It seems we are in front of LittleBlackBook! Who says adaptive general purpose movement and targeting can't match hard-coded behaviour? =)
Well, I'm curious to see where an updated LBB would rank... :-) But still, amazing work, huge congrats dudes!
Yeah, considering it was at about 80% in micro, I think 82% in nano wouldn't be too hard to believe. Oh well, I guess we'll have to keep improving Yatagan then =)
I had a quick look at the archives. In July 2011 when the current LBB 1.69e version came out it was at 84.19% APS!
So that is how good LBB is when its tables are up to date. Scary that after 2 years of inattention and with 30 bots showing a > 20 KNNPBI it is still at #3 with a 76.49% APS.
So even without any code updates just adding those missing bots to its tables would most likely see it reclaim the throne.
And if LBB 2.0 is based on Yatagan, then you'll really have your work cut out for you... :-P
But seriously, a 2 APS lead on the next non-LBB bot is crazy awesome.
It looks like Yatagan broke into the 80's! It now holds the nano throne by a margin of over 1.6% APS. And, it has less than 2% PWIN between it and Fuatisha.
Just for fun, let's try a version with
integer initialized to 64, and with division-based reduction instead of decrementation. I'm pretty sure it would hurt APS, but hey, it's worth a shot.