So I got 5 and was very close 4 times.. About 90 minutes of playing..
Nice. This is way better than I have done, so I am still bad! Good players should have no problem making some blocks.
I have a strange idea: maybe we need to create a coefficient of warp.. About 1.2 - 1.5.. I mean that playing with some reverse take for human about 1.2-1.5*time to reach coin.. But bot public the solutions fast and antiwarp will arrive to point slowler and slowler..
Not sure if I followed any of this. I understand what you are saying about a human taking some multiple of time because of rewind, but not what you are proposing we do. Can you clarify?
Do you mean that instead of targeting anti-warp to be "1:1 with TT" we should have it target to be some value like "1.5:1 with TT"? If this is what you are proposing I am not so sure if it is a good idea, since it re-introduces the delta between the clock synchronizations that makes a difficulty time warp possible in the first place. If we did such a "non 1:1" target I think it might defeat the *intended* purpose of the anti-warp for the sake of a better anti-bot, which could potentially be a serious mistake. Remember that the anti-warp is first and foremost a security fix, and nothing to do directly with bots. The fact that it does serve to counter this bot (because this bot is still dependent on a large map iteration step) is a nice side effect, but not the primary intent. I don't want to do something that puts the primary intent of anti-warp (that being mitigating the warp effect) at risk by what would be effectively an intentional reintroduction of an explicit warp!
I hope I am just misunderstanding what you mean.