Trying a 4Ghz stable OC on my E8400

[Zenoth]

I've decided to go for it again, to reach it in the first place (I tried before, of course, but it miserably failed, even just at 3.8Ghz failed too), since I've just received my brand new PSU last morning, namely a Corsair TX750W, and it's running smooth and very silently (140mm fan for the win). Now the concerns for a potentially faulty and/or insufficient (mainly on the +12V rail and the low amperage on it) Power Supply is out of the equation (my previous one could have caused issues, I've often suspected, but it's now something of the past).

And I'm here to seek help, of course.

First of all, my system is in my signature. And more precisely, here are the current BIOS settings, for my current and stable over-clock:

MB Intelligent Tweaker (MIT)

Robust Graphics Booster : Auto
CPU Clock Ratio : 9x
CPU Host Clock Control: Enabled
CPU Host Frequency(Mhz) : 416
PCI Express Frequency (Mhz) : 100
C.I.A.2: Disabled
Performance Enhance : Standard
System Memory Multiplier (SPD) : 2.40B (5 for 6)
Memory Frequency(Mhz) : 998
DRAM Timing Selectable : Manual

Standard Timing Control

CAS Latency Time: : 5
DRAM RAS# to CAS# : 5
DRAM RAS# Precharge : 5
Precharge Delay(tRAS) : 18

Advanced Timing Control

ACT to ACT Delay(tRRD) : 4
Rank Write to READ Delay : 4
Write To Precharge Delay : 7
Refresh to ACT Delay : 56
Read To Precharge Delay : 4
tRD : 7
tRD Phase Adjustment : Auto
Command to #CS Timing: : 2

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 : +0.35v (2.18v)
PCI-E OverVoltage Control : Normal
FSB OverVoltage Control : +0.20v
(G)MCH OverVoltage Control : +0.200v
MCH Reference Voltage Control : Normal
DDR Reference Voltage Control : Normal
DDR Termination Voltage Control : Normal
CPU GTLREF1 Voltage : Normal
CPU GTLREF2 Voltage : Normal
CPU Voltage Control : 1.41250v (1.392v in CPU-z)
Normal CPU Vcore : 1.22500v

These settings, so far, never failed me in Prime95, Orthos and CCCT, along with both Memtest86 and HCI Memtest, all of which I ran plenty of times, sometimes for more than 20 hours (such as OCCT), without getting a single issue. With that said, and now with my new PSU, I tried a little "quicky OC" just to see if I saw any differences over my previous PSU under particular settings and I did notice some.

Indeed, with my previous PSU when I tried to set it at 445x9 it wouldn't even boot at all, just a booting and restarting loop, and I had to reset the CMOS. But with my new PSU, the 445x9 setting sent me up to the Windows loading screen, and it crashed back to the BIOS, but at least it wouldn't constantly loop and restart itself. The thing is the vCore is already high for this CPU, and I don't want to try getting any higher for now even if I do have a third party cooler (and it does perform greatly, but it's still not liquid cooling either).

By the way, the GA-X38-DQ6's Memory Straps are summarized in this picture: http://i228.photobucket.com/al...Q6/bios/StrapChart.jpg

I think that my instability problem at 445x9 right now may well be related to the Memory strap I've selected (Memory/FSB ratio now set at 5 for 6). But I'm not entirely sure about that. I know that the GA-X38-DQ6 isn't owned by everyone, but I just wanted to get some suggestions. But as I said the voltages at the moment are already high, I think, for instance my BIOS' PC Health Status reports the following voltages (under my current OC settings of course):

VCore 1.396V
DDR18V 2.136V
+3.3V = 3.296V
+12V = 12.302V

I'm not sure what can be done, but I should have some room to play with. So what would you guys suggest me to adjust to at least get past the Windows loading screen at 445x9? Or perhaps at another setting getting me to 4.0Ghz?

Thanks.

 

[myocardia]

Your RAM speed is too high. Lower it to 2.0x, and take it out of the equation.

 

[Tempered81]

1:1 memory, run it at 445 x 2 aka 890mhz, default or spec timings too.

 

[Zenoth]

Well, bad news, 1:1 Memory doesn't work, it goes up to the on-board peripherals initialization screen, right before Windows starts to load, and then it restarts. I've tried both the 2.00D and the 2.40B Straps, no go on either of them (both giving 890Mhz but at different ratios of course, 2.40B Strap being 333Mhz FSB and 2.00D being 400Mhz FSB as indicated in the BIOS). I even increased the vCore voltage just one notch to see if it'd help, but still no go with that. I left the DRAM voltage the same, and then I reduced it, still no go. I tried 4-4-4-15 (default) and 5-5-5-18 (current) and no go either.

 

[Phew]

If that is a C0 E8400, that's par for the course. When I owned a C0, anything above 3.8 GHz just took more voltage than I was comfortable with (1.4+). You still have a 25%+ OC, which is nothing to sneeze at.

E0 E8400s, on the other hand, routinely push 4.3+ GHz within the Intel recommended voltage range (<1.3625).

It may just be your chip, not your BIOS settings.

 

[Zenoth]

Indeed, I'm also suspecting that it could be the case. If so it's fine, I'm contempt with my currently stable OC, getting to 4.0Ghz stable would just have been "for the heck of it", out of curiosity mostly. And yes indeed it's a C0 stepping and I bought it only two weeks after its release last year, it may have been a bad batch too.

 

[Tempered81]

well here is a C0 E8400 on a Gigabyte GA-X38-DQ6 running at 4.7Ghz. I doubt it's the chip or board.

http://img401.imageshack.us/my...age=everest4700jh3.jpg

I'm looking for some more information on overclocking with your motherboard, zenoth.

 

[Tempered81]

here's one at 4.3ghz, but it does take a lot of Vcore. maybe that is your issue.

http://img212.imageshack.us/im...6/everest434685dx4.jpg

If you lower your vcore while staying with your 3.74ghz settings, what is the lowest vcore you can reach before instability occurs?

some bios pictures:
http://i105.photobucket.com/al...s_2006/04022008193.jpg
http://i105.photobucket.com/al...s_2006/04022008190.jpg

And finally, someone with the exact same rig as yours:
http://www.xtremesystems.org/f...2751742&postcount=1625

:laugh: Sorry I couldn't be of much help. I was just searching through the gig x38 thread at XS and saw that you were already posting in it.

 

[Zenoth]

Thanks a lot for your time jaredpace, I do appreciate it. I've noticed that the guy OC'ing it at 4.3Ghz and 4.7Ghz went for a 1.5v+ vCore, which in my book is suicidal, I don't want to risk that much.

I barely want to go above 1.4, and if that is indeed the universal cause of my problems then I'll just deal with it, I don't want to fry this baby, it's the only one I have and I couldn't possibly afford another one if something goes wrong. And in the BIOS pictures I see that there's a 0.35v extra voltage on the PCI-E over-voltage setting, I hope that the guy knows what the risks are by doing that.

Anyhow, last night I've let mine run with Prime95 at 417x9, just to get 3.75Ghz and it's stable so far (9 hours, no issues). I'll keep doing this with very small increments until I reach the "wall" under my current vCore. Then I will increase the vCore a little, maybe to get around 1.410 or so, and then I'll stop. If I can get 3.80Ghz stable I'll stay there. But, yeah, thanks for your time man, it does give me some hope at OC'ing more... but I always thought that the E8400's OC'ed very high with low amounts of vCore, it's what has made their popularity over time, but indeed, exceptions always exist.