[Shootout-list] Bigloo matrix-norm #2

[Will M. Farr]

Hello,

Two notes about the bigloo matrix-norm benchmark:

1. You can just delete Bigloo #1.  It uses a sub-par algorithm 
(different from the one used in the problem description), which is 
*unbelievably* slow (I wrote it; I won't complain to see it go).

2. Bigloo #2 reports an error in the output, but it gives

1.2742241159529

while the gcc version gives

1.274224152

I think these are close enough that the bigloo program should be 
counted as succeeding; is there a way to check that the matrix norms 
are accurate to 5 digits or something, and call it good?

Thanks!
Will