| 1 |
Markus Dittrich wrote: |
| 2 |
> -----BEGIN PGP SIGNED MESSAGE----- |
| 3 |
> Hash: SHA1 |
| 4 |
> |
| 5 |
> On Fri, 18 May 2007, M. Edward (Ed) Borasky wrote: |
| 6 |
>> Yeah ... I just got an Athlon64 X2 4200+ and the first thing I did |
| 7 |
>> when I got the machine stabilized was build "blas-atlas" and |
| 8 |
>> "lapack-atlas" (3.7.30). New versions of Atlas generally show up in |
| 9 |
>> Portage or the science overlay within a day after Clint releases |
| 10 |
>> them, so I don't usually test the upstream source. |
| 11 |
>> |
| 12 |
>> The best I got out of my machine was something like 7 GFLOPS on a |
| 13 |
>> 32-bit test with 3.7.30, and there were some cases that looked like |
| 14 |
>> they should have done better. So I definitely want to test this 160M |
| 15 |
>> setting. |
| 16 |
>> -- |
| 17 |
> |
| 18 |
> Nice! Please let us know how your benchmarking goes. |
| 19 |
> The new version should hit portage |
| 20 |
> by tomorrow and I plan to go with 160M by default |
| 21 |
> unless I notice something strange during testing. |
| 22 |
> |
| 23 |
> Best, |
| 24 |
> Markus |
| 25 |
> |
| 26 |
> - --- |
| 27 |
> |
| 28 |
> Markus Dittrich (markusle) |
| 29 |
> Gentoo Linux Developer |
| 30 |
> Scientific applications |
| 31 |
> -----BEGIN PGP SIGNATURE----- |
| 32 |
> Version: GnuPG v1.4.6 (GNU/Linux) |
| 33 |
> |
| 34 |
> iD8DBQFGTg2nxlRwCwb7k40RAjOYAJ4/Va9/W8UM/v7EqmOgXLlH120cfgCdEIpQ |
| 35 |
> J4hpzeIHuRE+TgR8/Vo8rEU= |
| 36 |
> =1RAJ |
| 37 |
> -----END PGP SIGNATURE----- |
| 38 |
I just installed 3.7.31. It looks like there are no significant |
| 39 |
performance differences -- I've attached the two "SUMMARY.LOG" files |
| 40 |
from 3.7.30 and 3.7.31. |
Archives