Steamhammer is top zerg on SCHNAIL
So, SCHNAIL believes I successfully updated Steamhammer back on 11 July, but it is still running the old version. It’s easy to tell, because of the orthogonal movement pattern and the jittering of Watch squad zerglings before burrow is researched. I’ve been too lazy to diagnose it and try again. Presumably I got something wrong in the zip file.
Nevertheless, I was surprised to see that Steamhammer is the #1 zerg on SCHNAIL as of now. Well, counting active zergs. The top zergs by elo are
1472 Steamhammer 1463 Chris Coxe (aka ZZZKBot) 1452 ZurZurZur 1440 Crona 1432 Monster 1384 Killerbot (aka Marian Devecka)
Chris Coxe is up there because of its rushy habits. I’m not sure why Monster and Killerbot are so far down; I would have expected them to do well against humans. Microwave is ranked above Steamhammer on BASIL, but below on SCHNAIL at elo 1280. In some cases, bots on SCHNAIL may be older versions. I know some bot authors update often on BASIL and rarely on SCHNAIL. Maybe they’ll read this post, update, and try to push Steamhammer back down where it belongs.
In other news, AIIDE 2021 registration is open. Steamhammer will be competing.
Trackbacks
No Trackbacks
Comments
MarcoDBAA on :
But I think that the large number of openings and learning and good play at the start of games helps SH vs humans. Less probable that someone finds an effective way to win all games by exploiting the AI.
Conversely SHs weaknesses in comparison to some top zerg bots later on (although I did not watch many games in general this year, i believe this to be true) might be less relevant because many human players just cannot find a way to defeat bot macro in general, if they do not find themselves in a good position after the early game.
As you said in the other post, you have to improve the mid and late game indeed I think. Early game, including variability, is strong. And the bot is bottlenecking less vs humans, because it can overcome some problems with bot macro.
Jay Scott on :
I guess that bots with 1500 elo despite having played ranked games is due to elo being reset. It would make sense to reset elo to 1500 when the bot is set to practice only. Or maybe all elos got reset at some point, and game counts didn’t.
MarcoDBAA on :
If that is the case, there could be a problem with comparability, because many bots (not only Zur³ for example) might just not have enough games after the reset, whenever it happened.
MicrDK on :
I have pointed this out many times, but there are apparently other priorities.
MicroDK on :
Also, as long as bot developers don't have access to our bot learning data, it's difficult to improve learning parts or improve builds. Only way is to watch all replays, which I don't have time for.
Actually, I started out to pick 7-8 openings for each matchup which are normal in human play just so the bot would play the normal meta. Maybe that plan should be changed. Maybe that is too easy for humans. Maybe I should add the new Crazy Zerg. :D