ChessX is better than Scid and Scid_vs_PC in my opinion. http://chessx.sourceforge.net/
SCID vs PC
Sort:
ChessX handles giant databases poorly at the moment
Mar 30, 2017
0
#784
I have a decent size database ( all TWIC issues) and ChessX handles it fine. There was a new version out not too long ago. UI looks much better now.
I tried a 2.2-million PGN with it and it baulked. I didn't bother trying with my 7.8-million DB.
Hello, may I ask whether chess programs like fritz 15 and chessbase14 can go to a computer with AMD processor
SamuilGeorgiev έγραψε:
Hello, may I ask whether chess programs like fritz 15 and chessbase14 can go to a computer with AMD processor
Yes. Why shouldn't they?
Hello, may I ask whether chess programs like fritz 15 and chessbase14 can go to a computer with AMD processor
Mar 31, 2017
0
#789
SamuilGeorgiev wrote:
Hello, may I ask whether chess programs like fritz 15 and chessbase14 can go to a computer with AMD processor
Yes, they can.
I've exported 10,000 games annotated with evaluation score using Analysis Engine feature of SCID. Now I want to apply Linear Regression over a move's evaluation score, result of the game and the move winning score. The last one is the percentage of games won when the move is played. This information comes from the Tree window feature of SCID. Now what I want is to automate this for every move for each game that is I want to create a text file that contains the evaluation score for each move followed by a space, then the winning score of the move followed by another space and then the result of the game. I've read the documentation of SCIDvsPC (using that at the moment) and in its [programmer's reference][1], Tcl is mentioned. Since I'm new to this, can anyone guide me how can I automate this method of extracting move tree and the winning score of all of the played moves from it? I think [`sc_tree`][2] is the thing I'm finding but I dont know how to put this is in the script. Actually I dont know how do I proceed with getting started with the scipt. I had a networking course on tcl language in my last semester so I have some basic knowledge from some other field. Any links to learning anything new would also do. Suggestion would really help me. If you have time please go through the hyperlinks once as I think you may get the hint of what I should do if you have any knowledge about these stuff.
[1]: http://scidvspc.sourceforge.net/doc/progref.html
[2]: http://scidvspc.sourceforge.net/doc/progref.html#tree
I may have found the answer to my question. This from the online help file. "Games with no result do not contribute to the percentage drawn, but as half-won/half-loss."
Can someone explain how to read the information in the tree window? Compare the Score % with the graphs for moves 1 & 3. I don't understand. Move 3 has a higher score yet the graph appears to have let white. I was thinking the white = wins for white, grey = draws and black = wins for black. I also thought the Score % was success for white.
Apr 18, 2017
0
#792
Your basically right, but the Score % is a combination of 1 point for each White win and 1/2 point for each draw. Move 3 has fewer wins than 1, but has very little losses. Since x + .5y = Score, where x is % wins by white and y is % draws, then for move 3: x = Score - .5y x = 54.4 - (.5)x74, x = 17.4 (round to 17 since draws are only given in whole percents) so White wins 17% of games and since 100 - 74 - 17 = 9, then 9% of move 3 games are lost by White.
For move 1. x = 51.7 - (.5)x34 = 34.7%, so white wins 35% of the time, draws 34% of the time, and loses 31% of the time.
Thanks MickinMD. I will look at the ratings a little differently from now on. The graph itself is of more value to me than the score.
Yes and the free download chessbase reader can do the conversion from chessbase to pgn.
Scid vs PC 4.18 is out
Downloads http://scidvspc.sourceforge.net/#toc3
(The OS X version requires El Capitan... does it work on Sierra, High Sierra ?)
Changes http://scidvspc.sourceforge.net/#toc7
ScidvsMac Vers. 4.18 don't work on my IMac with macOS Sierra. Only start till the Welcomescreen and then ends.
stevenaaus wrote:
Scid vs PC 4.18 is out
Downloads http://scidvspc.sourceforge.net/#toc3
(The OS X version requires El Capitan... does it work on Sierra, High Sierra ?)
Thanks I downloaded and installed on my desktop. Unfortunately due to an unrelated issue my desktop computer is crashing. I think it's the power supply, So I'll put it here on my laptop. I haven't seen what are the improvements?
doxob wrote:
ScidvsMac Vers. 4.18 don't work on my IMac with macOS Sierra. Only start till the Welcomescreen and then ends.
It works for some people i'm told. Try removing $HOME/.scidvspc
I will post info when i get an Sierra to try out.
Removing $HOME/.scidvspc helps. But when I change the Language to German the next try fails the same way as before. I think that changing the configuration leads to break down.
"Depth per move" is definitely preferable, but different engines handle/report the depth in different ways. On my Core2Quad 2.4ghz, for annotation i use Critter at depth 15 for a decent graph, and 16 if i have a little more time. Toga on my Mac is slow to depth so i give it 12 or so. Stockfish, 17 or so... but trial and error is the way to go i think.