Building backwards for maze algos?

After looking at a few maze algos playing, I have noticed that they build filters behind the line that their EMPs follow across. Note the 5 filters on line 18 that sawtooth builds on round 7.
What is the point/advantage of doing this? I feel like it would just be a waste of cores that could be saved or used for early encryptors

1 Like

I’m fairly certain that the purpose of those filters is to redirect any enemy information units that might make it through, so that they stay in range of the destructors for a bit longer and don’t immediately score.

Jup, I’m pretty sure they are meant to decrease vulnerability to ping cannons. This is actually an important tactical idea that I forgot to put in my KeKroepes_2.0 algorithm, so it keeps losing to ping cannons :""").

Interesting, makes sense. Wouldnt placing the filters in the front work well too, because they’d still take longer to get around the wall, or they’d go across your wall that has destructors on it?

There are several things to consider here. First of all building on the 1st row will not have the same effect as a backwards wall, in that it doesn’t necessarily result in the info units being in range of destructors longer. Then there is the issue of fighting and defending on the border: it is very susceptible to EMP lines, and encrypted attacks as shield have less time to decay when defending on the border.

There are probably many pros and cons that I didn’t mention, but the underlying point is that it is actually really difficult to find the appropriate response to a problem your algo has. I’ve had countless times where I was trying to patch a vulnerability, and as a side-effect created a whole bunch of new vulnerabilities :stuck_out_tongue:.

I’m not sure whether the backward wall is the best option, maybe @n-sanders knows as he currently has the top algo which happens to be based of a maze :wink: