TalkChess.com

Time forfeits are handled by inserting [Termination "timeout"] in the PGN header. However, CCRL scripts do not recognize them as such so we will have to edit all Banksia PGN files and locate time forfeits manually. For the scripts to recognize time forfeits, the information will have to be at the en...

This game was adjudicated as a win for white. Maybe a bit prematurely, I am not sure white would have won left on his own. I suggest adjudicating on tablebases only if drawn. And it would be good if adjudication methods would be separated, e.g. adjudicate a draw after 200 moves but do not adjudicate...

AdminX wrote: ↑

Sat Dec 14, 2019 7:51 am

The 1st sign of trouble was that it no longer detected any previous installed engines.

Same for me on a Win10 64-bit home edition.

Zenmastur wrote: ↑

Thu Dec 12, 2019 6:30 pm

a net ELO loss.

We lost Elo a very long time ago now...

Guenther wrote: ↑

Wed Dec 11, 2019 5:53 pm

BTW shouldn't the bmi compiles be quite a tad faster normally?

But they are a bit faster, aren't they? Anyway, I don't know, must depend upon the engine architecture.

Compiled a few flavours for 1.65, with a big caveat, because I had never built versions before, I cannot run here myself, which is a bit awkward. Would be nice if someone could try the bmi or bmi2 version, if those run at all and with what speed. The other two are are tested here and behave very si...

There is a small, new feature I would like to specially introduce to everyone here. This feature is used to extract and display logs of engines for tournament games. To use that, from the list of tournament games, right-click to popup menu and select "View engine-log of game n". This is a wonderful...

Terje wrote: ↑

Fri Dec 06, 2019 3:26 pm

Oh, I forgot the O3 flag I've reuploaded new versions with it. Hopefully these will be comparable to Günther's

That's more like it. Now the pext compile is about 4200 knps.

Terje wrote: ↑

Fri Dec 06, 2019 2:04 pm

I've added 3 windows compiles to the tag. Feedback on whether they work or not would be appreciated

Thanks Terje. All three work for me. Under Shredder 13 I measure the following approximate speeds:

nopop: 1250 knps

pop: 1360

pext: 1430

Günther's compile: 3420 knps

Weiss 0.7 released. This version mostly improves search by better move ordering and adding pruning heuristics. More details in the tag. https://github.com/TerjeKir/weiss/releases/tag/v0.7 That's fine. Unfortunately I am still not trained to compile it myself. May I ask for some Windows executables?...