Page 7 of 9 FirstFirst ... 23456789 LastLast
Results 121 to 140 of 177
  1.    #121  
    Actually, now that I think of it, the server is probably gonna have to be reset a bunch of times during development anyway. I'm gonna try to keep it to a minimum but it's definitely going to have to happen at some point, I'm sure.

    As for the quitting/drawing, I am still figuring out exactly how I want to implement draws, turn undoing, and resignation. But I still think you should be able to quit a game as well. And by default that would result in the other player getting a win, I suppose. At least if you were playing competitively. Lots to think about.
  2. TRgEOff's Avatar
    Posts
    589 Posts
    Global Posts
    591 Global Posts
    #122  
    Quote Originally Posted by Brennan View Post
    As for the quitting/drawing, I am still figuring out exactly how I want to implement draws, turn undoing, and resignation. But I still think you should be able to quit a game as well. And by default that would result in the other player getting a win, I suppose. At least if you were playing competitively. Lots to think about.
    great to hear you are working on those features. resigning seems the same as (quit=loss ). I agree that one shouldn't be able to quit to get out of losing, and on ICC, the equivalent of "quitting" after ONE move only, by one or both players, counts as a "Game aborted". After that a "quit" registers as a loss to the quitting player. On ICC, players can agree at any stage to either a draw or to "Game aborted".

    Quitting games against chesscomp without registering a loss might be good however, since it allows you to try out features/ideas.

    I was thinking of a way that players could currently set up a draw. If they could agree via IM or TC that they wanted a draw, then they could both just shuffle pieces to get a draw by triple rep. I tried this vs chesscomp, just moving knights back and forth at the start of the game, but it didnt work. Are there some extra stipulations before triple rep comes into effect? I know it has worked for me before.

    cheers
    geoff

    Treo 650 1.20 LAP.
  3. #123  
    Quote Originally Posted by Brennan View Post
    I'm working on logging of game win/loss statistics. I had a couple ideas I wanted to run by the ChessChumps for feedback:

    1. Should I reset all the games when tracking goes live? Or can you guys live with some early games that don't go on your record but still exist in the server? I might be able to write a tool someday to update the old ones. But a clean break can be nice.

    2. What about quits? Should a quit mean the other side wins? Maybe if the quit is after X number of turns? What should X be then? 2, maybe 3?

    Let me know what you think!
    1. Don't reset all the games. I have too much at stake (even though I may be on the retreat on some games.)
    Yes, we could live with the early games that will not go on record.
    Clean break is ok but only when the new version is used. Force everyone to upgrade to new version when starting new games. Only those games should be considered official count. I would not worry about updating the old games. Consider those as pre-official game plays.

    2. Quits are resignation. Please implement a draw option. No points counted or 1/2 point per player. What's x number of turns? Confused about the x number of turns.

    <Sorry - moved the last couple of post to this thread as the other one refers to the games itself and members joining the games, or at least that's how I understood Trgeoff meant it to be>
  4. #124  
    Quote Originally Posted by TRgEOff View Post
    ... <snip> ...

    After I managed one great early move, leading to a promising position, I came up with a new variety of "mouse error" and lost a rook.

    I selected one rook (the "right" one) to move and then I thought I selected its correct destination square. But I was using my clumsy thumb instead of the stylus, so in fact I actually re-selected the other "wrong" rook that was sitting right next to my destination square. Without realising this (I was watching an exciting cricket match) I then tapped the destination square again, hit "Make move" and watched in horror as the wrong rook moved into the "right" place.

    I have had counselling but I am told it could be years before I fully recover.

    My therapist strongly urges the clean break to speed my return to normal life .

    cheers
    Haha, the horrors of clicking make move and seeing that you have move the wrong piece. Done that! That's why I was requesting for a way to back out your move. Maybe, after 20 seconds (unfortunate if you have bad connection or maybe, even if you had bad connection it should not commit the change, it should allow you to back it out.)
  5. TRgEOff's Avatar
    Posts
    589 Posts
    Global Posts
    591 Global Posts
    #125  
    Quote Originally Posted by ronbo2000 View Post
    <Sorry - moved the last couple of post to this thread as the other one refers to the games itself and members joining the games, or at least that's how I understood Trgeoff meant it to be>
    Yep that's the idea, but I bow humbly to any of Brennan's whims and fancies! (but I think he originally put the question on the other thread because if was so chess specific)

    "X turns" means the same as the more (mathematically) common "n turns" where n can stand in for any number (1,2,3....n). He is asking: if a game is really brief, should a QUIT therefore NOT count as a loss, and if so what is the cut-off limit for "really brief"? Should be it be one move, two moves, three moves etc.

    One extra piece of info on this: the infamous "Fool's mate" occurs after only TWO moves by each player, one of the reason that ICC has the "really brief" cut-off set so low, i.e. ONE move by each player.

    Quote Originally Posted by ronbo2000 View Post
    Haha, the horrors of clicking make move and seeing that you have move the wrong piece. Done that! That's why I was requesting for a way to back out your move. Maybe, after 20 seconds (unfortunate if you have bad connection or maybe, even if you had bad connection it should not commit the change, it should allow you to back it out.)
    Actually I think the scope for this type of error with Wifight is so low, that I personally don't think any accommodation is necessary.

    As to "undo" or "takeback" moves for "stupid" moves, in general I don't like them, but if implemented I think that it has to be something that both players "agree" to before the game. From my years of experience on ICC, I found it gets too complicated trying to decide what counts as a "stupid" move, and how many moves can be undone etc etc. It's better to use the pain of one's stupid move as a spur to playing better next time.

    But, to help us all in these situations, I think there should be a web link in the About tab to an online grief counsellor!






    cheers
    Last edited by TRgEOff; 09/18/2007 at 02:57 AM.
    geoff

    Treo 650 1.20 LAP.
  6. #126  
    Quote Originally Posted by TRgEOff View Post
    Yep that's the idea, but I bow humbly to any of Brennan's whims and fancies! (but I think he originally put the question on the other thread because if was so chess specific) ...<snip>...
    Ok, should I move these back then?

    Quote Originally Posted by TRgEOff View Post
    ...<snip>..."X turns" means the same as the more (mathematically) common "n turns" where n can stand in for any number (1,2,3....n). He is asking: if a game is really brief, should a QUIT therefore NOT count as a loss, and if so what is the cut-off limit for "really brief"? Should be it be one move, two moves, three moves etc.

    One extra piece of info on this: the infamous "Fool's mate" occurs after only TWO moves by each player, one of the reason that ICC has the "really brief" cut-off set so low, i.e. ONE move by each player....<snip>...
    If I understand this correctly, Quit not counted as loss, only applies to really short games as per your definition.

    Quote Originally Posted by TRgEOff View Post
    ...<snip>...Actually I think the scope for this type of error with Wifight is so low, that I personally don't think any accommodation is necessary.

    As to "undo" or "takeback" moves for "stupid" moves, in general I don't like them, but if implemented I think that it has to be something that both players "agree" to before the game. From my years of experience on ICC, I found it gets too complicated trying to decide what counts as a "stupid" move, and how many moves can be undone etc etc. It's better to use the pain of one's stupid move as a spur to playing better next time.

    But, to help us all in these situations, I think there should be a web link in the About tab to an online grief counsellor!....<snip>...
    I have done em, a lot. I knew why I should not have damn moved the piece but too much thinking or just getting distracted gets me to pick them. Touch move as you say it. Oh well, no harm in asking.
  7. TRgEOff's Avatar
    Posts
    589 Posts
    Global Posts
    591 Global Posts
    #127  
    Quote Originally Posted by ronbo2000 View Post

    1.Ok, should I move these back then?

    2.If I understand this correctly, Quit not counted as loss, only applies to really short games as per your definition.

    3. I have done em, a lot. I knew why I should not have damn moved the piece but too much thinking or just getting distracted gets me to pick them. Touch move as you say it. Oh well, no harm in asking.
    1. Brennan's move on this one! (but I dont think it matters much at this stage, at least until some other games get implemented-- not to say there isnt a thriving Tic-Tac-Toe community out there thats just keeping a low profile!)

    2. Exactly

    3. It's my personal choice, some people have very different views on this. But even in tournaments for kids at elementary school level they enforce "touch move" etc, otherwise things get difficult to manage (there's always the jokes about announcing that you are "adjusting", ie moving one of your pieces back to the centre of it's square ). For the same reasons, I think basic tournament "style" should be the default mode of play, and this will be even more important as, hopefully, wifight's chess playing base gets bigger and bigger.

    cheers
    geoff

    Treo 650 1.20 LAP.
  8.    #128  
    I can stick to this thread for game mechanic type discussion. I did think of it being chess-specific but all kinds of games would probably benefit from a move-take-back option. It will definitely be something both sides have to agree to.

    I do want to allow people to quit with no penalty if it's almost immediate. Good point on Fool's mate taking only 2 moves.

    The triple repetition rule isn't implemented yet, but it will be.
  9.    #129  
    I just posted the 'official' 0.10 prc. It now handles REM sleep mode correctly so you won't lose your connection to the server when your Treo goes to sleep. This is now the official current version of the prc so older versions might start bugging you to upgrade soon...

    http://get.wifight.com/ just in case anybody forgot...
  10. #130  
    1) quit after only 1 move (both sides) is no penalty. quit after that it is the same as resign....wouldn't have to even change menu option, maybe if quit & move count > 1 then message can be other player won instead of player quit message.

    2) no take backs....way too cumbersome aggreeing to imo & I would much rather see programming effort go towards other areas

    3) to help with bumbled moves maybe when you click on a piece it lights up squares as it does now, but when you click where u want the piece to go instead of an X have the piece actually move there. it isn't official/uploaded to server untill you click on the move button like now. solves the mouse, errr, thumb error problem. could even have the moved piece prior to clicking on move button be red or inversed or ghosted or something making it stand out as not committed/moved officially yet.

    4) throw out past games cus u may have to boot the server? you mean to tell me you don't have the wifight game database(s) on a raid 5 or sans with redundent real time mirroring at a secure facility 60 miles away, backed up nightly with mag tapes being couriered to a third offsite storage facilty under guard? gheeesk, what kind of outfit you running here anyway? yeah, save previous games if EASY to do so, otherwise clean break it is!

    5) if a person hasn't moved in x days (suggest 30) & it is past the first move, they forfeit the game, if only 1 move quit w/o penalty

    6) if going point system 1 pt win, 1/2 pt draw, 0 points loss.
    Last edited by boberang; 09/20/2007 at 08:56 PM.
  11. #131  
    Quote Originally Posted by boberang View Post
    6) if going point system 1 pt win, 1/2 pt draw, 0 points loss.
    I think it would be easier if the point system was 2 pts win, 1 draw and 0 loss...if he decides to do it that way.
    Sprint Palm Pre
  12. TRgEOff's Avatar
    Posts
    589 Posts
    Global Posts
    591 Global Posts
    #132  
    I'm suddenly unable to submit any turns. I get a message saying 'Script abort: coroutine: (string '?'): 10: attempt to compare nil with number'.
    I don't know what triggered it. I dl'd the new 0.10 but still the same. restoring preferences - no effect.

    games are on hold!
    cheers
    geoff

    Treo 650 1.20 LAP.
  13. TRgEOff's Avatar
    Posts
    589 Posts
    Global Posts
    591 Global Posts
    #133  
    actually its only for new games that I start myself. ongoing games r ok.
    geoff

    Treo 650 1.20 LAP.
  14. #134  
    Actually for me this problem happens on two of my on going games (one for Trgeoff and ssrjazz). The other older games are fine. I started noticing this yesterday afternoon Pacific time sometime after Brennan said he posted the 0.10 version.
    Last edited by ronbo2000; 09/21/2007 at 10:59 AM.
  15. #135  
    I just created a game vs chessbot & same issue (vs .10) with not being ale to submit a move.
  16. #136  
    hey, if going the route of non-point for wins & 1/2 points for draw what about:

    win: 2 * op win %
    draw: 1 * op win %
    loss: 0 pts

    this way quality of win comes into play. player 1 is 22-4 (0.846) & player 2 is 10-11 (0.476) so:

    player 1 wins: gets 2 * .476 or .952 points.
    player 2 wins: gets 2 * . 846 or 1.692 points.
    draw: plyr 1 gets .476 & player 2 gets .846 points.

    quality of win (or draw) is then taken into consideration for total points.

    courae, could implement however they do chess ratings too.
  17.    #137  
    Quote Originally Posted by boberang View Post

    4) throw out past games cus u may have to boot the server? you mean to tell me you don't have the wifight game database(s) on a raid 5 or sans with redundent real time mirroring at a secure facility 60 miles away, backed up nightly with mag tapes being couriered to a third offsite storage facilty under guard? gheeesk, what kind of outfit you running here anyway? yeah, save previous games if EASY to do so, otherwise clean break it is!
    Well actually the server is running on RAID 1 and the gamedb is backed up to Amazon S3 offsite storage periodically. So the games don't go away if I reboot the server, don't worry. But sometimes when you're programming something, data formats change and it's easier to start again than try to migrate old data. I'll (try to) be sure to announce any resets, and make them few and far between. Ideally it will be never but I can't promise that, at least until beta or something.
  18.    #138  
    Quote Originally Posted by TRgEOff View Post
    I'm suddenly unable to submit any turns. I get a message saying 'Script abort: coroutine: (string '?'): 10: attempt to compare nil with number'.
    I don't know what triggered it. I dl'd the new 0.10 but still the same. restoring preferences - no effect.
    It's a server-side thing; it's not you. I'm debugging it now. I think I fixed it but it only appears sporadically so I'm not 100% sure. Should be good now though.
  19. #139  
    well, got a new game invite from jazz so tried my game vs chessbot & was making moves again so good.....untill chessbot castled. king moved over two squares but rook didn't budge. now when I move it says piece not on square, script aborted.

    hey, good move on amazon s3 btw, that is an amazing & cheap service for what you get.
  20. TRgEOff's Avatar
    Posts
    589 Posts
    Global Posts
    591 Global Posts
    #140  
    yes as boberang says. if u play versus yourself, in any type of castling the rook just sits there.

    turn submission ok.
    geoff

    Treo 650 1.20 LAP.
Page 7 of 9 FirstFirst ... 23456789 LastLast

Posting Permissions