Any news of a Komodo update in sight?

Discussion of anything and everything relating to chess playing software and machines.

Moderators: hgm, Dann Corbit, Harvey Williamson

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
gordonr
Posts: 138
Joined: Thu Aug 06, 2009 6:04 pm
Location: UK

Re: Any news of a Komodo update in sight?

Post by gordonr » Tue Dec 12, 2017 1:36 pm

shrapnel wrote:rendering Alpha Beta type Chess Engines obsolete
Consider how the *vast* majority of chess players currently use a chess engine and tell me why that is now obsolete? Why do millions of amateur chess players suddenly need Alpha Zero to help point out their amateur mistakes? I look foward to the day when Stockfish is no longer good enough to find mistakes in my games! :)

abulmo2
Posts: 314
Joined: Fri Dec 16, 2016 10:04 am
Contact:

Re: Any news of a Komodo update in sight?

Post by abulmo2 » Tue Dec 12, 2017 2:06 pm

Laskos wrote:
abulmo2 wrote:
Leo wrote:
Werewolf wrote:Sorry that was a typo - I meant 1 GB of Hash.

And why did they use such a tiny amount?
To limit the strength of SF.
This have to be proved. I would like to see a SF 8 vs SF 8 (or Komodo/Houdini) match with different hash sizes to see the impact of the hash size to the elo, on the machine used by the alphaZero team. Unfortunately, I do not have a computer powerful enough to do any test, so I cannot do anything meaningful myself.
See here, for example:
http://www.talkchess.com/forum/viewtopi ... t&start=20

In the conditions of the match with A0, probably the necessary Hash was 32 or 64GB, and by looking at the link, it seems, SF8 had some 10% time-to-depth disadvantage. How it translates in Elo points there I don't know, Elo becomes a bit obsolete when the score is 28-0 with 72 draws. Maybe with 64GB Hash SF8 would have gotten 1-2 more draws. This Hash issue is a bit irrelevant, when we are talking of such hard to compare hardware used by A0 and SF8.
I am not convinced by the previous tests. For obvious reasons, they are very far from the conditions of the AlphaZero vs Stockfish match.
Richard Delorme

Werewolf
Posts: 1349
Joined: Thu Sep 18, 2008 8:24 pm

Re: Any news of a Komodo update in sight?

Post by Werewolf » Tue Dec 12, 2017 4:19 pm

shrapnel wrote:
Laskos wrote:How it translates in Elo points there I don't know, Elo becomes a bit obsolete when the score is 28-0 with 72 draws. Maybe with 64GB Hash SF8 would have gotten 1-2 more draws. This Hash issue is a bit irrelevant, when we are talking of such hard to compare hardware used by A0 and SF8.
This is the part which seems to be evading Werewolf's attention.
He is just simply desperately looking for excuses to explain the Rout of Stockfish. He will accept everything except the relevant FACT that there has been a quantum jump in computer chess, rendering Alpha Beta type Chess Engines obsolete.
To be fair, its been too sudden a shock for him and others like him and he will take time to adjust to the fact that there is a new Alpha in Town !
Of course, DeepMind not releasing more information and games since then for what ever reason, will only encourage his type.
You are my favourite comedian!

But you can't answer my questions, how come?

- why did they run with HT?
- why did they keep the hash far too small?
- why that time control?
- why no opening book?
- why are only the wins shown? Was SF close to winning some of the draws, perhaps?
- And have you programmed??! :wink:

shrapnel
Posts: 1340
Joined: Fri Nov 02, 2012 8:43 am
Location: New Delhi, India

Re: Any news of a Komodo update in sight?

Post by shrapnel » Tue Dec 12, 2017 4:39 pm

I don't have to be a Programmer to see that Stockfish was completely outclassed.
Run any Engine you like under whatever conditions you want and can you find 19. Re1 ?
[d]rnb2r2/p3bpkp/1ppq3N/6p1/Q7/2P3P1/P4PBP/R1B2RK1 w - - 0 19
Or any of the other simple but brilliant moves found by AlphaZero in the 10 example games, with all the Hash and Time you like.
Like I said earlier, just find the time to go through those 10 games, with any Engine you like and then maybe, just maybe you'll see the light ; instead of questioning my qualifications, as if they have any relevance to the brilliance of AlphaZero.
You can call me all the names you like, but your silly questions just confirm my suspicion that you are desperately looking for excuses to explain why Stockfish got its a$$ kicked.
Obviously, you just believe what you want to believe, no matter how far it is from the facts.
i7 5960X @ 4.1 Ghz, 64 GB G.Skill RipJaws RAM, Twin Asus ROG Strix OC 11 GB Geforce 2080 Tis

Leo
Posts: 996
Joined: Fri Sep 16, 2016 4:55 pm
Location: USA/Minnesota
Full name: Leo Anger

Re: Any news of a Komodo update in sight?

Post by Leo » Tue Dec 12, 2017 4:53 pm

Werewolf wrote:
shrapnel wrote:
Laskos wrote:How it translates in Elo points there I don't know, Elo becomes a bit obsolete when the score is 28-0 with 72 draws. Maybe with 64GB Hash SF8 would have gotten 1-2 more draws. This Hash issue is a bit irrelevant, when we are talking of such hard to compare hardware used by A0 and SF8.
This is the part which seems to be evading Werewolf's attention.
He is just simply desperately looking for excuses to explain the Rout of Stockfish. He will accept everything except the relevant FACT that there has been a quantum jump in computer chess, rendering Alpha Beta type Chess Engines obsolete.
To be fair, its been too sudden a shock for him and others like him and he will take time to adjust to the fact that there is a new Alpha in Town !
Of course, DeepMind not releasing more information and games since then for what ever reason, will only encourage his type.
You are my favourite comedian!

But you can't answer my questions, how come?

- why did they run with HT?
- why did they keep the hash far too small?
- why that time control?
- why no opening book?
- why are only the wins shown? Was SF close to winning some of the draws, perhaps?
- And have you programmed??! :wink:

They did those things to weaken SF. They wanted an impressive score. Good point. For all we know they didn't show some losses by AZ
Advanced Micro Devices fan.

shrapnel
Posts: 1340
Joined: Fri Nov 02, 2012 8:43 am
Location: New Delhi, India

Re: Any news of a Komodo update in sight?

Post by shrapnel » Tue Dec 12, 2017 5:00 pm

Leo wrote:they did those things to weaken SF. They wanted an impressive score. Good point. For all we know they didn't show some losses by AZ
LOL, another Conspiracy Theory guy. You can join the company of Tsvetkov,Szabo and werewolf.
Pathetic.
Obviously none of you have played through those 10 Games.
Also, go through these Articles, after you finish putting down AlphaZero.
ttps://www.extremetech.com/extreme/260215-alph ... w-world-ai
https://www.pcgamesn.com/chess/chess-go ... -stockfish.
I particularly like the Comment " Stockfish reduced to fish stock " :D
i7 5960X @ 4.1 Ghz, 64 GB G.Skill RipJaws RAM, Twin Asus ROG Strix OC 11 GB Geforce 2080 Tis

Werewolf
Posts: 1349
Joined: Thu Sep 18, 2008 8:24 pm

Re: Any news of a Komodo update in sight?

Post by Werewolf » Tue Dec 12, 2017 5:38 pm

shrapnel wrote:
Leo wrote:they did those things to weaken SF. They wanted an impressive score. Good point. For all we know they didn't show some losses by AZ
LOL, another Conspiracy Theory guy. You can join the company of Tsvetkov,Szabo and werewolf.
Pathetic.
Obviously none of you have played through those 10 Games.
Also, go through these Articles, after you finish putting down AlphaZero.
ttps://www.extremetech.com/extreme/260215-alph ... w-world-ai
https://www.pcgamesn.com/chess/chess-go ... -stockfish.
I particularly like the Comment " Stockfish reduced to fish stock " :D
Played through them and actually I agree with you Shrapnel - really impressive from AlphaZero.

I just wished SF wasn't weakened.

shrapnel
Posts: 1340
Joined: Fri Nov 02, 2012 8:43 am
Location: New Delhi, India

Re: Any news of a Komodo update in sight?

Post by shrapnel » Tue Dec 12, 2017 5:54 pm

Werewolf wrote:Played through them and actually I agree with you Shrapnel - really impressive from AlphaZero.

I just wished SF wasn't weakened.
Fair enough.
But I also read that even if latest Stockfish was used and given a good Opening Book, it may have got a draw or two more, but the end-result would have been the same.
What really impressed me about AlphaZero more than the moves themselves, was its almost total disregard for material value.
If it saw a potential advantage for itself, whether Tactical or Positional, it simply sacrificed/threw away material to achieve it.
Now, if a human did that, even a slight miscalculation would have been disastrous, but AlphaZero being AlphaZero that didn't happen.
As Hassabis said “It doesn’t play like a human, and it doesn’t play like a program, “It plays in a third, almost alien, way.”
Using a tweaked Houdini/Komodo one may be able to find one or two of the AlphaZero brilliancies, but not all.
Last edited by shrapnel on Tue Dec 12, 2017 6:07 pm, edited 1 time in total.
i7 5960X @ 4.1 Ghz, 64 GB G.Skill RipJaws RAM, Twin Asus ROG Strix OC 11 GB Geforce 2080 Tis

royb
Posts: 524
Joined: Wed Mar 08, 2006 11:53 pm

Re: Any news of a Komodo update in sight?

Post by royb » Tue Dec 12, 2017 6:04 pm

shrapnel wrote:I don't have to be a Programmer to see that Stockfish was completely outclassed.
Run any Engine you like under whatever conditions you want and can you find 19. Re1 ?
[d]rnb2r2/p3bpkp/1ppq3N/6p1/Q7/2P3P1/P4PBP/R1B2RK1 w - - 0 19
Or any of the other simple but brilliant moves found by AlphaZero in the 10 example games, with all the Hash and Time you like.
Like I said earlier, just find the time to go through those 10 games, with any Engine you like and then maybe, just maybe you'll see the light ; instead of questioning my qualifications, as if they have any relevance to the brilliance of AlphaZero.
You can call me all the names you like, but your silly questions just confirm my suspicion that you are desperately looking for excuses to explain why Stockfish got its a$$ kicked.
Obviously, you just believe what you want to believe, no matter how far it is from the facts.
OK, I tried Komodo 11.2.2 with a tweak to the Dynamism parameter (set it to 200; an increase over the default 130 value for this parameter) and it sees Re1 immediately on my laptop (I used a single core):

FEN: rnb2r2/p3bpkp/1ppq3N/6p1/Q7/2P3P1/P4PBP/R1B2RK1 w - - 0 19

K-11.2.2-Tactical-Dyn200:

1 00:00 3 3k -2.47 a3 Qxh6
1 00:00 7 7k -1.96 h3 Qxh6
1 00:00 14 14k -0.47 Ng4
1 00:00 31 31k +0.65 Re1 Kxh6
2 00:00 100 100k +0.48 Re1 b5 Qd4+
3 00:00 258 258k +0.48 Re1 b5 Qd4+ Qxd4 cxd4
4 00:00 533 417k +0.48 Re1 b5 Qd4+ Qxd4 cxd4
5- 00:00 778 487k +0.42 Re1 Kxh6
5- 00:00 944 562k +0.31 Re1 Kxh6
5- 00:00 2k 826k -0.74 Re1 Kxh6
5+ 00:00 2k 873k -0.40 Re1 Kxh6 Qe4
5+ 00:00 2k 890k +0.80 Re1 Be6
5 00:00 3k 1,017k +0.28 Re1 Be6 Ng4 f5 Ne3
6- 00:00 3k 1,018k +0.22 Re1 Be6
6- 00:00 3k 1,061k +0.11 Re1 Be6
6- 00:00 4k 1,158k -0.54 Re1 Be6
6 00:00 4k 1,188k -0.41 Ng4 f5 Ne3 b5 Qd4+ Qxd4 cxd4
7+ 00:00 5k 115k -0.35 Ng4 f5 Ne3
7 00:00 5k 122k -0.27 Ng4 f5 Ne3 b5 Qc2 f4 Rd1
7+ 00:00 6k 133k -0.18 Re1 Be6 h4
7+ 00:00 6k 144k +0.01 Re1 Be6 Qe4
7+ 00:00 6k 147k +0.26 Re1 f5
7+ 00:00 7k 156k +0.55 Re1 f5 h4
7 00:00 7k 168k +0.81 Re1 Be6 Qe4 Kxh6 h4 Kg7 Bxg5 Bxg5
8 00:00 9k 205k +0.80 Re1 Be6 Qe4 Kxh6 h4 Kg7 Bxg5 h6
9- 00:00 11k 240k +0.74 Re1 Be6
9- 00:00 12k 263k +0.63 Re1 Be6
9 00:00 16k 326k +0.61 Re1 Be6 Qe4 Kxh6 h4 b5 Kh2 Kg7 Bxg5
10- 00:00 20k 375k +0.55 Re1 Kxh6
10- 00:00 25k 448k +0.46 Re1 Kxh6
10- 00:00 29k 232k +0.18 Re1 Kxh6
10+ 00:00 38k 285k +0.33 Re1 Kxh6 Qe4
10 00:00 39k 293k +0.37 Re1 Kxh6 Qe4 Be6 h4 Kg7 hxg5 Kg8 Bf4 Qd5 Rad1 Qxe4
11- 00:00 42k 309k +0.31 Re1 Kxh6
11- 00:00 45k 332k +0.22 Re1 Kxh6
11- 00:00 50k 358k -0.06 Re1 Kxh6
11+ 00:00 57k 401k +0.08 Re1 Bf6 Ba3
11+ 00:00 58k 405k +0.38 Re1 Kxh6 Qe4
11 00:00 59k 414k +0.59 Re1 Kxh6 Qe4 Be6 h4 Kg7 hxg5 Kg8 Bf4 Qd5 Qb1 Qc5
12- 00:00 62k 425k +0.53 Re1 Kxh6
12- 00:00 64k 440k +0.44 Re1 Kxh6
12- 00:00 76k 495k +0.16 Re1 Kxh6
12+ 00:00 86k 537k +0.30 Re1 f5 h4
12+ 00:00 96k 499k +0.61 Re1 Kxh6 h4
12 00:00 119k 572k +0.64 Re1 Kxh6 h4 f6 hxg5+ fxg5 Qh4+ Kg7 Bxg5 Bxg5 Qxg5+ Qg6 Qh4 Qf6 Re7+ Rf7 Rxf7+ Qxf7 Re1 Be6
13- 00:00 127k 596k +0.58 Re1 Kxh6
13- 00:00 134k 619k +0.49 Re1 Kxh6
13- 00:00 144k 653k +0.21 Re1 Kxh6
13+ 00:00 159k 694k +0.36 Re1 Be6 h4
13- 00:00 169k 721k +0.16 Re1 Kxh6
13 00:00 187k 767k +0.01 Re1 Kxh6 h4 f6 hxg5+ fxg5 Qh4+ Kg7 Bxg5 Bxg5 Qxg5+ Qg6 Re7+ Rf7 Rxf7+ Kxf7 Qd8 Be6 Qc7+ Kg8 Rd1 Nd7 Bxc6 Rf8 Bxd7
14+ 00:00 205k 807k +0.07 Re1 Kxh6 h4
14+ 00:00 217k 828k +0.16 Re1 Kxh6 h4
14+ 00:00 243k 872k +0.44 Re1 Kxh6 h4
14 00:00 257k 898k +0.46 Re1 Kxh6 h4 f6 hxg5+ fxg5 Qh4+ Kg7 Bxg5 Bxg5 Qxg5+ Qg6 Re7+ Rf7 Rxf7+ Kxf7 Qd8 Be6 Qc7+ Kg8 Rd1 Nd7 Bxc6 Rf8 Bxd7
15+ 00:00 288k 951k +0.52 Re1 Kxh6 h4
15+ 00:00 358k 1,024k +0.61 Re1 Kxh6 h4
15+ 00:00 467k 1,115k +0.89 Re1 Kxh6 h4
15- 00:00 484k 1,129k +0.74 Re1 Kxh6
15 00:00 514k 1,150k +0.58 Re1 Kxh6 h4 f6 hxg5+ fxg5 Qh4+ Kg7 Bxg5 Bxg5 Qxg5+ Qg6 Qh4 Bf5 Re7+ Kh8 Rd1 Na6 Qd4+ Qf6 Bxc6 Qxd4 Rxd4 Rad8 Rxa7 Rxd4
16+ 00:00 599k 1,184k +0.64 Re1 Kxh6 h4
16- 00:00 632k 1,207k +0.57 Re1 Kxh6
16- 00:00 913k 1,341k +0.10 Re1 Kxh6
16+ 00:00 1,013k 1,360k +0.28 Re1 Kxh6 h4
16- 00:00 1,063k 1,373k +0.03 Re1 Kxh6
16 00:00 1,185k 1,392k 0.00 Re1 Kxh6 h4 f6 hxg5+ fxg5 Qh4+ Kg7 Bxg5 Bxg5 Qxg5+ Qg6 Re7+ Kg8 Qe3 Qf6 Re1 Bf5 Qf3 Bg6 Qe3 Bf5
17 00:00 1,403k 1,422k 0.00 Re1 Kxh6 h4 f6 hxg5+ fxg5 Qh4+ Kg7 Bxg5 Bxg5 Qxg5+ Qg6 Re7+ Kg8 Qe3 Qf6 Re1 Bf5 Qf3 Bg6 Qe3 Bf5
18+ 00:01 2,062k 1,493k +0.06 Re1 Kxh6 h4
18+ 00:01 2,197k 1,498k +0.14 Re1 Kxh6 h4
18+ 00:01 2,429k 1,502k +0.25 Re1 Kxh6 h4
18- 00:01 2,699k 1,514k +0.14 Re1 Kxh6
18 00:01 3,007k 1,520k +0.14 Re1 Kxh6 h4 f6 Be3 Bf5 Rad1 Qa3 hxg5+ fxg5 Qh4+ Kg6 Qh1 Kg7 Be4 Bxe4 Qxe4 Rf7 Bd4+ Bf6 Bxf6+ Rxf6 Re2 Qc5 Rd8 Rf8 Rxf8 Qxf8
19- 00:02 3,451k 1,563k +0.08 Re1 Kxh6
19- 00:02 3,957k 1,582k 0.00 Re1 Kxh6
19+ 00:02 4,196k 1,592k +0.08 Re1 Kxh6 h4
19+ 00:02 4,428k 1,602k +0.23 Re1 Kxh6 h4
19 00:02 4,674k 1,614k +0.38 Re1 Kxh6 h4 f6 Be3 Bf5 Rad1 Qa3 hxg5+ fxg5 Qh4+ Kg6 Qh1 Kg7 Be4 Bxe4 Qxe4 Rf7 Bd4+ Bf6 Bxf6+ Rxf6 Re2 Qc5 Rd8 Rf8 Rxf8 Qxf8
20- 00:03 5,073k 1,628k +0.32 Re1 Kxh6
20- 00:03 5,514k 1,649k +0.24 Re1 Kxh6
20- 00:03 5,940k 1,658k +0.13 Re1 Kxh6
20+ 00:04 6,884k 1,685k +0.23 Re1 Kxh6 h4
20+ 00:04 7,457k 1,697k +0.44 Re1 Kxh6 h4
20- 00:04 7,572k 1,699k +0.25 Re1 Kxh6
20 00:04 7,868k 1,707k +0.15 Re1 Kxh6 h4 f6 Be3 Bf5 Be4 Qe6 Bc2 Kg7 Bxg5 Bxc2 Bh6+ Kh8 Qxc2 Qf7 Bxf8 Bxf8 Qe4 Na6 Qxc6 Rd8 Rad1 Rxd1 Rxd1 Nc5 Qd5 Qg7 Qa8
21+ 00:05 9,480k 1,705k +0.21 Re1 Kxh6 h4
21+ 00:06 11,622k 1,725k +0.29 Re1 Kxh6 h4
21- 00:08 14,064k 1,748k +0.20 Re1 Kxh6
21- 00:08 15,035k 1,755k +0.05 Re1 Kxh6
21 00:09 17,281k 1,763k 0.00 Re1 Kxh6 h4 f6 Be3 Bf5 Be4 b5 Qc2 Bg4 hxg5+ fxg5 Bxh7 Bh5 Bf5 Qd5 Be4 Qd6 Bf5
22+ 00:10 18,667k 1,764k +0.06 Re1 Kxh6 h4
22+ 00:10 19,217k 1,768k +0.14 Re1 Kxh6 h4
22+ 00:11 19,567k 1,770k +0.25 Re1 Kxh6 h4
22- 00:12 21,808k 1,786k +0.14 Re1 Kxh6
22+ 00:13 23,951k 1,784k +0.28 Re1 Kxh6 h4
22+ 00:14 26,707k 1,785k +0.57 Re1 Kxh6 h4
22- 00:16 29,155k 1,785k +0.32 Re1 Kxh6
22+ 00:16 29,790k 1,785k +0.65 Re1 Kxh6 h4
22- 00:17 31,799k 1,793k +0.19 Re1 Kxh6
22 00:18 33,281k 1,793k +0.37 Re1 Kxh6 h4 f6 Be3 Bf5 Rad1 Qa3 Qd4 Bg6 Qg4 Kg7 h5 Bf7 Bd4 h6 Qf5 Rh8 Rd2 Bd8 Rde2 Qa5 Qg4 Qb5 Be4 Na6 Bf5
23+ 00:19 35,250k 1,796k +0.43 Re1 Kxh6 h4
23+ 00:20 37,351k 1,800k +0.51 Re1 Kxh6 h4
23- 00:21 39,583k 1,807k +0.42 Re1 Kxh6
23- 00:23 43,462k 1,814k +0.27 Re1 Kxh6
23 00:25 45,865k 1,819k +0.32 Re1 Kxh6 h4 f6 Be3 Bf5 Rad1 Qa3 Qd4 Nd7 g4 Bc5 Qc4 Bxe3 Rxe3 Ne5 hxg5+ fxg5 Rxe5 Bg6 Bxc6 Rad8 Rxd8 Rxd8 Bd5 Rf8 Re3 Qc5 Qxc5 bxc5 Re7 Rf4 Rxa7 Rxg4+
24- 00:28 51,794k 1,833k +0.26 Re1 Kxh6
24+ 00:31 56,872k 1,831k +0.33 Re1 Kxh6 h4
24+ 00:33 62,301k 1,839k +0.59 Re1 Kxh6 h4
24- 00:36 66,912k 1,839k +0.45 Re1 Kxh6
24+ 00:37 68,403k 1,839k +0.63 Re1 Kxh6 h4
24- 00:40 74,580k 1,832k +0.38 Re1 Kxh6
24 00:42 77,040k 1,821k +0.38 Re1 Kxh6 h4 f6 Be3 Kg7 Rad1 Qe6 Bd4 Qf7 hxg5 Bf5 g4 b5 Qa5 Be6 Qc7 Kg8 Qg3 Nd7 Bxc6 Rad8 Qf4 Ne5 gxf6 Bxf6 Rxe5 Rxd4 cxd4 Bxe5 Qxf7+ Kxf7 dxe5
25+ 00:44 80,203k 1,818k +0.44 Re1 Kxh6 h4
25+ 00:45 82,646k 1,818k +0.52 Re1 Kxh6 h4
25+ 00:49 89,776k 1,818k +0.78 Re1 Kxh6 h4
25- 00:54 98,991k 1,805k +0.64 Re1 Kxh6
25 01:00 107,611k 1,787k +0.61 Re1 Kxh6 h4 f6 Be3 Bf5 Rad1 Qa3 Qd4 Na6 hxg5+ fxg5 Qe5 Rad8 Rxd8 Bxd8 Bd4 Rf7 Be4 Qe7 Bxf5 Qxe5 Rxe5 Nc7 Bc2 c5 Be3 Bf6 Rf5 Ne8 Bxg5+ Kg7 Kg2 Nd6 Rd5 Bxg5 Rxd6
26- 01:06 117,424k 1,778k +0.55 Re1 Kxh6
26- 01:09 124,010k 1,774k +0.47 Re1 Kxh6
26- 01:14 132,076k 1,771k +0.36 Re1 Kxh6
26+ 01:20 142,557k 1,765k +0.46 Re1 Kxh6 h4
26+ 01:32 161,556k 1,752k +0.67 Re1 Kxh6 h4
26- 01:39 173,183k 1,740k +0.48 Re1 Kxh6
26+ 01:48 187,949k 1,737k +0.73 Re1 Kxh6 h4
26 01:52 194,730k 1,736k +0.66 Re1 Kxh6 h4 f6 Be3 Bf5 Rad1 Qa3 Qc4 b5 Qd4 Qa4 hxg5+ Kg6 Qd2 Nd7 Bd4 Rae8 Bxc6 fxg5 Be4 Qc4 Qc2 h5 Bxa7 Qe6 Bxf5+ Qxf5 Qxf5+ Kxf5 Rxd7 Bf6 Rd5+ Kg4 Rxe8 Rxe8 Kg2 Re2 Rxb5 Bxc3 a4 h4 Be3 h3+ Kh2
27- 02:00 209,269k 1,741k +0.60 Re1 Kxh6
27+ 02:15 234,446k 1,729k +0.67 Re1 Kxh6 h4
27- 02:21 246,501k 1,736k +0.58 Re1 Kxh6
27- 02:27 257,152k 1,740k +0.43 Re1 Kxh6
27 02:35 270,007k 1,735k +0.39 Re1 Kxh6 h4 f6 Be3 Bf5 Rad1 Qa3 Qc4 b5 Qd4 Rd8 hxg5+ Kg6 Qh4 fxg5 Qh1 Rxd1 Rxd1 Bf6 g4 Bc2 Rd2 Qc1+ Bf1 Kg7 Rd7+ Nxd7 Bxc1 Rc8 f4 h6 Be3 Bxc3 fxg5 hxg5 Qh5 Bf6 Bxg5 Bg6 Qh6+ Kf7 Bxf6
28+ 02:49 294,296k 1,732k +0.45 Re1 Kxh6 h4
28+ 02:59 310,590k 1,730k +0.53 Re1 Kxh6 h4

shrapnel
Posts: 1340
Joined: Fri Nov 02, 2012 8:43 am
Location: New Delhi, India

Re: Any news of a Komodo update in sight?

Post by shrapnel » Tue Dec 12, 2017 6:13 pm

royb wrote:
OK, I tried Komodo 11.2.2 with a tweak to the Dynamism parameter (set it to 200; an increase over the default 130 value for this parameter) and it sees Re1 immediately on my laptop (I used a single core):
So did I with a tweaked Houdini, read my Post above.
But using your tweaked Komodo you will never find all the brilliancies of AlphaZero.
What is more important is the Style of AlphaZero, which none of the Top 3 conventional engines can emulate.
i7 5960X @ 4.1 Ghz, 64 GB G.Skill RipJaws RAM, Twin Asus ROG Strix OC 11 GB Geforce 2080 Tis

Post Reply