Unofficial World Championship Of Computer Programs In International Draughts 2021

[Krzysztof Grzelak]

And of course the number of points in the tournament. Do you have to write understandable things on the website.

[clp]

And of course the number of points in the tournament. Do you have to write understandable things on the website.

Yes of course. That is the whole point of writing. In the Netherlands for tie-break regulations "the understandable thing" is SB for Round Robin tournaments. For non Round Robin tournaments it is WP (=Buchholz / Solkoff) followed by SB.

It is perfectly possible to have these settings in Arbiter Pro.

In the distant past (1900) it was considered to replace the score with SB.

In chess (Fide) the "understandable thing for Round-Robin tournaments is "Direct encounter".

To avoid confusion, list all tie-break criteria in order of priority.

[Krzysztof Grzelak]

Yes of course. That is the whole point of writing. In the Netherlands for tie break regulations "the understandable thing" is SB for Round Robin tournaments. For non Round Robin tournaments it is WP (=Buchholz / Solkoff) followed by SB.

It is perfectly possible to have these settings in Arbiter Pro.

In the distant past (1900) it was considered to replace the score with SB.

For you and others, the page has changed regarding the place of the program in the tournament.

[clp]

Hello Krzysztof,

For you and others, the page has changed regarding the place of the program in the tournament.

Do not forget to inform the participants. Finally, you change the conditions afterwards.

[Joost Buijs]

Hi Krzysztof,

There is no rule that states when you have only one move (a capture) that you have to perform this move immediately. In this case I let the engine calculate with reduced time to get a reasonable ponder move.

I agree that a GUI could make life more easy, but writing a GUI for just one tournament per year is out of the question. A tournament like this could be run fully automatically without human intervention (for instance look at TCEC). Of course you need a universally accepted interface standard to do this. Unfortunately there is no standardization at all in the computer draughts world, so this will never happen.

Joost

I mean, then the engine has less time to think about the rest of the movements. If you calculate all the movements the engines needs to make immediately, it would save about 2-3 minutes. And that's a lot.

It's is not as simple as this, you are right when an engine doesn't ponder and doesn't use a hash-table. By letting the engine calculate on these "single moves" the quality of the ponder-move will be better and the hash-table gets filled with useful information, both these things can save time for a subsequent search.

I give the engine 25% of the average move-time when there is only a single move, this means that 4 of these moves take the same time as 1 "normal" move, I can't imagine that this will cost an extra 2-3 minutes with the time-controls used by your tournament.

Of course for a human player it will look strange when you start thinking about a move when there is only one move to choose from.

[clp]

Hello Krzysztof,

For you and others, the page has changed regarding the place of the program in the tournament.

The page reads:

the place of the program in the tournament is determined by the greatest number of points scored in the tournament and Sonneborn - Berger

However, the order reported is not according to score, S.B.:

Place No. Program                        Total  Wins S.B.
1       4 NED Damage 16.1 NNUE             54    20  1589
2       6 NED Dragon Pro 4.6.2             50    17  1398
3      15 NED Sjende Blyn (August24th2021) 50    16  1448
4      10 USA Kingsrow 1.63                50    16  1444
5      14 FRA Scan 3.1                     50    16  1437
6       1 NED Ares 1.43c NNUE              50    16  1433

[Krzysztof Grzelak]

Which is wrong again

[clp]

Which is wrong again

Hello Krzysztof,
Is it possible to teach me the correct sequence and explain the logic of that sequence?
Help is appreciated.

[Krzysztof Grzelak]

Sorry, what's going on, because I don't really understand.

[Jelle Wiersma]

I personally think SB is the best, nevertheless, it is clear that Damage has "gold". Congratulations to Bert, and thanks to Krzysztof for running this 34-rounds tournament!

[hendrikv]

Hi Krzysztof,

You mention Points and SB in the rules, but you do not mention the number of victories as a criteria.
This means that according to your rules Sjende Blyn is second and Dragon third, because Dragon has lower SB then Sjende Blyn.

You should also mention number of wins as a criteria in the rules.

Greetings,
Hendrik

[Krzysztof Grzelak]

This is what the DraughtsArbiter Pro program showed me.

[clp]

This is what the DraughtsArbiter Pro program showed me.

Hello Krzysztof,

The settings of DraughtsArbiter Pro are inconsistent with the tournament regulations.

DraughtsArbiter: score, number of wins, S.B.
Your regulations: score, S.B.

In my opinion changing the regulations after completions of the tournament, as suggested by Hendrik, is not a good idea. Unless all participants agree.

[Krzysztof Grzelak]

Hello Krzysztof,

The settings of DraughtsArbiter Pro are inconsistent with the tournament regulations.

DraughtsArbiter: score, number of wins, S.B.
Your regulations: score, S.B.

In my opinion changing the regulations after completions of the tournament, as suggested by Hendrik, is not a good idea. Unless all participants agree.

I'm sorry but I don't really understand your suggestion.

[clp]

I'm sorry but I don't really understand your suggestion.

Hello Krzysztof,

DraughtsArbiter Pro program showed you the wrong order. How to correct it?

I am not an expert on DraughtsArbiter Pro, but it must be doable.