eugene@ames.UUCP (Eugene Miya) (12/19/84)
[I'm just sitting here waiting on a Cray...]
Before taking this job I, like many, had an aura of mystery around
Crays [those friends at lll-crg, lanl, hao, and other Cray sites
can pardon me for a sec]. Just for those without: I am working with a
fluid dynamics code which under our test conditions runs 0.27 cray seconds
including I/O. The code is vectorized. By turning on subscript range
checking (and turning off vectorization by default), the code now takes
over 60 cray seconds to run. I wonder how much I spent?
--eugene miya
NASA Ames Research Center
{hplabs,ihnp4,dual,hao,vortex}!ames!aurora!eugene
emiya@ames-vmsb.ARPA