Overclocking Help!!!

gochoo

Distinguished
Nov 3, 2007
6
0
18,510
First off let me tell you what I got:

Coolermaster 1000 Case
Antec Real Power 850w PSU
eVGA 680i LT MB but I was able to flash the BIOS to 680i P31
Q6600 G0
Thermalright Ultra Extreme 120 HSF
G.Skill 4GB PC6400 5-5-5-15 RAM
eVGA 8800GT x 2 SLI
WD Raptor 150X


Second, this is what I have my system overclocked and stable at:

3.3Ghz with the following setup:
9 multiplier with a 1480QDR linked and synced with my RAM
Core Voltage = 1.36
FSB Voltage = 1.3
RAM Voltage = 1.8
SPP Voltage = 1.3
MCP Voltage = 1.5
HT SPP <-> MCP Voltage = 1.2


OK, the past couple of days I have been experimenting with trying to get my system to overclock higher.

I can say that whenever I try 1500QDR - 1599QDR the computer wont post.

When I try 1600QDR - 1710QDR the system will post and almost reach into Windows.

Now when I try 1711QDR and higher, the system will get into windows with the following voltages:

Core = 1.46
FSB = 1.3
RAM = 1.95
SPP = 1.5
MCP = 1.65
HT SPP <-> MCP = 1.35

I can run the computer for about 2 minutes before it crashes.

Here are a few questions that I have:

Should I keep my current stable overclock or should I try and attempt to reach this new QDR of over 1700 (a whopping 3.8GHz+)?

Which voltages should I be messing with?

Is there such a thing as giving too much voltage?


What are the max idle and load temps that I should be reaching?

Any suggestions would be great!!!
 

cnumartyr

Distinguished
Nov 3, 2007
2,287
0
19,780
When you get bus speeds that high try increasing the FSB Termination a bit.

Part of your problem is that chipset as FSB holes. It just won't post at certain speeds, obviously you've run into that some. It's a quirky chipset with a Q6600 over 3.0 GHz.
 

cnumartyr

Distinguished
Nov 3, 2007
2,287
0
19,780


The 1.3v FSB is also known as FSB Termination, or CPU VTT. It can assist in getting a higher stable FSB. This might help or it might not.
 

cnumartyr

Distinguished
Nov 3, 2007
2,287
0
19,780


It's worth it to try different voltages honestly. Quirky chipset.

I'd see where I could get (like you said 2 minutes in windows).

Then change 1 voltage at a time and see how it effects it.
 

gochoo

Distinguished
Nov 3, 2007
6
0
18,510
So i take it that giving too much voltage may be a problem and I should try going up and down on each voltage settings?
 

cnumartyr

Distinguished
Nov 3, 2007
2,287
0
19,780



Mostly up.. honestly reading through again I think vCore should be up to around 1.4ish or so. Give that a shot and see if it'll boot.

Sorry I'm a bit tired and wasn't thinking straight.. Basically you hit a FSB hole and haven't really increased vCore since 3.3 GHz. Moving up to 3.6+ GHz requires a lot more vCore and if it gets into Windows and just randomly restarts I'd try to bump the vCore.

My CPU personally needs 1.4vCore to get stable in Windows at 3.6 GHz. At 3.3 GHz it only needed something like 1.312. Basically the voltage needs go through the roof as you get that high.
 

cnumartyr

Distinguished
Nov 3, 2007
2,287
0
19,780


Have you tried more vCore than 1.46 to see if it would get stable for more than 2 minutes?

Other than that..not sure on 680i's.