| Home
| Calendar
| Mail Lists
| List Archives
| Desktop SIG
| Hardware Hacking SIG
Wiki | Flickr | PicasaWeb | Video | Maps & Directions | Installfests | Keysignings Linux Cafe | Meeting Notes | Linux Links | Bling | About BLU |
Is the current state of BLADE solid enough to do useful work?
I've read the 0.3 notes and it looks like a fairly complete package, but
I'd like to hear from people who've tried it.
I have an application which does lots of serious numbercrunching --
written in fortran 77, takes 24-48 hours to converge on a pentium 66.
f2c should handle it easily, as far as I can tell -- will be testing that
on my home machine.
I will be putting together a machine to run it on at work and am considering
an Alpha, if the performance win justified the extra cost.
What sort of speed win could I hope to get over a pentium 66?
(Presumably I can scale within pentium family by clock frequency, so I would
expect a factor of 2 or so from a 133 mhz, etc.)
Thanks
Steve
horne at pfc.mit.edu