G
Guest
Guest
Archived from groups: alt.comp.hardware.overclocking.amd (More info?)
Hi,
Still tying to bring my NF7-S/Mobile 2600+ up to speed and started to
experiment with BIOSes. Only tried 2 now, but found something very
strange.
With all parameters being the same, one BIOS clears Prime, but also
runs much higher CPU temps.
I did 11x194 at Vcore 1,65 (being 1,568-1,552), all other voltages
standard, same room temp.
Overall power use was 124 watt.
Power use and voltages were exactly the same, running both BIOS 27 an
BIOS D1.0 (which I found at
http://www.logan.eclipse.co.uk/downloads.htm )
The differences are (temps in celsius):
- The 27 BIOS runs Prime just over 4 hours max. Case 25, CPU 42 degree
- The D1.0 BIOS runs Prime OK, case temp staying 25, but the CPU gets
to 49
degree
I don't understand: how can temps get up so much, while overall power
use is the same and Vcore variations are the same also (indicating the
same load for the PSU)? The extra heat has to come from somewhere ...
It could be one of the BIOSes reads CPU temp wrong, but I don't think
so (?)
The only logical explanation I can think of is the D1.0 BIOS stresses
the CPU more. But how can that be while power consumption is the same
and it runs even more stable?!
And my problem with this: D1.0 runs stable at this speed, but I can't
go up much, because of CPU temp :-/
Anyone else seen this?
Anyway, back to getting a good BIOS:
I've read a lot about BIOS issues posted in 2003, SATA drives being
corrupted etc. I'm not using SATA drives, switched off in BIOS.
Are there still major issues with NF7-S BIOSes?
Is there some consensus about good and bad ones?
Thanks,
Boyd Noorda
Hi,
Still tying to bring my NF7-S/Mobile 2600+ up to speed and started to
experiment with BIOSes. Only tried 2 now, but found something very
strange.
With all parameters being the same, one BIOS clears Prime, but also
runs much higher CPU temps.
I did 11x194 at Vcore 1,65 (being 1,568-1,552), all other voltages
standard, same room temp.
Overall power use was 124 watt.
Power use and voltages were exactly the same, running both BIOS 27 an
BIOS D1.0 (which I found at
http://www.logan.eclipse.co.uk/downloads.htm )
The differences are (temps in celsius):
- The 27 BIOS runs Prime just over 4 hours max. Case 25, CPU 42 degree
- The D1.0 BIOS runs Prime OK, case temp staying 25, but the CPU gets
to 49
degree
I don't understand: how can temps get up so much, while overall power
use is the same and Vcore variations are the same also (indicating the
same load for the PSU)? The extra heat has to come from somewhere ...
It could be one of the BIOSes reads CPU temp wrong, but I don't think
so (?)
The only logical explanation I can think of is the D1.0 BIOS stresses
the CPU more. But how can that be while power consumption is the same
and it runs even more stable?!
And my problem with this: D1.0 runs stable at this speed, but I can't
go up much, because of CPU temp :-/
Anyone else seen this?
Anyway, back to getting a good BIOS:
I've read a lot about BIOS issues posted in 2003, SATA drives being
corrupted etc. I'm not using SATA drives, switched off in BIOS.
Are there still major issues with NF7-S BIOSes?
Is there some consensus about good and bad ones?
Thanks,
Boyd Noorda