View Full Version : G0 Q6600 overclocking
snowmanwithahat
06-09-09, 01:53 PM
So, my sig lies, I haven't had this overclocked since I instlaled windows 7 and now that it seems stable (as stable as a non release version of windows can be) I'm going to start overclocking again... but this time foregoing the conservative 333 fsb and hoping for 400 which would be about 3.6ghz
Thing is though... is this realistic for the C2Q series? I see it alot on the dual cores (and up to 4ghz) but i don't see many people pushing the quad to those speeds.
And before you have to ask, yes it has pretty good cooling. I'm still using a Zalman 7700cu (remember those)... the same one that graced my Opteron 165 (1.8ghz @ 2.6ghz) for more than 3 years.
Anyway, I'm excited to try this out again.... what's the key stuff you guys use for testing stability? is prime95 + ati tool the defacto stability testers still?
and one more thing... what's a "high" voltage for these? 1.325 is stock and I've seen people push theirs up to 1.4 but it just seems quite high to me....
Realistic, yes. On that heatsink? probably not.
The VID of G0s varies, the one I had was 1.31 ish
$n][pErMan
06-09-09, 10:08 PM
My VID is 1.2500V and my Q6600 G0 is currently OC'ed to the speed in my sig! :D It is very relaistic... but keep in mind your memory might hold you back... I had to move to the RAM in my sig from the 4 sticks of DD2-800 I had before. Could have been to much stress on the NB or just crappy ram. Either way... I managed 3.7Ghz stable on this rig, but it ran pretty hot so I took it down to the speed I have listed now :) Good luck and keep us posted! I run Orthos and max out all cores that way by setting the affinity of that program instance for the 4 cores on the CPU.
They are a truely great chip the Q6600 G0. The good thing about the 65mm chips, is they can handle more volts than there 45mm counterparts.
I used to run mine at 3.6Ghz 1.45 volts prime95 stable. It still runs fine today, in a mates PC.
Make sure you try OCing with an 8x multiplier. I had much better luck with that :). Also the higher FSB will help make it faster, too.
snowmanwithahat
06-10-09, 10:27 AM
so after playing with it 2 things are apparent.
400x9 is not stable, i haven't investigated if it's voltage or what but I think that's what it would be unless the processor can't handle that fsb speed because 400x8 wasn't very stable either.
intel's speedstep is crap....
and @ the comment about my memory... it should be fine if it's in a 1:1 ratio at 400mhz right.... I see no reason why that would hold me back if it's at the "stock" memory speed unless there's a relationship I'm not aware of. But yes, I do plan on going to a 1:1 relationship for the memory with the 400 fsb.
so it sounds like 1.45v is the high end of the spectrum for voltage?
Also I'm pretty confident in this heatsink... is there any reason I shouldn't be? I know the 95xx series from zalman was the "new high end" back in like '05 after I bought this... but have things really changed that much since then?
snowmanwithahat
06-10-09, 09:34 PM
3.2 (400x8) on 1.375 v.... seems kind of high but not horrible. I might push for 3.6 but I"m not sure if I want to try that high of a voltage
3.2 (400x8) on 1.375 v.... seems kind of high but not horrible. I might push for 3.6 but I"m not sure if I want to try that high of a voltage
1.4 volts is nothing as long as you have a good after market CPU heatsink. The 65mm cores can handle that no problems at all.
1.4 volts is nothing as long as you have a good after market CPU heatsink. The 65mm cores can handle that no problems at all.
Yup, anything up to 1.5vCore with that CPU is fine an long as it's below 71*C..
http://processorfinder.intel.com/Details.aspx?sSpec=SLACR
Xion X2
06-11-09, 10:54 AM
Yup, anything up to 1.5vCore with that CPU is fine an long as it's below 71*C..
http://processorfinder.intel.com/Details.aspx?sSpec=SLACR
I ran mine for 1.57 for over a year. It was a Xeon, though.
snowmanwithahat
06-11-09, 01:41 PM
Well it's pretty decent cooling.... better than stock I'm sure but I'm not quite sure how it stacks up now-days against the big name heatsinks
Zalman 7700cu
all copper - flowery heatsink... not heatpipes but before this tower - heatpipe era dawned it was top dog. it's also the big 120mm fan version so it pumps a good amoutn of air...
Just one thing though... what are you guys using to monitor cpu temps? I used to use motherboard monitor because it had the nice little tray feature but now that I'm running 4 cores I'd like to see the breakdown of the actual core temps... I was using speedfan before but the fluctation in core temps made me weary of it's accuracy.
[AK]Zip
06-11-09, 01:59 PM
Use real temp. I don't think the Zalman 7700cu will keep up on higher clocks speeds due to temps.
snowmanwithahat
06-11-09, 02:55 PM
Zip;2027263']Use real temp. I don't think the Zalman 7700cu will keep up on higher clocks speeds due to temps.
what do you guys suggest? I dno't wnat to dump too much money into a system like this because it's not exactly cutting edge even though it still performs well... tuniq tower?... also considering lapping the heatsink since I heard the core 2 s and onward all the way up to the i7s have horribly concave or convex heatspreaders
[AK]Zip
06-11-09, 07:05 PM
I would suggest you first try and see the best you can do on what you have. If you feel that you are being limited due to temps then upgrade. Otherwise just stick with it.
snowmanwithahat
06-11-09, 09:00 PM
yeh i'm in the process of doing that right now.
temps are
73/73/71/70
with max of
77/75/73/71
on all 4 cores (0-3) after about 30 min of "max heat" on prime 95...
... hta'ts at 3.2ghz 1.385v so yes... my heatsink appears to definetly falter here... I originally got it when I had a p4.. then got a dual core opteron... both cpus weren't nearly as hot as a quad core.
I'm thinking of lapping first and then seeing where that gets me... I'm not too interested in goin gwith more exotic cooling unless that involves water + CM cosmos s..... I've been wanting that case for a while but idk...
$n][pErMan
06-11-09, 10:26 PM
Try running a higher multiplier and lower FSB... will take less voltage and you will probably get a higher clock with ease. You will also lower your temps. Not sure how your motherboard works... but on mine 400Mhz FSB is the highest stress point... at which point 401Mhz goes into different timings and the NB becomes more stable. However, it runs faster below 400Mhz FSB unless I somehow manage to get it stable at a FSB above 420Mhz.
Eliminator
06-11-09, 10:34 PM
[pErMan;2027595']Try running a higher multiplier and lower FSB... will take less voltage and you will probably get a higher clock with ease. You will also lower your temps. Not sure how your motherboard works... but on mine 400Mhz FSB is the highest stress point... at which point 401Mhz goes into different timings and the NB becomes more stable. However, it runs faster below 400Mhz FSB unless I somehow manage to get it stable at a FSB above 420Mhz.
interesting how you can even boot between 350-400.... there's a nice fsb hole there and mine won't even boot in that range... i run 401x9 24/7 at 1.5V (actual) on an ocz vendetta 2 cooler
$n][pErMan
06-12-09, 12:32 AM
interesting how you can even boot between 350-400.... there's a nice fsb hole there and mine won't even boot in that range... i run 401x9 24/7 at 1.5V (actual) on an ocz vendetta 2 cooler
Mine boots without issue... although between 391-400Mhz it takes quite a bit more voltage. 100% stable at 390Mhz on the FSB for me though. I was also surprised because I read about that FSB hole on this motherboard :) The system is faster and more stable for me at my current setting that when I was running 3.6Ghz and above 401Mhz on the FSB. Go figure? lol. I am only running the NB voltage at 1.25V too!!! :D Maybe I just got lucky...
[pErMan;2027595']Try running a higher multiplier and lower FSB... will take less voltage and you will probably get a higher clock with ease. You will also lower your temps. Not sure how your motherboard works... but on mine 400Mhz FSB is the highest stress point... at which point 401Mhz goes into different timings and the NB becomes more stable. However, it runs faster below 400Mhz FSB unless I somehow manage to get it stable at a FSB above 420Mhz.
+1
snowmanwithahat
06-12-09, 02:03 PM
Thanks for all the comments. I'll definetly keep them in mind when I start to explore other overclocks. I didn't know some boards had a hard time at the 400mhz mark.
I guess the whole reasoninig behind that was to get the memory running 1:1 and try and get it stable at that fsb and hopefully step up to a 9x multipler without having too much issue. I'm probably going to try what you guys said though and start backing off on the fsb and go back to a 9x multipler... thinking I might back off on the voltage too.
I started a stress-test on prime95 this morning before I left for work.... it's been on 5 hours so far and I haven't had a chance to see if it's still alive. I did notice the core temps were peaking very high so more than likely a lapping or new heatsink is in order.
I really appreciate the comments though! I'll report back when I get home if it's stable or not. (3200, 400x8 @ 1.385v)
$n][pErMan
06-12-09, 06:35 PM
You might gain more speed running at 1:1 ratio at a lower FSB with a higher Multiplier anyway as your memory timings will tighten ;)
snowmanwithahat
06-12-09, 10:29 PM
Good point... I've never had good luck with my memory but i've been on KVR DDR400 for the past 2 builds which really really held me back alot.
So far I'm 13 hours prime95 stable at these settings (400x8 1.385v).... but I don't think this is where I want to stay anyway... ultimately the goal is 400x9 @ whatever voltage is comfortable but I'll try what some of you suggested here.
Probably dropping down to 377 fsb and 9x multipler for 3.4 and try that out next.
btw, 1.2 is my current NB voltage (i think) does that sound about appropriate?
$n][pErMan
06-13-09, 01:30 AM
Good point... I've never had good luck with my memory but i've been on KVR DDR400 for the past 2 builds which really really held me back alot.
So far I'm 13 hours prime95 stable at these settings (400x8 1.385v).... but I don't think this is where I want to stay anyway... ultimately the goal is 400x9 @ whatever voltage is comfortable but I'll try what some of you suggested here.
Probably dropping down to 377 fsb and 9x multipler for 3.4 and try that out next.
btw, 1.2 is my current NB voltage (i think) does that sound about appropriate?
1.2 is pretty low for that FSB... then again you are P45 vs. my P65 so maybe it runs on less voltage that high and its normal. Maybe someone else could answer that one :p I happen to be running my FSB really high on only 1.25V but most on my mobo this high run 1.35V or 1.45V.
Eliminator
06-14-09, 05:23 PM
[pErMan;2027702']Mine boots without issue... although between 391-400Mhz it takes quite a bit more voltage. 100% stable at 390Mhz on the FSB for me though. I was also surprised because I read about that FSB hole on this motherboard :) The system is faster and more stable for me at my current setting that when I was running 3.6Ghz and above 401Mhz on the FSB. Go figure? lol. I am only running the NB voltage at 1.25V too!!! :D Maybe I just got lucky...
nice... my NB voltage is 1.45V... it's been like this for more than 2.5 years since I bought it
snowmanwithahat
12-25-09, 04:08 AM
This G0 q6600 has been dissapointing.
I just got an OCZ Vendetta 2 cooler and the temps went down alot! I was seeing 77-80 C under full load at 3ghz, and it managed to go down to 55 C under full load at the same settings.
So with this new cooler power I figured I'd try and overclock it a bit... well it's been very dissapointing. I am at 3.5ghz but at a very very high vcore of 1.5375v. I can't seem to hit 3.6ghz either no matter what I do. It's booted, and been relatively stable, but that's at 1.55v and one core would usually fail at p95 after abou 15 - 17 minutes.
Oh well... case closed with this one. I'll settle for 3.5ghz but I'd really love to replace this with a Q9xxx cpu at some point, I'm sure ~4ghz would be possible with that.
vBulletin® v3.7.1, Copyright ©2000-2012, Jelsoft Enterprises Ltd.