Thanks for the tests, Patrick. I think the author reported it on his own page that the newer version is atleast around 30 elo better than the previous one, even in faster bullet time controls.
It must be something wrong in the time management, Colossus lose 3 games on time. It's never happend with 2007d.
Regards,
Patrick
It is very slow and inevitably gets into time trouble in longer games without increment.
The author only tests with bullet time controls and probably didn't notice.
It's not the speed of the game as such but the lack of any Fischer time increment that's causing the problem. Unfortunately I always test with time increments so didn't notice the problem when there isn't any.
The driver behind using the time more aggressively was that I was sick of seeing Colossus lose still having maybe 30%+ of its clock time remaining.
Hence I biased it to use much more time early (and hence play better early). It then leaves less time for the endgame, but at least you've survived that far! With even a small time increment it isn't so much of a problem, but with no time increment... Ooops!
first CEGT results with 40/3 (repeated time control) show an improvement of 35 points compared with 2007d.
But - as usual - much tooooo less games for a (halfway) reliable statement.
Till our next update next weekend 500 games will be played giving a better comparison (hopefully....)