RealDebugMonkey
commented
9 years ago I'm assuming you're referring to the new grazer. (The old grazer probably won't be touched much now, and will probably be removed when/if we make the new grazer a little better at the early game.)
The new grazer should already be wary of standard split-kill situations.
The remerging scenario is a little trickier since the server doesn't actually tell us which blobs belong to which user. All blobs have their own ID, name, and color. The best we can do is to assume that cells that are the same color, have the same skin, and have the same name all belong to the same player. I'd hate to make the code overly complex for this, but I think it might be viable to artificially inflate the threat-level of a split blob.
Suggestion accepted. I'll try and make some improvements around this when I get some spare time.
Szigerd
Gjum
I am not a programmer or even speak Javascript, but from the code I see that you create a potential and then the grazer is riding on the Gradient. It seems that you handle all the blobs not ours the same way. This results imho in an underestimation of the threat of split blobs So it happen that the grazer is ignoring lets say the 2 95% blobs from the some opponent and then dying when the blobs reunite. or chasing the 60% green blob of 1 player where the other blob of that player is e.g. 200% and the grazer should not chase it. Perhaps you can chenge this. Thanks