SCID vs PC

Except that the computer opponent doesn't claim 50-move draws, and there's no "claim draw" button for the human to press. So the game goes on until checkmate, stalemate, or until the human aborts the game. The program will claim 3-position rep draws and stalemates. (I'm using version 4.11, but I suspect the latest version works the same. Also, I only tried this using an endgame starting position; I didn't try it by playing a full game.)

I don't know if this feature was overlooked or if he hasn't had a chance to add it yet. (Scid vs. PC is mainly a database program, and the playing features are only recent additions.)

By the way, Arena allows you to set the option for checking the 50-move rule to either "always", "never", or "only games from starting position".

Then that is the fault of the computer opponent, not svpc.

The simple solution is for the player to claim the draw by aborting the game an modifying the result to reflect that.

The program is primarily a chess DB and not a program for playing games (even though there are features to do that).

I don't think so. I think the GUI should control those options. At least I don't remember ever seeing the 50-move draw option ever specified in an engine configuration file.

I have tried Arena and other programs, but I have adopted SvPC for classroom and home use, to demonstrate, review, analyze, and practice games, endings, etc. SvPC has extensive, intuitive graphic functionality that the other programs do not, features that are easily grasped by elementary and middle school children.  I have even tailored the chess board so that it looks much like the green and white vinyl tournament boards that we use in our elementary and middle school chess clubs. The program has helped us to get 'on the radar' in the Dallas Independent School District, where we took third place (out of 28 schools) in a recent tournament, with a whole new team that was only six weeks old. I hope that someone continues to develop this program because it has an excellent feature set.

I am certainly not complaining about Scid's 50-move 'issue'.  I only mentioned it because I happened to notice it, and thought perhaps that the developer might want to be aware. I do not consider it a showstopper. But if kids play the queen vs knight ending on offense, the program should properly claim a draw (on the defense engine's behalf) whenever the kid on the strong side fails to checkmate in 50 moves.

Another feature that I would like to see addressed is that when I back up in the course of playing against the computer to try another move I have to click Play > UCI engine > Play, to resume.  I would like to be able to resume play without having to reset the game.  Again, if that's not an easy code change, I am still exuberant to have a program that meets 'almost' all the features on my wish list. And the impoverished kids in my Title One School chess clubs here in Texas appreciate the opportunity to be able to train by running endgame exercises with ScidVsPD.  I have given them each a three-dollar flash drive with the SvPC program and thousands of exercises on it. This allows them to practice chess on any PC, whether or not they have access to the internet - for free.

Thanks

The engine is a player.  The GUI is facilitating the game.  Claiming a draw is the responsibility of the player, by rule.

Thanks for the feedback. We should probably acknowledge the 50 move draw. Checkout subversion, or replace checkRepetition with this once should do the trick... though i havent tested it.

proc checkRepetition {} {

  # Only show draw dialog once
  if {$::drawShown} {
    return 0
  }

  set fen [sc_pos fen]
  if {[lindex $fen 4] > 99} {
    set ::drawShown 1
    pauseGame
    sc_game tags set -result =
    sc_pos setComment "50 move rule"
    tk_messageBox -type ok -message $::tr(Draw) -parent .main.board -icon info
    catch {::game::Save}
    return 1
  }

  set elt [lrange [split $fen] 0 2]
  if {$elt == [lindex $::lFen end]} {
    return 0
  }

  lappend ::lFen $elt
  if { [llength [lsearch -all $::lFen $elt] ] >=3 } {
    set ::drawShown 1
    pauseGame
    sc_game tags set -result =
    sc_pos setComment "3 fold repetition"
    tk_messageBox -type ok -message $::tr(Draw) -parent .main.board -icon info
    catch {::game::Save}
    return 1
  }

  return 0
}

I guess you're right that the engine has to claim the 50-move draw, but if the GUI doesn't listen to the engine, I guess it doesn't happen.

I'm not competent to edit and recompile the program and presume that I am not expected to do so.  I will be ecstatic if and when the changes are incorporated into a new rev.  Thank you for considering my suggestion. I am overwhelmed that response to this idea has been so rapid, or that anyone would even note or care about the 'issue'.

I presume that my seemingly-recalled 'autoflip' option in #705 was merely a figment of my imagination, right?

Yeah, #705 - i havent thought about that before, but it sounds ... I don't think so.

The Scid gui does not need recompiling. Just find and edit the file "scid.gui" , searching for "proc checkRepetition" , replace that procedure, and save scid.gui and restart. This change did not make it into ScidvsPC 4.15, which was released last saturday.

Download links    http://scidvspc.sourceforge.net/index.html#toc3
Changelog    http://scidvspc.sourceforge.net/index.html#toc7

I'm using Linux and to lazy to compile. Its running great with Wine.

Thanks, i like it very much!

Steven

Thanks.  I have installed v. 4.15 and I modified scid.gui. I have also verified that the 50-move draw properly triggers and displays the disappointing news to the 'strong side'.  Now the kids will have to prove that they can play dodgeball against the opponent knight and win the game within 50 moves.  Queen vs Knight is probably the most challenging ending that I will try to make 'mandatory learning' for the club members.

I appreciate your help regarding this issue.

Under a few Linux distros (e.g. Archlinux, Manjaro etc) compiling from source can be done even by kids- using automated bash scripts. The application itself, the needed dependencies which are not in the official repositories and the build dependencies are installed by typing a very simple command- and it can also be done using a GUI.

I have Peppermint (like LUbuntu), on my old HP Mini  ;-)

Thats not easy to compile ScidvsPC. Scid is no big problem, because the old version is really installt.

The annotate button on my scid vs. pc doesn't work. When I click on it, it opens a window "Configure Annotation". I select the choices I want and click "Ok" and nothing happens. Does anyone else have this problem and know how to fix it? I tried upgrading to the latest version of scid vs. pc and it didn't help. 

I am trying to use the batch annotate feature to analyze 100 positions that I have in a database. Is there another program that can do this? 

This is pretty weird. It always worked for me (I've just verified this with SCID vs PC 4.15), I've been using Komodo 6/7 and Stockfish 6 with SCID vs PC. Does an engine start for you when you press the little "train" (?) icon (Analysis Engines) in the toolbar? Do you use only the default engines that came preinstalled with SCID?

I always start the engine first, before annotating. Pretty sure that is the way it has to be done. Maybe that is the problem idoun is seeing.

I can run any analysis engine fine.

What's really odd is that I did a bunch more tests on different databases and the ones I created a long time ago, with games already in them, the batch annotation works fine. It only does not work for any new databases I create. It looks like I can use the "Maintenance" window to delete all games, I can then compact the database, and then hopefully loading the new positions I want into that old database will work. Although I will be losing old games. 

I have no idea why it does not work with new databases, maybe there is some limit to how many databases one can have? 

Shouldn't be a limit, unless maybe you are loading a lot of them at once.

I updated last week and when I get on the computer that has that version on it I will test mine on a new database.

Also, you are on a database and not just a PGN file?

Yes, I create a new database. Then I open a PGN file containing the positions and drag it onto the database in the Gamelist window, which copies the positions into the database. 

Another thought, which I doubt, but I am grasping at straws - maybe I have not been deleting old databases properly. I didn't find a way to do it in the program so I had been deleting the actual files outside of the program (when the program was not running). But I don't see why this would affect new databases. I can try to make a database in a new directory and see if it works.