Engine release: weiss

[Terje]

Hello, first time posting here.

I'm fairly new to the scene; I got into chess programming late this summer after TCEC's twitch stream popped up in my recommendeds. After chatting with some of the viewers it seemed like an interesting challenge so I got to watching the VICE youtube series and started working from there with VICE as a base (I didn't feel like reinventing the wheel I had just seen built). Most parts have changed drastically since then, and it has gotten a fair bit stronger. If I had to guess it's somewhere around 2500 ccrl elo at fast TC given TBs.

https://github.com/TerjeKir/weiss/releases/tag/v0.6

I'd like to thank Andrew Grant for allowing me to join his testing framework (http://chess.grantnet.us/index/), without his help weiss would not be anywhere close to where it is. Thanks also to everyone I've interacted with in TCEC chat, your help as well as comraderie has been a major boost as well.

It's been a lot of fun so far and I look forward to continuing to improve on weiss!

(I have only tried compiling with gcc, would not be surprised if others don't work. There are very rare crashes on linux.)

[Guenther]

Hello, first time posting here.

I'm fairly new to the scene; I got into chess programming late this summer after TCEC's twitch stream popped up in my recommendeds. After chatting with some of the viewers it seemed like an interesting challenge so I got to watching the VICE youtube series and started working from there with VICE as a base (I didn't feel like reinventing the wheel I had just seen built). Most parts have changed drastically since then, and it has gotten a fair bit stronger. If I had to guess it's somewhere around 2500 ccrl elo at fast TC given TBs.

https://github.com/TerjeKir/weiss/releases/tag/v0.6

I'd like to thank Andrew Grant for allowing me to join his testing framework (http://chess.grantnet.us/index/), without his help weiss would not be anywhere close to where it is. Thanks also to everyone I've interacted with in TCEC chat, your help as well as comraderie has been a major boost as well.

It's been a lot of fun so far and I look forward to continuing to improve on weiss!

(I have only tried compiling with gcc, would not be surprised if others don't work. There are very rare crashes on linux.)

Thanks for the release. Will try a gcc compile tomorrow.

[Dann Corbit]

It has an impressive branching factor and very high NPS for a single thread.
I can't compile for PEXT here, so I muddled with the makefile.

I also added this for ucinewgame:

Code: Select all

        else if (BeginsWith(line, "ucinewgame"))
        	{
        		ParseFen(START_FEN, pos); /* DRC: I put this here. */
                        ClearHashTable(pos->hashTable);
                }

Which I admit is naughty of me, but I can't help myself, since I hate doing "position" all the time.

Code: Select all

E:\chess\weiss-0.6\bin>weiss-sse-pgo.exe
HashTable init complete with 2097152 entries, using 32MB.
uci
id name weiss 0.5
id author LoliSquad
option name Hash type spin default 32 min 4 max 16384
option name Ponder type check default false
option name SyzygyPath type string default <empty>
uciok
ucinewgame
setoption name SyzygyPath value c:\chess\syzygy;
setoption name Hash value 16384;
HashTable init complete with 1073741824 entries, using 16384MB.
go infinite
info score cp 105 depth 1 seldepth 1 nodes 24 nps 24000 tbhits 0 time 0 hashfull 0 pv e2e4
info score cp 0 depth 2 seldepth 4 nodes 106 nps 106000 tbhits 0 time 0 hashfull 0 pv e2e4 e7e5
info score cp 75 depth 3 seldepth 7 nodes 349 nps 349000 tbhits 0 time 0 hashfull 0 pv e2e4 d7d5 d2d3
info score cp 5 depth 4 seldepth 9 nodes 1556 nps 1556000 tbhits 0 time 0 hashfull 0 pv e2e4 d7d5 b1c3 e7e6
info score cp 67 depth 5 seldepth 14 nodes 5482 nps 322000 tbhits 0 time 16 hashfull 0 pv b1c3 d7d5 e2e4 d5e4 c3e4
info score cp 7 depth 6 seldepth 18 nodes 11624 nps 683000 tbhits 0 time 16 hashfull 0 pv b1c3 d7d5 e2e4 g8f6 e4d5 f6d5
info score cp 65 depth 7 seldepth 18 nodes 28901 nps 1700000 tbhits 0 time 16 hashfull 0 pv b1c3 e7e5 e2e4 b8c6 d2d3 d7d6 c1d2
info score cp 15 depth 8 seldepth 21 nodes 62880 nps 1905000 tbhits 0 time 32 hashfull 0 pv b1c3 d7d5 d2d4 g8f6 c1f4 c8f5 e2e3 b8c6
info score cp 55 depth 9 seldepth 21 nodes 127924 nps 1998000 tbhits 0 time 63 hashfull 0 pv g1f3 d7d5 d2d4 g8f6 c1d2 b8c6 b1c3 c8f5 e2e3
info score cp 27 depth 10 seldepth 25 nodes 263653 nps 2375000 tbhits 0 time 110 hashfull 0 pv e2e4 d7d5 e4d5 g8f6 d2d4 f6d5 f1b5 c8d7 c2c4 c7c6 g1e2
info score cp 55 depth 11 seldepth 25 nodes 409244 nps 2365000 tbhits 0 time 172 hashfull 0 pv e2e4 d7d5 e4d5 e7e6 b1c3 e6d5 d2d4 b8c6 c1e3 c8e6 g1f3
info score cp 15 depth 12 seldepth 25 nodes 719180 nps 2413000 tbhits 0 time 297 hashfull 0 pv e2e4 d7d5 e4d5 e7e6 b1c3 e6d5 d2d4 g8f6 c1g5 c8e6 f1b5 b8c6 g1f3
info score cp 34 depth 13 seldepth 26 nodes 826771 nps 2396000 tbhits 0 time 344 hashfull 0 pv e2e4
info score cp 54 depth 14 seldepth 26 nodes 1788136 nps 2429000 tbhits 0 time 735 hashfull 0 pv e2e4 d7d5 e4d5 e7e6 f1b5 c7c6 d5c6 b8c6 b1c3 c8d7 b5c6 d7c6 g1f3 f8c5 d2d3
info score cp 45 depth 15 seldepth 30 nodes 4043519 nps 2509000 tbhits 0 time 1610 hashfull 1 pv e2e4 e7e5 b1c3 b8c6 g1f3 g8f6 d2d4 e5d4 f3d4 f8b4 c1g5 b4c3 b2c3 h7h6 g5f6 d8f6
info score cp 26 depth 16 seldepth 31 nodes 6102099 nps 2551000 tbhits 0 time 2391 hashfull 1 pv e2e4 e7e5 b1c3 b8c6 g1f3 g8f6 d2d4 e5d4 f3d4 f8b4 c1g5 h7h6 g5f6 d8f6 d4b5 e8g8
info score cp 29 depth 17 seldepth 31 nodes 8809404 nps 2550000 tbhits 0 time 3453 hashfull 3 pv e2e4 e7e5 b1c3 b8c6 g1f3 g8f6 d2d4 e5d4 f3d4 f8b4 c1g5 h7h6 g5f6 d8f6 d4b5 b4c3 b5c3 d7d6
info score cp 38 depth 18 seldepth 35 nodes 17580886 nps 2717000 tbhits 0 time 6469 hashfull 4 pv e2e4 e7e5 g1f3 b8c6 d2d4 e5d4 f3d4 g8f6 b1c3 f8b4 c1g5 d8e7 g5f6 e7f6 a2a3 f6d4 a3b4 d4b4
info score cp 16 depth 19 seldepth 35 nodes 28227418 nps 2871000 tbhits 0 time 9828 hashfull 4 pv e2e4 e7e5 g1f3 b8c6 b1c3 g8f6 f1b5 a7a6 b5c6 d7c6 h2h3 f8d6 d2d4 e5d4 d1d4 c8d7 c1g5 c6c5 d4d3
info score cp 25 depth 20 seldepth 35 nodes 39849732 nps 2982000 tbhits 0 time 13360 hashfull 6 pv e2e4 e7e5 g1f3 b8c6 b1c3 g8f6 f1b5 f8b4 e1g1 e8g8 d2d3 b4c3 b2c3 d7d6 b5c6 b7c6 c1e3 c8a6 a1b1 d8d7
info score cp 23 depth 21 seldepth 37 nodes 65557035 nps 3126000 tbhits 0 time 20969 hashfull 9 pv e2e4 e7e5 g1f3 b8c6 d2d4 e5d4 f3d4 g8f6 d4c6 d7c6 d1d8 e8d8 c1g5 c8e6 b1d2 f8e7 f1e2 h7h6 g5e3 e7b4 e4e5 f6d7
info score cp 31 depth 22 seldepth 41 nodes 102209378 nps 3267000 tbhits 0 time 31282 hashfull 12 pv e2e4 e7e5 g1f3 b8c6 d2d4 e5d4 f3d4 g8f6 d4c6 d7c6 d1d8 e8d8 c1g5 c8e6 b1d2 f8e7 f1e2 h7h6 g5e3 e7b4 e4e5 f6g4 e3d4
info score cp 30 depth 23 seldepth 41 nodes 158619145 nps 3442000 tbhits 0 time 46078 hashfull 16 pv e2e4 e7e5 g1f3 b8c6 d2d4 e5d4 f3d4 g8f6 d4c6 d7c6 d1d8 e8d8 c1g5 f8d6 b1d2 d8e7 f1e2 c8e6 e1g1 a8d8 c2c3 h8e8 g5f6 e7f6 f2f4
info score cp 27 depth 24 seldepth 41 nodes 243786129 nps 3579000 tbhits 0 time 68110 hashfull 31 pv e2e4 e7e5 g1f3 b8c6 d2d4 e5d4 f3d4 g8f6 d4c6 d7c6 d1d8 e8d8 c1g5 f8d6 b1d2 d8e7 g5f6 e7f6 f1c4 c8d7 e1c1 h8e8 h2h3 a7a6 a2a3 c6c5
info score cp 25 depth 25 seldepth 41 nodes 372739329 nps 3684000 tbhits 0 time 101157 hashfull 50 pv e2e4 e7e5 g1f3 b8c6 d2d4 e5d4 f3d4 g8f6 d4c6 d7c6 d1d8 e8d8 c1g5 f8d6 b1d2 d8e7 g5f6 e7f6 f1c4 b7b5 c4e2 h8e8 e1c1 a7a6 e2f3 c8e6 b2b3
info score cp 25 depth 26 seldepth 41 nodes 637934635 nps 3763000 tbhits 0 time 169500 hashfull 82 pv e2e4 e7e5 g1f3 b8c6 d2d4 e5d4 f3d4 g8f6 d4c6 d7c6 d1d8 e8d8 c1g5 f8e7 b1c3 h7h6 g5e3 c8d7 e1c1 e7b4 f1d3 d8e7 h2h3 d7e6 e3f4 b4c3 b2c3

E:\chess\weiss-0.6\bin>

Here is a link to the SSE version I made, but I did make the one change I mentioned and also had to edit the makefile so the profile would not crash.


Edit: Note that the binary is in the .\bin folder

[Terje]

It has an impressive branching factor and very high NPS for a single thread.
I can't compile for PEXT here, so I muddled with the makefile.

...

Which I admit is naughty of me, but I can't help myself, since I hate doing "position" all the time.

...

Recently added LMR which did wonders for the branching factor. The first month or two practically all I did was try to improve speed after adding bitboard representation

I might make that change to ucinewgame in a future patch. I don't do much UCI communication manually so it didn't occur to me that it might be useful.

Edit: Is there a reason you added -mtune? I was under the impression that as long as -march is available it implies -mtune?

[Dann Corbit]

Edit: Is there a reason you added -mtune? I was under the impression that as long as -march is available it implies -mtune?

An oversight.
march is stronger than mtune and covers both.
I always do mtune so that my builds are optimized for the machine I built them on to some degree at least.
But instead of PEXT, I chose march=native to just use whatever it could.

Your engine is really, really fast for a single thread.
Often, things like LMR actually work against the node count, because you spend more time figuring out where to search and where to skip.
I did not delve into it yet, but I imagine your evaluation must not be very complex.
Either that, or you are really darn clever.

[Terje]

Edit: Is there a reason you added -mtune? I was under the impression that as long as -march is available it implies -mtune?

An oversight.
march is stronger than mtune and covers both.
I always do mtune so that my builds are optimized for the machine I built them on to some degree at least.
But instead of PEXT, I chose march=native to just use whatever it could.

Your engine is really, really fast for a single thread.
Often, things like LMR actually work against the node count, because you spend more time figuring out where to search and where to skip.
I did not delve into it yet, but I imagine your evaluation must not be very complex.
Either that, or you are really darn clever.

Eval is indeed extremely simple :p

Edit: Not only is it simple, nothing has been tuned or anything, so it's probably also bad at what little it tries to do

[Dann Corbit]

I think your approach is a good one.
By that, I mean focus on search.
Every trim of the branching factor is an enormous speedup.

Of course, your eval is helping you decide what to trim, so eventually it will need some attention.

[Gabor Szots]

The engine introduces itself as v0.5, is that right? The package name is v0.6.

[Terje]

The engine introduces itself as v0.5, is that right? The package name is v0.6.

Oops, I forgot to update that. It should say 0.6.

[Guenther]

I compiled a non popcount version today from current 0.7 dev master w/o problems.
If someone needs it I will add it for download.

First I thought though something was wrong with my compile because it sent only a1a2 or a1b1 pvs in cmd
until I realized it needs the 'position startpos' command too ;-)
Oh and it did not work with HGMs UCI2WB adapter in Winboard (reason not checked yet), but with Polyglot.

BTW it does not tell it has correctly found the syzygy tables, if given, as most programs do.
I needed a few games to see tb hits in the log. The score is also unusually low if syzygy tables are already on the board.
In the little test game which finally showed tb hits, the position after 54...Kxa5 leaves a 5men position (syzygy 5 were used),
but the eval remained around the same as before for several moves between +15 and +16.

The raw search speed is ofc tremendous as Dann already said, this was visible even on my old quadcore w/o popcount
and just SSE4.1.

[pgn]
[Event "Test_2019 Winboard 4.9.182601"]
[Site "CAPPUCCINO"]
[Date "2019.11.23"]
[Round "-"]
[White "Weiss_07-64"]
[Black "Abrok 5.0"]
[Result "1-0"]
[TimeControl "40/60"]
[Annotator "1. +0.45 1... -0.24"]

1. e4 {+0.45/13} Nc6 {-0.24/12 2.2} 2. d4 {+0.96/15 1.4} d5 {-0.26/11 1.9}
3. exd5 {+0.60/13 1.5} Qxd5 {+0.10/9 1.7} 4. Nf3 {+0.45/14 1.5} Bf5
{-0.14/9 1.5} 5. Nc3 {+1.33/12 1.5} Qe6+ {-0.26/9 1.4} 6. Be3
{+1.27/13 1.5} O-O-O {-0.50/10 2.5} 7. d5 {+1.67/13 1.5} Qf6 {-0.44/10 1.9}
8. Bc4 {+1.35/13 1.5} Nb4 {-0.64/9 1.4} 9. Bb3 {+1.28/13 1.5} Bg4
{-0.48/8 1.1} 10. a3 {+1.94/14 1.5} Nc6 {-0.82/9 1.9} 11. h3 {+1.99/15 1.5}
Bxf3 {-0.78/10 1.1} 12. Qxf3 {+1.69/17 1.5} Qxf3 {-0.82/11 1.1} 13. gxf3
{+1.57/17 1.5} Na5 {-1.00/11 1.0} 14. Ba2 {+1.50/13 1.5} b6 {-1.18/11 1.3}
15. Bf4 {+1.35/13 1.5} a6 {-1.16/10 1.4} 16. O-O-O {+1.37/13 1.5} g6
{-1.00/10 0.9} 17. Rhe1 {+1.15/12 1.5} Bg7 {-0.96/11 1.4} 18. Ne4
{+1.00/11 1.5} h6 {-0.86/10 1.4} 19. Bh2 {+0.75/12 1.5} f5 {-0.64/10 1.3}
20. Nd2 {+0.76/15 1.5} g5 {-0.68/10 1.1} 21. f4 {+0.78/14 1.5} Bd4
{-0.70/9 1.4} 22. Re2 {+1.39/13 1.5} Bf6 {-1.18/9 1.7} 23. c4
{+1.14/15 1.5} b5 {-0.70/9 1.3} 24. Bb1 {+1.79/13 1.5} bxc4 {-0.96/9 1.4}
25. Bxf5+ {+1.60/15 1.5} Kb7 {-1.10/10 1.4} 26. Nf3 {+1.65/13 1.5} Rf8
{-1.10/9 1.3} 27. Kb1 {+1.65/12 1.5} Nb3 {-1.30/9 1.4} 28. Be4
{+1.76/12 1.5} Ka7 {-1.40/8 1.4} 29. Ne5 {+2.08/14 1.5} Bxe5 {-1.66/10 1.6}
30. fxe5 {+2.18/17 1.5} Nc5 {-1.94/10 1.2} 31. Rc1 {+2.58/16 1.5} Nxe4
{-2.08/10 1.0} 32. Rxe4 {+2.84/17 1.5} Rd8 {-2.06/11 1.4} 33. Rexc4
{+3.03/17 1.5} Rxd5 {-2.10/11 1.5} 34. Rxc7+ {+3.03/15 1.5} Kb6
{-3.06/11 1.8} 35. f3 {+3.78/14 1.5} Rd1 {-2.10/11 2.6} 36. Rxd1
{+4.93/18 1.6} Kxc7 {-4.96/11 1.0} 37. e6+ {+5.27/16 1.6} Kc6
{-5.10/13 1.5} 38. Rd8 {+5.12/18 1.6} a5 {-5.10/12 1.1} 39. b3
{+5.19/17 1.6} Kc5 {-5.30/10 0.7} 40. Bc7 {+5.71/19 1.7} a4 {-5.32/8 0.7}
41. bxa4 {+6.40/18 1.5} h5 {-6.24/11 3} 42. Be5 {+6.32/18 1.4} Rh6
{-6.06/11 2.0} 43. Rxg8 {+6.35/17 1.5} g4 {-6.20/11 4} 44. hxg4
{+6.57/17 1.5} Rxe6 {-6.08/10 2.5} 45. Bc3 {+6.66/15 1.5} hxg4
{-6.20/10 2.2} 46. Rxg4 {+6.67/15 1.5} Kd5 {-6.26/11 2.6} 47. Rd4+
{+7.01/17 1.5} Kc6 {-6.16/12 4} 48. Re4 {+7.51/16 1.5} Rxe4 {-6.36/12 1.8}
49. fxe4 {+16.70/18 1.5} Kb6 {-6.58/14 1.5} 50. a5+ {+25.95/16 1.5} Kc6
{-6.54/14 1.6} 51. a6 {+25.98/16 1.5} Kb6 {-5.50/14 1.3} 52. a4
{+27.37/15 1.5} Kxa6 {-5.86/15 1.6} 53. a5 {+17.43/16 1.5} Kb5
{-5.84/15 1.6} 54. Bd4 {+17.43/15 1.5} Kxa5 {-4.28/15 1.2} 55. Kc2
{+15.98/18 1.5} Kb5 {-4.32/15 1.1} 56. Bc3 {+15.99/18 1.5} Kc4
{-4.58/17 1.4} 57. Kd2 {+16.04/19 1.5} Kb5 {-4.48/16 0.9} 58. Kd3
{+15.99/18 1.5} Kc5 {-5.00/16 1.4} 59. Bg7 {+15.88/18 1.5} Kd6
{-5.00/16 1.1} 60. Bf8 {+15.90/17 1.6} Kd7 {-5.40/15 0.9} 61. Kc3
{+15.88/19 1.5} Ke8 {-5.22/17 1.1} 62. Bh6 {+16.03/20 1.5} Kf7
{-5.26/16 1.0} 63. Bf4 {+15.90/19 1.5} Kg6 {-5.28/15 0.9} 64. Bh2
{+15.88/18 1.5} Kh5 {-5.08/15 0.9} 65. Bb8 {+16.04/17 1.5} Kg4
{-5.24/16 1.1} 66. e5 {+15.96/19 1.5} Kf5 {-5.40/17 1.1} 67. Kd4
{+16.18/20 1.5} Ke6 {-5.76/17 1.0} 68. Ba7 {+16.07/20 1.5} Kd7
{-5.70/17 1.2} 69. Kd5 {+15.90/16 1.6} e6+ {-5.80/17 0.9} 70. Kc5
{+16.21/20 1.5} Kc7 {-6.12/19 1.0} 71. Bb8+ {+16.06/20 1.5} Kd7
{-6.28/22 1.3} 72. Kb6 {+16.17/22 1.5} Ke7 {-6.16/18 1.2} 73. Kc6
{+16.18/21 1.5} Ke8 {-7.06/18 2.3} 74. Kd6 {+16.07/20 1.5} Kf7
{-7.28/17 0.7} 75. Kd7 {+15.97/19 1.5} Kf8 {-7.12/15 2.1} 76. Bd6+
{+16.17/18 1.6} Kf7 {-9.56/17 1.3} 77. Bc7 {+16.12/18 1.6} Kf8
{-9.56/16 1.5} 78. Kxe6 {+16.12/18 1.5} Ke8 {-10.88/13 0.4}
{Xboard adjudication} 1-0[/pgn]

Code: Select all

1574465097.312 Adapter->GUI: 15 +1743 123 2604478 Bd4 e6 a6 Kc6 a7 Kb5 a8=Q Kc4 Qa2+ Kb5 Qb3+ Kc6 Qxe6+ Kc7 Kc2 Kb7 Qd7+ Kb8 Be5+ Ka8
1574465097.542 Engine->Adapter: bestmove c3d4 ponder e7e6
1574465097.542 Adapter->GUI: move c3d4
1574465097.542 POLYGLOT MOVE Bd4
1574465097.542 POLYGLOT THINK -> WAIT
1574465097.572 Engine->Adapter: 
1574465098.724 GUI->Adapter: time 3942
1574465098.724 GUI->Adapter: otim 3178
1574465098.724 GUI->Adapter: usermove b5a5
1574465098.724 POLYGLOT MOVE Kxa5
1574465098.724 POLYGLOT WAIT -> THINK
1574465098.724 POLYGLOT START SEARCH
1574465098.724 POLYGLOT Not sending option "Ponder" since it already has the correct value.
1574465098.724 POLYGLOT FEN 8/4p3/8/k7/3BP3/8/8/1K6 w - - 0 55
1574465098.724 Adapter->Engine: position startpos moves e2e4 b8c6 d2d4 d7d5 e4d5 d8d5 g1f3 c8f5 b1c3 d5e6 c1e3 e8c8 d4d5 e6f6 f1c4 c6b4 c4b3 f5g4 a2a3 b4c6 h2h3 g4f3 d1f3 f6f3 g2f3 c6a5 b3a2 b7b6 e3f4 a7a6 e1c1 g7g6 h1e1 f8g7 c3e4 h7h6 f4h2 f7f5 e4d2 g6g5 f3f4 g7d4 e1e2 d4f6 c2c4 b6b5 a2b1 b5c4 b1f5 c8b7 d2f3 d8f8 c1b1 a5b3 f5e4 b7a7 f3e5 f6e5 f4e5 b3c5 d1c1 c5e4 e2e4 f8d8 e4c4 d8d5 c4c7 a7b6 f2f3 d5d1 c1d1 b6c7 e5e6 c7c6 d1d8 a6a5 b2b3 c6c5 h2c7 a5a4 b3a4 h6h5 c7e5 h8h6 d8g8 g5g4 h3g4 h6e6 e5c3 h5g4 g8g4 c5d5 g4d4 d5c6 d4e4 e6e4 f3e4 c6b6 a4a5 b6c6 a5a6 c6b6 a3a4 b6a6 a4a5 a6b5 c3d4 b5a5
1574465098.724 Adapter->Engine: go wtime 39420 btime 31780 movestogo 26
1574465098.734 Engine->Adapter: info score cp 455 depth 1 seldepth 1 nodes 24 nps 24000 tbhits 0 time 0 hashfull 635 pv b1c2
1574465098.734 Adapter->GUI: 1 +455 0 24 Kc2
1574465098.734 Engine->Adapter: info score cp 445 depth 2 seldepth 3 nodes 75 nps 75000 tbhits 2 time 0 hashfull 635 pv b1c2 e7e6
1574465098.734 Adapter->GUI: 2 +445 0 75 Kc2 e6
1574465098.734 Engine->Adapter: info score cp 455 depth 3 seldepth 4 nodes 173 nps 173000 tbhits 3 time 0 hashfull 635 pv b1c2 e7e6 c2d3
1574465098.734 Adapter->GUI: 3 +455 0 173 Kc2 e6 Kd3
1574465098.734 Engine->Adapter: info score cp 445 depth 4 seldepth 5 nodes 408 nps 408000 tbhits 9 time 0 hashfull 635 pv b1c2 e7e6 c2d3 a5b5
1574465098.734 Adapter->GUI: 4 +445 0 408 Kc2 e6 Kd3 Kb5
1574465098.734 Engine->Adapter: info score cp 569 depth 5 seldepth 7 nodes 1104 nps 1104000 tbhits 23 time 0 hashfull 635 pv b1b2 a5a4 b2c2 e7e5 d4e5
1574465098.734 Adapter->GUI: 5 +569 0 1104 Kb2 Ka4 Kc2 e5 Bxe5
1574465098.734 Engine->Adapter: info score cp 561 depth 6 seldepth 7 nodes 1964 nps 1964000 tbhits 47 time 0 hashfull 635 pv b1b2 a5a4 b2c2 e7e5 d4c3 a4b5
1574465098.734 Adapter->GUI: 6 +561 0 1964 Kb2 Ka4 Kc2 e5 Bc3 Kb5
1574465098.734 Engine->Adapter: info score cp 573 depth 7 seldepth 9 nodes 3846 nps 3846000 tbhits 87 time 0 hashfull 635 pv b1b2 a5a4 b2c3 e7e5 d4e5 a4b5 c3d4
1574465098.734 Adapter->GUI: 7 +573 0 3846 Kb2 Ka4 Kc3 e5 Bxe5 Kb5 Kd4
1574465098.734 Engine->Adapter: info score cp 578 depth 8 seldepth 11 nodes 14129 nps 14129000 tbhits 325 time 0 hashfull 635 pv d4c3 a5a6 c3b2 a6b5 b1c2 b5a5 c2d3 e7e5 b2e5
1574465098.734 Adapter->GUI: 8 +578 0 14129 Bc3+ Ka6 Bb2 Kb5 Kc2 Ka5 Kd3 e5 Bxe5
1574465098.754 Engine->Adapter: info score cp 614 depth 9 seldepth 12 nodes 24188 nps 24188000 tbhits 564 time 0 hashfull 635 pv d4c3 a5a6 b1c1 a6a7 c1c2 a7b8 c3e5 b8a8 e5d4 e7e5 d4e5
1574465098.754 Adapter->GUI: 9 +614 0 24188 Bc3+ Ka6 Kc1 Ka7 Kc2 Kb8 Be5+ Ka8 Bd4 e5 Bxe5
1574465098.754 Engine->Adapter: info score cp 614 depth 10 seldepth 14 nodes 43515 nps 2559000 tbhits 1056 time 16 hashfull 635 pv d4c3 a5a6 b1c1 a6a7 c3b4 e7e6 c1c2 e6e5 b4c3 a7a8 c3e5
1574465098.754 Adapter->GUI: 10 +614 2 43515 Bc3+ Ka6 Kc1 Ka7 Bb4 e6 Kc2 e5 Bc3 Ka8 Bxe5
1574465098.764 Engine->Adapter: info score cp 1021 depth 11 seldepth 15 nodes 82721 nps 2506000 tbhits 2014 time 32 hashfull 635 pv d4g7 a5a6 e4e5 a6a7 g7f6 a7a8 e5e6 e7f6 e6e7 a8b8 e7e8q b8c7
1574465098.764 Adapter->GUI: 11 +1021 3 82721 Bg7 Ka6 e5 Ka7 Bf6 Ka8 e6 exf6 e7 Kb8 e8=Q+ Kc7
1574465098.791 Engine->Adapter: info score cp 1145 depth 12 seldepth 16 nodes 123037 nps 1922000 tbhits 3174 time 63 hashfull 635 pv d4g7 a5a4 e4e5 a4b4 g7f6 b4a3 e5e6 e7f6 e6e7 a3a4 e7e8q a4b4 e8e7 b4c3
1574465098.791 Adapter->GUI: 12 +1145 6 123037 Bg7 Ka4 e5 Kb4 Bf6 Ka3 e6 exf6 e7 Ka4 e8=Q+ Kb4 Qe7+ Kc3
1574465098.816 Engine->Adapter: info score cp 1560 depth 13 seldepth 17 nodes 192918 nps 2030000 tbhits 5139 time 94 hashfull 635 pv d4c3 a5a6 b1c1 e7e5 c3e5 a6b5 e5d4 b5a6 e4e5 a6b7 e5e6 b7a6 e6e7 a6b5
1574465098.816 Adapter->GUI: 13 +1560 9 192918 Bc3+ Ka6 Kc1 e5 Bxe5 Kb5 Bd4 Ka6 e5 Kb7 e6 Ka6 e7 Kb5
1574465098.866 Engine->Adapter: info score cp 1560 depth 14 seldepth 18 nodes 293310 nps 2065000 tbhits 7771 time 141 hashfull 635 pv d4c3 a5a6 b1c1 e7e5 c3e5 a6b5 e5d4 b5a6 e4e5 a6b5 e5e6 b5a6 e6e7 a6b5 e7e8q b5c4
1574465098.866 Adapter->GUI: 14 +1560 14 293310 Bc3+ Ka6 Kc1 e5 Bxe5 Kb5 Bd4 Ka6 e5 Kb5 e6 Ka6 e7 Kb5 e8=Q+ Kc4
1574465099.006 Engine->Adapter: info score cp 1575 depth 15 seldepth 20 nodes 579802 nps 2056000 tbhits 15768 time 281 hashfull 635 pv d4c3 a5a6 b1c1 e7e5 c3e5 a6b5 e5d4 b5a6 e4e5 a6b5 e5e6 b5a6 e6e7 a6b5 e7e8q b5c4 e8e4
1574465099.006 Adapter->GUI: 15 +1575 28 579802 Bc3+ Ka6 Kc1 e5 Bxe5 Kb5 Bd4 Ka6 e5 Kb5 e6 Ka6 e7 Kb5 e8=Q+ Kc4 Qe4
1574465099.226 Engine->Adapter: info score cp 1598 depth 16 seldepth 21 nodes 1045460 nps 2086000 tbhits 26929 time 500 hashfull 635 pv b1c2 a5a4 e4e5 a4a3 d4b2 a3a2 b2a3 a2a1 e5e6 a1a2 a3e7 a2a1 e7c5 a1a2 e6e7 a2a1 e7e8q
1574465099.246 Adapter->GUI: 16 +1598 50 1045460 Kc2 Ka4 e5 Ka3 Bb2+ Ka2 Ba3 Ka1 e6 Ka2 Bxe7 Ka1 Bc5 Ka2 e7 Ka1 e8=Q
1574465099.476 Engine->Adapter: info score cp 1578 depth 17 seldepth 22 nodes 1587150 nps 2116000 tbhits 41494 time 749 hashfull 635 pv b1c2 a5a4 e4e5 a4a3 d4b2 a3a2 b2a3 a2a1 e5e6 a1a2 a3e7 a2a1 e7c5 a1a2 e6e7 a2a1 c5d4 a1a2 e7e8q
1574465099.476 Adapter->GUI: 17 +1578 75 1587150 Kc2 Ka4 e5 Ka3 Bb2+ Ka2 Ba3 Ka1 e6 Ka2 Bxe7 Ka1 Bc5 Ka2 e7 Ka1 Bd4+ Ka2 e8=Q
1574465099.906 Engine->Adapter: info score cp 1598 depth 18 seldepth 24 nodes 2569016 nps 2164000 tbhits 66341 time 1186 hashfull 635 pv b1c2 a5a4 e4e5 a4a3 d4c5 a3a4 c5e7 a4b5 c2d3 b5a5 e5e6 a5a4 e7c5 a4b3 e6e7 b3a2 d3c2 a2a1 e7e8q
1574465099.906 Adapter->GUI: 18 +1598 119 2569016 Kc2 Ka4 e5 Ka3 Bc5+ Ka4 Bxe7 Kb5 Kd3 Ka5 e6 Ka4 Bc5 Kb3 e7 Ka2 Kc2 Ka1 e8=Q
1574465100.186 Engine->Adapter: bestmove b1c2 ponder a5a4
1574465100.186 Adapter->GUI: move b1c2