Two Computer Match | 2-Core | i5-5200U

Discussion of computer chess matches and engine tournaments.

Moderators: hgm, Rebel, chrisw

User avatar
Nordlandia
Posts: 2821
Joined: Fri Sep 25, 2015 9:38 pm
Location: Sortland, Norway

Two Computer Match | 2-Core | i5-5200U

Post by Nordlandia »

Image

Image

As demonstrated, hardware is almost identical.

White: i5-5200U
Black: i5-5200U
Hash: 2048 Mb allocated for engine
EGTB: 6-men Syzygy SSD
GUI: Fritz 15 [last patch]
OS: Windows 10
Ponder: On
TC: 30m+30s | ~ 1.5 hours +/- per game



[pgn][Event "DESKTOP-DD1B2QU, Rapid 30m+30s"]
[Site "DESKTOP-DD1B2QU"]
[Date "2017.03.11"]
[Round "1"]
[White "Komodo 10.3 64-bit"]
[Black "asmFishW_2017-03-09_popcnt"]
[Result "1/2-1/2"]
[Annotator "0.09;0.00"]
[SetUp "1"]
[FEN "rnbqk2r/4bppp/p2ppn2/1p6/3NP3/1BN2Q2/PPP2PPP/R1B2RK1 b kq - 0 1"]
[PlyCount "98"]
[TimeControl "1800+30"]

{Intel(R) Core(TM) i5-5200U CPU @ 2.20GHz 2195 MHz W=34.1 plies; 2 627kN/s;
31 687 987 TBAs B=36.0 plies; 2 726kN/s; 25 850 750 TBAs} 1... Qb6 {0.00/30
109 Both last book move} 2. Be3 {0.09/29 72} Qc7 {0.06/30 60 (Qb7)} 3. Qg3 {
0.19/29 97} O-O {0.00/30 23} 4. f3 {0.23/30 88} Nc6 {0.07/32 305 (Bd7)} 5. Nxc6
{0.28/26 50} Qxc6 {0.00/29 15} 6. Rfd1 {0.23/30 179} Bd7 {0.00/32 52} 7. a4 {
0.26/29 42 (a3)} b4 {0.00/31 45} 8. Ne2 {0.22/29 18} e5 {0.08/30 4 (Rac8)} 9.
Qe1 {0.19/29 92 (Rac1)} Be6 {0.00/31 181} 10. Qxb4 {0.18/33 0} Rab8 {0.00/35 61
} 11. Qd2 {0.10/34 41 (Qc3)} Bxb3 {0.00/33 51} 12. cxb3 {0.14/33 1} Rxb3 {
0.00/34 12} 13. Nc3 {0.10/33 113} Qb7 {0.00/39 0} 14. Ra2 {0.15/32 105} Rc8 {
0.00/39 0 (Rd8)} 15. a5 {0.10/31 77 (Qe2)} Rc6 {0.00/33 58} 16. g3 {0.00/34 45
(Bb6)} Bd8 {0.00/40 65 (h6)} 17. Rc1 {0.00/32 100 (Qe2)} h6 {0.00/34 63 (h5)}
18. Rd1 {0.00/34 93 (Rc2)} Ne8 {-0.22/31 51} 19. h4 {0.00/35 22} Nc7 {-0.14/32
89} 20. Qc2 {-0.02/34 90 (Rc1)} Rb4 {-0.08/33 216} 21. Qf2 {-0.10/34 0} Qc8 {
-0.24/31 97 (Nb5)} 22. Kg2 {-0.06/30 115 (Bb6)} f5 {-0.08/31 105} 23. Qd2 {
0.00/30 20 (exf5)} Rb8 {-0.04/30 110 (Rbc4)} 24. Ba7 {0.00/32 108 (Bb6)} Rb7 {
0.00/30 88} 25. Bf2 {0.00/33 0} fxe4 {-0.08/33 65} 26. Nxe4 {0.00/33 0} d5 {
0.00/35 142} 27. Nc5 {0.00/34 0} Rb8 {0.00/34 80 (Rb5)} 28. Re1 {0.00/31 156}
Bf6 {0.00/38 0} 29. Nd3 {0.06/29 47 (b4)} Qf5 {-0.09/27 35 (e4)} 30. Nb4 {
0.00/28 76 (Qe2)} Rd6 {-0.18/29 51} 31. Qd3 {-0.19/28 71 (Nd3)} Qe6 {-0.21/29
67} 32. Bc5 {-0.07/29 26} Rd7 {-0.08/31 70} 33. Bb6 {-0.07/31 36 (b3)} e4 {
-0.18/31 87} 34. fxe4 {-0.22/31 0} dxe4 {-0.30/29 10} 35. Qc2 {-0.35/29 42} Re8
{-0.12/32 57} 36. b3 {-0.44/30 71} Be7 {-0.26/33 0} 37. Ra4 {-0.56/31 63} Bxh4
{-0.16/34 17} 38. Bxc7 {-0.56/33 110} Bg5 {-0.21/36 3} 39. Nd3 {-0.47/34 159
(Rd1)} exd3 {-0.31/33 38} 40. Rxe6 {-0.50/38 8} dxc2 {-0.41/34 3} 41. Rxe8+ {
-0.40/37 27} Kf7 {-0.79/38 7} 42. Bf4 {-0.40/41 76} Kxe8 {-0.77/46 0} 43. Bxg5
{-0.14/40 42} hxg5 {-0.77/49 12} 44. Rc4 {-0.13/42 33} Rd2+ {-0.77/50 0} 45.
Kh3 {-0.12/45 68 (Kf3)} Ke7 {0.00/51 111 (Kd7)} 46. Rc5 {-0.01/47 52} g4+ {
0.00/56 10 (Kd7)} 47. Kxg4 {0.00/45 26} g6 {0.00/58 28 (Kd6)} 48. Kf4 {0.00/46
55 (Kh3)} Kd7 {0.00/54 58 (Ke6)} 49. Ke3 {0.00/50 43 (b4)} Rg2 {0.00/63 58} 50.
Kd3 {0.00/51 0 (b4) Draw accepted} 1/2-1/2

[/pgn]

[pgn][Event "DESKTOP-DD1B2QU, Rapid 30m+30s"]
[Site "DESKTOP-DD1B2QU"]
[Date "2017.03.11"]
[Round "2"]
[White "asmFishW_2017-03-09_popcnt"]
[Black "Komodo 10.3 64-bit"]
[Result "0-1"]
[Annotator "0.20;0.00"]
[SetUp "1"]
[FEN "rnbqk2r/4bppp/p2ppn2/1p6/3NP3/1BN2Q2/PPP2PPP/R1B2RK1 b kq - 0 1"]
[PlyCount "81"]
[TimeControl "1800+30"]

{Intel(R) Core(TM) i5-5200U CPU @ 2.20GHz 2195 MHz W=29.6 plies; 2 723kN/s;
964 667 TBAs B=30.0 plies; 2 497kN/s; 1 190 805 TBAs} 1... Qc7 {0.00/30 122
Both last book move} 2. Qg3 {0.20/27 40} O-O {0.13/29 72 (Nc6)} 3. Bh6 {
0.00/29 138} Ne8 {0.07/30 0} 4. Rad1 {0.00/32 81 (Rfd1)} Bd7 {0.07/28 57 (Bf6)}
5. Nce2 {0.17/25 42 (Be3)} Nc6 {0.00/28 52} 6. c3 {0.19/28 26} Kh8 {0.01/32 190
} 7. Bc1 {0.08/31 129 (Be3)} Nf6 {-0.02/29 87} 8. Qh4 {0.00/31 33 (f3)} Ne5 {
-0.14/27 190 (Ng8)} 9. Bc2 {0.00/28 142} a5 {-0.10/28 40 (Ng6)} 10. f4 {
0.19/25 72} Ng6 {-0.07/29 0} 11. Qh3 {0.18/28 279 (Qe1)} e5 {-0.10/26 114} 12.
Nf5 {0.12/29 0} Be6 {-0.24/25 31 (Rfd8)} 13. g4 {0.13/26 94} Bxa2 {-0.17/29 65
(b4)} 14. Nxe7 {0.29/30 183 (g5)} Nxe7 {-0.24/28 49} 15. g5 {0.11/28 1} Nfg8 {
-0.57/29 297 (Nd7)} 16. f5 {0.14/25 83} f6 {-0.78/27 31 (b4)} 17. g6 {-0.64/26
163} h6 {-0.86/29 0} 18. Qf3 {-0.87/28 363} Rfd8 {-1.01/32 0} 19. Rf2 {-1.12/
28 279 (Rfe1)} d5 {-1.05/28 49} 20. Re1 {-1.24/29 154} dxe4 {-1.31/32 0 (b4)}
21. Bxe4 {-1.38/29 70} Bd5 {-1.49/30 0} 22. Ng3 {-1.45/29 48} a4 {-1.55/29 0}
23. h3 {-1.39/27 42 (Ref1)} Bxe4 {-1.59/31 151} 24. Nxe4 {-1.58/29 30} Nd5 {
-1.39/28 18} 25. Rg2 {-1.74/25 13 (Rd2)} b4 {-1.88/26 36 (Rac8)} 26. cxb4 {
-1.57/24 32} Nxb4 {-1.90/29 31} 27. Be3 {-2.03/26 15} a3 {-1.78/29 42} 28. bxa3
{-1.89/28 31} Rxa3 {-1.84/31 35} 29. Rge2 {-1.91/29 24 (Nc5)} Ne7 {-2.36/29 112
} 30. Qg4 {-1.92/31 0} Nbd5 {-2.31/31 147 (Ned5)} 31. Rc1 {-1.94/26 40 (Kh2)}
Nc6 {-2.58/30 90 (Qb7)} 32. Bd2 {-2.73/30 57} Qb6+ {-2.85/32 34} 33. Kh2 {
-2.85/31 6} Nd4 {-2.89/32 29} 34. Rf2 {-2.89/32 18} Ne3 {-3.06/33 100} 35. Bxe3
{-2.97/37 0} Rxe3 {-2.93/34 67} 36. Rc3 {-3.05/40 0 (Rcf1)} Nf3+ {-3.92/34 152}
37. Qxf3 {-3.31/41 0} Rxf3 {-4.04/34 27} 38. Rcxf3 {-3.68/40 110} Kg8 {-4.14/
35 0 (Qb4)} 39. Rc3 {-4.10/32 89 (Rc2)} Kf8 {-4.29/33 38} 40. Rcc2 {-4.44/32 7
(Nc5)} h5 {-5.28/30 38} 41. Rb2 {-5.35/33 55} Qe3 {-5.83/33 0 (Qc6) Time} 0-1

[/pgn]

https://nb.lichess.org/rQGUFneP

https://nb.lichess.org/iY2RfwqU
Paul Bedrey
Posts: 1146
Joined: Thu Mar 09, 2006 11:46 am
Location: Saratoga Springs New York

Re: Two Computer Match | 2-Core | i5-5200U

Post by Paul Bedrey »

How did you get the two computers to converse with each other? I've been playing computer versus computer for years making moves manually.
User avatar
Nordlandia
Posts: 2821
Joined: Fri Sep 25, 2015 9:38 pm
Location: Sortland, Norway

Re: Two Computer Match | 2-Core | i5-5200U

Post by Nordlandia »

Paul Bedrey wrote:How did you get the two computers to converse with each other? I've been playing computer versus computer for years making moves manually.
Computer A host private cload engine

Computer B select engine from cload.

This works if you have playchess subscription active, in addition to Fritz gui installed on both machines.

Also you need to compensate for transmission delay & ping by adjusting "Move Overhead" parameter. I recommend 200ms.
Hugo
Posts: 782
Joined: Tue Dec 01, 2009 11:10 am

Re: Two Computer Match | 2-Core | i5-5200U

Post by Hugo »

this is a nice idea with the CB cloud to play engine matches when each engine uses a seperate computer.

Older Fritz GUIs do also support the "two computer match" option for such games. But you need to connect the two computers with a NullModemCable. For that kind of connection, you need a RS232 port on your computer. Many Laptops have such port, Desktop computers have it very rarely, but you can use an adapter RS232 to USB.
Fritz 13 GUI can be upgraded with the "two computer match" option by placing the auto232 in the folder called "Devices" in "C:\Program Files (x86)\ChessBase".
http://clemens-keck.de/CB-Auto232.zip
its a little tricky, until it works. You have to choose the right com port, and one computer is master, one is slave.
http://www.stmintz.com/ccc/index.php?id=341379
regards, C.K.


Nordlandia wrote:
Paul Bedrey wrote:How did you get the two computers to converse with each other? I've been playing computer versus computer for years making moves manually.
Computer A host private cload engine

Computer B select engine from cload.

This works if you have playchess subscription active, in addition to Fritz gui installed on both machines.

Also you need to compensate for transmission delay & ping by adjusting "Move Overhead" parameter. I recommend 200ms.
User avatar
Nordlandia
Posts: 2821
Joined: Fri Sep 25, 2015 9:38 pm
Location: Sortland, Norway

Re: Two Computer Match | 2-Core | i5-5200U

Post by Nordlandia »

Hugo wrote:this is a nice idea with the CB cloud to play engine matches when each engine uses a seperate computer.
I discovered the possibility just an week ago!

But it had to be said, there is some issues and irritation involved.

For example, unless "Move Overhead" paramter is left unadjusted, - very often games is lost on time. Furthermore sometimes game is stopped for whatever reason (probably transmission error).

I have contacted Chessbase support staff and one of the technician responded and are aware of the issues i've experienced.

It is not funny to setup engine match overnight and wake up hours later with game stuck on move 12.

Maybe this can be fixed/patched soon.
User avatar
Guenther
Posts: 4606
Joined: Wed Oct 01, 2008 6:33 am
Location: Regensburg, Germany
Full name: Guenther Simon

Re: Two Computer Match | 2-Core | i5-5200U

Post by Guenther »

Hugo wrote: Older Fritz GUIs do also support the "two computer match" option for such games. But you need to connect the two computers with a NullModemCable. For that kind of connection, you need a RS232 port on your computer. Many Laptops have such port, Desktop computers have it very rarely, but you can use an adapter RS232 to USB.
Fritz 13 GUI can be upgraded with the "two computer match" option by placing the auto232 in the folder called "Devices" in "C:\Program Files (x86)\ChessBase".
http://clemens-keck.de/CB-Auto232.zip
its a little tricky, until it works. You have to choose the right com port, and one computer is master, one is slave.
http://www.stmintz.com/ccc/index.php?id=341379
regards, C.K.
Much better should be Bernhard Wallners NetChess or Connect by HG.
(the old autoplayer had several other limitations and bugs, eg. time sync problems etc.)
https://www.google.de/search?q=wallners ... Le8gewyoFY

There is plenty of information about it in the net, thus I took the first better search result here:
http://www.open-chess.org/viewtopic.php?f=7&t=2053
Hugo
Posts: 782
Joined: Tue Dec 01, 2009 11:10 am

Re: Two Computer Match | 2-Core | i5-5200U

Post by Hugo »

long ago I have tried Auto232 games. I had that 2.0 version from Matthias Feist (ChessBase). It should work better.
There is no need to adjust move overhead settings. You can add an operator bonus for each move 1sec or 2sec is enough.
CB-Support I think will not be given. Newer GUIs do not even have the two computer match option.

What I did not like with netchess was the limitation of RAM for hashtables.

C.K.
User avatar
Nordlandia
Posts: 2821
Joined: Fri Sep 25, 2015 9:38 pm
Location: Sortland, Norway

Re: Two Computer Match | 2-Core | i5-5200U

Post by Nordlandia »

The alternative is Arena 3.5.1 autoplayer.

Despite the fact that Arena is far from bug-free GUI.

Invest 2x RS232 to USB

Image
mjlef
Posts: 1494
Joined: Thu Mar 30, 2006 2:08 pm

Re: Two Computer Match | 2-Core | i5-5200U

Post by mjlef »

I am not sure you can turn hyperthreading on or off, but if you can, it would be very interesting to see the same program play itself, but with hyperthreading on on on machine and Thread set to the number of hyperthreads, and on the second machine hyperthreading off and Threads set to the number of cores (so half of the first machine.

We discuss which is better here a lot, but I do not think anyone has done the test using two identical machines.

Mark
jpqy
Posts: 550
Joined: Thu Apr 24, 2008 9:31 am
Location: Belgium

Re: Two Computer Match | 2-Core | i5-5200U

Post by jpqy »

Nordlandia wrote:The alternative is Arena 3.5.1 autoplayer.

Despite the fact that Arena is far from bug-free GUI.

Invest 2x RS232 to USB

Image
Why not directly with one cable USB to USB ?!

JP.