a time loss.

Thanks Graham. I'll look into it.

Edit: Looks like I have a stack overflow. I'll fix it.

[j.t.]

Nalwald 1.9 Release.
Biggest change is that evaluation parameters are now optimized using generic gradient descent. Additionally, piece square tables are now dependent on the enemy and own king position (one piece square table for each of the 64 squares the enemy king could be on and one piece square table for each of the 64 squares our own king could be on). I also added (and removed) some other evaluation features.

Self-play improvement for 40 seconds + 0.3 seconds on a fast CPU is around 140 Elo, few tests against two other engines (Raven 1.10, Loki 3.0) indicate that true improvement is closer to 90 Elo.