TalkChess.com

Myrddin took 90 seconds to give a Mate in 9 with Kg2. But, no surprise, the great old mate-finder The King from CM9000 finds Re3 and announces Mate in 8 in eight seconds. Time Depth Score Positions Moves 0:00 1/3 4.17 2247 1.Qh3 Kf8 2.Qh8+ Ke7 3.Qxg7+ Kd6 4.Qxb7 Bxb7 0:00 1/4 4.57 7786 1.Qh3 Kf8 2.Q...

Johan did the MultiPV method in The King (Chessmaster). There was a "randomness" setting for each personality, from 0-100. A randomness of 20, for example, gave +/- 10cp for each move. So strong (i.e. GM-level) personalities that had that level of randomness would often play moves that could be call...

Myrddin on 8 cores finds Rxd4 in 8 seconds, but then keeps waffling between Rxd4 and Rxe3. The score is never close to a draw, though, even after two minutes.

This is probably the easiest position I have. Your engine should avoid Qxa6 and Rxa6 as they both lead to an immediate 3-fold draw: [D]3r3N/R1p1kppp/n6n/8/2QPP3/5q2/5P1P/5RK1 w - - 0 24 Also, this is a classic position, but it is much harder for young engines to find: [D]1k6/4R3/1p6/p1p3p1/qnBn2Qp/5...

Dann Corbit wrote: ↑

Sun Aug 16, 2020 12:39 am

silentshark wrote: ↑

Sat Aug 15, 2020 5:13 pm

So, here's my lazy question, what's the easiest way to create multiple processes and share memory between them, in the world of Windows?

Use CreateProcess.
You can almost simulate a fork with it.

Yep, that's how I did it in my code sample above.

With the nasty legacy code I have (lots of global variables etc.), the multiple process approach might be the easiest thing for me to use. So, here's my lazy question, what's the easiest way to create multiple processes and share memory between them, in the world of Windows? And that's exactly the ...

Results against TSCP: 1 thread on i7-6700 3.40GHz (4 cores) -----------------Plankton64_03----------------- Plankton64_03 - TSCP181c : 84.0/100 75-7-18 (=11=111=101=11=1110=1011111111==0=11111110=11=11=1111=1111=1111111111110111110111==1111=1=1111111111) 84% +288 -----------------TSCP181c----------...

8/3r4/5p2/2k1nQ2/8/3p4/3K4/8 b - - This position came up in Halogen v Myrddin in a CCRL match. There is only one move that wins for Black. Myrddin failed to find it within the time control even with 5-man tablebases. :-( Stockfish-dev with pure NNUE eval (no hybridization) finds the correct move in...

This position came up in Halogen v Myrddin in a CCRL match. There is only one move that wins for Black. Myrddin failed to find it within the time control even with 5-man tablebases.

Even The King from CM9000 (of course, with only one thread) can find it in less than a minute: Time Depth Score Positions Moves 0:00 1/3 -3.52 1297 1.Kb1 Kg8 2.c3 0:00 1/3 -3.49 1523 1.a3 Nc6 2.Qxb5 Bxf2 0:00 1/3 -3.45 2262 1.c3 Qc6 2.Kb1 0:00 1/4 -3.26 4564 1.c3 Nxa2+ 2.Kb1 h6 3.Kxa2 hxg5 4.hxg5+ K...