She asks to do it :).Quote:
Originally Posted by raikkok
Printable View
She asks to do it :).Quote:
Originally Posted by raikkok
Hi there i have
Q6600 B3 and DFI LANParty UT P45-T3RS ram Corsair 1333C9DHX 4GB
i can't overclock it any more than 3.4Ghz the cooling is wc D-tek v2
powersupply corsair 850w
bios is(LP45D703) latest one
CORE QUAD Q6600 Stepping B3 VID : 1.3250
DFI LANParty UT P45-T3RS
Corsair XMS3 DHX CM3X2048-1333C9DHX /9-9-9-24/ 1.7V vers.3.2 4GB
Corsair TX850W - CMPSU-850TX
Asus 8800GTX clocked
GPU Shader Clock 1458
GPU memory Clock 999
Creative Audigy 2 ZS Platinum pro
6x500Gb Maxtor Diamond Max 22 RAID 5 SATA2
2x80Gb Maxtor RAID 0 SATA2
Windows XP Pro x64
here are my spec in the bios
i think a have a lot of wrong set parameters in the bios settings ...
Best Regards
Mitko
Genie BIOS Page BIOS Field
Exit Setup Shutdown……Mode 2
Shutdown After AC Loss……enable
AC Shutdown Free……disable
OC Fail Retry Counter……1
OC Fail CMOS Reload……disable
CPU Clock Ratio……9
//// DRAM:FSB ratio……20:10 ////
Target CPU Clock……3420
CPU Clock……380
Boot Up Clock…… auto
CPU Clock Amplitude……800mv
CPU Clock0 Skew……300ps
CPU Clock1 Skew……100ps
DRAM Speed……333/1333
Target DRAM Speed……1521
PCIE Clock……100 MHz
CPU Spread Spectrum……Disabled
PCIE Spread Spectrum……Disabled
CPU Feature Page
Thermal Management Control disable
PPM (EIST) Mode disable
Limit CPUID MaxVal disable
C1E Function disable
Execute Disable Bit disable
Virtualization Technologydisable
Core Multi Processing enable
DRAM Timing
Enhance Data Transmitting turbo
Enhance Addressing fast
T2 Dispatch disable
Flex Memory Mode……?
Cas Latnecy Time (tCL)……8
RAS# to CAS# Delay (tCRD)……8
RAS# Precharge Time (tRP)……8
Precharge Delay (tRAS)……22
All Precharge to Act……auto
REF to ACT Delay (tRFC)……auto
Performance LVL (Read Delay)……auto
Read delay phase adjust……auto
Write to PRE Delay (tWR)……auto
Rank Write to Read (tWTR)……auto
ACT to ACT Delay (tRRD)……auto
Read to Write Delay (tRDWR)……auto
Ranks Write to Write (tWRWR)……auto
Ranks Write to Read (tWRRD)……auto
Read CAS# Precharge (tRTP)……auto
All PRE to Refresh……auto
Clock Setting Fine Delay
DLL and RCOMP Settings……auto
CH1 DRAM Default Skew……Model auto
CH2 DRAM Default Skew……Model auto
RCOMP Setting…Model auto
Fine Delay Step Degree……70ps
Ch1 Clock Crossing Setting……more agressive
DIMM1 Clock Fine Delay……current 324
DIMM2 Clock Fine Delay……current 324
Ch1 Control0 Fine Delay……current 270
Ch1 Control1 Fine Delay……current 270
Ch1 Control2 Fine Delay……current 256
Ch1 Control3 Fine Delay……current 256
Ch1 Command Fine Delay….current 302
Ch2 Clock Crossing Setting……more agressive
DIMM3 Clock Fine Delay……current 102
DIMM4 Clock Fine Delay……current 232
Ch2 Control0 Fine Delay……current 124
Ch2 Control1 Fine Delay……current 124
Ch2 Control2 Fine Delay……current 256
Ch2 Control3 Fine Delay……current 256
Ch2 Command Fine Delay……current 170
Ch1Ch2 Common Clock Setting……auto
Ch1 RDCAS GNT-Chip Delay……auto
Ch1 WRCAS GNT-Chip Delay……auto
Ch1 Command to CS Delay……auto
Ch2 RDCAS GNT-Chip Delay…auto
Ch2 WRCAS GNT-Chip Delay……auto
Ch2 Command to CS Delay……auto
Read Delay Phase Adjust
Ch1 Read Delay Phase (4~0)……7-7-7-7-7
Channel 1 Phase 0 Pull-in……enable
Channel 1 Phase 1 Pull-in……enable
Channel 1 Phase 2 Pull-in……enable
Channel 1 Phase 3 Pull-in……enable
Ch2 Read Delay Phase (4~0)……7-7-7-7-7
Channel 2 Phase 0 Pull-in……enable
Channel 2 Phase 1 Pull-in……enable
Channel 2 Phase 2 Pull-in……enable
Channel 2 Phase 3 Pull-in……enable
Voltage Settings
CPU VID Special Add……1.4625
DRAM Voltage Control……1.69
SB Core/CPU PLL Voltage……1.55
NB Core Voltage……1.379
CPU VTT Voltage……1.375
Vcore Droop Control……disable
Clockgen Voltage Control……3.45
CPU GTL 0/2 REF Volt……0.63
CPU GTL 1/3 REF Volt……0.63
North Bridge GTL Ref Volt……0.63
FSB Vref……AUTO
Anyway im still pushing this pig Q6600, but my board's vdroop and vdrop is getting ridiculous, for example when I set in BIOS 1.6v, I get 1.580 idle and 1.52v load. like wtf! and that's with the load line calibration set to enabled and with the pencil mod done like a billion times.
Small ffs, 35 min pass, on cheap ass water, vmch and cpu pll mods
http://img686.imageshack.us/img686/3647/phprphcaqpm.jpg
Hi, I also have a relatively recent Q6600 that does quite good f.e. 9x400 1.35 Prime stable. I wanted to sell this CPU but before that I wanted to give it a push for the last time, this is why i am posting here.
Has anyoneone who is running above 1,5V ever heard their boardīs capacitors sing? It occurs when running Prime95 for me on a Ud3p and TP45 as well. I find this scary anybody noticed this as well?
Are you sure it's not your PSU? I had to replace mine, because it used to "whine" under high load, (video here http://www.youtube.com/watch?v=qmdRQ_qb_8c ). Also few months ago I replaced my P5B vanilla after 2.5 years of use because one of the capacitors became leaky.Quote:
Originally Posted by Sebastione
Pretty sure. One time i was using the TP45 sitting on its box with the foam under it. When i moved it away the foam was melted....
It doesnt matter anyway now as i will just sell the chip but i never understood how people were able to put 1,6 through one regularly. The sound i get resembles a mixture of a snake and steam. The higher the Vcore and the higher the load the worse it gets.
This is my poor P5B "vanilla", 1.63V while playing CS:S, ofc no cooling for mosfets either.
http://img44.imageshack.us/img44/7452/rszhpim5625.jpg
Kids, don't try running high volts on a single connection
I got a Q6600 a couple of months ago, it's primarily been used in my VM host machine@stock voltages. This is from an older 2007 batch with a very low VID of 1.2250v. 3.6GHz only took 1.360v on load using my UD3P; the 1.3250v Q6660 I had before needed 1.5v+.
http://img706.imageshack.us/img706/1...400x9prime.jpg
I'm a little late to the party!:rofl:
This one has been running my file server and the board died so I had to get a new one and have to play. This Q6600 is from black Friday 2007:up:
I'm sure it has more to give.....This is blend at 26 hr
http://i76.photobucket.com/albums/j7...1296v-26hr.jpg
What's the consensus on a G0 that has had 1.85v though for most of it's life. Thinking about getting one 2nd hand but that high voltage is causing me concern as I see most people using 1.6v max on H20.
This was on air (Thermalright Ultra 120A - not the TRUE, but the model down) and the temps were said to be 65-70c after a few hours of SETI.
It was running at 3.6Ghz but recently he says he's dropped it to 3.5Ghz.
Is this chip likely to die in the next year or so?
Wait what? 1.85V for 3.6GHz? Something is a bit wrong here. You sure you haven't misspelled anything?Quote:
Originally Posted by Wam7
Anyway CPUs don't die that easily, especially 65nm. And most people don't run higher than 1.6V because the temperatures get out of hand.
I double checked with him and he was pretty sure that's what he was using! (I am still unconvinced)Quote:
Originally Posted by NoRogeR
Anyway, I got it for a very good price so time to see how bad it is...
1.85V?.... That's crazy I'm running 1.5V at 3.6GHz with a VID of 1.325V
Anyways I was hoping to hit 3.8GHz (475*8) since I was running 3.4 fine with an ARCTIC COOLING Freezer 7 Pro, but it seems to demand over 1.6V. I'm running it with a NH-D14 now with aftermarket fans (Nocuta fans have too little static pressure for my taste but I am using them in my case)...
do you guys think it'll be safe with 1.6V+?
I know temperatures might be an issue at that voltage, but if they're not could I expect to have this running for a few more years (at least 2.5)?
Edit: btw the highest temp I've hit after hours of Prime95 is 61C... Don't think I've ever gone over the low 50s with normal applications
http://img63.imageshack.us/img63/2048/36ghz.png.
1.5v and only 61C tops!
Anyway my Q6600 keeps getting an error on the 4th core at 400x9 with 1.575 bios volts.. I already tried upping the NB voltage to 1.33 and SB to 1.60, VTT is at around 1.3xxxx and ref is at 67% anything else worth trying ?
Posted a pic of my voltage settings (note I'm running at 450*8 so you're NB/SB voltages should be lower maybe?)... also a word of caution: my mobo has major vDroop/Drop. The 1.5375 core voltage is actually 1.5V as shown in my previous post and some of the other voltages might also be a bit lower. I'm running it on a Gigabyte EP45-UD3P btw.
http://img38.imageshack.us/img38/514/biosj.jpg
sorry for posting this 3 years late
Here is my Q6600 B3 build from 07
24/7 setting, still running strong
I rushed to get this B3 when Q6600 was just released
http://img519.imageshack.us/img519/6139/q66z.jpg
Striker Extreme
G.Skill 4Gb
8800GTX
watercooled
here`s my O/C http://i141.photobucket.com/albums/r...7/Untitled.png
set it as 9x333 and the ram running at 800mhz and the cpu voltage is 1.288 not sure if i thats good ?
Yeah that's great, normally for a 3.0GHz overclock I don't touch the cpu voltage, but I guess that depends on your processor/mobo. You shouldn't run into any troubles however until you start hitting more than 1.5V though some people might tell you otherwise.:up:Quote:
Originally Posted by yatesy17
With a H50 you can probably get to 3.2 without much extra voltage.Quote:
Originally Posted by yatesy17
iv pushed it to 3.2ghz now . the cpu voltage has stayed the same but the ram is running at 879mhz now . here is the results http://i141.photobucket.com/albums/r...Untitled-1.png
I have a Abit IP35-E mobo , how much voltage can the NB take with its passive cooler, 1.33v ?
I'm running my mobo, the Gigabyte EP45-UD3P also with a passive heatsink with the NB at 1.5V, I'm sure it could take more, but the VCore is really limiting me from overclocking any further. What you want to do is check your temperatures mine hover in the 30s (C) according to Everest, I wouldn't get worried until I was hitting the 50s though.
I think my IP35-E has no sensor for the NB, even at stock speeds the heatsink feels scorching hot, can put my fingers on it for over 3 seconds. the list on my everest is mother,CPU1-4,PWM,GPU and the harddrive.
yeah it doesnt its also poo for vdroop lolQuote:
Originally Posted by Demo
get a fan on it as it limits your FSB clock quite a lot I found:up:
Who here feels the Q6600 still has a good amount of life left in it? when playing games my G15 says processor usage is around 60% and that includes other tasks in the background. Im running a fairly conservative 3.6 GHz overclock with my x48 (no board voltage changes, no VTT changes 1.44V on water) and I dont think I need to bump it up to 3.8-4 Ghz. I see that being done later on once it gets older and I feel its more expendable if it blows up.
What are your thoughts on this?
I agree there's no need to jump to i7 for gaming or most applications. The best you're really going to get out of an i7 is better memory performance, but you say you have an X48, so you should be able to get similar performance in most applications/games.
I however do want to make the jump for 1366 because my Q6600 refuses to go above 3.6GHz without going significantly over 1.5Vcore. I'm thinking of hitting 4+GHz on a W3520. I would have maybe gone with a Q9550/9650, but for being "outdated" they are very overpriced.
I certainly would agree with the Q6600 being a decent chip for average use even in the future. I still have one myself as well as two Q9550s which overclock very nice on water. I also have i7 920 systems on water for benching but all my main systems including the one I am typing this on is a Q9550 on water at 3850 and is more then enough for any software I am using, they are great chips.
Quote:
Who here feels the Q6600 still has a good amount of life left in it? when playing games my G15 says processor usage is around 60% and that includes other tasks in the background. Im running a fairly conservative 3.6 GHz overclock with my x48 (no board voltage changes, no VTT changes 1.44V on water) and I dont think I need to bump it up to 3.8-4 Ghz. I see that being done later on once it gets older and I feel its more expendable if it blows up.
What are your thoughts on this?
I've been running a Q6600 for the past year and half and there's nothing I've thrown at it that it can't handle. Outside of stress programs, there's not a game out that I've haven't been able to play at 1900x1200 with all the eye candy enabled. I'm running 9x430 on a 333 strap; my RAM runs at 1722 Mhz. I'm running a SSD for my OS and a pair of RaptorX in RIAD 0. The RAID is about as quick as the SSD in regard to sequential reads / writes. My bottleneck resides in my video cards. Looking at synthetic benchmarks, my system appears a bit lackluster compared to a x58 system. x58 is only about 5%-7% faster than what I have now. I've got a QX9650 waiting in the wings that I'll pop in soon. I hope I can clock the hell out of that chip. x48 is a good chipset, but it was cut short by x58. The C2duo quads are very good relevant chips. Judge for yourself by the longevity of this thread. It's close to 3 years old. I didn't even get that much longevity from my S939 board with a dual core Athlon 4400+!
The Q6600 @ 3.2Ghz or more is still very decent for gaming, despite what gaming benchmarks show between the Q6600 vs i920 the real world difference is not that noticeable.
Btw my computer crashes at 8x425 but not at 8x426, any thoughts ?
I read in an article a while ago that motherboards have certain frequencies at which they are completely unstable regardless of voltage. Can't seem to find it now but it was exactly what you mentioned: they couldn't post at a certain frequency but raising it [by one] they were able to get it stable.Quote:
Originally Posted by Demo
I really want to find that article, but can't find it right now if I do I'll post the link.
http://img690.imageshack.us/img690/6...beast37ghz.png
SLACR L744A996 - Still playing around with PLL, VTT and VCore. Hoping for 3.8GHz.
Nice OC are you on water??
Edit: never mind... just took a look at your sig :rolleyes:
For me with PLL set to 1.7 it reboots randomly after 30 mins. When set to 1.77 it's stable for days. I haven't yet tried 1.75 or 1.73. It's a work in progress...:cool:
Here's my system before everything else in it...Quote:
Originally Posted by Zayras
(I had to do a little dremmel work on the waterblock to make it clear the heatsinks)
Ok, I know I should really consolidate all my posts, but I have a question to all you:
What voltages are you using for Vcore, PLL, VTT, GTLref?
1.5 for Vcore (VID of 1.325 :mad:), 1.5 PLL, 1.32 VTT, .78 Ref (I think) @ 3.6GHz
not completely stable yet, but passed an hour of Prime95 so it needs minimal tweaking
Edit: nice job on the block dremel work, looks like a tight fit :up:
One more thing are you running 24/7 @ 3.7GHz with those voltages? I might give it a try I just got some new fans for my NH-D14 and it's having a hard time hitting 60 ^_^
I get random restarts if I set the Vcore to 1.5. It was running on as low as 1.425 volts, but wasn't very stable. I've never had prime95 give me an error with this chip. All it does it random restarts if I get the settings wrong. Right now I'm at the stage where I am lowering each voltage setting to see how low it can go. Lower vCore means lower temps, and it seems that while my chip can handle 85c for hours, when I do an 8+ hour run, I come back home and find my computer has restarted unexpectedly. That's why I keep the vCore the way it is...Quote:
Originally Posted by Zayras
@alexanderhuzar
Your PLL is too high, 1.57v should do :up:
Also playing with the MCH (and to a lesser effect CPU) GTL's helps a fair bit imo.
vcore: 1.53vQuote:
Originally Posted by alexanderhuzar
pll: 2.00v
vmch: 1.84v
vfsb: 1.5v
cpu gtl 0.63x
nb gtl 0.67x
What's the max vpll people put on a Q6600? and vmch on a P35?
not the lowest voltages i know.Running at 3.8GHz solid. I'm vdroop limited, anyone got a good P5K-E vdroop mod?
http://www.overclock.net/intel-mothe...p45-ud3-4.htmlQuote:
What's the max vpll people put on a Q6600? and vmch on a P35?
First 3 posts or so cover PLL Voltage, I would be scared of 2V, I'd say bump it down to at least 1.8V, but I'm no expert and it's your chip ;). For the MCH or NB voltage I would say as long as it's not at extreme temperatures I wouldn't worry too much.
They are talking about 45nm chips. My SLACR is a 65nm...Quote:
Originally Posted by Zayras
True, but even looking up "max Q6600 PLL" on google you get most people saying 1.6-1.8V are the "safe" voltages.
I've lowered my voltages to the following:Quote:
Originally Posted by Zayras
1.4325 Vcore
1.57 PLL
1.35 VTT
1.36 MCH (P55 northbridge at 463 MHz, but could go lower).
So far, so good. Passes Intel Burn Test - 10 runs at default settings.
Nice :up:
I've been playing with the GTL-REF from 61% to 67% anything lower than 67% gives me a random BSOD when I close some programs, the most prominent of them all is Yahoo Messenger.
Hello everyone. Found this page looking for info to oc a bit more my Q6600.
I'm using stable at 3.6GHz 24/7. The main board is a Gigabyte EP45-UD3P rev 1.1. Started with vcore 1.5 and lowering a notch couple of times.
Seems the CPU is getting used to :D Now it's working at 1.4875. Have to try a bit more though.
Glad to have you here Ivan :up: The giga board will up quads like no other board, you just have to find the refs / skews and volts that works for ya...Quote:
Originally Posted by ivanpacheco
for my quads always loved for 1,2VTT .785CPU ref and .80MCH, and if you wanna push mem .84MCH
so with skews i use 900/900/0/50
and bump MCH core, 1,4+ if you are using 4GB and OC' mem
Cheers :up:
Sergio
wow high vcore, but high clock:p:Quote:
Originally Posted by Bootup05
4 GH with v1.4 :up:
http://img25.imageshack.us/img25/7464/4ghhypi16mb.jpg
Good to know people are still having fun with the Q6600!
Hey guys, should I only see a small fluctuation with vcore?
BIOS, everest and cpuz all report the same, v1.224 -> v1.232. Even at 100% load with Prime95. So this is my total vdroop?
Just a small clock while I wait for the new cooler to arrive.
CPU Voltage: 1.2625v
CPU Voltage Reference: AUTO
CPU Voltage Damper: Enabled
CPU PLL Voltage: 1.50v
DRAM Voltage: AUTO
FSB Termination Voltage: 1.20v
North Bridge Voltage: 1.25v
North Bridge Voltage Reference: AUTO
South Bridge Voltage: AUTO
Yeah, that looks about right. In my experience, however, with the UD3P motherboard if you bump up the voltage one notch or so it shouldn't fluctuate anymore.
Ok thanks Zayras.
The voltages are fine at their lowest until I get up to 400+fsb right?
It depends on your mobo and chip, but generally you shouldn't have to raise them much, if at all... I'm running at 475 FSB right now lol, tried getting 480+ but it crashes after a few minutes of stress testing :(
My old score.
Zayras, what voltages do you run for 475? and are you running 1:1 with the lowest strap?
I found even at 333 I couldnt run 1:1 in the tightest strap, 3dmark11/vantage would crash.
Should I force the lowest setting for CPU Voltage Reference while i'm not pushing it?
Is your PCI-E set to 100MHz? Here's my settings:
http://img823.imageshack.us/img823/9852/475fsb.png
I might be able to get 8x multiplier, but that would require a lot of voltage from my chip.
http://img412.imageshack.us/img412/8...sbvoltages.png
Word of caution, my vCore and CPU PLL are set rather high, I need to find stable and more suitable voltages for them (my mobo also has a vDroop/drop of ~0.02 volts).
Oh memory is running at 1:1 with 5-5-4-15 timings, although I haven't really tried pushing it much.
Also a lot of boards can't do much more than 450MHz FSB with a quad.
Yea pcie is set to 100. Your volts do seem a bit high, what temps you getting under load?
No problem with 1:1 now, however I always fall back to running on a divider which lets the ram run closer to its stock 1066 as it performs better.
I tried 3.6ghz last night (9x400), though I had to set 1.5v in bios. Dropped and drooped to 1.452v -> 1.462v which was okay because temps swang from 38c -> 58c. Not Prime stable though...prolly help once I put a sink on the upper mosfets.
I'm trying to get at least 3.6 24/7 stable...though 3.8 would be nice, whats your highest stable?
Would PLL 1.6v help with getting mine to Prime?
Umm... I don't think I've gone over the low 50s while stress testing.
My Q6600 (VID 1.325) required a bit more than 1.5V to be completely stable I think, but I was running at 450x8.
Raising the PLL can help when OCing the FSB. By the way, what's your chips VID? Maybe you need more Vcore, also is CPU Termination/VTT set to Auto or have you set it yourself? Raising it could help.
Whats your full temp range @ 3.6.. and is that CoreTemp?
Mines also VID 1.325. Are you saying PLL only contributes to getting a high FSB, rather than getting cpu stable?
I dont have the VTT option, only the following:
CPU Voltage
CPU Voltage Reference
CPU Voltage Damper
CPU PLL Voltage
Hi guys, i need yor help, i would like push my q6600 as much as i can on air for now. Currently running 3.4 Ghz at 1.3V and vid is 1.2875V but even with this settings is not 100% stable it fails IBT at extreme settings in second linpack and temp was 62 C max at that test otherwise 28 - 30 idle with
IFX-14 and two gentle typhoons. Please have look at my bios and give me advice. Thank you
ASUS P5Q DLX Template : 3.4
JumperFree Configuration Settings
AI Overclock tuner: M
CPU Ratio Setting: 9x
FSB Strap to North Bridge: A
FSB Frequency: 380
PCI-E Frequency: 100
DRAM Frequency: 1013
DRAM CLK Skew on Channel A1: A
DRAM CLK Skew on Channel A2: A
DRAM CLK Skew on Channel B1: A
DRAM CLK Skew on Channel B2: A
DRAM Timing Control: M
1st Information :
CAS# Latency: 5
DRAM RAS# to CAS# Delay: 5
DRAM RAS# Precharge: 5
DRAM RAS# Activate to Precharge: 15
RAS# to RAS# Delay : 3
Row Refresh Cycle Time: 55
Write Recovery Time: 6
Read to Precharge Time: 3
2nd Information :
READ to WRITE Delay (S/D): 7
Write to Read Delay (S): 3
WRITE to READ Delay (D): 5
READ to READ Delay (S): 4
READ to READ Delay (D): 5
WRITE to WRITE Delay (S): 4
WRITE to WRITE Delay (D): 7
3rd Information :
WRITE to PRE Delay: 14
READ to PRE Delay: 5
PRE to PRE Delay: 1
ALL PRE to ACT Delay: 6
ALL PRE to REF Delay: 6
DRAM Static Read Control: D
DRAM Read Training: D
MEM. OC Charger: E
AI Clock Twister: Lighter
AI Transaction Booster: M10
CPU Voltage: 1.325
CPU GTL Voltage Reference (0/2): A
CPU GTL Voltage Reference (1/3): A
CPU PLL Voltage: 1.52
FSB Termination Voltage: 1.30
DRAM Voltage: 2.02
NB Voltage: 1.24
NB GTL Reference: A
SBridge Voltage: 1.10
PCIE SATA Voltage: 1.50
Load Line Calibration: E
CPU Spread Spectrum: D
PCIE Spread Spectrum: D
CPU Clock Skew : A
NB Clock Skew : A
Advance CPU Settings
CPU Ratio Setting: 9
CPU VID: 1.2875
C1E Suppport: D
Max CPUID Value Limit: D
IntelŪ Virtualization Tech: D
Vanderpool Technology:
CPU TM Function: E
Execute Disable Bit: E
@Chestnut: PLL won't really do too much if you're not pushing FSB, the factors you should focus on the most for now are getting the right Vcore, Termination/VTT (kinda weird if your mobo doesn't show it... I think it might be shown as FSB Termination, but am not sure), and NB voltage... try turning the Damper on, see if it helps with vDroop
For temps I think I was in the low-mid 30s idle, low-mid 50s gaming, low-mid 60s stress testing, but those are just approximations, I'll have to set it to 3.6GHz again to give you my temps.
I haven't tried getting it stable above 3.6GHz for fear of frying my chip lol... I'd be more willing to try it now.
@tommyk999: You probably want to raise you're Vcore quite a bit and try bumping up FSB Termination as well, see if it helps you get a lower Vcore.
What is highest FSB termination i can put 1.50?
It's 1.55V I believe, I have it at 1.52 with a Vdroop/drop of ~.02V it's probably around 1.5V.
You might also want to try increasing NB voltage a bit, see if it helps.
I have my Q6600 (VID 1.325) running at 450x8, 1.5375Vcore BIOS, 1.504 actual. Temperatures are (they vary since I have a fan controller, at times I go for performance, others for silence): low-mid 30s idle, low-mid 40s gaming (Civ V, which stresses the CPU and RAM quite a bit), high 40s to mid 50s stress testing. ...CPU fans are running on full, 2 Nanoxia 2000RPM fans (they actually aren't that loud for 2000RPM) and 1 Nocuta NF-P14.
Found out the cause of instability :)
Cooler not seated properly so lapped Cpu aswell and reseated and all good.
IBT maxstress 67C still Primeing and will push it to 3.8 later today.
CANT GET 3.8GHZ STABLE NEED HELP PLS
My settings:
3.8Ghz
Vcore 1.53
GTL A
CPU PPL 1.52
FSB Term 1.32
Ram 2.10
NB 1.24
NB GTL A
SB 1.10
Pcie sata 1.50
Asus pq5 dlx, Corsair CM2x2048-8500C5D, IFX-14 2xgentle typhoon, Corsair TX750W, 460GTX 1Gb
Your PLL is quite low for 3.8GHz, though not every chip can do 3.8GHz at 1.53v.Quote:
Originally Posted by tommyk999
Ok thank you i try play with pll and rise vcore a bit
Hello, I have Q6600 G0 VID 1,2625v, and a DFI LP JR P45-T2RS.
My core #2 is heavily limiting the overclock, it cannot even reach 3000MHz at def Vcore (it crashes in few minutes with prime95 Small FFT), while the others are stable even at 3400MHz!
CoreTemp says my temp are fine, around 50° in full load.
I need processing power essentially for games, so it would be better to have 3 cores at 3600MHz than 4 cores at 3000MHz, is there a way to achieve this, can I disable the weak core?
Post your bios settings
if you have your game runnin.... hit ctrl+alt+del
or simply start task manager
under the process tab... choose the application or game ...right click on it...then click on "set affinity" select which cores you want to run...
........
but really....do you really know if that core is not good?
in Genie BIOS Settings, I have set:
- CPU features (C1E, EIST etc.) all disabled except Multiprocessing
- CPU clock (fsb) to 430MHz
- CPU Ratio to 9x (3440MHz)
- DRAM Strap to 333/800 (1033MHz)
- PCI-E clock to 101MHz
- CPU VID +200mV (1.376v under load in Windows)
- PLL 1,75V (just to be safe, I saw it is needed only from 470MHz FSB)
- all other settings to default, including RAM to AUTO (2x2GB Corsair PC2-8500)
These settings are pretty much stable (8hours of prime95 small FFT).
But if I set a low vcore voltage of +50mV without touching the other settings, and I run prime95, the 3rd core gives immediately an error, while the others run for hours (but if the weak core causes troubles in windows, the system hangs).
If I go on, leaving the same +50mV but raising the FSB, in prime95 the third core gives obviously an immediate error, and within a pair of minutes it is followed by the fourth core.
If I raise FSB againg, without increasing voltage, the third core gives immediately the error and the fourth follows within seconds (then system often hangs).
By msconfig, I can set to start windows with only 3 cores instead of 4, but there is no option to choose which cores to use, so it powers off the last one.or even with only 2 cores, it powers off the third and fourth ones.
So I think it's a matter of dies, the first one, with core #0 and #1, can reach 3800MHz, the second die, with core #2 and #3, struggles even at 3400MhZ.
The two cores of the second die ar also run 5-6 °C cooler than the two first ones.
Is there a manner in BIOS to apply a superior voltage to die2, maybe with some parameters I don' understand like skews?
For the last 3 years I've never really refined the voltage settings that I have on my Q6600.
Im using an Abit IP35-E motherboard with a G0 Q6600 @ 3.4Ghz
My current settings are as follows
426x8 ( 425 seems to be really unstable and BSOD's alot! )
1.475 Bios Vcore, thats like 1.360 when fully loaded by Prime95
I have the GTL Ref at 67% << Dunno if this needs tweaking
vDIMM at 1.95 ( using 8Gb 5-5-5-18-2T )
vNB 1.29v
PCIe 101Mhz
VTT 1.27xx ( other options are 1.23xx and 1.2000 , this might also need tweaking ? )
My CPU cant do 400x9 even at 1.5xx voltages.
What stepping do u have? G0 or B3? Check with CPU-Z (under CPU Revision should be either B3 or G0).
Also did you say 1.475 Vcore in Bios, but 1.36 while stress testing with Prime95? Can you double check that with CPU-Z... if you haven't already?
If after double-checking you confirm that that is the case, try Enabling LLC (Load Line Calibration) in BIOS, chances are it's disabled.
Also what PSU do you have?
Oh... yeah, you might want to raise VTT and NB Voltage, see if that helps.
G0, The Abit IP35-E has no LLC option. The Vdrop for this mobo is normally that high when I go higher with the Vcore, Yes I've double checked that under CPU-Z for the last 3years and 1 month. I have a silverstone 750ZF.
Raising VTT and NB voltage will help me with exactly achieving what ?
Raising NB Voltage can help you with the higher FSB clock.
Raising VTT can help you lower your Vcore once you get a more or less stable voltage. It's also called FSB Termination Voltage since it's related to the FSB...
...But I had actually confused it with PLL, so it might not be as important for your overclock :rolleyes:Quote:
System Bus Input Supply Voltage, often referenced as FSB termination voltage (VFSB) simply the voltage the fsb is running on. logic levels on the fsb are relative to this voltage. higher vtt results in higher logiclevels and thus in increased noise ratios. increasing vtt might help overclocking.
My next step on the NB voltage from 1.29 is 1.33.. at 1.29 the NB heatsink is so hot that I cant keep my fingers on it for more than 2 seconds.Quote:
Originally Posted by Zayras
My VTT increase is bound to the NB voltage.
Heh, in that case you might not want to raise the voltage. Also, some chips can't do 3.6GHz (that's a 50% overclock after all). Mine needs a lot of voltage to stay at 3.6GHz (more than 1.5V, but then again my chip's VID is 1.325)
My NB voltage goes up to 1.5V I believe, but I also believe Gigabyte has some pretty good heatsinks on this mobo
IIRC, my VID is also 1.325.. Would you happen to know if 378FSBx9 or 400x8 would perform better, thats 3.2Ghz vs 3.4Ghz.
Well, 378x9 would be a lower FSB, which means less heat for your mobo.
400x8 means higher FSB and certain programs might benefit from that, but I believe overall you would see more performance with the extra 200MHz on the CPU.
Im getting mixed results with different games though, its a 50-50 on the result, I guess ill settle with the one that requires lesser voltage.
OLD SCORE
http://i50.tinypic.com/2mi2rgy.jpg
Great! Had this one too... mine did 3.92 ghz max on IP35E :p
Quote:
Originally Posted by Gaul