Minic version 3

[Frank Quisinsky]

Hi Vivien,

sorry that I used your time in reading my bad English.

I know, normally I should not changed any default settings by the programmers for eng-eng tourneys.
Programmers know here baby better ...

But if I looking in all the long games in detail (without resign-mode more clear) ... I saw very rarely a game where contempt in endgame helps an engine for a better result. In neural-network times I think contempt for endgames isn't necessary if I study the game results. For some years, before NN, the situation has turned completly. Contempt for endgames helps a lot vs. weaker engines for a better game results and it can give +15 Elo for stronger engines. Tested out in selfwork for some years with Stockfish.

Contempt for endgames for stronger engines are maybe 1-3 Elo more today ... and the move average or calculating / computing time is no longer in any relation for 1-3 Elo more.

Contempt for mid-games is fine for me!
If more as around 18 pieces on board ... contempt = on is right ... if the engine is super-strong.

And each intelligent idea to find out the playing strength of the opponent during the game is still running with a mainipulated contempt is in my opinion very nice.

Best
Frank

For engines around 200 Elo or more weaker as Stockfish it can be interesting to forced draw, also in mid-games if over a longer time (example first 20 own calculated moves) the own eval goes not higher. To forced a draw give in most of cases more Elo as to avoid a draw. Easy, the super-strong engines can't exploit the endgame power. Very important for engines weeker in endgames or engines with a more speculative and aggressive playing-style. If I am looking in detail in game stats from our engines ... different programmers do that or somewhat like this. Explain the lesser move-average ... more clearly to see with resign=off games. Loud thinking ...

[matejst]

Thanks, Vivien!

[xr_a_y]

Frank,

if you plan to run engine versus engine games for comparison of engines, I would indeed suggest to keep most of the default parameters given by engine authors.

But on the other hand, parameters are also here to experiment, tune and have fun with, so feel free to try stuff with contempt.

A fun thing to do with Minic is to use UCI_Opponent (name of the opponent) or UCI_RatingAdv (Elo difference Minic-opponent) so that Minic adapt itself to try to win hard or avoid loosing.

I'm not sure I'm able to give an advice leading to smaller game length between even engines sorry.

[Frank Quisinsky]

Vivien,

You wrote:
A fun thing to do with Minic is to use UCI_Opponent (name of the opponent) or UCI_RatingAdv (Elo difference Minic-opponent) so that Minic adapt itself to try to win hard or avoid loosing.



Thinking the same ... after your release.
That's my plan for tomorrow after work.
To late for today.

And to your engine:
All is fine with Minic, stats are OK with move-average. Minic is a very hard defender if other engines like to try it with storming pawns. Minic - Wasp is one of my favorite eng-eng constellation. A long time I saw not a fast lost game. Often Minic are able to find a way in light poorer position to draw. A bit Karpov style you created, often I am thinking if Minic try to search better fields with a full board for bishops and knights in the beginning of the game. That is good, all the available engines must not produced the same style of chess. So I like Minic a lot. I am observe Minic results all the time, since the beginning you know. It have to do with all the good and very nice contacts I had in Winboard times with French programmers. May sound is funny!



Thank you and good luck in my tourney.
Around Nov. 03rd the new Minic will go on start ... after round 10 from my still running tourney.

Best
Frank

[Frank Quisinsky]

Hi Vivien,

what you are doing is working well, I am trying it out on one of my PCs.
Congratulations ... a nice-to-have feature!

By the way:
After a bad start with Minic 3.39 NN in my tournament the results are getting better and better ... 84 played 90 minutes games only.
Move average = 88 (improved) ... contempt works well.

Everything is wonderful with the new Minic!

With longer time controls there are other laws on the chessboard.
30 Elo more is a lot in this case ... but possible!

I am looking at the weekend in the games from the 5 updates that are still running.
Minic is indeed a hard fighter and often found the way to draw in significantly worse positions. It seems he comes from very intelligent exchanges of pieces when the endgame is near. Not sure ... just impressions!

Best wishes
Frank

[Rubinus]

Hi all,

long time no see ! I was quite busy this year professionally and personnally, so I didn't invest much time in chess programming.

Anyway a new Minic version is finally available (Minic 3.39 https://github.com/tryingsomestuff/Mini ... s/tag/3.39). This release includes a new net named "nyctophobic narwhal", some good search patches and some minor bug fixes. Overall it shall play near 30Elo better than last release in self test at short TC.

Binaries shall be found here : https://github.com/tryingsomestuff/Mini ... inic3/3.39

Have fun !

Ugh. Somehow I can't figure out which version belongs on my computer. This one I'm typing on has AMD Ryzen, 6 cores. I use avx2 for most engines, avx512 doesn't run me. But I'm getting lost in the names here.I'm running zver3, but which is optimal for me, I don't know. For version 3.17 I had sandybridge downloaded and it worked. Version 3.39 sandybridge doesn't work for me.

[Ciekce]

This one I'm typing on has AMD Ryzen, 6 cores.

Which CPU specifically? Ryzen is a brand

[xr_a_y]

Hi all,

long time no see ! I was quite busy this year professionally and personnally, so I didn't invest much time in chess programming.

Anyway a new Minic version is finally available (Minic 3.39 https://github.com/tryingsomestuff/Mini ... s/tag/3.39). This release includes a new net named "nyctophobic narwhal", some good search patches and some minor bug fixes. Overall it shall play near 30Elo better than last release in self test at short TC.

Binaries shall be found here : https://github.com/tryingsomestuff/Mini ... inic3/3.39

Have fun !

Ugh. Somehow I can't figure out which version belongs on my computer. This one I'm typing on has AMD Ryzen, 6 cores. I use avx2 for most engines, avx512 doesn't run me. But I'm getting lost in the names here.I'm running zver3, but which is optimal for me, I don't know. For version 3.17 I had sandybridge downloaded and it worked. Version 3.39 sandybridge doesn't work for me.

Ok, if you are using avx2 for other engine, then probably the skylake or one of the zen onez shall work. It's a bit worring though that the sandybridge one (AVX) is not working ...

I'm suspecting there is another issue here. Have you checked the size of the downloaded file ?

[Rubinus]

Yes, that's where the problem will be. It downloaded 130 KB, while znver3 version has 146MB. After downloading again, I have the sandybridge version also 146MB and it works. Thanks. My configuration of this computer is AMD Ryzen 5 4500U with Radeon Graphics 2.38 GHz.
I have other machines with Intel i7 and i9, so I will install this elsewhere later.

[xr_a_y]

Yes, that's where the problem will be. It downloaded 130 KB, while znver3 version has 146MB. After downloading again, I have the sandybridge version also 146MB and it works. Thanks. My configuration of this computer is AMD Ryzen 5 4500U with Radeon Graphics 2.38 GHz.
I have other machines with Intel i7 and i9, so I will install this elsewhere later.

If your hardware is AVX2, you definitly shall take advantage of this. The AVX2 build is +40Elo versus the AVX one.

Your file is only 130k probably because you downloaded Minic-Dist as a submodule without activating git-lsf.

I'd suggest you simply download executable by hand on this page instead https://github.com/tryingsomestuff/Mini ... inic3/3.39