Voltage seems too high for 3.0Ghz Q6600 + EP45-UD3R...

[SCND30]

Hello all,
I seem to be having an issue, new to overclocking but trying to absorb as much as possible and successfully OC my system. I'm trying to hit 3.2ghz with my Q6600 and I'm hoping my problem is caused by my inexperience. I'm currently at 3.0ghz (67 degrees at load) and from what I've read others online say I should have lower voltage than what's necessary to get 3.0ghz. I tried 400x8 but to no avail, I could get booted into windows but failed small FFt with prime95 within seconds. Here are the settings I used to achieve 3.0ghz:

MB Intelligent Tweaker(M.I.T.)
Robust Graphics Booster -------------- : [Auto]
CPU Clock Ratio ---------------------- : [9]
Fine CPU Clock Ratio ----------------- : [0.0] <-- ignore if not available
CPU Frequency ---------------------: 3.00 <-- this is set by your bios

Clock Chip Control
Standard Clock Control
CPU Host Clock Control --------------- : [Enabled]
CPU Host Frequency (Mhz) --------------: [333]
PCI Express Frequency (Mhz) ---------- : [100] <--- Set to 100, never leave this on Auto

C.I.A.2 ------------------------------ : [Disabled]

Advanced Clock Control ------- [Press Enter]
CPU Clock Drive ---------------------- : [800]
PCI Express Clock Drive -------------- : [900]
CPU Clock Skew (ps) ------------------ : [0]
MCH Clock Skew (ps) ------------------ : [0]

DRAM Performance Control
Performance Enhance ------------------- : [Standard] <--- set to STANDARD when overclocking
Extreme Memory Profile --- (X.M.P.) --- : [Disabled] <--- Leave this Disabled
(G)MCH Frequency Latch ---------------- : [Auto]
System Memory Multiplier -------------- : [2.40B] <-- this is the setting that you are using
Memory Frequency (Mhz) ---------------- : [800 800]
DRAM Timing Selectable ---------------- : [Auto] <-- set to MANUAL for overclocking

Standard Timing Control
CAS Latency Time -------- 5 ---------- : [Auto]
tRCD ------------------- 5 ---------- : [Auto]
tRP --------------------- 5 ---------- : [Auto]
tRAS ------------------- 15 ---------- : [Auto]

Advanced Timing Control
tRRD ------------------- 3 ---------- : [Auto]
tWTR ------------------- 3 ---------- : [Auto]
tWR --------------------- 4 ---------- : [Auto]
tRFC ------------------- 52 ---------- : [Auto]
tRTP ------------------- 3 ---------- : [Auto]
Command Rate (CMD) -------------------- : [2] <--- Auto or 0 is OK

Driving Strength Profiles
Driving Strength --------- : Press Enter <--- press ENTER to modify memory sub-timings when overclocking

Channel A
Static tRead Value --------------------- : [Auto]
tRD Phase0 Adjustment ------------------ : [Auto] <-- leave this and below set to AUTO
tRD Phase1 Adjustment ------------------ : [Auto]
tRD Phase2 Adjustment ------------------ : [Auto]
tRD Phase3 Adjustment ------------------ : [Auto]
Trd2rd(Different Rank) ----------------- : [Auto]
Twr2wr(Different Rank) ----------------- : [Auto]
Twr2rd(Different Rank) ----------------- : [Auto]
Trd2wr(Same/Diff Rank) ----------------- : [Auto]
Dimm1 Clock Skew Control --------------- : [Auto] ps
Dimm2 Clock Skew Control --------------- : [Auto] ps

Channel B
Static tRead Value --------------------- : [Auto]
tRD Phase0 Adjustment ------------------ : [Auto] <-- leave this and below set to AUTO
tRD Phase1 Adjustment ------------------ : [Auto]
tRD Phase2 Adjustment ------------------ : [Auto]
tRD Phase3 Adjustment ------------------ : [Auto]
Trd2rd(Different Rank) ----------------- : [Auto]
Twr2wr(Different Rank) ----------------- : [Auto]
Twr2rd(Different Rank) ----------------- : [Auto]
Trd2wr(Same/Diff Rank) ----------------- : [Auto]
Dimm1 Clock Skew Control --------------- : [Auto] ps
Dimm2 Clock Skew Control --------------- : [Auto] ps

Motherboard Voltage Control
CPU
Voltage Types------------ Normal -------- Current -----
Load-Line Calibration ----------------- : [Enabled]
CPU Vcore --------------- 1.3500V ---- : [1.40000V]

CPU Termination --------- 1.200V* ----- : [1.240V]
CPU PLL ----------------- 1.500V* ----- : [1.500V]

CPU Reference ----------- 0.760V* ----- : [Auto]
CPU Reference2 ---------- ?? ----- : [Auto]

MCH/ICH
MCH Core ---------------- 1.100V ------- : [1.100V]
MCH Reference ----------- 0.760V ------- : [Auto]
MCH/DRAM Ref ------------ 0.900V ------ : [Auto]
ICH I/O ----------------- 1.1500V ------- : [1.500V]
ICH Core ---------------- 1.100V ------- : [1.100V]
DRAM
DRAM Voltage ------------ 1.800V ------- : [1.900V]
DRAM Termination -------- 0.900V ------- : [Auto]
Channel A Reference ----- 0.900V ------- : [Auto]
Channel B Reference ----- 0.900V ------- : [Auto]

Advanced Settings
Limit CPUID Max. to 3 ------------------ : [Disabled]
No-Execute Memory Protect -------------- : [Enabled]
CPU Enhanced Halt (C1E) ---------------- : [Disabled]
C2/C2E State Support ------------------- : [Disabled]
x C4/C4E State Support ----------------- : [Disabled]
CPU Thermal Monitor 2 (TM2) ------------ : [Enabled]
CPU EIST Function --------------------- : [Disabled]
Virtualization Technology -------------- : [Disabled]

Integrated Peripherals
Legacy USB Storage Detect ------------- : [Disabled]

[Chike]

Re: Voltage seems too high for 3.0Ghz Q6600 + EP45-UD3R...

67c are core temperatures or CPU? Under what kind of load?
What is vcore in CPU-Z, under load and idle?
Try 1.55-1.61v CPU PLL. 1.16-1.24v MCH Core. Up to 1.32v CPU Termination.
SMM 2.00D or 2.00BB whichever is more stable and yield better performance.
Also set tRD(Static tRead Value in BIOS, Performance Level in MemSet) to 9 for both channels, once you are stable you can try to lower it.
Test it with 400x7 first to eliminate bus/memory problems, then 400x8.

The SpinQ is a bit weak, temps might be in the way.

[Qu4k3r]

[B][U]GIGABYTE GA X48T DQ6 Clock Settings[/U][/B] 
 
[B][U]MB Intelligent Tweaker (MIT)[/U][/B] 
 
Robust Graphics Booster...........:auto 
CPU Clock Ratio...................:8 
CPU Frequency.....................:3200 
CPU Host Clock Control............:enable 
CPU Host Frequency(Mhz)...........:400 
PCI Express Frequency (Mhz).......:100 
C.I.A.2...........................:disable 
Performance Enhance...............:standard 
System Memory Multiplier (SPD)....:4.00D 
Memory Frequency(Mhz).............:1600 
DRAM Timing Selectable ...........:manual 
 
[B]Standard Timing Control[/B] 
CAS Latency Time (tCL)............:8 
DRAM RAS# to CAS#  (tRCD).........:8 
DRAM RAS# Precharge (tRP).........:8 
Precharge Delay (tRAS)............:24 
 
[B]Advanced Timing Control[/B] 
ACT to ACT Delay(tRRD)............:auto 
Rank Write to READ Delay (tWTR)...:auto 
Write To Precharge Delay (tWR)....:auto 
Refresh to ACT Delay (tRFC).......: auto 
Read To Precharge Delay (tRTP)....: auto 
Static tREAD Value (tRD)..........:7 
Static tREAD Phase Adj............: auto 
Command Rate (CMD)................:1T 
 
[B]Clock Driving and Skew Control[/B] 
CPU/PCIEX Clock Driving Control...:normal 
CPU Clock Skew Control............: normal 
(G)MCH Clock Skew Control.........: normal 
 
[B]System Voltage Control[/B] () = Base Voltages 
DDR3 OverVoltage Control...(1.50v):+0.35 
PCI-E OverVoltage Control.,(1.50v):normal 
FSB OverVoltage... 65nm = 1.25v..+:normal 
(G)MCH OverVoltage Control,(1.25v):normal 
CPU GTLREF1 Voltage*.......(67%)..:normal 
CPU GTLREF2 Voltage*.......(61%)..:normal 
Loadline Calibration..............:enable 
CPU Voltage Control...............:1.4325v 
Normal CPU Vcore..................:1.3250v 
Note: * Hidden Features Press [CTRL] + [F1] from main Menu to Access + FSB 45nm= 1.15v  

[profJim]

Re: Voltage seems too high for 3.0Ghz Q6600 + EP45-UD3R...

Please post a screen-shot that includes the information in step #4 from http://forums.tweaktown.com/gigabyte...tml#post295784, post #5.

Your bios template looks like it was copied from a DDR3 system!!!
Command rate = 1T?
Memory frequency = 1600 (4.00D x 400)?
Your memory is rated at 800MHz, 5-5-5-15 with 1.8 - 1.9 volts.

What is the normal room temperature?
What computer case do you have? How many case fans?

[Chike]

Re: Voltage seems too high for 3.0Ghz Q6600 + EP45-UD3R...

Your bios template looks like it was copied from a DDR3 system!!!

X48T is indeed a DDR3 motherboard.

[profJim]

Re: Voltage seems too high for 3.0Ghz Q6600 + EP45-UD3R...

Oops, I overlooked the fact that Qu4k3r isn't the original poster that was using an EP45-UD3R.

[SCND30]

Re: Voltage seems too high for 3.0Ghz Q6600 + EP45-UD3R...

67c are core temperatures or CPU? Under what kind of load?
What is vcore in CPU-Z, under load and idle?
Try 1.55-1.61v CPU PLL. 1.16-1.24v MCH Core. Up to 1.32v CPU Termination.
SMM 2.00D or 2.00BB whichever is more stable and yield better performance.
Also set tRD(Static tRead Value in BIOS, Performance Level in MemSet) to 9 for both channels, once you are stable you can try to lower it.
Test it with 400x7 first to eliminate bus/memory problems, then 400x8.

The SpinQ is a bit weak, temps might be in the way.

This is what it reads when running Prime95 small FFt test... and Vcore in CPU-Z is 1.36. Going to use the settings you suggested after I get any other helpful settings so I can do it all at once.



Did that answer your questions about load and idle? under idle the temps are about 40 degrees, sorry if I'm not answering correctly, still trying to wrap my head around everything relative to OCing in the past 48 hours or so...I'll be more than glad to try and provide the right info if I'm getting it wrong.

Please post a screen-shot that includes the information in step #4 from http://forums.tweaktown.com/gigabyte...tml#post295784, post #5.

Your bios template looks like it was copied from a DDR3 system!!!
Command rate = 1T?
Memory frequency = 1600 (4.00D x 400)?
Your memory is rated at 800MHz, 5-5-5-15 with 1.8 - 1.9 volts.

What is the normal room temperature?
What computer case do you have? How many case fans?

Here are the screens Jim, and normal room temperature is about 70 degrees, the case I'm using is this one and I have the stock fans in ... Side 25 cm - ( 1 installed) Front 12 cm - ( 1 installed) Rear 12 cm - ( 1 installed) Rear 9 cm - ( 1 installed) Top 9 cm - ( 1 installed)

[Chike]

Re: Voltage seems too high for 3.0Ghz Q6600 + EP45-UD3R...

Try the PLL first.

[SCND30]

Re: Voltage seems too high for 3.0Ghz Q6600 + EP45-UD3R...

So I tried your settings Chike, kept bumping my vcore up everytime prime95 failed the small fft test, each time it took longer before a worker had a hardware failure but I had to stop around 1.46 vcore in BIOS because idle my cpu temp was around 40, but when I start the torture test is instantaneously jumps to around 63 degrees and climbs to around 70-71 right when a worker is about to fail. The last test around 4-5 mins before failure. I'm assuming that my temps are what's getting in the way and if I want to OC to 3.2 I'm going to have to invest in water cooling? Or would it be a better idea now that the CPU is a few years old to just upgrade my mobo + cpu + water cooling (less desirable option due to the cost).

[Qu4k3r]

Re: Voltage seems too high for 3.0Ghz Q6600 + EP45-UD3R...

For 400MHz FSB, I think (G)MCH Frequency Latch should be 400MHz, or am I wrong? First time I did OC, I had an E7200 and made a table to fill in with data only palying with 3 volatges:

• Vcore
• MCH/NB
• FSB/VTT/CpuTermination.

Lowering ram frec first with relaxed latencies, to aviod errors produced by ram issues, in that way is 100% sure that any problem is only caused by cpu. I know it'll take so long but is safer to reach any oc goal. Use prime95 with blend FFTs. I hope it will be helpfull.

[profJim]

Re: Voltage seems too high for 3.0Ghz Q6600 + EP45-UD3R...

Several general guidelines:

• Running with cpu host frequency = 400 sometimes causes problems, but if you run at 398 - 399 or 401 - 402 you might not have stability problems.
• At 400MHz & higher, using the B-strap will often let you use tRD (Static tRead Value) one step tighter (lower) than using the D-strap, up to about 460MHz. Some find it easier to get a stable setup when using the D-strap. It's not recommended to use the memory D-strap below 400MHz.

Increasing VTT (Cpu Termination) can help with stabilizing a system. Ideally this setting should be lower than Cpu Vcore by at least 0.05 - 0.10 volts. Some overclocked systems are stable with VTT near the stock setting, others require VTT to be close to Cpu Vcore. Your goal is to set VTT to the lowest stable value.

1. I suggest lowering Cpu Vcore and increasing VTT at the same time to see if this setup is stable.
2. With your 2x2GB PC2-6400 memory, try running with MCH Core = 1.16 -1.20 volts. I've seen several Gkill 8500 memory systems that are only stable with MCH Core = 1.05 (sometimes lower).

Start with step #1. Don't do both steps at the same time, as there are too many variables being changed at the same time.

I'm not a fan of enabling LLC (LoadLine Calibration) but I've seen a fair number of setups in the forum where the only way to get a stable system was to enable LLC when performing high overclocks. With LLC enabled, the vdroop is less and you can then lower CPU Vcore several steps. You can compute the vdroop value by checking the Vcore value in the bios PC Health Status screen and the Core Voltage value displayed in the Cpu-Z screen. Vdrop is the difference between the Cpu Vcore setting in the M.I.T. bios and the Vcore value in the bios PC Health Status screen.

Which SpinQ cpu cooler are you using, the older "horizontal" setup or the newer "tower" VT version?

[SCND30]

Re: Voltage seems too high for 3.0Ghz Q6600 + EP45-UD3R...

Several general guidelines:

• Running with cpu host frequency = 400 sometimes causes problems, but if you run at 398 - 399 or 401 - 402 you might not have stability problems.
• At 400MHz & higher, using the B-strap will often let you use tRD (Static tRead Value) one step tighter (lower) than using the D-strap, up to about 460MHz. Some find it easier to get a stable setup when using the D-strap. It's not recommended to use the memory D-strap below 400MHz.

Increasing VTT (Cpu Termination) can help with stabilizing a system. Ideally this setting should be lower than Cpu Vcore by at least 0.05 - 0.10 volts. Some overclocked systems are stable with VTT near the stock setting, others require VTT to be close to Cpu Vcore. Your goal is to set VTT to the lowest stable value.

1. I suggest lowering Cpu Vcore and increasing VTT at the same time to see if this setup is stable.
2. With your 2x2GB PC2-6400 memory, try running with MCH Core = 1.16 -1.20 volts. I've seen several Gkill 8500 memory systems that are only stable with MCH Core = 1.05 (sometimes lower).

Start with step #1. Don't do both steps at the same time, as there are too many variables being changed at the same time.

I'm not a fan of enabling LLC (LoadLine Calibration) but I've seen a fair number of setups in the forum where the only way to get a stable system was to enable LLC when performing high overclocks. With LLC enabled, the vdroop is less and you can then lower CPU Vcore several steps. You can compute the vdroop value by checking the Vcore value in the bios PC Health Status screen and the Core Voltage value displayed in the Cpu-Z screen. Vdrop is the difference between the Cpu Vcore setting in the M.I.T. bios and the Vcore value in the bios PC Health Status screen.

Which SpinQ cpu cooler are you using, the older "horizontal" setup or the newer "tower" VT version?

Thanks again Jim, the spinq I'm using is this one. I've seen mixed reviews about the cooler but my guess is it's not the greatest for what I'm doing. I changed the VTT and Vcore like you suggested. While trying to boot windows using 401x8 I had to up my vcore last night to around 1.46 in BIOS. So now I'm going to try backing off the vcore a bit(around 1.45 now) and upping my VTT to around 1.3 and see if that helps stability. I'm able to boot windows and I'm going to see if I can run Prime95 small ffts without any issues. It usually fails relatively quick or my temps hit 71 full load within 15 mins or so.

and Quaker thanks for your info I'm going to mess around with the settings you supplied as well and see if they match the table you provided. Unfortunately I have to go out of town until wednesday night so I can't mess around with it tonight or tomorrow but I will be checking this thread via my phone and any more suggestions will be most appreciated.

EDIT: Couldn't boot into windows without LLC enabled with these settings: vcore - 1.45, PLL - 1.57, VTT - 1.26. So I enabled LLC and upped the vcore just slightly to 1.476 until I was able to boot windows and try running Prime95. Had a hardware failure in worker #2 within seconds of starting the test while temps were approximately 65 degrees, cant imagine how high they would be if the test were to continue :/

[profJim]

Re: Voltage seems too high for 3.0Ghz Q6600 + EP45-UD3R...

Your SpinQ fan needs to be replaced. I've recently read several reviews where your model or the newer SpinQ VT were mediocre at best, with cpu load temperatures that are typically 8 - 10 degrees higher than the better mid priced cpu coolers.

Check out http://forums.tweaktown.com/gigabyte...tml#post389150 post #10 where you can buy a decent Xigmatek cpu cooler for less than $25 after rebate. The Newegg rebate is good through March 31st. I like the idea of being able to easily add a second fan (if needed) for additional cooling. The Dark Knight model offers similar performance and has several very positive Q6600 user reviews. The only possible downside with either of these units is that they come with an 89 cfm fan that is loud when the fan speed is above 1600 rpm. Maximum fan speed is 2200 rpm.

**edit**
I just found the attached information from XbitLabs:

[SCND30]

Re: Voltage seems too high for 3.0Ghz Q6600 + EP45-UD3R...

So is it safe to say that with this setup and the spinq that achieving 3.2ghz is not possible because of the temps? Just clarifying for sure so I can explore the other CPU cooking alternatives like you suggested.

[Chike]

Re: Voltage seems too high for 3.0Ghz Q6600 + EP45-UD3R...

So is it safe to say that with this setup and the spinq that achieving 3.2ghz is not possible because of the temps? Just clarifying for sure so I can explore the other CPU cooking alternatives like you suggested.

Probably so.
Check out Gerardz settng for 3.2GHz. He seems to have a great chip, but although his default vcore is lower than yours and he made it to 3.2GHz with "normal" settings, he had to upgrade his CPU cooler (from stock) and doing quite well with Xigmatek Dark Knight.

But please post full BIOS settings for 3.2GHz, and lets see if you can run with lower vcore.