First off, system specs:
P6N SLI-FI, 2.6 BIOS, 60mm fan blowing on NB
E6550 (65nm, G0 stepping, default speed of 7x333 for 2.33GHz)
Scythe Ninja RevB with a home built back plate and bolted to the board
2x1GB Geil DDR2 800MHz C4, supposed to be D9 chips GX22GB6400UDCA=
http://www.newegg.com/Product/Product.aspx?Item=N82E16820144062
8800GT 512MB
Antec TP3-650W PSU
I've had the board for over a year and did some initial overclocking putting it around 3.3GHz using 1800 QDR FSB. I finally decided to really spend some time wringing every last bit out of it. It is 24 hours OCCT stable at these settings:
500MHz FSB (2000 QDR), unlinked
900MHz memory 4-4-4-12-2T at 2.00V (10:9 FSBRAM ratio)
CPU volts +0.0375V
NB 1.45V
SB 1.60V
FSB VTT at 16%
I'm right on the bleeding edge of what the NB can do as bumping the bus speed up much higher always results in a quick memtest86 failure despite dropping the CPU down to the 6x multiplier. There's not much left in the CPU though either. It crashed in Vista with 0.000V in the BIOS very quickly. It failed OCCT after 1.5hours with 0.05V, so I had to back it down to 0.0375V.Idle temps run around 40C and load peaks at 63C with Core Temp 0.99.3 which lists my Tjunction as 100C (case panels off). I was getting weird messages and BSODs during reboots until I bumped the VTT all the way up to 16%. VTT becomes pretty important when you start pushing the FSB to the limits, but based on my research, I wouldn't recommend the 20% setting for anything but short benchmarking runs and even 16% might be too much for 24/7 operation.
I don't know why, but on my board with the 2.6 BIOS, I still have to press F4 to get it to show the C1E Support option in the BIOS.
P6N SLI-FI, 2.6 BIOS, 60mm fan blowing on NB
E6550 (65nm, G0 stepping, default speed of 7x333 for 2.33GHz)
Scythe Ninja RevB with a home built back plate and bolted to the board
2x1GB Geil DDR2 800MHz C4, supposed to be D9 chips GX22GB6400UDCA=
http://www.newegg.com/Product/Product.aspx?Item=N82E16820144062
8800GT 512MB
Antec TP3-650W PSU
I've had the board for over a year and did some initial overclocking putting it around 3.3GHz using 1800 QDR FSB. I finally decided to really spend some time wringing every last bit out of it. It is 24 hours OCCT stable at these settings:
500MHz FSB (2000 QDR), unlinked
900MHz memory 4-4-4-12-2T at 2.00V (10:9 FSBRAM ratio)
CPU volts +0.0375V
NB 1.45V
SB 1.60V
FSB VTT at 16%
I'm right on the bleeding edge of what the NB can do as bumping the bus speed up much higher always results in a quick memtest86 failure despite dropping the CPU down to the 6x multiplier. There's not much left in the CPU though either. It crashed in Vista with 0.000V in the BIOS very quickly. It failed OCCT after 1.5hours with 0.05V, so I had to back it down to 0.0375V.Idle temps run around 40C and load peaks at 63C with Core Temp 0.99.3 which lists my Tjunction as 100C (case panels off). I was getting weird messages and BSODs during reboots until I bumped the VTT all the way up to 16%. VTT becomes pretty important when you start pushing the FSB to the limits, but based on my research, I wouldn't recommend the 20% setting for anything but short benchmarking runs and even 16% might be too much for 24/7 operation.
I don't know why, but on my board with the 2.6 BIOS, I still have to press F4 to get it to show the C1E Support option in the BIOS.