Help to oc my E8400 on Lanparty X38 Lt @4Ghz

[johnp1983]

Hello boyz,

I'm having strange problems with the mainboard and the cpu stability in oc @4Ghz.

My system fails Super PI 1M... I have tried raising up the voltage of cpu and the northbridge, but it's a strange situation. The os crash after some second of being in idle or when i start some stressing test like super pi1m as i said before. In particular the system freezes and it restart itself with the last bios settings.

Bios settings is:
E8400 Q746A366 @ 9x445, v-core set to 1.49 (1.52 in Windows in Idle)
Nb-vcore @ 1.45v
Cpu VTT @ 1.46v
Strap, I've tried both 333\667 and 333\800... but the situation won't change

It's normal that my cpu at those voltages is still unstble?! I read of people that with air cooling can be stable @ 445x9 with 1.33\1.37 cpu v-core

Can you give me some tweaks to solve this situation?



Dfi Lanparty X38 Lt (2008\11\1 bios)
Zalman 9700 Cnps (fan @ 2000 RPM)
Crucial Ballistix 2x1GB DDR2 1066Mhz at 5-5-5-15-2T @ 2.25V)
front 2*120mm Nanoxia Fx12 2000rpm
back 1*120mm Nanoxia Fx12 2000rpm
Xp Pro Sp2

Bye!!

[Garrett]

Mighty high vcore and VTT you're running there... Try running 1.35v load (calculating droop and all)
I have the VTT @ 1.20v... works flawlessly on the DFI P35 DK...

[johnp1983]

Hello Garrett, thanks for the answer...

If i set cpu v-core to 1.35 in full load (in bios 1.32), the system restarts before the Win Xp loading screen's coming out.

This happens also by lowering the cpu vtt and v-nb core; I've tried some settings... but the systems inesorably crash and restart itself.

I don't know what to do, considering that i payed a lot for this high-end motherboard.

[yokomo]

whats it like at slightly lower clocks ? also whats your temps like ?
and agin your giving that 65nm alot of juice ...... im runing mine @1.25 in bios ! (yes im on water but my load temp is still 55c )

[johnp1983]

I think that cpu T° sensors are wrong... this is the situations @ default with C1E & EIST activated (obviously I disabled them in oc)

[Garrett]

Hmm.... you might have a poor overclocker too... I don't think it's the board...

and yokomo, I think you mean 45nm?
the load temps are totally screwed up, something's up with the temp sensor on the e8x00 cpu's, it's all over the net

[johnp1983]

Hmm.... you might have a poor overclocker too... I don't think it's the board...

and yokomo, I think you mean 45nm?
the load temps are totally screwed up, something's up with the temp sensor on the e8x00 cpu's, it's all over the net

So you think it's simply a bad cpu for 445x9?

[yokomo]

yeah 45nm woops
looksliek your running ram faster than cpu fsb you shoudl really be runing them both in synch till your sure your cpu is topped out and stable.

[johnp1983]

Even if I use @ 1:1 ratio it won't change...

the cpu requires an allucinant voltage to be topped @ 4ghz; what a damned cpu... others e8400 are stable @ 4Ghz with very less v-core on the cpu.

[G4h4o8s6T]

Have you tried tweaking the GTL settings at all?? These 45nm chips like low GTLs, for 4ghz I would try 70/70/70 and work your way down. Not only should this add stability to your current vcore/vtt setting, but you should be able to lower them significantly.

With my E8400 at 500x8, on my DFI P35 board, I only use VTT at 1.1v and GTLs at 60/60/60.

[Ron 61]

Try this, clock it down to were its benchmark stable and slowly work back up until you get to the point were it starts getting unstable. At that point is were you need to individually adjust the voltages and test with a short benchmark. Hopefully it will respond to one of the voltage adjustments. Wouldn't hurt to clear cmos either. These cpu's are finicky to say the least. Also, CPU VTT of 1.46v IMO is torture.

[johnp1983]

I tried to use GTL in various settings like 62\62\62, 65\65\65\, 68\68\68 ect.

with

1.35v-core in bios
1.30v-nb core
1.13v-fsb vtt

But all that it happens is that the system stuck on a blank screen after the Raid configuration is being loaded by te Bios.

Also by raising the v-core to 1.39 in bios the situation won't change.

I'm seriously thinking there's a problem on board, or in the psu.

I dunno what to do...

[G4h4o8s6T]

Im guessing you've updated to the latest Bios as well?? Have you looked through the DFI X38 thread to see if whatever Bios your using works well with these 45nm CPUs??

Other than that, Im with you and have no idea what else to do, and it could possibly be a bad board.

Though maybe someone more knowledgeable will chime in and have some more suggestions, dont give up yet!

[johnp1983]

Yes I give a look to the Dfi X38 thread and E8400\E8500 official thread, but no one answer to my posts...

I have flashed some time ago the last bios for my board: 1\11\2008, and at thise time there are no other bios on the dfi official site.

I also have problem with 0.5 multi... the board doesn't let me activate them. So I'm thinking that this problems are due to a crappy bios...

BTW, i writed a mail to DFI support Europe... describing them my problem: hoping that they will answer as soon as possible.

[Ron 61]

Might try DFI Club.

[zsamz_]

set your gtl setting to 68-69%
and the voltage you using is way too high on the cpu
gtl settings are verry important 1 wrong value n you wont boot
did you try setting them on auto?

[RGone]

Works fine if you drop the multiplier to 8 then enable N2 multiplier and it will set 8.5 multipier. There is no 9.5 multiplier so it will not set when you have the highest multiplier already set since it will up by 0.5 for the half multi.

I have one of the largest POS overclocking E8400 I have seen. 1.47 Vcore just to run a couple of hours at 4050Mhz and I have tried it in 3 different boards before you ask. There are a few of these crappy overclockers around and I "know" I have one. When you get it up to about 1.45 Vcore it will usually prime a few hours at 3870Mhz or my POS clocker does.

Yes I give a look to the Dfi X38 thread and E8400\E8500 official thread, but no one answer to my posts...

I have flashed some time ago the last bios for my board: 1\11\2008, and at thise time there are no other bios on the dfi official site.

I also have problem with 0.5 multi... the board doesn't let me activate them. So I'm thinking that this problems are due to a crappy bios...

BTW, i writed a mail to DFI support Europe... describing them my problem: hoping that they will answer as soon as possible.

[Mav451]

Something is seriously strange. GTL tweaking should not be necessary for such a mild OC - and I'm speaking from experience (I didn't touch it). All you need to OC is vcore - that's it. Raising VTT or NB at such a low FSB is only going to create more unnecessary heat.

*Btw: are these new sticks of RAM? Have you tested them - individually and per slot (to eliminate both slots and each stick, respectively)?

[johnp1983]

Might try DFI Club.

Yes... I started @ thread on dfi club forum just 3 days ago.

set your gtl setting to 68-69%
and the voltage you using is way too high on the cpu
gtl settings are verry important 1 wrong value n you wont boot
did you try setting them on auto?

As I said before... I tried also to using gtl settings with various value like 68\68\68 or 70\70\70 with voltage settings like

1.35v-core in bios
1.30v-nb core
1.13v-fsb vtt

but the system crash befoure windows loading logo will appear on screen.

Works fine if you drop the multiplier to 8 then enable N2 multiplier and it will set 8.5 multipier. There is no 9.5 multiplier so it will not set when you have the highest multiplier already set since it will up by 0.5 for the half multi.

I have one of the largest POS overclocking E8400 I have seen. 1.47 Vcore just to run a couple of hours at 4050Mhz and I have tried it in 3 different boards before you ask. There are a few of these crappy overclockers around and I "know" I have one. When you get it up to about 1.45 Vcore it will usually prime a few hours at 3870Mhz or my POS clocker does.

For the fractional multiplier is my BIG mystake... I tried N2 with 8x multi and the 0.5x appear magically.

As i write some minutes ago on dfi club forum (assuming you're the same Rgone of that board), I don't understand the meaning of "POS"... since I'm not English madrelingua.

Something is seriously strange. GTL tweaking should not be necessary for such a mild OC - and I'm speaking from experience (I didn't touch it). All you need to OC is vcore - that's it. Raising VTT or NB at such a low FSB is only going to create more unnecessary heat.

*Btw: are these new sticks of RAM? Have you tested them - individually and per slot (to eliminate both slots and each stick, respectively)?

The Ram sticks aren't new... prior to use them with Wolfy E8400 I tested them a lot with an E6600 @ 3.4\3.6\3.8Ghz also with Dfi X38 board... and they perormed well.

BTW I'm runnging Orthos blend with "6" on priority settings... just to test if the systems is stable @ default. Next bench I'll do is a double memtest for windows on the ram. I'll post the screens in the next 2\3 hours.

[HDCHOPPER]

"POS" = piece of Shi*

[johnp1983]

"POS" = piece of Shi*

Oooh... now I've understand!!

[johnp1983]

So, here's the screens:

Orthos blend priority 6@def.


Memtest 4 Windows @def

[johnp1983]

EDIT

And here's a screen with a little tights timings on Ram:


Dram timings section was:
- Enhance Data transmitting: FAST
- Enhance Addressing: FAST
- T2 Dispatch: Enabled

Ch1 Clock Crossing Setting: More Aggressive
Ch2 Clock Crossing Setting: More Aggressive
Ch1Ch2 CommonClock Setting: More Aggressive

Ch1 Read delay phase (4~0)
- Channel 1 Phase 0 Pull-In: Enabled
- Channel 1 Phase 1 Pull-In: Enabled
- Channel 1 Phase 2 Pull-In: Enabled
- Channel 1 Phase 3 Pull-In: Enabled
- Channel 1 Phase 4 Pull-In: Enabled

Ch2 Read delay phase (4~0)
- Channel 2 Phase 0 Pull-In: Auto
- Channel 2 Phase 1 Pull-In: Auto
- Channel 2 Phase 2 Pull-In: Auto
- Channel 2 Phase 3 Pull-In: Auto
- Channel 2 Phase 4 Pull-In: Auto

The remaining timings, like clock fine delay was on AUTO.

I can assume it's not a memory instability problem right?!