| 1 |
On Fri, Mar 14, 2008 at 3:45 PM, Filipe Sousa <natros@×××××.com> wrote: |
| 2 |
> -----BEGIN PGP SIGNED MESSAGE----- |
| 3 |
> Hash: SHA1 |
| 4 |
> |
| 5 |
> Mark Knecht wrote: |
| 6 |
> |
| 7 |
> | On Fri, Mar 14, 2008 at 11:23 AM, Dan Farrell <dan@×××××××××.cx> wrote: |
| 8 |
> | |
| 9 |
> | It's an interesting question and one I've not tried to test. Does an |
| 10 |
> | AMD64 machine running a 32-bit or 64-bit install run faster or slower |
| 11 |
> |
| 12 |
> | with one or the other. For all of my everyday work - Gnome, Firefox, |
| 13 |
> | web browsing, email, MythTV, etc., it's been my assumption that there |
| 14 |
> | wouldn't be any noticeble difference. I run 64-bit but assume I'd run |
| 15 |
> |
| 16 |
> | at more or less the same speed if I ran 32-bit. I may be wrong. Anyone |
| 17 |
> | have any measured data? Same machine, two installs? |
| 18 |
> |
| 19 |
> I have a few numbers from genlop -t |
| 20 |
> |
| 21 |
<SNIP> |
| 22 |
> |
| 23 |
> These times are from the same machine: |
| 24 |
> abit ip35 pro, c2quad q6600 (2.4ghz), 4gb ram, (320+320) sataII raid0 (155-MB/s) |
| 25 |
> |
| 26 |
> - -- |
| 27 |
> Filipe Sousa |
| 28 |
|
| 29 |
Nice numbers and nice machine. Must have set you back a bit. ;-) |
| 30 |
|
| 31 |
It seems when I match up exact revisions you're getting something |
| 32 |
between a 10-15% speed increase. Quite nice. |
| 33 |
|
| 34 |
Do we know that the amount code compiled is identical? I imagine |
| 35 |
everything you show is correct since the speed increase is pretty |
| 36 |
consistent from app to app. |
| 37 |
|
| 38 |
Great stuff. Thanks. |
| 39 |
|
| 40 |
Cheers, |
| 41 |
Mark |
| 42 |
-- |
| 43 |
gentoo-user@l.g.o mailing list |
Archives