Sudden death in UCI?

[Peter Fendrich]

In the UCI spec I found the following under the go-command

* movestogo <x>
there are x moves to the next time control, this will only be sent if x > 0, if you don't get this and get the wtime and btime it's sudden death

What is sudden death?

/Peter

[Laurens Winkelhagen]

Hi Peter,

Sudden death is the time control where the mps (moves per session) are infinite, in effect that means that there is only one session. e.g. 40 minutes per side for the whole game.

This as opposed to 40 minutes for 40 moves;-)

I'm actually not sure of incremental time controls fit in this classification of sudden death. (e.g. 2 minutes per side for the whole game + 2 seconds per move).

Regards, Laurens.

[Peter Fendrich]

Hi Peter,

Sudden death is the time control where the mps (moves per session) are infinite, in effect that means that there is only one session. e.g. 40 minutes per side for the whole game.

This as opposed to 40 minutes for 40 moves;-)

I'm actually not sure of incremental time controls fit in this classification of sudden death. (e.g. 2 minutes per side for the whole game + 2 seconds per move).

Regards, Laurens.

Thanks,
it's what I thought.

Your example with 2 minutes + 2 seconds per mover is actually the Fisher clock, isn't it...
But there other strange combinations in the go command that doesn't make sense. For instance "go mate 6 depth 5" is nonsense to me and "go movetime x wtime y" is another one. I hope that the GUIs never will send such commands!
/Peter

[Ingo Bauer]

Hello

> For instance "go mate 6 depth 5" is nonsense to me and "go >movetime x wtime y" is another one. I hope that the GUIs never will send >such commands!

You are right, these combination are not very usefull and there are others (e.G.: "go ponder ... wtime ..." ...). I checked with the Shredder Classic GUI and it is impossible there to get these commands in one string.

But do not forget these sentence out of the UCI description:

* if the engine receives a command which is not supposed to come, (...) it should also just ignore it.

So after a "mate 6" or a "movetime x" ignore any other depth or time command.

Bye
Ingo

[Peter Fendrich]

So after a "mate 6" or a "movetime x" ignore any other depth or time command.

Hello Ingo,
For me "go depth x movetime y" makes sense! The search will stop if any of them is reached.
This is a good example of the confusion created by lack of protocol standard. Either you or me can say what's right or wrong here.
/Peter

[Ingo Bauer]

So after a "mate 6" or a "movetime x" ignore any other depth or time command.

Hello Ingo,
For me "go depth x movetime y" makes sense! The search will stop if any of them is reached.
This is a good example of the confusion created by lack of protocol standard. Either you or me can say what's right or wrong here.
/Peter

Hello

I was refering to your two examples:

For instance "go mate 6 depth 5" is nonsense to me and "go movetime x wtime y" is another one.

These two examples definately do not make sence.

Now you combined these two, which is something completly different.
If a GUI sends "go depth movetime y" and your engine can deal with it your are free to accept it. The sentence is: "if the engine receives a command which is not supposed to come...". I do not see a problem or any confusion with these two commands because these two are not contrary while the others are. (Ich lasse mich gerne eines besseren belehren!)

Something different is to find a GUI that is doing it like it is done in your example. Without making a complete search I would say, that right now there is NO UCI GUI that can send any of your examples.

Btw: Talking about " a good example of the confusion created by lack of protocol standard" is a real irony in the WB-forum!

Bye
Ingo