P6X58S-E & 930 build and overclock log

[crazyDAJT]

recently updated my Q6600 rig with the flowing parts after finding a new home for the old stuff.

my old mini build log can be found @ now with Thermochill PA120.3

so the new part that are went in.

i7 930
Asus P6X58D-E
6GB of Corsair XMS3
Corsair HX750
EK Supreme HF

I also got a HD4870X2 (with XSPC water block) off the members market.

the remaining items are as follows:
ArmorLCS case
2 x WD Raptor 74GB in Raid0
1 x Barracuda 500GB for storage
Thermochill PA120.3
Laing 18W with plexi top

two issues that i came up-against very quickly. firstly i got all this stuff and realised my D-TEK FuZion v1 was not compatible with LGA1366 motherboard, i had to fork out for a new block. also during the initial leak test i shorted out my old pump so had to get a new one

so I have almost spent more on my water loop in the past week then it cost me to get it in the first place.

so the case built up during 5hr leak test, loop order is: pump-CPU-GFX-RAD-REZ-pump

and during running up, still need to sort the cables up and tidy up, but i want to see how good the new PC was.

more to follow....

 

[crazyDAJT]

so after the build a and installing windows, (i needed to get my bluray drive as i couldn't get started with my one). installed windows, updated everything and fired up Prime95, Core Temp and CPUZ.

was a bit surprised that my stock Vcore under load was as high as 1.176V! and I was pushing 60C on water. at this point was was thinking A) I got a very poor chip B) the HD4870X2 was dumping far to much heat into the loop.

after a good 5Hr small FFT I was happy that that everything was stable. so thought it was a good time to start OC'ing.

starting slowly i set a target of 3.3GHz. 160X21 keeping the Vcore at stock levels of 1.18V in BIOS. ruining small FFT for 2Hr, CPUz reported Vcore at 1.168V. temps peeked at 58C, but what is nice is how even the temperatures are never more then 3-4C out.

then using some settings I found on the internet I went for 3.8GHz.

181X21 and a Target Vcore of 1.2.

1024x548

I have read up on how good these chips are for clocking but this is just stupid easy. fist time 8-12Hr stable in Small FFT and Large FFT with temperatures barely elevated from stock settings. the room my PC is located was very very warm when running these settings probably 25C easy. When I checked it at 9am this morning temps where 59-57-59-56.

will be pushing the chip to 4GHz later today

 

[crazyDAJT]

reserved.

 

[crazyDAJT]

4GHz 1.08V?

so i have now taken the chip to smack on 4GHz (190X21)

I set the QPI/Dram core to 1.3V (will lower this when i'm done and 100% stable)

I set the CPU voltage to 1.2225V in BIOS. this initial read as 1.216V in CPUZ but looks to have dropped to 1.208V this morning.

temp's peeking at 65C in my warm room but stabilise around 60-62C.

im surprised buy the low CPU voltage to get this far? and that it changed overnight? any ideas.

This was a first time stable run of small FFT for 10hr.

1024x550

 

[Greebo]

Looking good so far. It's after 4Ghz that things start getting fun. I think with mine I whacked it on 1.25v core and 1.25v QPI/dram and 4Ghz and it flew through prime and Intel burn test.

I know have it on 4.4Ghz but needs 1.385V and quite high QPI/dram to be stable but not really tweaked it yet to see if I can get the voltages lower.