Your network blocks the Lichess assets!

lichess.org
Donate

Tofiks v1.0.0. is finally here.

Is source code available under a free software license?

Is source code available under a free software license?

@Toadofsky said in #2:

Is source code available under a free software license?
Sorry missed your comment @Toadofsky, yes the source is available and is licensed as GPL https://github.com/likeawizard/tofiks

@Toadofsky said in #2: > Is source code available under a free software license? Sorry missed your comment @Toadofsky, yes the source is available and is licensed as GPL https://github.com/likeawizard/tofiks

Could you explain, since you are obviously intent on sharing not only raw source code as its own documentation, but also sharing the concepts and design trajectory (that is a compliment), how does the transposition table changes which nodes along the candidate subtrees are going to be tested for prior presence in TT. Does this happen only at leaf nodes, or could it happen anywhere? Can interior nodes trigger a TT hit? things like that.

And now a long standing concern of mine about chess engine, and the opportunity that your very open design, thinking and source code project could allow:

I am curious about getting the most information out of a single root search. I understand that is not in the usual engine tournament specifications and might be detrimental to speed, and hence ELO (given the pool of engine which are compete under the same specifications, some of which are about chess tournaments with clocks, others from history of engine development).

I understand also, that the temptation would be great, and the pressure high, not to put too much of you work hours toward full node typology statistics being possible output per some mode of you engine.. I do cherish the hope that you information sharing concern (all the way to concept and design, which is manifest in your blog series), will be maintained in the engine verbose mode output, but I understand that this is not code optimization practice, if the specifications do not require it.... priorities being what they are.

So to what degree would you push by yourself, without having any specification requirement from tournament about it, the development of such an analytic or chess land debug mode for your engine?

if there was an engine out there that could show the way to some chess-meaningful example output transparency level, allowing proper analysis by users in the context of chess position analysis (it could be another realm of competitoin, which is best analytical tools, with margins of errors display, etc.... thinking of any engineering development of measure instruments type of features)... That would be great innovation.

Could you explain, since you are obviously intent on sharing not only raw source code as its own documentation, but also sharing the concepts and design trajectory (that is a compliment), how does the transposition table changes which nodes along the candidate subtrees are going to be tested for prior presence in TT. Does this happen only at leaf nodes, or could it happen anywhere? Can interior nodes trigger a TT hit? things like that. And now a long standing concern of mine about chess engine, and the opportunity that your very open design, thinking and source code project could allow: I am curious about getting the most information out of a single root search. I understand that is not in the usual engine tournament specifications and might be detrimental to speed, and hence ELO (given the pool of engine which are compete under the same specifications, some of which are about chess tournaments with clocks, others from history of engine development). I understand also, that the temptation would be great, and the pressure high, not to put too much of you work hours toward full node typology statistics being possible output per some mode of you engine.. I do cherish the hope that you information sharing concern (all the way to concept and design, which is manifest in your blog series), will be maintained in the engine verbose mode output, but I understand that this is not code optimization practice, if the specifications do not require it.... priorities being what they are. So to what degree would you push by yourself, without having any specification requirement from tournament about it, the development of such an analytic or chess land debug mode for your engine? if there was an engine out there that could show the way to some chess-meaningful example output transparency level, allowing proper analysis by users in the context of chess position analysis (it could be another realm of competitoin, which is best analytical tools, with margins of errors display, etc.... thinking of any engineering development of measure instruments type of features)... That would be great innovation.