I am glad that it really got better in your rating list! Which I suposse used the same conditions for amyan 1.597 and 1.6.
I say this also because I did a mistake in my testing. I have to figure why yet but should be easy. Amyan runs a bit slower as uci engine under Arena. Faster as winboard engine. (Perhaps too much engine output considering the fast games I used!? or the hashfull % calculation is slow!? tonight I will see.)
When I tested I mixed winboard-or uci mode even for the same version of amyan. I have to repeat my games.
I am glad that it really got better in your rating list! Which I suposse used the same conditions for amyan 1.597 and 1.6.
I say this also because I did a mistake in my testing. I have to figure why yet but should be easy. Amyan runs a bit slower as uci engine under Arena. Faster as winboard engine. (Perhaps too much engine output considering the fast games I used!? or the hashfull % calculation is slow!? tonight I will see.)
When I tested I mixed winboard-or uci mode even for the same version of amyan. I have to repeat my games.
please have a very very nice day. My best wishes.
Hi Antonio,
1.597 played many games in the past. So I'm not sure that the rating can be compared yet. In my tournament I use always UCI mode.
Hi. Amyan 1.61 is only up to 6% faster during a game than 1.6 depending on hash size. If hash is very big then the diff. is close to zero. Testing 1.61 for separate when already you tested 1.6 may be a little waste of cpu.. but you know what you do.
I had to realease 1.61 because analisis mode under winboard was not working.
adieguez wrote:Amyan runs a bit slower as uci engine under Arena. Faster as winboard engine. (Perhaps too much engine output considering the fast games I used!? or the hashfull % calculation is slow!? tonight I will see.)
When I tested I mixed winboard-or uci mode even for the same version of amyan. I have to repeat my games.
Hi Antonio,
under ChessGUI, the uci version of Amyan 1.61 doesn't seem to work correctly. I've reinstalled as a winboard engine and it runs fine.
for the record, the hahsfull calculation was slow and I never noticed before as I most of the time I use winboard mode.
I remember when amyan 1.59 was a bit weaker than amyan 1.57 in your list... also in another list it got weaker.. this is so weird, so much time has passed and only now the little mistery was resolved. I even at that time asked help of testers, but didn't get to a conclusion, off course I was confused.