Gigabyte GA X38 DQ6 Blog

[lokeey]

i'm trying to get a stable 3.2GHz OC, but i still continue to get BSOD when i kick off small fft test using prime95.

I've tried 400x8 using 2.66D (1066) and 355x9 using 3.20A (1065).

Here are my current settings. I've changed the strap from 2.66D to 3.20A (1200) and I'm still getting a BSOD when i kick off prime95 small fft test. any suggestions on what i should change?

MB Intelligent Tweaker (MIT) BIOS F7
Robust Graphics Booster: [Turbo]
CPU Clock Ratio: [8X]
CPU Host Clock Control: [Enabled]
CPU Host Frequency(Mhz) : [400]
PCI Express Frequency (Mhz) : [100]
C.I.A.2: [Disabled]
Performance Enhance : [Standard]
System Memory Multiplier (SPD) [3.20A]
Memory Frequency(Mhz): 1200
DRAM Timing Selectable : [Manual]
********Standard Timing Control ***********
CAS Latency Time: [5]
DRAM RAS# to CAS# :[5]
DRAM RAS# Precharge : [5]
Precharge Delay(tRAS) : [11]
******** Advanced Timing Control ***************
ACT to ACT Delay(tRRD) : [3]
Rank Write to READ Delay : [3]
Write To Precharge Delay : [6]
Refresh to ACT Delay : [30]
Read To Precharge Delay : [3]
Static tRead Value : [6]
Static tRead Phase Adjust : 5 [Auto]
Command Rate (CMD) : [2T]
******** Clock Driving and Skew Control **********
CPU/PCIEX Clock Driving Control : [1000mV]
CPU Clock Skew Control : [Normal]
(G)MCH Clock Skew Control : [Normal]
System Voltage Control: [Manual]
DDR2 OverVoltage Control : [+0.40v]
PCI-E OverVoltage Control : [+0.35v]
FSB OverVoltage Control : [+0.25v]
(G)MCH OverVoltage Control : [+0.175v]
MCH Reference Voltage Control : [Normal]
DDR Reference Voltage Control : [Normal]
DDR Termination Voltage Control : [Normal]
CPU Voltage Control : [1.35000v]
Normal CPU Vcore: 1.32500v

[supershanks]

Check your running with the latest version Death's Overclocker's Tool Kit. is now a great place to visit

We've talked about TRD being to tight so i'd change this

Static tRead Value : [6]

---> 7

If it's failing on small ffts it tends to imply cpu issues.
Try increasing cpu voltage a notch & rerunning.
FSBOvervoltage control (vtt) can also be raised a notch. (if possible can't remember if 0.25v is max


luck

[lokeey]

We've talked about TRD being to tight so i'd change this
---> 7

for some reason i kept thinking 6, i don't know why.

If it's failing on small ffts it tends to imply cpu issues.
Try increasing cpu voltage a notch & rerunning.

1.375? or should i go a tad bit higher than that?

FSBOvervoltage control (vtt) can also be raised a notch. (if possible can't remember if 0.25v is max

i think 0.35v is the max, but i'll raise it some as well.

thanks! i'll make these changes and see how it goes.

[supershanks]

1.375? or should i go a tad bit higher than that?

just up it a notch , if it fails do it again.

luck

[lokeey]

okay, i went a lil slower this time. instead of going to 355x9 or 400x8, i eased it up to 345x9 (3.1GHz). went up on the vcore a tad and ran prime95 for 30 mins. no issues. i know it was only 30 mins, but i just wanted to see if it would bsod when starting blend or small fft test and it didn't. so i went up again to 355x9 upped the vcore again, just a notch kicked off prime95 and BSOD! ugh!!!

i also noticed that it got pretty hot. didn't reach the 70C mark, but was very very close to doing so. i may just leave it at 333x9 for now until i get an alternate cooling solution like water.

[Marcus]

Originally Posted by lokeey

okay, i went a lil slower this time. instead of going to 355x9 or 400x8, i eased it up to 345x9 (3.1GHz). went up on the vcore a tad and ran prime95 for 30 mins. no issues. i know it was only 30 mins, but i just wanted to see if it would bsod when starting blend or small fft test and it didn't. so i went up again to 355x9 upped the vcore again, just a notch kicked off prime95 and BSOD! ugh!!!

i also noticed that it got pretty hot. didn't reach the 70C mark, but was very very close to doing so. i may just leave it at 333x9 for now until i get an alternate cooling solution like water.

When I was o'cing with the stock heatsink I was hitting a wall between 2.9-3.1Ghz. Now, since I'm water cooling the proc :-) it's around 3.5Ghz....I cannot get 3.6 working even with 1.525v !!

This week-end I want to try a new thermal paste arctic silver 5 instead of the ceramic one to see if there is a big diff in temps.... I might also try to bow the cpu block :-)
It's crazy how much time we can spend trying to tune a pc. If I compute all my hours it would have been a lot cheaper just buy the new QX9650 :-) lol

Question about crucial ballistix 8500...what is the max voltage I can use...I know it says 2.20v but...with a dominator memory fan is it dangerous to try pushing it to 2.35v?


cheers,
Marcus

[supershanks]

lokeey,
In 99% of casrs you can't go from 345x9 to 355x9 without raising vcore. Your 30min prime success shows that you had the right, or very close , voltage for 345. If you want to do 355 you have to raise the voltage again it's noy always a linear relationship 1 notch of vcore = +10 fsb. your best trying as you did if it bsods then lower the FSB by 1 until it doesn't. You might find your stable on say 351 , the point is you'll have found the exact point of stabilty, rather than just taking an arbitary increase of 10.

Marcus,
you can use 2.35v ion short bursts for benches if you need to but i wouldn't go much above 2.2v in long term usage.

There is reason for this shown here:-
MUltimeter Measurments
Although for the abit it shows like most motherboards, when memory is on load a 0.04v overvolt cuts in.
Now this is when your memory is all ready running at it's hottest. I found that i never needed over 2.175v to support stable memory speeds over 1175mhz.
That means that my ram on load just about hits 2.2v. I've always actively cooled ram, but am on my 3ed set, in about 18months. I've not tested the gigabyte but from reading around have heard it does overvolt ram.

luck

[Cosmic]

Hi All,

After A full 24 hrs of fiddling I have finally got a moderately respectable O/C so I thought I would post my settings and results. Also A few questions at the bottom.

I was trying to get a O/C of about 3.6GHz without having the temps go over 65 degrees on full load, with an ambient room temperature of about 28C. I didn’t quite get there but pretty close.

A Couple Of Notes; I’m Running Air Cooling, I haven’t made any significant effort to tighten my memory timings yet, and pointing a pedestal fan at the side of my open case dropped my temps by about 5 degrees (so obviously I still have some work to do on case airflow hehe)

BIOS Revision: F7

Advanced Bios Features
----------------------

CPU Enhanced Halt (C1E) ................... [Disabled]
CPU Thermal Monitor 2(TM2) ................ [Enabled]
CPU EIST Function ......................... [Disabled]


PC Health Status
---------------

CPU Warning Temperature ................... [70C/158F]
CPU Smart Fan Control ..................... [Disabled]


MB Intelligent Tweaker (M.I.T.)
-------------------------------

Robust Graphic Booster ..................... [Turbo]
CPU Clock Ratio ............................ [ 9 x]
CPU Frequency .............................. 3.50GZh (389x9)
CPU Host Clock Control ..................... [Enabled]
x CPU Host Frequency(MHz) .................. 389
PCI Express Frequency(MHz) ................. [Auto]
C.I.A.2 .................................... [Disabled]
Performance Enhance ........................ [Standard]
System Memory Multiplier ................... [2.66C]
Memory Frequency(MHz) ............. 800 .... [1037]
DRAM Timing Selectable ........... (SPD) ... [Manual]

********** Standard Timing Control **********

CAS Latency Time .................    5  ... [5]
DRAM RAS# to CAS# Delay ..........    5  ... [5]
DRAM RAS# Precharge ..............    5  ... [5]
Precharge Delay(tRAS) ............   18  ... [16]

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

ACT to ACT Delay(tRRD) ...........    3  ... [4]
Rank Write To READ Delay .........    3  ... [4]
Write To Precharge Delay .........    6  ... [8]
Refresh To ACT Delay .............   42  ... [56]
Read To Prechard Delay ...........    3  ... [4]
Static tRead Value ...............    7  ... [8]
Static tRead Phase Adjust ........    1  ... [Auto]
Command Rate(CMD) ................    2  ... [2T]

******* Clock Driving & Skew Control ********

CPU/PCIEX Clock Driving Control ............ [1000mV]
CPU Clock Skew Control ..................... [Normal]
(G)MCH Clock Skew Control .................. [Normal] 

********* System Voltage Optimized **********

System Voltage Control ..................... [Manual]
DDR2 OverVoltage Control ................... [+0.35V]
PCI-E OverVoltage Control .................. [+0.35V]
FSB OverVoltage Control .................... [+0.15V]
(G)MCH OverVoltage Control ................. [+0.050]
MCH Reference Voltage Control .............. [Normal]
DDR Reference Voltage Control .............. [Normal]
DDR Termination Voltage Control ............ [Normal]
CPU GTLREF1 Voltage Ratio .................. [Normal]
CPU GTLREF2 Voltage Ratio .................. [Normal]
Loadline Calibration ....................... [Enabled]
CPU Voltage Control ........................ [1.41250V]
Normal CPU Vcore ........................... 1.28750V

PRIME95 2Hr Load Test - 

  ; Intel Core2 Quad Q6600 2.40Ghz O/C to 3.50GHz 
  ; 4 x Corsair CM2X1024-6400C4 (800MHz) O/C to 1037 MHz (5,5,5,16 2.15v)

|======|==================================|==================================|==================================|
| Room |         Idle Temperature         |   Full Load Average Temperature  |     Full Load Max Temperature    |
|------|----------------------------------+----------------------------------+----------------------------------|
|||||||| CPU  |  C0  |  C1  |  C2  |  C3  | CPU  |  C0  |  C1  |  C2  |  C3  | CPU  |  C0  |  C1  |  C2  |  C3  |
|======|======+======+======+======+======|======+======+======+======+======|======+======+======+======+======|
| 29 C | 34   | 39 C | 36 C | 40 C | 36 C | 57 C | 63 C | 59 C | 62 C | 59 C | 59 C | 65 C | 61 C | 65 C | 61 C |
|------|------+------+------+------+------|------+------+------+------+------|------+------+------+------+------|
| 24 C | 27 C | 32 C | 27 C | 33 C | 27 C | 50 C | 54 C | 51 C | 55 C | 50 C | 51 C | 57 C | 52 C | 56 C | 51 C |
|------|----------------------------------|----------------------------------|----------------------------------|

* Load Temps Via Prime95 Torture Test "In-place large FFTs" run 2hrs 17min

Q1: I seemed to hit a wall at 3.50GHz. Even if I increased the vcore to 1.42560V I still couldn’t get any further OC by upping the "CPU Host Freq" (Also 1.42560 was taking me over 65C which was my hard cut off). Any suggestions on how I could get anything more ?

(I also tried increasing the (G)MCH further but that didn’t make any difference either)


Q2: For some reason other than 1:1 none of the x.xxA multipliers would work stably for me, even at very modest timings and O/C's ?


Q3: Any suggestions on Tightening my memory timings ?



Thanks guys... hope the info is useful


Cosmic.

[supershanks]

Hi cosmic ,
good going m8

1) Try upping

FSB OverVoltage Control .................... [+0.15V]

This is the equivalent of Vtt that yuou see on the Abit & Dfi Boards. It's scalable with FSB, so if you get a wall as you describe, nudging this up a notch can help.
I've found in the past that increasing this slightly may enable a slightly lower vcore. So it's worth experimenting using the 2 together.

2) I've only really used 2.5 on the A strap. but I did find that 2.66 despite being a higher strap, is also far more stable. Strange but true.

3)

********** Standard Timing Control **********

CAS Latency Time .................    5  ... [5]
DRAM RAS# to CAS# Delay ..........    5  ... [5]
DRAM RAS# Precharge ..............    5  ... [5]
Precharge Delay(tRAS) ............   18  ... [16] try 10~12

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

ACT to ACT Delay(tRRD) ...........    3  ... [4]
Rank Write To READ Delay .........    3  ... [4]
Write To Precharge Delay .........    6  ... [8]
Refresh To ACT Delay .............   42  ... [56]  Try 30~35
Read To Prechard Delay ...........    3  ... [4]
Static tRead Value ...............    7  ... [8] You might get away with 7 ,but 8 is safe
Static tRead Phase Adjust ........    1  ... [Auto]
Command Rate(CMD) ................    2  ... [2T]

I tend run large ffts if i want to check memory, small ffts if I'm concerned with CPU stability , but blend for an overall test.

luck

[Cosmic]

Originally Posted by supershanks

Hi cosmic...

1) Try upping ... FSB OverVoltage Control ...

This is the equivalent of Vtt that yuou see on the Abit & Dfi Boards. It's scalable with FSB ...

luck

Thanks alot for that SuperShanks ... thats a real little gem ... I wasnt able to get my O/C up any further but I was able to lower my vcore by 2 notches which has made a noticable difference to my temps ... !THANKS!

Also got the ram settings a little tighter ... will test it all properly tomorow when temps get back towards 28 - 30C in my office (any hotter than that and I put my AirCon on ).

Cosmic