ok, got my system up and running - using Speedfan the core temps at idle are 19 - 22C at ambient 25C so the liquid cooler seems to be working. Used Coretemp and same set showed 32C??????Unlinked the memory, disabled various suggested items in BIOS. Left memory at auto timings, left all voltages at auto. Increased FSB to 1300 - orthos runs two cores at 50% for an hour with no error and temps on the two cores got to 32C. Ran Prime95 torture test - used all 4 cores at 25% but crashed after 9 minutes altho temps stayed around 22 on all cores. Trying to go over 1300 gets me a BSOD from XP. Going over 1400 gets me a CMOS reset.Only way to get max usage of the cores was to run several different programs at the same time - got up to 95%.I've read about lowering the multiplier, increasing Vcore and/or tweaking the memory but I don't fully understand their relationship. What should be the next step? Should I manually enter ram timings? Thanx, Vic System: Q6600 G0 CPU P5N32-E SLI 680i MoBo 2G XP2-8500 DDR2 1066FSB Mushkin 996535 RAM 320G 7200 HD partitioined for XP/Vista/Programs 2 - 74G Raptors in RAID0 500G 7200 HD for backup SATA DVD burner Evga 8800GTS 640 PCIx Kandalf LCS case w/ built in liquid cooling 850W Thermaltake power supply

Hi Vic -This is going to be a fun thread to watch. Your nVidia based board allows you to ramp up the FSB independently from the RAM clock - I checked since last commenting about it.Give me a bit of time today and I'll find some overclocking guides, but they all told me to keep RAM relaxed while you work out what the best deal is for your CPU. I'll get back to you, but others will probably jump in before me. Have fun.-Seadog

We need to see what the resultant settings are once you run windows/Download and run two instancse of CPU-Z from here http://www.cpuid.com/cpuz.php. With your system running at 1300FSB, post some screen shots of the CPU and memory tabs selected, like this:http://forums.avsim.net/user_files/176753.jpgGary

Alllrighteee, Vicster, here's a few links to guides, starting with nVidia's own for the exact chipset. Enjoy!http://www.nvidia.com/docs/CP/45121/nforce...verclocking.pdfhttp://www.xbitlabs.com/articles/cpu/displ...e-oc-guide.htmlhttp://www.overclockersclub.com/pages/overclock_faq/http://www.techarp.com/showarticle.aspx?artno=331&pgno=0http://www.overclock.net/overclock.php?fil...cking-guide.htmhttp://www.techwarelabs.com/guides/misc_mo...rclockingguide/http://www.pcstats.com/articleview.cfm?articleID=1804and finally an index to articles at Toms Hardware. I liked the series of three articles that began last December; scroll down to find Part I.http://www.tomshardware.com/cpu/cpuoverclocking/index.html-Seadog

Thanx for the links - had a few but extra reading can't hurt. Didn't really expect much from the first outing. Got some info from Mushkin on proper settings for my RAM so I'll plug them in tonight and see what I can come up with.The temps were confusing me also as TAT, Speedfan & Coretemp all show differing temps. From what I've been able to find out Speedfan reads about 15C low so I'm idling at 37C not 22C.Sam - thanx for the headdsup on dual instances of CPU-Z - after I make the memory changes suggested by Mushkin I'll take some readings.VicSystem: Q6600 G0 CPU Stock 2.4 (for now)P5N32-E SLI 680i MoBo 2G XP2-8500 DDR2 1066FSB Mushkin 996535 RAM 320G 7200 HD partitioined for XP/Vista/Programs 2 - 74G Raptors in RAID0 500G 7200 HD for backup SATA DVD burner Evga 8800GTS 640 PCIx (on the way)Kandalf LCS case w/ built in liquid cooling 850W Thermaltake power supplyVisit the Virtual Pilot's Centerwww.flightadventures.comhttp://www.hifisim.com/Active Sky V6 Proud SupporterRadar Contact Supporter: http://www.jdtllc.com/

You're welcome, Vic, and that's great news. It means you are officially way ahead of me and I get to sit back and just watch the fun. :-cool -Seadog

Sam - This is after first try.edit: obviously screen caps are not among my hidden talents. :)BTW, the color mess is a function of the program not what I see on the screen.Vic

Vic,Actually it's Gary, not Sam, but that's OK as we both haunt this forum together ;-)The main thing I wanted to check was that your memory was indeed not limiting your overclock. I see that you have it anchored to 800MHz, which should be more low enough for now for you to find your OCing limit.Given that your temps are so low (even with the +15C factor), I would be bumping your core voltage up to 1.4V and see if that improves your chances at 3.0GHz. From what I have read of G0 cores, you should easily achieve 3.0GHz and perhaps even higher. Gary

LOL! I was looking at your message and reading another that I had printed out from Sam.I've attached a second screen cap. This is still with all voltages at auto. Temps are up at 51C on first two cores because I was running Orthos - shut it down after 40 minutes.Someone mentioned that setting the voltages manually might help a bit also, seems like they think the P5N32 board might scale the voltages.I'll see if I can kick it up a notch w/o increasing voltage.Tried setting the Memory FSB to 1066 which is what the memory supports and couldn't get out of POST so I set it back to 800.Appreciate the guidance *GARY*....:)VicVisit the Virtual Pilot's Centerwww.flightadventures.comhttp://www.hifisim.com/Active Sky V6 Proud SupporterRadar Contact Supporter: http://www.jdtllc.com/

OK, at the end of the night this is what I've found. With memory set at 800 and timing of 5-5-5-4-12 2T and all voltages set to auto I can get stable 1334 which is 3.001G with load temps at 50C.If I try 1335 - I get BSOD. Just before I quit I tried manually setting the Vcore to 1.40 but still couldn't get above 1334.Also tried lowering the multiplier to 8 with no success.Gotta do some more reading to understand what happens next.VicVisit the Virtual Pilot's Centerwww.flightadventures.comhttp://www.hifisim.com/Active Sky V6 Proud SupporterRadar Contact Supporter: http://www.jdtllc.com/

Here's the EVGA forum. They are working the Q6600 on 680i boards. http://www.evga.com/forums/tt.asp?forumid=13And here's Coolaler's Q6600 run (This guy's in the loop and has benched an ES Penryn already). It's in Chinese, but all we need are the pictures. http://www.coolaler.com/~coolaler/forum/pr...?t=157070&pp=10He's using an old(er) 965 chipset, but the basics are the same. That's the Thermalright CPU air cooler. Single fan, nothing special. Notice he's got a fan blowing on his ram. You're Ok because you can decouple your ram from the FSB and run your ram lower if you have to. However Intel guys, look closely. He's using DDR2-667 ram with a 445mhz FSB (=DDR-990). Make a Note. What's going on here? Notice he gets to 4.0, but has to use 1.56 CPU volts and it's not stable. At 4.0, single SuperPi runs are all he can get. Nice, though. This means the PMDG 744 might make it to final, but as all the autogen kicked in, it'd crash the 'puter. At 3.6, CPU volts are at 1.4 and we see Orthos running (further down). That's the stresser-outer. The Q's running strong. Where are the temps? Who cares. He knows these chips can't hurt themselves. If it's runnin' stable, that's the test.

Vic, looks like you've hit a motherboard FSB hole or wall as they call it, especially if it is happening at 8 x 1334. Try 8 x 1600 and at stock voltage and see if you can push past the hole - 3.2GHz should be no sweat for your CPU.Gary

Gee! I must be learning something - I came to the same conclusion. I've read about walls and straps so now I see what might be going on. I'll try 8x1600 tonight when I get home.I am getting VERY adept at clearing the CMOS and re inputting settings. :)VicVisit the Virtual Pilot's Centerwww.flightadventures.comhttp://www.hifisim.com/Active Sky V6 Proud SupporterRadar Contact Supporter: http://www.jdtllc.com/