Help getting Q9550 stable over 3.6Ghz

K

keldegar

Limp Gawd
Joined
Dec 11, 2006
Messages
339
Hey everyone, trying to figure out why I can't get my Q9550 stable over 3.6Ghz.

Here is my config:

Gigabyte GA-EP45-UD3P
Intel Core 2 Quad Q9550
G-Skill DDR2-1066 rated at (6-6-6-15) w/ 2.0V 8GB (4x2GB)

Currently Prime95 4 threads "Blend" 24 hour stable @ 425 FSB with 8.5 divider, = 3612Mhz

RAM is at 2.4 divider, 333Mhz strap. so it's running at 1022Mhz, with 6-6-6-15 timings @ 2T.

Idle temps: 37C
Load temps: 55C

Voltages:
Load-Line Calibration...........................[ Enabled ]
CPU Vcore [ 1.21250 ]...................[ 1.25 ]
CPU Termination [ 1.20 ]...................[ 1.20 ]
CPU PLL [ 1.50 ]...................[ 1.50 ]
CPU Reference [ 0.760 ]...................[ 0.863 ]

MCH Core [ 1.100 ]...................[ 1.30 ]
MCH Reference [ 0.760 ]...................[ 0.863 ]
MCH/DRAM Refernce [ 0.900 ]...................[ 0.960 ]
ICH I/O [ 1.500 ]...................[ 1.500 ]
ICH Core [ 1.100 ]...................[ 1.100 ]

DRAM Voltage [ 1.800 ]...................[ 2.0 ]
DRAM Termination [ 0.900 ]...................[ 0.960 ]
Channel A Ref [ 0.900 ]...................[ 0.960 ]
Channel B Ref [ 0.900 ]...................[ 0.960 ]


I tried to boot it at 440~450FSB (in 1mhz increments) and it just reboots back into BIOS resetting to default saying overclock failed (such a nice BIOS feature).

I've set the CPU vcore as high as 1.350 with no effect. I also tried changing the RAM to 2.0, so running at slower speed, same problem. What's holding me back here?
 
another trick.. take 2 of the sticks of ram out and see if it works.. known intel and amd issue with overclocking and having all 4 memory banks full.. might help.. might not.. though it sounds like your doing everything right.. ill let the intel guys help you fine tune it.. if the memory idea doesnt pan out..
 
Did you try taking the CPU Termination voltage to 1.30? This is VTT. I had to raise it on mine. Also try taking your MCH up to 1.4 just to see if it changes anything. I think I have mine at 1.32V
 
Have you tried different dividers, including a 1:1? My UD3R is really picky about dividers, even at a mild overclock.
 
Tried different dividers, no help.

Upping MCH allowed me to boot, but soon got BSODs. I'm thinking its the 8GB of RAM, but I need the RAM since I run multiple Virtual Machines for testing. I guess I can live with 3.61 ghz 24/7 stable @ 1.25 Vcore.

thanks for the advice guys... i think the NB voltage helped, but I dont want to raise it too high since there's no NB temp sensor on this board.
 
Okay I added a fan to the northbridge and intake in my 5.25" bay to blow cooler air over the CPU.

I want to try to make this work.

What settings should I use?

Here my current settings to get 3.61Ghz stable 24hour prime blend.

******Motherboard Intelligent Tweaker (M.I.T.)******

Robust Graphics Booster.........................[ Auto ]
CPU Clock Ratio.................................[ 8 ]
Fine CPU Clock Ratio............................[ + 0.5 ]
CPU Frequency 3.4GHz............................[ 425 x 8.5 ]

******Clock Chip Control******

CPU Host Clock Control..........................[ Enabled ]
CPU Host Frequency (Mhz)........................[ 425 ]
PCI Express Frequency (Mhz).....................[ 100 ]
C.I.A 2.................................................... [Disabled]

******Advanced Clock Control******

CPU Clock Drive.................................[ 800mv ]
PCI Express Clock Drive.........................[ 900mv ]
CPU Clock Skew..................................[ 0ps ]
MCH Clock Skew..................................[ 0ps ]

******DRAM Performance Control******

Performance Enhance.............................[Standard]
(G) MCH Frequency Latch.........................[333Mhz]
System Memory Multipler.........................[ 2.4B]
Memory Frequency 1066...........................[ 1020 ]
DRAM Timing Selectable..........................[ Manual ]

CAS Latency Time................................[ 6 ]
tRCD............................................[ 6 ]
tRP.............................................[ 6 ]
tRAS............................................[ 18 ]

******Advanced Timing Control******

tRRD............................................[ Auto ]
tWTR............................................[ Auto ]
tWR.............................................[ Auto ]
tRFC............................................[ 54 ]
tRTP............................................[ Auto ]
Command Rate (cmd)..............................[ 2 ]

******Channel A Timing Settings******

Static tRead Value ...............[ 8 ]
tRD Phase0 Adjustment ................[ Auto ]
tRD Phase1 Adjustment ................[ Auto ]
tRD Phase2 Adjustment ................[ Auto ]
tRD Phase3 Adjustment ................[ Auto ]

tRD 2rd (Differnt Rank) ................[ Auto ]
tWR 2wr (Differnt Rank) ................[ Auto ]
tWR 2rd (Differnt Rank) ................[ Auto ]
tRD 2wr (Same/Difft Rank) ................[ Auto ]

Dimm 1 Clock Skew Control.......................[ Auto ]
Dimm 2 Clock Skew Control.......................[ Auto ]
DDR Write Training..............................[ Auto ]

******Channel B Timing Settings******

Static tRead Value ...............[ 8 ]
tRD Phase0 Adjustment ................[ Auto ]
tRD Phase1 Adjustment ................[ Auto ]
tRD Phase2 Adjustment ................[ Auto ]
tRD Phase3 Adjustment ................[ Auto ]

tRD 2rd (Differnt Rank) ................[ Auto ]
tWR 2wr (Differnt Rank) ................[ Auto ]
tWR 2rd (Differnt Rank) ................[ Auto ]
tRD 2wr (Same/Difft Rank)................[ Auto ]

Dimm 1 Clock Skew Control.......................[ Auto ]
Dimm 2 Clock Skew Control.......................[ Auto ]
DDR Write Training..............................[ Auto ]

******Motherboard Voltage Control******

Load-Line Calibration...........................[ Enabled ]
CPU Vcore [ 1.25]..............................[ 1.25V ]
CPU Termination [ 1.20 ]...................[ 1.20V ]
CPU PLL [ 1.50 ]...................[ 1.500V ]
CPU Reference [ 0.760 ]...................[ Auto ]

MCH Core [ 1.100 ]...................[ 1.24V]
MCH Reference [ 0.760 ]...................[ 0.863V ]
MCH/DRAM Refernce [ 0.900 ]...................[0.960V ]
ICH I/O [ 1.500 ]...................[ 1.500V ]
ICH Core [ 1.100 ]...................[ 1.100V]

DRAM Voltage [ 1.800 ]...................[ 2.0V]
DRAM Termination [ 0.900 ]...................[0.960V]
Channel A Ref [ 0.900 ]...................[ 0.960V]
Channel B Ref [ 0.900 ]...................[0.960V]
 
Did you try MCH ref of .80 or .74-6? set it with VTT at 1.2v then raise VTT to a little higher.

You may need a tad more Vcore
 
Wow this is so weird, but at the suggestion from googling my exact board and chip, I decided to raise CPU PLL one tick. (Have no idea what that does btw)

This is what I set my BIOS for 444FSB X8.5 = 3774Mhz, DDR2-1066 (444x2.4 = 1066):

******Motherboard Voltage Control******

Load-Line Calibration...........................[ Enabled ]
CPU Vcore [ 1.25 ]...................[ 1.275V ]
CPU Termination [ 1.20 ]...................[ 1.3 ]
CPU PLL [ 1.50 ]...................[ 1.57 ]
CPU Reference [ 0.760 ]...................[ Auto ]

MCH Core [ 1.100 ]...................[ 1.320 ]
MCH Reference [ 0.760 ]...................[ 0.863 ]
MCH/DRAM Refernce [ 0.900 ]...................[ 0.970 ]
ICH I/O [ 1.500 ]...................[ 1.500 ]
ICH Core [ 1.100 ]...................[ 1.100 ]

DRAM Voltage [1.8]...................[2.000]
DRAM Termination [0.9]...................[0.970]
Channel A Ref [0.9]...................[0.970]
Channel B Ref [0.9]...................[0.970]


Now it boots!! I'm current running prime95 blend, 4 threads. So far it's hasn't errored (only been about 15 minutes), going to leave it overnight. Idle still 37C, load 55/51/50/52 (the 4 different cores).

I think I broke through my barrier by raising the PLL slightly.. Going to see if I can do 450 FSB next. Then I'll try dropping the voltages on Vcore and MCH Core... get it running cool. Thanks for everyone's help!! 3.4ghz was as simple as upping the fsb to 400... 3.6 took a little more work but was simple.. 3.8 was quite difficult! Hopefully I can get 3.8 stable and run that 24/7.. 1Ghz overclock! :)
 
funny thing bout that pll setting it was what i needed to change to get my e7400 stable @ 4.0 GHZ as well just one bump was all it needed.

anyone know what it does?
 
I think I left my CPU PLL at auto. I need to check that. I get wierd issues where if I try to eject the cd drive before the machien posts it will knock it back to stock clock. Seems picky on the post but passes every stability test I throw at it.
 
bigwooly said:
funny thing bout that pll setting it was what i needed to change to get my e7400 stable @ 4.0 GHZ as well just one bump was all it needed.

anyone know what it does?
Click to expand...

So prime95 currently 7 hours stable and running... going to let it keep going. I did some research on CPU PLL, this is what I found:

All clock circuits have a certain amount of jitter. Increasing the clock speed results in the jitter increasing also. By raising the PLL voltage the noise margin is increased effectively lowering the perceived jitter. It just helps to gain stability with high FSB.

Operating spec for PLL, Intel quotes 1.5 volts +/- 5% (so 1.425 to 1.575)

So I'm right near the max with 1.57 in BIOS. Does EasyTune show PLL voltage in Windows? I heard the UD3P might overvolt CPU PLL slightly.. maybe I should change it back to Auto?
 
Running Intel Burn Test, I hit 69C max on Core 0 in RealTemp.

Passed 10/10 runs @ 3774Mhz!

I think this is good enough for me... I don't want to go over 70C. The voltage is 1.275 in BIOS and 1.248 in CPU-Z.

Prime95 load @ 55C (57C Max)
IBT load @ 67C! (69C Max)
 
So was the PLL voltage needed for the FSB or the CPU clock? Sounds like FSB, since you tried lower multipliers with same FSB and still didn't work, right?
 
keldegar said:
Running Intel Burn Test, I hit 69C max on Core 0 in RealTemp.

Passed 10/10 runs @ 3774Mhz!

I think this is good enough for me... I don't want to go over 70C. The voltage is 1.275 in BIOS and 1.248 in CPU-Z.

Prime95 load @ 55C (57C Max)
IBT load @ 67C! (69C Max)
Click to expand...

Sounds like your stable. Maybe hit up an overnight run of memtest also.
 
Varmint said:
So was the PLL voltage needed for the FSB or the CPU clock? Sounds like FSB, since you tried lower multipliers with same FSB and still didn't work, right?
Click to expand...

Yeah, it was the PLL to get the FSB that high. With 6 multi still would get BSODs.
 
Okay, still having issues... at 3.77Ghz it's stable 24 hours w/ Prime95, 10/10 Intel Burn test... yet if I reboot a lot, sometimes it has trouble posting and resets to default settings.

Same issue as Copyright... anyone have any ideas about that? I've bumped it back down to 3.61Ghz because I tend to reboot a lot (dual-boot machine).
 
keldegar said:
Okay, still having issues... at 3.77Ghz it's stable 24 hours w/ Prime95, 10/10 Intel Burn test... yet if I reboot a lot, sometimes it has trouble posting and resets to default settings.

Same issue as Copyright... anyone have any ideas about that? I've bumped it back down to 3.61Ghz because I tend to reboot a lot (dual-boot machine).
Click to expand...

I notice this same problem as well. I had my Q9450 at 3.92 ghz with Prime95 stable, sometimes it has trouble posting and keep doing the 2 secs restarts. This motherboard is kind of pissing me off.
 
bail_w said:
I notice this same problem as well. I had my Q9450 at 3.92 ghz with Prime95 stable, sometimes it has trouble posting and keep doing the 2 secs restarts. This motherboard is kind of pissing me off.
Click to expand...

I usually leave my machine on, its not a huge issue, but I tend to reboot at least once a day, and just started noticing it resetting to defaults (not every time).

It doesn't do this when I have it on my 3.61Ghz stable settings. (Notably MCH voltage, CPU vCore, and PLL voltages are all stock, compared to 3.77Ghz stable I had to raise each of those).
 
I have an evga 780i FTW board, and bumped my PLL up to 1.7 (it goes from 1.6 to 1.7, nothing in between). Seems to help me be stable at 3.8, with the same chip. You guys think that's too high?
 
gweminence said:
I have an evga 780i FTW board, and bumped my PLL up to 1.7 (it goes from 1.6 to 1.7, nothing in between). Seems to help me be stable at 3.8, with the same chip. You guys think that's too high?
Click to expand...

i would stay under 1.6v for a q9550.
 
Agreed, my PLL was 1.57 in BIOS to get 3.77Ghz stable (450 FSB). It wouldn't boot without upping PLL past 425 FSB.
 
I am at 490FSB with 1.57v PLL, so i doubt you need to go any higher than 1.6.
 
I had no trouble taking mine to 445mhz with 1.5v PLL. I still get the odd issue when I restart sometimes it loses my clock. Then again the machine is on 24/7 so I rarely reboot. Only takes a split second to load my OC profile so not a huge issue. I did have one lock up in windows the other day but it was directly related to my Bit Defender antivirus. I had to hold the power button to shut it down. When it rebooted I lost my OC. I only seem to lose my OC with wierd issues like that. I rebooted today atleast 5 times with no issues. *shrug* My MCH is 1.30V and my CPU Termination (VTT) is 1.30V. I have the ram @ 1068mhz with 8 gigs of OCZ Reaper @ 2.2V.
 
You must log in or register to reply here.
Back
Top