Hello all,
Short summary on my o/c results with the AW9D Max (I'm using the final BIOS 1.2, posted on the German forum hardwareluxx)
1) E6400 @ 3.2GHz (FSB 400 1:1), 1.375V VCore, MCH 1.76V, OCZ Plat or Corsair C4 4-4-4-15, VDim 2.1
It is Prime95 stable for several hours (more than 12, I had to stop the test due to the CPU change).
Super Pi 1M was around 17.5s, 0.6 seconds faster on the AW9D compared to the DQ6, all settings equal.
2) E6600 Week28
CPU - 3.4GHz (FSB 378 1:1) 1.375V VCore, MCH 1.65V, Corsair C4 4-4-4-15 (have to try also the Platinum). Running 2x Prime95 since 3 hours.
SuperPi 1M around 14.8s
This is really a sweet spot, because CPU VCore is rather low and so does the MCH. I think this is the 24/7 setup for me. Core temp shows 60-62C (Thermalright SI-128 and good case airflow), room temperature is 26C at the moment.
It is simply not possible to run above FSB400 stable, with both CPUs. For FSB400 the Max needs already 1.76 on the MCH. I tried using a low CPU multiplier to rule out the CPU limitations, but even so memtest fails the test 5 at just 5%. I tried MCH all the way to 1.90V, it helps to boot and start testing, but it never finishes memtest5: it freezes or corrupts the screen.
Here are some of my findings not related to overclock, call it a mini-review if you like:
a) Holes for heatsink attachment are a bit bigger than on the Gigabyte board, they attach much easier on the Max. On the Gigabyte, it has some interference and it is much harder to lock (using the Intel attachment solution or Thermalright SI-128).
b) Max bios boots much faster than the DQ6, specially if using AHCI for the Intel Sata ports.
c) The Max seems to like OCZ Platinum better than the OCZ Gold - Corsair C4 is fine also. On the DQ6 it was the opposite, Gold was easier to run.
d) AW9D Max does not turn off when changing BIOS settings. Thats a big plus on my book. If it fails, it resets again and goes back to the last good boot setup. Or hold insert while pressing reset. Very nice !
5) Board comes with a SLI BRIDGE
I'm keeping the Max and the E6600. That's more than enough !!
I have to sell either my DQ6/E6400 or DS4/E6300 - the idea is to sell the DQ6/E6400, as the DS4/6300 runs very well at 3.15/450 4-4-4-15.
Some weeks ago I would recommend the E6400 with a P965 board, but after experimenting the E6600/975 I would have to say now this is a better bet IMO.
For high FSB speeds on the P965 you need good ram and DDR2 prices are just too high a the moment, so it is better to have a higher multiplier CPU that can run its full overclock potential under FSB400.
PS:
There is a discrepancy between the VCore set on BIOS, uGuru reported and measured on the board.
On uGuru the board seems to be overvolting, on real life it is undervolting.
It was measured by W1zzard (techpowerup master), check here. It's in a German forum but he posted in English
http://www.forumdeluxx.de/forum/showpost.php?p=4492847&postcount=1811
Curious isn't it?
Short summary on my o/c results with the AW9D Max (I'm using the final BIOS 1.2, posted on the German forum hardwareluxx)
1) E6400 @ 3.2GHz (FSB 400 1:1), 1.375V VCore, MCH 1.76V, OCZ Plat or Corsair C4 4-4-4-15, VDim 2.1
It is Prime95 stable for several hours (more than 12, I had to stop the test due to the CPU change).
Super Pi 1M was around 17.5s, 0.6 seconds faster on the AW9D compared to the DQ6, all settings equal.
2) E6600 Week28
CPU - 3.4GHz (FSB 378 1:1) 1.375V VCore, MCH 1.65V, Corsair C4 4-4-4-15 (have to try also the Platinum). Running 2x Prime95 since 3 hours.SuperPi 1M around 14.8s
This is really a sweet spot, because CPU VCore is rather low and so does the MCH. I think this is the 24/7 setup for me. Core temp shows 60-62C (Thermalright SI-128 and good case airflow), room temperature is 26C at the moment.
It is simply not possible to run above FSB400 stable, with both CPUs. For FSB400 the Max needs already 1.76 on the MCH. I tried using a low CPU multiplier to rule out the CPU limitations, but even so memtest fails the test 5 at just 5%. I tried MCH all the way to 1.90V, it helps to boot and start testing, but it never finishes memtest5: it freezes or corrupts the screen.
Here are some of my findings not related to overclock, call it a mini-review if you like:
a) Holes for heatsink attachment are a bit bigger than on the Gigabyte board, they attach much easier on the Max. On the Gigabyte, it has some interference and it is much harder to lock (using the Intel attachment solution or Thermalright SI-128).
b) Max bios boots much faster than the DQ6, specially if using AHCI for the Intel Sata ports.
c) The Max seems to like OCZ Platinum better than the OCZ Gold - Corsair C4 is fine also. On the DQ6 it was the opposite, Gold was easier to run.
d) AW9D Max does not turn off when changing BIOS settings. Thats a big plus on my book. If it fails, it resets again and goes back to the last good boot setup. Or hold insert while pressing reset. Very nice !
5) Board comes with a SLI BRIDGE
I'm keeping the Max and the E6600. That's more than enough !!
I have to sell either my DQ6/E6400 or DS4/E6300 - the idea is to sell the DQ6/E6400, as the DS4/6300 runs very well at 3.15/450 4-4-4-15.
Some weeks ago I would recommend the E6400 with a P965 board, but after experimenting the E6600/975 I would have to say now this is a better bet IMO.
For high FSB speeds on the P965 you need good ram and DDR2 prices are just too high a the moment, so it is better to have a higher multiplier CPU that can run its full overclock potential under FSB400.
PS:
There is a discrepancy between the VCore set on BIOS, uGuru reported and measured on the board.
On uGuru the board seems to be overvolting, on real life it is undervolting.
It was measured by W1zzard (techpowerup master), check here. It's in a German forum but he posted in English
http://www.forumdeluxx.de/forum/showpost.php?p=4492847&postcount=1811
Curious isn't it?
Facebook
Twitter