• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Asus Commando+2GPU+4x1Gb blue screen

H

HighYield

Gawd
2FA
Joined
Feb 2, 2008
Messages
944
I fold on a pair of old Asus Commando (P965 chipset) motherboards and have recently started getting the dreaded blue screen since doing a ram exchange.

System 1 had a Q6600@3.2GHz, 4x1GB Corsair DDR2 800 4-4-4-12 running at 5-5-5-15@2.1V (within spec) and a GTX260 gpu.

System 2 had a Q6600@3.2GHz, 2x1GB of the same ram, and 2 8800GT cards

Both boards have been folding just fine for about 10 months 24/7. I got some of the new "cheap" 2x4Gb ram kits and loaded system 1 up with 4x2Gb. No problems at the same settings (although this ram runs at 1.8volts). I then took out the matched pairs that had been running in system 1 and put them in system 2 since they were known to work together and in the same type of motherboard at the same speed. Once I got back to folding all hell broke loose and it wouldn't fold long at all (<10 min) without getting the dreaded blue screen.

I pulled back the OC from 9x356 to 9x333 and increased the north bridge voltage from auto (1.40v according to the bios monitor) to 1.40v (1.456 according to the bios monitor). This got me 2-4 hours of stability vs the 10 min of before.

The question is how much north bridge voltage can be applied safely to get this machine stable again? Do I need to overvolt something else: FSB termination voltage, memory voltage, DDRII channel ref voltage?

I increased the north bridge 1.425 (1.488 according to the bios monitor) and will see how that does. Edit- This only lasted about 2 hours before blue screen.

 
Make sure your system has the patch for the over 3GB ram usage.


I had a similar issue.



 
The memory was remapped in bios and 64 bit vista has no problem seeing it. I think that patch is for the problem that was related to vista not booting if you had more than 3 gig installed.... is there another patch for >3gig of ram usage?
 
Have you run Memtest86+ to test that the memory is error free / /

Its the first thing I run, even before I install the OS, just to find out if the memory works with the MoBo.

Luck ............ :D
 
Tigerbiten said:
Have you run Memtest86+ to test that the memory is error free / /

Its the first thing I run, even before I install the OS, just to find out if the memory works with the MoBo.

Luck ............ :D
Click to expand...

Ran memtest for about 90 min (at least through 1 full pass) - no errors.
 
Well since everything was running fine before switching to the cheaper 1.8v memory I would say that your motherboard settings were and are fine and do not need changed.
What does seem very odd is the 1.8v that you say those memory modules run at.
I think that those voltages must be in error. I would do some research on the correct upper limit voltage for those memory.

 
Razor_FX_II said:
Well since everything was running fine before switching to the cheaper 1.8v memory I would say that your motherboard settings were and are fine and do not need changed.
What does seem very odd is the 1.8v that you say those memory modules run at.
I think that those voltages must be in error. I would do some research on the correct upper limit voltage for those memory.

Click to expand...

I have two of the same mother board folding. The one with the 4x2Gig 1.8v corsair ram is running fine and at 3.2GHz. The other board is running the 4x1Gig 2.1V corsair (the 2.1V ram is CAS4 the 1.8V ram is CAS5) and CAS4 ram worked fine in the other motherboard.
I clocked the system down to an anemic 290X9=2.6 GHz and it has been running stable for 4 hours. I think it just needs a bit more voltage to get it stable again at 3.2GHz. I'm just not sure where to apply it.
 
HighYield said:
I have two of the same mother board folding. The one with the 4x2Gig 1.8v corsair ram is running fine and at 3.2GHz. The other board is running the 4x1Gig 2.1V corsair (the 2.1V ram is CAS4 the 1.8V ram is CAS5) and CAS4 ram worked fine in the other motherboard.
I clocked the system down to an anemic 290X9=2.6 GHz and it has been running stable for 4 hours. I think it just needs a bit more voltage to get it stable again at 3.2GHz. I'm just not sure where to apply it.
Click to expand...
I believe your NB limit on that board is 1.5v so I don't think thats it.
Try upping your CPU back to were you know it runs stable prior to the new memory but strap your memory at 1:1 5/5/5/15 and see if they will stabilize there.
 
What is the stop code, etc, etc of the BSOD ??
That could well be the first place to start looking.

Can you load all the CPU cores with Prime/Orthos and just let it run ??
Will it run a Video card stress test without error ??
I tend to stress each physical sub-system in turn to help find out whats wrong.
If they all work error free then it time to think it could be a software/driver error.

Luck ........... :D
 
I'll try the pencil mod if nothing else works.... Doesn't sound like that is reversible if there is trouble. :eek:

Tigerbiten: Sorry, I did not write down the stop codes at the blue screen. I will the next time. The machine has been at 100% CPU load and 100%x2 GPU load for about 6 hrs now with the Q6600@2.6GHz. If it can go 24hrs like this I'll just keep upping the FSB until there is trouble and then move from there. I had run prime 95 for few hours at 3.2GHz and it seemed stable until I added the 2 GPU folding clients. What is killing me here is that this boxen was a rock with 2 GPU and 2 SMP on 64 bit vista with only 2 pieces of ram @3.2GHz. Now with 4 ram sticks it is a pain in ass.
 
HighYield said:
I'll try the pencil mod if nothing else works.... Doesn't sound like that is reversible if there is trouble.
Click to expand...
Just like with anything in pencil, if you wan't it to go away just use the eraser.
You can allways erase it off the resistor and reapply it anytime.
 
Tigerbiten said:
What is the stop code, etc, etc of the BSOD ??
That could well be the first place to start looking.

Can you load all the CPU cores with Prime/Orthos and just let it run ??
Will it run a Video card stress test without error ??
I tend to stress each physical sub-system in turn to help find out whats wrong.
If they all work error free then it time to think it could be a software/driver error.

Luck ........... :D
Click to expand...

This system ran stable for 30hrs or so @2.6GHz. Confirming the software is ok and that the ram that was swapped in is ok. (The rest of the system had been running fine for a very long time). Not being content to run at 2.6GHz I tried it at 2.8 GHz with no north bridge adjustment and it blue screened within 10 min once the CPU and 2 GPU clients were running.

The error info follows:

Problem signature:
Problem Event Name: BlueScreen
OS Version: 6.0.6001.2.1.0.768.3
Locale ID: 1033

Additional information about the problem:
BCCode: 1000007e
BCP1: FFFFFFFFC0000005
BCP2: FFFFF80001CC6DF2
BCP3: FFFFFA60005AFA08
BCP4: FFFFFA60005AF3E0
OS Version: 6_0_6001
Service Pack: 1_0
Product: 768_1


Does anyone know what that stuff means ??
 
Does it have the newer 1801 BIOS? I have the same Motherboard and it was first distributed with poor memory voltage settings.
 
the only thing id say is once you start overclocking it again.. try upping the memory voltage to 1.9 or 1.95 if you have the option to do so.. this was the only way i could overclock with my pos generic kingston memory even though the rated voltage is 1.8v @ 5-5-5-15.. atleast in my case i think my issue was with the vdroop on the ram.. since 1.8v in bios was actually 1.78v to the memory.. raised it to 1.9v which was the lowest setting i had outside of auto(which did the default 1.8v and it raised my voltage to 1.88v on the memory and everything was stable up to the current overclock i have on my be-2350 in my signature.. so it might be worth taking a shot at.. it wont kill the memory and if it doesnt work just put it back at 1.8v.. but it might give you an idea on whats causing it to BSOD..
 
HighYield said:
This system ran stable for 30hrs or so @2.6GHz. Confirming the software is ok and that the ram that was swapped in is ok. (The rest of the system had been running fine for a very long time). Not being content to run at 2.6GHz I tried it at 2.8 GHz with no north bridge adjustment and it blue screened within 10 min once the CPU and 2 GPU clients were running.

The error info follows:

Problem signature:
Problem Event Name: BlueScreen
OS Version: 6.0.6001.2.1.0.768.3
Locale ID: 1033

Additional information about the problem:
BCCode: 1000007e
BCP1: FFFFFFFFC0000005
BCP2: FFFFF80001CC6DF2
BCP3: FFFFFA60005AFA08
BCP4: FFFFFA60005AF3E0
OS Version: 6_0_6001
Service Pack: 1_0
Product: 768_1


Does anyone know what that stuff means ??
Click to expand...


googling the errors everything leads to the memory size being 4gb.. but im still leaning toward the voltage..
 
You must log in or register to reply here.
Back
Top