Don Stewart wrote: ...
There may well have been changes to the strictness analyser that make some of the bangs (or most) unnecessary now. Also, its very likely I didn't check all combinations of strict and lazy arguments for the optimal evaluation strategy :)
I suspect the optimum details will change again when we get to GHC 6.8. Yes, I got bored trying different combinations too. A genetic algorithm that knocks out different combinations might be fun. The ones in rand seem to make the most difference and I decided the code was easier to read if I took them all out.
If it seems to be running consitently faster (and producing better Core code), by all means submit. I don't think this is a ghc bug or anything like that though: just overuse of bangs, leading to unnecessary work.
-- Don
It was consistently faster on my machine, but it would be interesting to compare with the run-times on Sterling's PPC machine. I'll have a play with Sterling's new program and report. Richard.