How high can I go on my Q6600?

Boostage

Limp Gawd
Joined
Sep 25, 2008
Messages
327
I just got a G0 from newegg, but unfortunately had a VID of 1.325 :(

I got it up to 3.5ghz but it took 1.49 Volts to get there. should I even consider going to 3.6 or 3.7ghz? obviously I would need to exceed 1.5v.. is that safe? My temps are actually fine I idle at 33c according to Uguru (aBIT ip35 pro board) although the Bios says its idling at 39c

Cooler is a Xigmatek s1283, ocz freeze paste ram is Corsair xms2 twin2048-6400 running 1:1.25
 
Use Real Temp, Speedfan or CoreTemp to get a much more accurate reading of your temperatures. And please post your load temps with those temperature programs open. Are you stress testing your overclock in anyway? If so, for how long?
 
What are your load temps? I also have a 1.325 VID Q6600 which makes me sad. Anyways I have the same cooler and I've overclocked to 3.3ghz and under load my temps are in the 60s!!! My main concern is the temperature discrepancy between the cores. Like a 10* difference. I've remounted the heatsink already but it didn't do anything for my temps. It seems like people are getting way better temps with this cooler than I am.
 
What are your load temps? I also have a 1.325 VID Q6600 which makes me sad. Anyways I have the same cooler and I've overclocked to 3.3ghz and under load my temps are in the 60s!!! My main concern is the temperature discrepancy between the cores. Like a 10* difference. I've remounted the heatsink already but it didn't do anything for my temps. It seems like people are getting way better temps with this cooler than I am.

Heatsink and/or Processor might need lapping.
 
Ok this is BS I primed it and it BSOD's right away. kept on messing with it anfd right now i have it t 3.3ghz with 1.45 volts and its priming just fine as I type this, so far so good but temps are 12 degrees between cores

Core 0 68
Core 1 68
Core 2 56
Core 3 56

before I even consider going thru all this BS of pulling the board out and then attepmting to lap the cpu. is there a noticeable gain going from 3.3 to 3.6? considering hats a total of 1.2ghz when you add all 4 cores?

****EDIT***** i took a weird gamble and it kinda worked I had an extra case fan This one http://www.newegg.com/Product/Product.aspx?Item=N82E16835185054 it blows 133cfm. well i replaced the xigmatek cpu fan with it and now my new temps are according to core temp..

Core 0 62
Core 1 62
Core 2 51
Core 3 51

This is at 100% load in prime. but my other question still stands. is it worth it to lap it?
 
To me i say no, some will say yes. But i don't see going through all that trouble for 300 more mhz. Not to mention when you lap your cpu say bye-bye to the warrenty.

Boostage said:
considering hats a total of 1.2ghz when you add all 4 cores?

You don't do quads cores like that. If that was true they would sell a Q6600 and try and pass it off as a 9.6gighrz cpu.

Are you running a 32bit os as you only have 2 cores working. I would use a 64bit os and get the whole deal.

Good Day
 
Are you running a 32bit os as you only have 2 cores working. I would use a 64bit os and get the whole deal.

....I think all 4 cores are working and he's just seeing a heavy temperature variance due to poor HSF mounting and possibly bad mounting surface. Regardless (and this is coming from a 64bit OS advocate) please don't spread misinformation that 32bit OSes can't support quad core processors properly -- they do without issue, assuming it's something reasonably modern and patched up to date.
 
No. 300Mhz is not worth the effort for the very very minimal gains.

Thanks. and an update. it seems to be stable at 3.44ghz running @ 1.55 volts I know Intel rates the Chip at 0.85-1.5v is that tiny bit of voltage over bad?
 
You don't do quads cores like that. If that was true they would sell a Q6600 and try and pass it off as a 9.6gighrz cpu.

Are you running a 32bit os as you only have 2 cores working. I would use a 64bit os and get the whole deal.

I disagree with you if a application supports smp and can scale up to 4 processors then it really is like having 9.6ghz to run that particular program, although there are few applications written to accomplish this but they are coming and personally I think windows does a great job a spreading the load.

32 bit is still faster than 64 bit, the only reason to goto 64 bit is to remove the 4GB barrier, that being said if you look at M$'s development line and their support they are not even going to support 2008 in 32bit, oh and 32 bit can quite happily see and use all cores mine does (vista 32)
 
I disagree with you if a application supports smp and can scale up to 4 processors then it really is like having 9.6ghz to run that particular program, although there are few applications written to accomplish this but they are coming and personally I think windows does a great job a spreading the load.

They've been talking about "they're coming!" for years now and not much has surfaced. First it was about full quad core utilization, now it's parallel processing w/ the GPU(?). Guessing the tremendous amount of work and resources needed to code such stuff (or just an excuse) is a major deterrent.
 
I disagree with you if a application supports smp and can scale up to 4 processors then it really is like having 9.6ghz to run that particular program, although there are few applications written to accomplish this but they are coming and personally I think windows does a great job a spreading the load.

32 bit is still faster than 64 bit, the only reason to goto 64 bit is to remove the 4GB barrier, that being said if you look at M$'s development line and their support they are not even going to support 2008 in 32bit, oh and 32 bit can quite happily see and use all cores mine does (vista 32)

SMP does not scale 1:1 even in ideal situations, and day to day you get nowhere near ideal istuations, so your argument here is moot.

Also ...32bit software is in no way 'faster' than 64bit software doing the same thing on a 64bit platform. At all. Ever. I don't know where you got this idea, but it's incorrect.
 
SMP does not scale 1:1 even in ideal situations, and day to day you get nowhere near ideal istuations, so your argument here is moot.

Also ...32bit software is in no way 'faster' than 64bit software doing the same thing on a 64bit platform. At all. Ever. I don't know where you got this idea, but it's incorrect.

OK SMP does not scale 1:1 even so you could argue that you are effetivly getting more cycles as compared to a single processor model, as long as the code supports it.

I stand corrected on the 32/64bit thing, This was old information from when the AMD64's started getting popular, but as i have found out newer processor are now nativley 64bit with 32 bit registers which would obviously give a major advantage to 64 bit.
 
TS. How hard was it to remove the rubber nipple fan adapters on the xigmatek fan to put on the new fan? I been contemplating getting a new fan for my heatsink but I'm thinking it'd be a pain in the ass to remove those rubber fan mounts.
 
TS. How hard was it to remove the rubber nipple fan adapters on the xigmatek fan to put on the new fan? I been contemplating getting a new fan for my heatsink but I'm thinking it'd be a pain in the ass to remove those rubber fan mounts.

It wasn't that difficult. the rubber slides between the fins U just pull them off. attach them to the new fan and re-attach the rubbers one at a time with the new fan. I didn't have to pull my mobo. heck I didn't even have to pull out any ram sticks. max load in prime is like 60-61* 3.4ghz 1.460v oh it helps to use a needle nose pliers to pull the rubbers thru the fan.
 
Thanks. and an update. it seems to be stable at 3.44ghz running @ 1.55 volts I know Intel rates the Chip at 0.85-1.5v is that tiny bit of voltage over bad?

Bit more explanation - yes it's bad, but not in the sense that your system will die in a week. These processors haven't been out long enough for people to really get sense of how overclocking affects their longevity. There are claims that running your procs at stock allows them to go for 5-8 years.

So yes, it will reduce your proc's lifespan, but by how much is unknown.

As for the temp differences, usually it's because of miscalibrated/bad sensors. Realtemp has a test for stuck sensors that can sort of test for this.
 
My same G0 1.325 vid does 3.7 on water but at 1.6vcore
1.55vcore 3.6ghz 24x7 for me.
 
....I think all 4 cores are working and he's just seeing a heavy temperature variance due to poor HSF mounting and possibly bad mounting surface. Regardless (and this is coming from a 64bit OS advocate) please don't spread misinformation that 32bit OSes can't support quad core processors properly -- they do without issue, assuming it's something reasonably modern and patched up to date.


What the heck are you talking about spreading misinformation? Your trying to make a arguement because you think your words are better then someone elses?

You do get more benefit from a 64bit os then a 32bit using a quad core. You can fly be here posting b/s post all you want. But that crap don't work on me. And all i said was to get the most out his cpu use 64bit os. So you may want to slow down a bit before you post. Nothing i gave him was wrong or misinformationed!

Wouldn't be any different in him buying 4 gig of ram and running a 32bit os. Same example he would still need a 64bit os you run it like it suppose to be used.

Good Day
 
Actually killer_K you did say I would only be able to use 2 core with a 32-bit OS. Silent was only trying to correct you. although he was a little rough. I think you confuse Physical sockets with cores. XP can only see two physical cpu's but I believe it can recognize up to 8 cores.
 
Are you running a 32bit os as you only have 2 cores working. I would use a 64bit os and get the whole deal.

In your pic you have two cores being used correct. I simply said use the 64bit os to get the whole deal. Meaning the best use out your cpu. Perhaps they took it the wrong way maybe? As i went from vista 32bit to vista 64bit and it is a night and day difference using the quad core in it.

Good Day
 
It wasn't that difficult. the rubber slides between the fins U just pull them off. attach them to the new fan and re-attach the rubbers one at a time with the new fan. I didn't have to pull my mobo. heck I didn't even have to pull out any ram sticks. max load in prime is like 60-61* 3.4ghz 1.460v oh it helps to use a needle nose pliers to pull the rubbers thru the fan.

I mean removing the rubber things from the fan itself...they are made so they like pop into the fan and looks like they'd be hard to find out. Like you say use pliers to pull the rubbers through the fan...but I wanan get them out of the fan. Did you just use pliers to pull them out?
 
Back
Top