Hi,
your implementation runs about 1000 times faster than what I tried before. That's great.
I have started playing around with a very nice wooden puzzle, a pentacube
with 125 voxels to be filled with 25 identical 5-voxel pieces,
see https://glocq.github.io/en/blog/20250428/.
I have extracted the code from there and played around with it, see: https://github.com/mdrslmr/wpp
Later I found your implementation and created a Demo according to the one you provided only for the pentacube. This runs about 1000 times faster. Then I have integrated your algorithm into my
version. This is still about 300 times faster. It looses some time for the output creation, I guess.
It would be cool, if your code could be provided as a package on hackage, if you could be bothered?
Hi,
your implementation runs about 1000 times faster than what I tried before. That's great.
I have started playing around with a very nice wooden puzzle, a pentacube
with 125 voxels to be filled with 25 identical 5-voxel pieces,
see https://glocq.github.io/en/blog/20250428/.
I have extracted the code from there and played around with it, see: https://github.com/mdrslmr/wpp
Later I found your implementation and created a Demo according to the one you provided only for the pentacube. This runs about 1000 times faster. Then I have integrated your algorithm into my
version. This is still about 300 times faster. It looses some time for the output creation, I guess.
It would be cool, if your code could be provided as a package on hackage, if you could be bothered?