Hi guys, I would like to compare OCs so if you have a QX6700 please post your OCs and vcore. Thanks.
Some like:
320x12 = 3845GHz 1.55vcore (set in BIOS) FPO/BATCH#: L644B747
http://valid.x86-secret.com/show_oc?id=169838
Printable View
Hi guys, I would like to compare OCs so if you have a QX6700 please post your OCs and vcore. Thanks.
Some like:
320x12 = 3845GHz 1.55vcore (set in BIOS) FPO/BATCH#: L644B747
http://valid.x86-secret.com/show_oc?id=169838
Can't get mine any higher than the default 2.66Ghz without stability issues. :( I am running an Intel Bad Axe 2 board with 4GB of Corsair Dominator 8500. I get can my machine to boot into Windows at 3.2, and it appears to be P95 and Orthos stable, but if I try to do any gaming or work with 3DSMax I get random nv4disp.dll BSOD's. I was running the default 1.35V, as with any thing higher my CPU temps instantly jump to well over 90C with my Tuniq Tower.
Some weekend I am going to try reseating my heatsink and see if I can get it to at least 3Ghz with slightly more voltage.
Quote:
Originally Posted by nicepun
has asus fixed the p5w dh deluxe bios yet?
Or is it just 3845x2 and 3200x2
I hate that bug, download CPUZ 1.35, you can read all 4 cores that way.
either way, nice week 44 chip ya got there, the only retail chip i've played with is a week 40. Does 3.4x10= 340FSB on xbx2 no problem on air. Big vDroop, vcore set to 1.4375 in bios though w/ enhanced power slope. Didnt bother with multimeter, but vcore reported around 1.35 under full load w/ intel tools.
Cooling = scythe infinity w/ 1300RPM fan in a case. Temps ~70c-80c logged from coretemp after 40+ hours of prime and various looping 3D benchmarks.
The rest dont really count because they are ES
Sorry, I don't follow!Quote:
Originally Posted by irev210
Additional Comments: I do hate that I have to use 1601 bios in order to use 11 or higher multipliers.....:slapass: otherwise 3200-3400 is the highest.
he is trying to say THAT when you OVERCLOCK with asus QUAD core ONLY 2 of the cores are OVERCLOCKED and the other two stays at the original speed. Thats the reason why i moved to BX2
That not your qx6700 you'll be glad to hear, that's your graphics card causing problems(nv4disp.dll). What graphics card are you using?Quote:
Can't get mine any higher than the default 2.66Ghz without stability issues. I am running an Intel Bad Axe 2 board with 4GB of Corsair Dominator 8500. I get can my machine to boot into Windows at 3.2, and it appears to be P95 and Orthos stable, but if I try to do any gaming or work with 3DSMax I get random nv4disp.dll BSOD's. I was running the default 1.35V, as with any thing higher my CPU temps instantly jump to well over 90C with my Tuniq Tower.
Some weekend I am going to try reseating my heatsink and see if I can get it to at least 3Ghz with slightly more voltage.
I've had similar issues with my 8800 gtx and applications like lightwave.
Two things. Firstly with ntune or rivatuner, setting the fan speed to 85-90%, instead of auto. I found helped a lot. Suggesting to me that these BSOD are in part temperature related.
Secondly you need to possibly look at global settings for your card(3d Applications), and see if you can find the best configuration for it. In addition I believe that people have more joy with max if the viewport is set to directx, rather than opengl.
I'm presuming that you have something like an 8800 GTX or GTS?
RLM
Quote:
Originally Posted by BobyTT
I see...well I just checked with CUPZ 1.35 and Sandra and all cores are OCed.
Quote:
Originally Posted by irev210
This piece of magnificent engineering can only be found on ABIT boards: Click
its a driver problem lol, i even have it on my geforce 4Quote:
Originally Posted by RLM
Quote:
Originally Posted by Chosen.
nah, actually a lot of boards suffered from that problem.
even dfi infinity had/has that problem.
Glad to see Asus fixed it.
Yeah, I am running an 8800GTX card and the even at idle the temps are fairly high at around 73C. Most of my BSOD are when running OpenGL with 3DSMax, as the screen will temporarily freeze, my sound will start looping, and eventually the system will halt with a BSOD error. The strange thing is the crashes seem much less frequent if I run my machine at the stock speeds as opposed to even a mild overclock. Very frustrating, as when I was running at 355x9 things were super snappy, but the random nv4displ.dll errors were really starting to frustrate me. I am kinda at a loss as to why I get so many nvidia related errors when running things even slightly out of spec.Quote:
Originally Posted by RLM
QX6700, 8x450 on the abit QuadGT at 1.40V.
311x15 = 4671,48GHz 1.65vcore (set in BIOS) FPO/BATCH#: L632A923
http://valid.x86-secret.com/show_oc.php?id=156088
wow...nice fsb for the quad on that board...Quote:
Originally Posted by bingo13
may i ask what fsb that chip has been hitting on other mobo's?
480x8 (1.47V) on the Commando, 465x8 (1.43V) on the RD600. The chip just goes in the tank after 3600 unless voltage goes up in a hurry. I have found any voltage over 1.47 does not improve overclocking, even at higher or lower multipliers. It hits a brick wall around 3840, cooling does not matter. I will take 3600 with 1.4V any day though. ;)Quote:
Originally Posted by flytek
Quote:
Originally Posted by bingo13
interesting, what type of cooling have you used?
All the qx6700's ive played with loved more voltage, sadly more voltage = way more heat. Single stage couldnt even keep it in the negatives :(
bingo, you should get some xeon x3220's. You would have some fun.
For a good comparison, you guys should really post the actual Vcore under load.
DDTUNG:cool:
Early air cooling test.
Mobo: BFG 680i P24
Cooling: Tuniq Tower 120 with two additional fans either side
CPU speed: 3000Mhz 250x12
Stablity test: Prime95 v25.1 10mins in so far
Max CoreTemps under SmallFFT test: 67C
Vcore idle: 1.24
Vcore load: 1.20
Ambient temps: 26C
http://slidog.bigblog.com.au/data/2/...0221194642.png
I've been able to boot a Q6600 at 3.2 easily on both P5B Deluxe and P5N32-E, but it's not stable. Most stable I could get was about 2.9ghz.
//all with 9x multi.
Same experience here. On my QX6700 I can get 4200 MHz stable easy for benching and everyday use as well as most games that don't use more than one/two core(s).Quote:
Originally Posted by irev210
Priming (dual orthos) 4 cores @ 3700 Mhz and 1.45V in bios (about 1.40V real) is as much heat as my Mach II GT cooler can possibly handle.
I've given it up to 1.75V in another motherboard and got I've had it up to 4.6 Ghz, but the heat is just too much for the mach II GT for anything other than max 2-3 mins of superpi... :(
I hope the 45 nm's are coming out sooner rather than later. The QX's are really nice overclocking chips, they respond really well to voltage as well, however they just get really, really hot when you go over 1.4V (real)
319x11 in xbx2 @ 1.575 in bios
300x12 in aw9d-max @ 1.525 in bios
292x12 in P5B-dlx @ 1.625 in bios
so actual voltage ~1.48-1.52 = 3.5 prime 25 stable under blend/small/large fft for 12+ hrs and all settings used 24/7 for DC apps like CPDN, QMC, and SETI
340x10@1.42V bios (1.341V under load) using XBX2
Temps hit 57-58C with all four cores loaded under Prime95. Rig is Prime stable for 8+ hours at those settings..
I'm priming 2.93 GHz at 1.2 vcore idle/1.17 vcore load right now. Air cooling with 26*C ambient.
i know that a mach2gt (modded) could hold a qx6700 at 500fsb ore 4ghz at stock voltage.Quote:
Originally Posted by irev210
Huh? Link to the numbers/screen of the person doing that please?Quote:
Originally Posted by Drag
Asus P5b Deluxe
Core 2 Extreme 6700 (kentsfield) 1,33v
2gb Gskill pc6400 GA c3
Mach II cooling
http://valid.x86-secret.com/cache/127648.png
http://82.173.172.10/Rol-Co/1m4g.png
http://rol-co.nl/images/langer.jpg
I see. It's an engineering sample. Not really indicative of actual retail chips.Quote:
Originally Posted by Drag
But impressive.
But can you hold that with sustained load, or just a quick superpi 1m?
QX6700, Asus P5WDH, 2GB TeamXtreme D9GMH, Watercooled (PA120.3 rad, 1/2", Swiftec 655 pump)
3.5GHz - 48degC idle and 65degC load after 8 hours prime
Currently now priming small 8K fft @ 2.66 GHz with 1.14 vcore under load.
This was a test i did i while ago.
Cooling was my overklokk singlestager which doesnt get very cold but holds load pretty well.
Temp in bios is about +10C lol
Max fsb and clock
http://i107.photobucket.com/albums/m...anuar07002.jpg
100% stable speed
http://i107.photobucket.com/albums/m...eng_QX6700.png
Didnt do any screens of loadtesting @ 4,14GHz, but I thought it wasnt to bad for an singlestager.
Quote:
Originally Posted by loafer87gt
I just can not believe, that f***n nv4disp.dll BSOD
is from ti4200 prehistorical eon!!!!!
try google>nv4disp.dll fix etc....
VIA used to have some fix for that problem, it did work on some nonVIA based mobos.
ti4200, 5900,6600...and even 8800 have same problem?
Check nV+VIA forums for that fix.
Advice, try ti use D3D instead OGL, it is even more faster in MAX7,8,9.
regards!
http://img213.imageshack.us/img213/3...x4startnd3.jpg
P5B DLX
Scythe Infinity + SanAce 260CFM
Room Temp->10C
Bios set -> 1.63v
Windows idle-> 1.53v
load-> 1.45v
Hi, What is the real vcore for this frequency ? and is it stable or not ?Quote:
Originally Posted by nicepun
3350 mhz Quad prime @ 1.4v on a STORM.
3970 mhz on 1m/wprime 32m @ 1.55v on a STORM.
3910 mhz wprime 1024m
I will more testing but so far 1.55vcore for 320x11 (3.5Ghz) stable as hell. I will measure real vcore and let you know.Quote:
Originally Posted by pepinorang
http://www.hwbot.org/signature.img?iid=4300&thumb=false
vcore was 1.625v
I just got done building my rig a few hours ago and am still messing around, but so far, 3.6GHZ, 1.4 vcore. Temps from 30ish to 48ish so far but have not run anything heavy other than 4 instances of P95 for about 30 minutes, and 500% memtests. This is definitely foreign territory for me because that last intel board I had was the IC7-Max3. Also, when I went for 3.7, I don't know if it was ram, fsb, may both, but it would BSOD before Vista completed loading. PWM temps also shoot up pretty fast after 3.6/1.4vcore, so I think I'll hang around this OC with the CPU and concentrate on RAM now.
This bios on this IN9 is a little scary!
By the way, running 4 sticks of:
http://www.newegg.com/Product/Produc...82E16820227159
Think I'll get much more from it?
Propably, if you get rid of this Abit (and any Abit) board and buy something else, like Evga, Asus, Gigabyte, DFI, etc.Quote:
Originally Posted by Slammin
Have a look here ;)
Vapochill LS results.
Just the CPU dependant benchmarks.
Pifast - hwbot comparison
http://slidog.bigblog.com.au/data/2/...0220230901.png
PCMark05 - hwbot comparison
http://slidog.bigblog.com.au/data/2/...0220230930.png
wPrime 32M - hwbot comparison
wPrime 1024M - hwbot comparison
Also, 3DMark06
http://slidog.bigblog.com.au/data/2/...0227212121.png
Quote:
Originally Posted by Chosen.
Hmm,
I gues the latest bios must have taken care of it because I am 100% certain, at least by every measure (Everest, benchmarks, etc.) I can make that all cores are clocking together.
I never even booted into an OS with the shipping bios, but I did save it to floppy. If I ever get bored, I may load it just to play, but I doubt I get that bored anytime soon!
Edit:
I just checked with RMClock and sure enough, two cores with multi 11, and two with multi 10! I stand corrected! I hope abit fixes this!
qx6700 on xbx2 with latest bios..
266x11 = 3ghz, all voltages at default.. stable dual orthos, gaming, 3d benchies, without problem.
266x12 = 3.2ghz, all voltages at default.. unstable orthos for core 3 n 4, but stable gaming, 3d benchies..
will try higher when i got time to play with this new processor :)
Download an older version of CPUz, adjust the multiplier upwards (try 11, 12 or 13 for example) and then run CPUz 1.33-1.35, which allows you to see each of the 4 cores separately and get ready for a surprise :)Quote:
Originally Posted by Slammin
http://img459.imageshack.us/img459/793/ourtkw4.th.jpg
easy'r way :D
run s-pi 1m and set affinity to core 1 only and look at the result
run s-pi 1m second time and set affinity to core 4 and look at the result.
you be shokked :eek:
Quote:
Originally Posted by Rol-Co
er.. that's exactly what my screenshot is showing: one Spi1M @core1 and one @core3, 3 whole seconds difference, 12x300 from BIOS, 2 cores stuck at 10x300.
See My Sig. I'm sure others can do better, But I'm stuck with slow ram, So I'll make do until I can replace It. :D
Vcore for the OC in my sig is 1.60 on Striker
Im sure i can go much higher but I I am waiting for a new phase unit before I go higher..
http://valid.x86-secret.com/show_oc?id=169633
You can also use the -core parameter in the shortcuts for the latest cpu-z to see what individual cores are set at. RMClock is easier though.Quote:
er.. that's exactly what my screenshot is showing: one Spi1M @core1 and one @core3, 3 whole seconds difference, 12x300 from BIOS, 2 cores stuck at 10x300.
A fix to this is use CrystalCPUID. Example I have my multiplier set to 12x275 right now. If I run CrystalCPUID, go through each core and from functions select 'Intel Enhanced Speedstep Control' one of them shows up with a multiplier of 6. If I then change that to 12, all cores are then overclocked evenly.
I tested this with cinebench and my rendertime dropped from 15 seconds down to 12 seconds, so it does work.
RLM
i played with 3dmax this morning and yes, the nv driver problem appeared! :(
Not a QX6700. but i been messing with a Q6600.
And i have to say if you want to test stability encoding movies with MeGUI into h.264. Tests very quickly
Ive had to clock back down from 3.4Ghz to 3Ghz just to get it to run for more than 2 hours (still going so hopefully dun crash) But Orthos + Prime 95 were running for hours on end before :S
Same here - I was happy with Prime95 running "endlessly" without errors on my QX6700 at 3.3 GHz, utntil I started real work rendering HDV in Vegas. Crahes and errors, that's what I got! Since I din't want to increase voltages any more due to the temps, I backed down to 3.1 and am very happy with it.
*See Below*
What are your temperatures? I used to have a Ninja Plus revB, but it must have been flawn as CoreTemp showed up to 90C at full load! I reverted to the stock cooler for time being, which gives me some 20C lower temps. However, I'd like to install a better CPU cooling - is you Scythe OK?Quote:
Originally Posted by SSXeon
My highest OC so far was 3.2 GHz (12x267) stable I think. I've run SuperPI 1m,2M,4M,8M,16M, Aquamark 3.0, all 3DMarks except 2001 using the default bios which came with the board (701 I guess). Now my question is what bios would give me higher OC (I'm using an an Scythe Infinity with a 120mm fan on it)?
Air results
Tuniq Tower 120 with two Antec Stealth fans on either side to assist with air flow in 23C ambients.
Max CoreTemps after around 8minutes of Prime95 Small FFT test.
Core#0 65C
Core#1 65C
Core#2 61C
Core#3 61C
8 hours Prime95 blend stable at 3Ghz with 1.248v / 1.208v (windows Idle / Prime vdroop)
http://slidog.bigblog.com.au/data/2/...0223074259.png
Swiftech ultra now installed with w/c on the cpu and northbridge.
A quick prime test on blend. CPU 315x11 (3465mhz) 1.45v in bios Memory 2:3 950 mhz 4-4-4-12-42 2.2v.
Tried 400x9, but core 3 is weak and I need 1.55v in bios to get stable.
Ambient 30c ish
Note: for some messed up reason coretemp shows I have a Q6800. Yesterday it was a Q6900:confused:
http://www.pixel-shack.com/bench1.jpg
RLM
Well go figure, after running @ 3.41 for 30mins, i lost a core. Im guessing its my room temp, as its 7* warmer in here then it was 1 week ago. Stable so far @ 3.33GHz.
Max Temps:
Core#0 69*C
Core#1 65*C
Core#2 69*C
Core#3 65*C
http://img213.imageshack.us/img213/2483/primeaj7.jpg
Hey guys. Great thread. One request. Don;t make reference to your sig, cos in time to come, when peeps are looking in on this thread, they won't know wtf your talking about when your sig is changed.
Oh and all you 266FSB huggers... crank it UP!!!!! even if you have to drop the multi, you'll get better perf.
running a qx6700 on badaxe 2. 333x12 vcore=1.55 . This chip dies at 380+ fsb. I was using x6800 yesterday at 460fsb no problems. 333x12 200\667 4.3.4.8 seems to be best performance for me. System on chilled liquid cooling.
still working out the max OC of the chip...but my biggest issue is the heat load. Running @ 1.4v with all cores running they'll get to 90+ real quick, on the zalman LED. I use TAT for 2 cores and quatro for the other two. Yikes! Air cooling just won't cut it. Got a nice h20 system just need to install it. Petra has some great customer service :D
board will do 380fsb pretty reliably in the testing ive done so far.
CPU will do one mult over on stock V, and one more with just a little more juice.
not going to spend much more time trying to eek out every bit of performance on air, but so far seems like a very nice chip.
I'm having problems with the temps of my QX6700, as well. In fact, I had to revert from Scythe Ninja RevB to the stock cooler, which does much better! I RMA'd the Ninja, and will be giving next try when a new one arrives.
Anyway, for the 24/7 work, I'd like to keep the power savings setting in BIOS (SpeedStep, C1) so that the CPU only kicks when needed (right now it's working at some 1.14V and 1.8GHz, with fans almost stopped; when I do some HDV rendering it goes up to 3.10 GHz and the temps never exceed 75C). FSB at 310 MHz is the max I can get without increasing the Vcore from Auto (which is the only setting possible when using C1/SpeedStep). I have overclocked the RAM to some 466 MHz (5:5:5:18) at 2.20V (Geil DDR2-800). What I'd like to do is increase the FSB a bit higher, so that the overall system performance is better - if the CPU temps will get too high, I can go down with the multiplier to 9 or 8 (currently 10).
My question is: which voltage settings (apart from RAM already at 2.20, and Vcore that I need to leave at Auto for the above reasons) should I change in order to assure stability with FSB = 333 or higher?
Hope i dont get trouble for asking this here but i have one problem when i overclock over 3.8 i have problems with video. Playing a divx or xvid or dvd playback and dots appear on screen. 3D gaming all is fine only video playback and dots appear. pci set to 101mhz in bios. Seems if i bring it up to 105mhz the dots get worse so that may have something to do with it. Im using 2 ati 1950 crossfire. did not have this problem with my x800 only came when i swapped it for this qx6700 chip. Did a format but no change.
qx6700
intel badaxe 2
pc8000 OCZ alpha titanium
1950xtx crossfire
I'm testing my new qx6700 on p5wdh, p5bdlx and commando.
But how do you guys get prime to run 4 threads (and thus stress all cores)?
What version are the guys with the prime screenshots above running?
I never get more then 1 thread!
QX6700
Vcore: 1.24V (load)
Quadprime, Custom + small FFTs in place:
http://img211.imageshack.us/img211/9...rimeca5.th.jpg
QX6700, 3480Mhz: 1.352V...
http://img181.imageshack.us/img181/7...rimekg3.th.jpg
QX6700, 3600Mhz, 1.416V..
http://img219.imageshack.us/img219/7...rimebx0.th.jpg
QX6700, 3691Mhz, 1.440V...
http://img390.imageshack.us/img390/3...rimejl6.th.jpg
Nice? Isn't it? :cool:
what board is that?
I wouldnt go that far, as i dont like anything over 70* ;) Defently install the H20 system, i wana see the results!
Nice, and sux that your scythe is messed up, mine runs perfect. :D As for the FSB, good question, when i get to 333, it wont boot. And i have tried different volts on the FSB, with no luck.
http://www.octeam.fr/telechargement/...95-v.25.1.html
Very nice.
P5B Deluxe Wifi/AP
CPU watercooled and very hot :D
As far as I know it's one of the best clocking boxed QX6700 in europe with watercooling...
And I don't need it anymore -> to much power, a normal core2duo cpu has enough power for games. So pm me when you're looking for a good qx6700 (bill from 15.1.2007)
Twords the end of the year, you will be using a good amount of those 4 cores. Just saying, only reason i built this system is for longitivity. You will lose money on a highend purchase, IMHO i would keep it. Supreme commander uses quad cores, not fully but they do use them. Valve is even using quad cores for their next game, and possibly HL2:Ep2. Its your choice, but i just feel future proofing is the ultimate, esp with a cpu.
BTW what clocks did you get out of the quad? :)
Pedro isn't a normal user... pretested hardware from the manufacturer doesn't count, same as ES cpus... they are not comparable to normal boxed/retail CPUs...
I haven't tried the new prime yet, thanks for sharing.
testing my new QX6700 ES week33
air cooled zalman 9500 , crappy cheap bord , could not set any lower vcore or higher FSB or multie at this time :p:
bord is foxconn P965 , Evga 680i died yestarday :( , still have to rma P5W DH and a 2 P5B :p:
http://i3.photobucket.com/albums/y90...zqaudprime.jpg
EDIT : not to all user's plz post week also ;)
My 2 results both L647 F085 and both on XBX2 boards.
One in a Koolance case which just doesn't have the cooling power. 3.33Ghz(10x333) at 1.4V 75C load.
Rig pic
http://home.comcast.net/~msibios/DCP02445.JPG
The other is cooled by Fuzion block, PA120.3, and MD-20. 3.8Ghz(11x346) at 1.575V 70C load. It was 10hrs prime stabe but I didn't like it for my 24/7 machine, so I backed down to 3.66Ghz(11x333) at 1.5V at 65C
Rig pic
http://home.comcast.net/~msibios/DCP02447.JPG
EDIT: meant 11x346 to 246
Where can i get the new prime95? Been looking for it but can only find older version? Anyway, 4x orthos works as you can choose core there as well.
I have my rig running at 3.4 24/7 now, works very well, tried 3.6 but that isn't 8 hours orthos stable yet. Still trying to figure out what to do on the commando voltages. everything but vcore on auto seems to work just as well as much higher voltages!
Shame the p5w dh didn't like to oc the qx6700, that board is so much easier to work with. But that wouldn't even give me 3.4 stable.
I can't boot into windows from 3.9 onwards. BSOD
I have the L647 F078 stepping
damn read first page off the thread !
btw gone post lowest temps on air :D
it hass 8°C in my room and got a shot off idle temps windows @ 2core 13°C and 2 others 18°C :D
Yes the math works better that way doesn't it. :p: I couldn't find a setting for the RAM(the Ballistix from you) at 346 that made me happy, but at 333 they run 500 4-4-4 at stock volts(8hours memtest)so that seemed to work better all around for the system another reason I backed down from 3.8Ghz.
@Catscit
What temps are you getting on your week 47, I'd like to compare.
Asus P5WDG2 WS Pro @ 400FSB
http://img180.imageshack.us/img180/1028/prime400iy6.jpg
Scythe Infinity + 120mm @ 55CFM
ftp://mersenne.org/gimps/p95v251.zip
EDIT: Just looked at the first page of the forum and found 25.2 but I haven't used it yet http://www.benchmarkhq.ru/fclick/fclick.php?fid=246
Finally figured out why i couldnt go 333x10. My memory was being oced to 378x2, so I locked it @ 1:1 and runs the 1333 strap perfectly. 340x10 is a no go, even with 1.38v. So ima stick with this speed, its good enough for now, or when Penryn comes out. :D :D
They do if you have faster memory. Im happy with this setup, i wanted cheap 667, as ima get 64-vista and another 2GB of it. Still i can live with 1333MHz fsb.
http://img501.imageshack.us/img501/3...our2222zn5.gif
All my cooling can handle. went to 71*C load on 2 chips, 65*C for the other 2. BTW i oced 666Mhz over stock, haha.
well ive got my h20 setup installed...and its running much cooler...but I seem to have hit a CPU limit @ 3.45 or so. It will run 3.45 nice and stable @ 1.4V, (266mhz * 13) but wont run 280 even with 1.6V....really odd.
That chipset heatsink gets *hot* though...I may slap a fan on it and see if that helps.
It is a big cooler, but its too hot to even touch...I may try reseating it with some arctic silver as well.
My QX6700 should be here by the end of the week hopefully my P5WDG2 WS Pro can live upto it im on a e6600 chip at 3.7ghz 1.1 1.65v with temps -46 idle on my ss phase unit i just hope to squeeze a bit more out of her :).
Just got this system together and can't seem to get it past 3.2Ghz (320 x 10). I've upped most of the voltages to 1.5 and mem at 2.2. Memory is set to 5,5,5,15 2T at 800mhz. Cooling seems pretty good at load ~44.
I'm pretty new to OCing but was hoping some of you pros could get me above this with a stable setting on the fsb and memory. I now realize that the 680i Asus boards aren't that great for ocing the Q which is a little disappointing now that I have the board.
I've had problems with temperatures of my QX6700 since the very beginning; even with the Ninja Plus revB, they reached 90C under full load of Prime95! For this reason, I could never exceed 3.0GHz (and even at this speed, was uneasy about temperatures). I replaced the Ninja with another unit, cause I thought the first one could be defective (and/or the thermal grease not properly applied), but it didn't help much.
I'm thinking now about the Zalman CNPS9700NT with ZM-STG1 grease; does anyone have experience with that one one the QX6700?
Currently running at 3.3 with 1.3V, FSB 1333~1.4V, Memory 1066~2.0V, CAS 5. Load temps are 47-49C which should get better as the AS5 breaks in. NB is 37C under load and it is all H20 cooled. I have had it stable to 3.6 with ease but my load temp was 58C and I have only had it a week so I backed off until everything breaks in. This is a very easy chip to OC with the eVGA 680i MB.
Since I'm after as quiet operation as only possible, I'm keeping the C1E auto and SpeedStep on (Asus P5W DH Deluxe). This leaves me with all voltages on auto, and the best super-stable result is 3.15 GHz (with RAM at some 395 MHz 4:4:4:12). Do you think it is a good result, considering default voltages?