also, tests are no longer executable files. I guess I no longer need
that particular hack...

I'm not using the ideal estimate of the integral length since I'm doing
the integral with the 1D spectrum, but I think this should be close
enough.

computation now takes place for all watching CPUs, not just computing
one.

and modified plain test to work with it. other tests still need to be
updated though.

i.e. remove some redundant bits