| 1 |
Thierry de Coulon posted <200602252104.35727.tcoulon@××××××××.ch>, |
| 2 |
excerpted below, on Sat, 25 Feb 2006 21:04:35 +0100: |
| 3 |
|
| 4 |
> So, I am wondering what causes the conflict: the board, the opterons, |
| 5 |
> Gainward's card? |
| 6 |
> Has anyone a 6600GT running fine ona double opteron? |
| 7 |
|
| 8 |
It could be the marginal timings between the card and the board (and the |
| 9 |
main memory). Sending signals of hundreds of megahertz thru not only a |
| 10 |
hard-wired (and factory tweakable) mainboard, and a similar expansion |
| 11 |
card, but thru the slot connector, is inherently an unstable situation. |
| 12 |
Some cards will simply cause problems with some boards, even tho both are |
| 13 |
within spec and "work" in general, and there's not much that can be done |
| 14 |
about it except switch things out to get something that works. |
| 15 |
|
| 16 |
Of course, timings on a dual CPU board are going to be even more finely |
| 17 |
tuned, and memory as well has signals at hundreds of megahertz, and a |
| 18 |
socket that introduces a slight bit of variability in capacitance and |
| 19 |
connectivity on each conductor -- that can make HUGE difference at |
| 20 |
hundreds of megahertz radio frequency. |
| 21 |
|
| 22 |
Power, as already mentioned, is probably a more common issue than this, |
| 23 |
but... |
| 24 |
|
| 25 |
-- |
| 26 |
Duncan - List replies preferred. No HTML msgs. |
| 27 |
"Every nonfree program has a lord, a master -- |
| 28 |
and if you use the program, he is your master." Richard Stallman in |
| 29 |
http://www.linuxdevcenter.com/pub/a/linux/2004/12/22/rms_interview.html |
| 30 |
|
| 31 |
|
| 32 |
-- |
| 33 |
gentoo-amd64@g.o mailing list |
Archives