Puzzle Description:

#1: Aurelian Ginkgo (AurelianGinkgo) on Jan 16, 2020

I still have no idea what's going on. I'm counting on #3, Brian.

Radiator temperature.

Hm. The site automatically marked the puzzle as "Definitely logically solvable with line logic alone," but it sure seems to me to need lookahead. I don't remember seeing the system get that wrong before.

Oh, this is a light you don't want to see go on while you are driving. Fun solve, Brian.

I agree, Valerie.

Turn the heater on. It will help cool the radiator fluid.

@Varlerie -- it's decidedly nontrivial line logic I think. The 2-3-3-2 lets you fill in 2 extra squares at the bottom. r19 c9/c12 are black from c9/c12 and the starting line logic, and it turns out c8/c13 have to be black as well, as otherwise you couldn't fit a 2-3-3-2 into the left or right 12.

My breakthrough was realizing that c10 or/and c11 in r20 must be black, since c10 and c11 in r19 cannot both be black (as otherwise there would be at least 4 black squares in a row in r19 that can only have at most 3 black in a row). I agree, doesn't seem complicated enough to call it moderate lookahead, but definitely less trivial than the usual LL.

Michael - Yours is definitely moderate lookahead as you use more than one row at a time.

I agree with Andrew.

After LL,
Look at the pixel at r19c12:
> If you crowd all the clues as far left as possible, it will be the leftmost pixel of the last 2 in that row, so c13 must be black.
> If it is not the last 2, then it must be one of the 3s. And if so, it must extend into c13 or it would run into the pixel in c9 making a block of four.
= So either way, r19c13 must be black.

Then simple LL to finish

The same works (mirrored) for c8.

I won't add anything beyond I concur on the logic.

Great puzzle Brian!

Goto next topic

You must register and log in to be able to participate in this discussion.