Problem Overclocking E8400 @ 3.6ghz

[chibanginpr]

Ive read numerous of threads and other sites with tutorials and walkthroughs on overclocking on a P35-DS3L. I've tried numerous of settings. I was told I could overclock to 3.6 without any CPU Voltage changes. Everytime I would boot up a weird windows error would appear saying windows encountered an error from lsass.exe and would have to terminate. So i even tried boosting the CPU Voltage up to 1.22500V to no avail same problems.

What am I doing wrong?

P35-DS3L
E8400
2Gb Crucial Ballistix PC6400 @ 800MHz
Artic Cooling Freezer 7 pro CPU Cooler
Seagate Barracuda 7200.10 320Gb 16 MB Cache
Corsair 550VX PS
Nvidia 8500GT 512mb

My Bios Settings were as follows:

Robust Graphic Booster [Auto]
CPU Clock Ratio [9 X]
CPU Frequency 3.60GHz (400x9)
CPU Host Clock Control [Enabled]
CPU Host Frequency(Mhz) [400]
PCI Express Frequency(Mhz) [100] Also tried [Auto]
C.I.A.2 [Disabled]
Performance Enhance [Standard]
System Memory Multiplier (SPD) [2.00]
Memory Frequency(mhz) 800 800
DRAM Timing Selectable (SPD) [Manual]
*******Standard Timing Control*********
CAS Latency Time 5 [4]
DRAM RAS# to CAS# Delay 5 [4]
DRAM RAS# Precharge 5 [4]
Precharge Delay(tRAS) 18 [12]
********Advanced Timing Control******
I didnt change any of these settings . I just left this on auto

System Voltage Control [Manual]
DDR2 OverVoltage Control [+0.4V] for 2.2V timings
PCI-E OverVoltage Control [Normal]
FSB OverVoltage Control [+0.1V] I Also tried [Normal]
(G) MCH OverVoltage Control [Normal]
CPU Voltage Control [1.22500V] Also Tried [Normal]

 

[rubix_1011]

Loosen your RAM timings a little...try 6-6-6-18/20 or set to AUTO for now.

Remember, not everyone gets to the level as 'all the forum postings'. A lot has to do with PSU quality, board setup, specific RAM modules (and their individual qualities) and CPU qualities. I'm not saying its not possible, but you might have to do more tweaking than someone else to achieve the same speeds. Also remember, the overclocks achieved with new hardware will probably not be as stable over time and you might have to scale back a little. What speeds are you seeing so far?

 

[ZOldDude]

http://www.google.com/search?hl=en&q=lsass.exe&btnG=Google+Search

Seems to be a software issue,try running chkdsk befor you OC.

 

[ctbaars]

I use:
5-5-5-15 t2 mem +0.1v setting
CPU voltage of 1.25V

 

[dechy]

Have you checked the temps in the BIOS? I must of installed a least a dozen aftermarket fans for my home PCs, and I don't think I've ever gotten it right on the first try lol.

Even with my e8400, my temps were REAL high (55+ idle) in the BIOS using a CNPS9700... had to take it off, clean all the thermal paste and re-seat everything. Finally worked.

Like Rubix mentionned, loosen to 5-5-5-18. Not sure why you're overvolting the FSB, even if it's just 0.1V, I didn't have to touch anything but the CPU voltage (and only a very tiny bit)

BTW, seeing as I don't have the same board, what's the Performance Enhance feature? Maybe even at standard it's trying to OC something by a tiny portion and throwing everything off (I hate the Asus equivalent)

And Robust Graphic Booster?

 

[hughyhunter]

Did you re-format when you installed new hardware? Yes... It does seem to be a software issue. Does the problem subside when you set FSB back to 333mhz? If it does it's not your overclock. You should really determine if that's the case.
I would have to say that if you are running your memory at stock timings, frequencies, and voltages than it's not the issue.
If you are impatient about this it's not going to be worth-while. You should take your time, re-format windows if you haven't, make sure you have all the latest drivers, and bios updates. If you still are encountering problems loosen timings up a bit so you KNOW it's not memory and give your CPU at least 1.325v. That's what my board give my E8400 by default in auto at 3.6ghz. But I lower it to 1.225 because obviously I dont need that. I have heard though people needing a bit more than that like closer to 1.275v. So if that's the case in your situation that's fine. Just try not to go to far over 1.4v if you want to play it safe. I have had mine at 1.38v for months without error with a 4.05ghz overclock.

 

[rubix_1011]

PCI Express...leave at 100/133...don't bump this up. (I don't remember what default is...)

Set your RAM timings to AUTO

Set your Memory Frequency to AUTO

Set your Voltages to AUTO to start

Once you get the speed you want, start with your chip voltage, move it down and see how stable it is. Chances are, AUTO will set it a little high.

Then, leaving your RAM voltages at AUTO, start toying with timings...try 6-6-6-18/20, then 5-5-5-18/15...see what you get. Leave your frequency at auto; I have never changed mine. Once you find a good spot, start working on your RAM voltage. Chances are, you shouldn't have to bump those much to keep your speeds.

 

[closed_deal]

I'd agree with hughyhunter sounds the more likely cause and best way to fix .

 

[hughyhunter]

http://www.google.com/search?hl=en&q=lsass.exe&btnG=Google+Search

Seems to be a software issue,try running chkdsk befor you OC.

I cant help but to agree... all other suggestions are good but the OP is a little vague. I would like to know if he was getting this error before he overclocked and if he get's it when he goes back to stock settings.

 

[hutt]

Well we have the same board, ram and proc as for me i started with 3.6 before going higher and had np getting it with NO volt changes. I agree with others as this sounds like software to me. I run the ram at 4-4-4-12 with no problems.

 

[closed_deal]

Could we have some more info if you get past POST without errors etc... does it fall over before the default windowz loading screen? What version of windowz?

*EDIT* When you reset your timings your RAM should be fine. Maybe try lowering your RAM speed and keeping the timings. It may work out slightly quicker.

 

[chibanginpr]

Thanks guys Ill keep trying. I dont have the lsass.exe problem when i set back to optimized defaults. So its something to do with my overclock. Everything runs fine when its at 333. Ill toy with the Voltage a little. When i relax my timings should everything still be on auto? or should i up the volts on that as well?

 

[hughyhunter]

^^ keep it on Auto... It depends for the memory though, I always set my memory voltage on the highest that the manufacturer will allow for warranty purposes. For me that's 2.2v.

 

[chibanginpr]

Ok well im partially stable right now on 3.6 @ 1.275v and 6-6-6-18 timings. sits idle at about 46-50c. But when i run Prime95 which i have no idea what it does nor does the help thing work. I open Prime95 and hit just stress test and pick Blend which does some tests of everything. It starts off with 2 worker threads. Worker thread #1 runs pretty hot of 70-75c but it keeps running. Worker thread#2 runs for less then a minute and says it gets an error and stops running. What do i do?

[May 2 13:27] Work thread starting
[May 2 13:27] Beginning a continuous self-test to check your computer.
[May 2 13:27] Please read stress.txt. Choose Test/Stop to end this test.
[May 2 13:27] Test 1, 4000 Lucas-Lehmer iterations of M19922945 using FFT length 1024K.
[May 2 13:28] FATAL ERROR: Rounding was 0.5, expected less than 0.4
[May 2 13:28] Hardware failure detected, consult stress.txt file.
[May 2 13:28] Torture Test ran 0 minutes - 1 errors, 0 warnings.
[May 2 13:28] Work thread stopped.

 

[hughyhunter]

CPU errors.... bump up voltage! Try 1.3v now. Also you are running a bit hot. Do you have the stock HSF... if you do you will just get hotter with more voltages. A good idea would be to get new HSF.

 

[chibanginpr]

Ok well i bumped it up to 1.28 worker thread #2 ran for 2 minutes then error. I have a artic cooling freezer 7 pro on. It hits about 70-76 under 100% load.

 

[hughyhunter]

That's a pretty good cooler. However it does seem like your temps are way to high. Maybe you have a bad mount or arent making contact to cpu properly. It just seems a bit high to me. Where are you getting your readings from?

 

[chibanginpr]

Hmm ok .. just ran @ 1.3v .. I didnt get an error but i turned it off considering my temps were hitting 80c. Im using Core Temp. Where should i be at with temps running 100%?

 

[monst0r]

A TJunc less than 70c would be OK (not for 24/7 use). I always like to keep my TJunc under 60c, seems safe. look at computronix's sticky for more info on temps here.

Scale 1: Duo
E4x00: Tcase Max 73c, Stepping M0
E2xx0: Tcase Max 73c, Stepping M0
E8x00: Tcase Max 72c, Stepping C0
E8x90: Tcase Max 72c, Stepping C0
E6x50: Tcase Max 72c, Stepping G0
E6540: Tcase Max 72c, Stepping G0

-Tcase/Tjunction-
--70--/--75--75-- Hot
--65--/--70--70-- Warm
--60--/--65--65-- Safe
--25--/--30--30-- Cool

 

[kpo6969]

I have the same cpu, mobo, and cooler
1.ram voltage gives mobo fits from what I read (I have GS 4x1 6400@800mhz NQ 1.8-2.0v)
I run 3.6ghz @ 1.216-1.232
FSB 400
What I did is start v-core @ 1.275 and go from there (mine is 1.2625 if I remember correctly)

Reseat your cooler also, I idle @ 35c and load @ 54c (using RealTemp-TjMax 95c)

 

[Shadow703793]

CAS Latency Time 5 [4]
DRAM RAS# to CAS# Delay 5 [4]
DRAM RAS# Precharge 5 [4]
Precharge Delay(tRAS) 18 [12]

The problem is the RAM timings. Try loosening them a bit and see.

Also see:
http://www.tomshardware.com/forum/forum2.php?config=tomshardwareus.inc&cat=29&post=245679&page=1&p=1&sondage=0&owntopic=3&trash=0&trash_post=0&print=0&numreponse=0&quote_only=0&new=0&nojs=0
if you haven't already seen it.

If interested:
http://www.tomshardware.com/forum/forum2.php?config=tomshardwareus.inc&cat=29&post=246979&page=1&p=1&sondage=0&owntopic=3&trash=&trash_post=&print=0&numreponse=0&quote_only=0&new=0&nojs=0#t1754889

 

[kpo6969]

CAS Latency Time 5 [4]
DRAM RAS# to CAS# Delay 5 [4]
DRAM RAS# Precharge 5 [4]
Precharge Delay(tRAS) 18 [12]

The problem is the RAM timings. Try loosening them a bit and see.

Also see:
http://www.tomshardware.com/forum/forum2.php?config=tomshardwareus.inc&cat=29&post=245679&page=1&p=1&sondage=0&owntopic=3&trash=0&trash_post=0&print=0&numreponse=0&quote_only=0&new=0&nojs=0
if you haven't already seen it.

If interested:
http://www.tomshardware.com/forum/forum2.php?config=tomshardwareus.inc&cat=29&post=246979&page=1&p=1&sondage=0&owntopic=3&trash=&trash_post=&print=0&numreponse=0&quote_only=0&new=0&nojs=0#t1754889

Alot of good info in those threads.

 

[vinman]

Just remount your cooler dude I'm running 3.6 ghz @23c in an antec 900 but your temps are too high, that's why it's erroring out HEAT makes it not run right 5-5-5-8 on memory will be fine also. You should be able to pretty much pick up your motherboard by the heatsink if it's on right.

 

[chibanginpr]

Wow 23c? Anyways .. My 8800GTS just arrived so i installed that and remounted my Cooler and now my comp stays around 60-63c Under 100% load. Thats with 1.29v and auto memory timings. Ill mess around with it more but im glad i got it down and stable .. Ill run Prime95 longer but i just ran it for 35 minutes with no errors.

 

[hughyhunter]

^^ looks good... you must have had a bad mount. Also you should look into lapping that sucker and using the "right" amount of arctic silver 5 thermalpaste.