CRoberson wrote:The 2011 Fifth Annual ACCA World Computer Rapid Chess Championships is tentatively scheduled for the weekend of July 23/24 2011. It is tentative only to give others a chance to tell me of a serious timing conflict. After looking around the usual tournaments, I couldn't find a conflict.
The time control and rules are mostly as they have been. The only change may be that all participants provide a binary of the version entering. This is for tracking purposes and ease of catching clones. I haven't completely decided on that rule yet. It is open for discussion.
I will have the web pages up this week. You can respond here or PM me if you have any questions, comments or requests.
Hi Charles,
Rule: "8.Participants are allowed 2 disconnects due to accidents"
(The underlining is mine)
May we assume that disconnecting on purpose for changing books/engine version etc is not allowed during a game?
BTW: Good luck!
I guess it is if the operator accidently forgot to set the book....
Nahh, it is very unlikely that this could happen... isn't it?
The idea is 2 disconnects, because the network connection dropped or PC failed or ....
However, it is allowed to use one of your disconnects to turn on a book that you forgot to turn on. You might disconnect to turn on kibitzing if you forgot to turn it on. In fact, you should do so to get kibitzing working.
As far as books go, you can change books between rounds. It would be very difficult to do so during a game: 19 moves will go by in 1 second with two books sticking to well known lines.
You can change engine versions between rounds.
You can never change engines: that is he that enters X has to play X the whole tournament. He can have different versions of X, but can not play A or B or C or .... Y or Z. If I were to register Telepath, then I must play some version of Telepath for the entire tournament. I would not be allowed to switch to NoonianChess or Ares during the tournament.
I would recommend not to adopt any rules that cannot be enforced. If I disconnect to change book, and say my connection dropped, how do you want to detect it?
I can add that for those using the latest WinBoard version there is no need to disconnect in order to:
1) Switch on or off kibitzing. Just tick or untick the 'Auto Kibitz' checkbox in the ICS Options menu dialog.
2) Switch on the GUI book. Just untick the checkbox 'First has own book' in the Common Engine Options dialog, tick 'Use book' there, and provide a valid filename for it (if that is already done). In principle you can also change book this way, but as Charles pointed out, you are likely to be too late, unless the wrong book gets you out of book really early, or the opponent is out of book early.
3) Switch on a book of your UCI engine using its book through Polyglot. The relevant settings (similar to those described above) are in the Polyglot section of the Engine #1 Settings dialog.
4) Switch on the book of your engine itself if it supplies the proper options to do so (such as Fruit, for example), through the Engine #1 Settings dialog. If your engine does manage its own book, whether it is WinBoard or UCI, it is of course highly recommended that you make that configurable through the GUI interface rather than through some obscure ini file only read at startup.
A small patch in WinBoard would also allow you to change engine (version) without disconnecting, between or during games. You can already switch engine during a local game. The only reason it does not work now is that I gray out the 'Load Engine' menu item in ICS mode. (Changing from Zippy to non-Zippy mode or back might be a bit harder.) I was not sure this was desirable.
And a completely different issue: ICS can have faults to, such as FICS during CCT. The only way to cure the ICS when it hung was to disconnect, and there is nothing I can do against it from WinBoard. Will such disconnects be counted against the participant? An early detect of 'ICS hanging' can be obtained by using the -keepAlive option, but it would be to no avail if the only way to cure it would make you forfeit the game anyway.
Due to the problems with FICS during the last CCT, we will not consider FICS as a host. We will continue to utilize ICC. We've not had problems there.
Disconnects between rounds do not count as penalties. Disconnects that don't interrupt a game are not penalized. There is an ICC command that allows us to track disconnects with time stamps.
If you can turn on kibitz or books without disconnecting that is fine. If it takes a disconnect to turn kibitz on then you should do so if it would be the first of the game for your engine.
Usually, people disconnect between rounds if they want to switch engine versions or reboot the computer ... I don't have a problem with a GUI option that allows changing versions without a disconnect.
If you can come up with a tool that allows a computer reboot without a disconnect, I am fine with that too.
CRoberson wrote:Due to the problems with FICS during the last CCT, we will not consider FICS as a host. We will continue to utilize ICC. We've not had problems there.
Disconnects between rounds do not count as penalties. Disconnects that don't interrupt a game are not penalized. There is an ICC command that allows us to track disconnects with time stamps.
If you can turn on kibitz or books without disconnecting that is fine. If it takes a disconnect to turn kibitz on then you should do so if it would be the first of the game for your engine.
Usually, people disconnect between rounds if they want to switch engine versions or reboot the computer ... I don't have a problem with a GUI option that allows changing versions without a disconnect.
If you can come up with a tool that allows a computer reboot without a disconnect, I am fine with that too.
You may want to put in some exception that deals with the situation where there are multiple rapid disconnects (which should count as a single one in my mind). This has happened to me a number of times on ICC, where some script error logs me in and out rapidly (as in several times in a second). Or just put in some wiggle room, such as making it clear the tournament director can ignore a disconnect under special circumstances if desired.
CRoberson wrote:Due to the problems with FICS during the last CCT, we will not consider FICS as a host. We will continue to utilize ICC. We've not had problems there.
Disconnects between rounds do not count as penalties. Disconnects that don't interrupt a game are not penalized. There is an ICC command that allows us to track disconnects with time stamps.
If you can turn on kibitz or books without disconnecting that is fine. If it takes a disconnect to turn kibitz on then you should do so if it would be the first of the game for your engine.
Usually, people disconnect between rounds if they want to switch engine versions or reboot the computer ... I don't have a problem with a GUI option that allows changing versions without a disconnect.
If you can come up with a tool that allows a computer reboot without a disconnect, I am fine with that too.
You may want to put in some exception that deals with the situation where there are multiple rapid disconnects (which should count as a single one in my mind). This has happened to me a number of times on ICC, where some script error logs me in and out rapidly (as in several times in a second). Or just put in some wiggle room, such as making it clear the tournament director can ignore a disconnect under special circumstances if desired.
-Sam
I agree. I suggest "number of turns in which you disconnected".
IMO, turning on kibitz w/o disconnecting, should still count as one disconnect. However, this is a never ending problem. Some engines choose not to fix their kibitz bugs and refuse to use winboard with auto kibitz.
CRoberson wrote:Due to the problems with FICS during the last CCT, we will not consider FICS as a host. We will continue to utilize ICC. We've not had problems there.
Disconnects between rounds do not count as penalties. Disconnects that don't interrupt a game are not penalized. There is an ICC command that allows us to track disconnects with time stamps.
If you can turn on kibitz or books without disconnecting that is fine. If it takes a disconnect to turn kibitz on then you should do so if it would be the first of the game for your engine.
Usually, people disconnect between rounds if they want to switch engine versions or reboot the computer ... I don't have a problem with a GUI option that allows changing versions without a disconnect.
If you can come up with a tool that allows a computer reboot without a disconnect, I am fine with that too.
You may want to put in some exception that deals with the situation where there are multiple rapid disconnects (which should count as a single one in my mind). This has happened to me a number of times on ICC, where some script error logs me in and out rapidly (as in several times in a second). Or just put in some wiggle room, such as making it clear the tournament director can ignore a disconnect under special circumstances if desired.
-Sam
The disconnect rule pertains to disconnects that actually interrupt the game and that is stated in the rules page. The TD's at ICC have access to info that tells if a disconnect actually interrupted the game.
Scenario 1: The game has started and you zoom through the book and it disconnects. This counts as a disconnect.
Scenario 2: In some attempt to reconnect, you disconnect again before the game can be resumed. This does NOT count as a disconnect.
Scenario 3: In some attempt to reconnect, you resume game and then loose connection again. This counts as a disconnect.
Scenario 4: The round starts and you disconnect and reconnect before your game is started. This does NOT count as a disconnect.
Scenario 5: The round starts and you are several minutes late for the round. This does NOT count as a disconnect.
I think this covers the bases. If anybody has a question/suggestion lets hear it.
CRoberson wrote:Due to the problems with FICS during the last CCT, we will not consider FICS as a host. We will continue to utilize ICC. We've not had problems there.
Disconnects between rounds do not count as penalties. Disconnects that don't interrupt a game are not penalized. There is an ICC command that allows us to track disconnects with time stamps.
If you can turn on kibitz or books without disconnecting that is fine. If it takes a disconnect to turn kibitz on then you should do so if it would be the first of the game for your engine.
Usually, people disconnect between rounds if they want to switch engine versions or reboot the computer ... I don't have a problem with a GUI option that allows changing versions without a disconnect.
If you can come up with a tool that allows a computer reboot without a disconnect, I am fine with that too.
You may want to put in some exception that deals with the situation where there are multiple rapid disconnects (which should count as a single one in my mind). This has happened to me a number of times on ICC, where some script error logs me in and out rapidly (as in several times in a second). Or just put in some wiggle room, such as making it clear the tournament director can ignore a disconnect under special circumstances if desired.
-Sam
The disconnect rule pertains to disconnects that actually interrupt the game and that is stated in the rules page. The TD's at ICC have access to info that tells if a disconnect actually interrupted the game.
Scenario 1: The game has started and you zoom through the book and it disconnects. This counts as a disconnect.
Scenario 2: In some attempt to reconnect, you disconnect again before the game can be resumed. This does NOT count as a disconnect.
Scenario 3: In some attempt to reconnect, you resume game and then loose connection again. This counts as a disconnect.
Scenario 4: The round starts and you disconnect and reconnect before your game is started. This does NOT count as a disconnect.
Scenario 5: The round starts and you are several minutes late for the round. This does NOT count as a disconnect.
I think this covers the bases. If anybody has a question/suggestion lets hear it.
I am thinking of scenario 3, which can happen multiple times in rapid succession because of ICC auto-restarting game when your script is logging you on and off due to some bug. In my opinion all those restarted games should just count as 1 disconnect, but you can of course count them however you want to.
CRoberson wrote:Due to the problems with FICS during the last CCT, we will not consider FICS as a host. We will continue to utilize ICC. We've not had problems there.
Disconnects between rounds do not count as penalties. Disconnects that don't interrupt a game are not penalized. There is an ICC command that allows us to track disconnects with time stamps.
If you can turn on kibitz or books without disconnecting that is fine. If it takes a disconnect to turn kibitz on then you should do so if it would be the first of the game for your engine.
Usually, people disconnect between rounds if they want to switch engine versions or reboot the computer ... I don't have a problem with a GUI option that allows changing versions without a disconnect.
If you can come up with a tool that allows a computer reboot without a disconnect, I am fine with that too.
You may want to put in some exception that deals with the situation where there are multiple rapid disconnects (which should count as a single one in my mind). This has happened to me a number of times on ICC, where some script error logs me in and out rapidly (as in several times in a second). Or just put in some wiggle room, such as making it clear the tournament director can ignore a disconnect under special circumstances if desired.
-Sam
The disconnect rule pertains to disconnects that actually interrupt the game and that is stated in the rules page. The TD's at ICC have access to info that tells if a disconnect actually interrupted the game.
Scenario 1: The game has started and you zoom through the book and it disconnects. This counts as a disconnect.
Scenario 2: In some attempt to reconnect, you disconnect again before the game can be resumed. This does NOT count as a disconnect.
Scenario 3: In some attempt to reconnect, you resume game and then loose connection again. This counts as a disconnect.
Scenario 4: The round starts and you disconnect and reconnect before your game is started. This does NOT count as a disconnect.
Scenario 5: The round starts and you are several minutes late for the round. This does NOT count as a disconnect.
I think this covers the bases. If anybody has a question/suggestion lets hear it.
I am thinking of scenario 3, which can happen multiple times in rapid succession because of ICC auto-restarting game when your script is logging you on and off due to some bug. In my opinion all those restarted games should just count as 1 disconnect, but you can of course count them however you want to.
-Sam
I don't recall any auto-resume. I recall having to type "resume" myself on ICC.
Nahh, it is very unlikely that this could happen... isn't it?
