|
Nonpython posted:June is for the Performance/Mid-Range. Think the line-equivalent to the /i[5|7]-2\d00\w/ models. (Wow, that was nerdy.) If this were facebook I'd like that pretty hard.
|
#
¿
Mar 5, 2011 23:12
|
|
|
|
| # ¿ Apr 25, 2024 05:57 |
|
|
FaustianQ posted:I only fanboy for VIA and Imagination because if my cause isn't hopeless and the perpetual underdog, what's even the point? The point is ethics in computer journalism
|
|
#
¿
Mar 2, 2017 04:28
|
|
|
Zero Gravitas posted:LIke I said, this machine is intended as a toy machine to test out some workflow processes and whether its possible to do this idea i have (however slowly) on a shitbox frankenstein machine thrown together from spare parts I have lying around and some from the bargain basement bin instead of dropping serious coin on it. I know from work that OpenFOAM doesnt play particularly nice with hyperthreading so I was trying to avoid that. I appreciate the nod towards ex-server hardware, but I'm really not looking for that kind of stuff right now. I went to one of OpenFOAM's trainings and they said it was fine with hyperthreading, it just didn't do anything except use a bit of overhead in parallelization. Although based on that I haven't bothered trying it and can't speak for any benchmarks. From what I've read, the optimum OpenFOAM computer minimizes threading and maximizes memory bandwidth. I dunno what physics you're simulating, but if you're doing transient, turbulent, or really any flow that takes longer than a day to converge, I would really recommend ECC RAM. Without ECC you will not get repeatable results, and the odds of experiencing an error increase proportionally to the length of calculation.
|
|
#
¿
Mar 14, 2017 01:14
|
|