Unstable computer with lots of error messages under memtest86

[Segvoia]

About 2 months ago I bought a brand new machine.

Specs:
Motherboard: GIGABYTE GA-G33M-DS2R
RAM: Crucial DDR2 BallistiX PC6400 4 x 1GB
CPU: INTEL Core 2 Duo E8400 3.0GHz LGA775 FSB1333 6MB

But now my computer is unstable from a long time. BSOD are random while the computer is in use, and sometimes I have to restart the computer up to 3-4 times in a row before it at all able to get into Windows.

The error messages which I get suggests that there is a memory problem. I tried using memtest86 v2.01 for 10 minutes. It found 226 errors.

Intel Core 2 3000MHz
L1 Cache: 64K 49,181 MBps
L2 Cache: 6144K 20,979 MBps
Memory: 4095M 4,043 MBps
Chipset: Intel P35/G33 (ECC: Disabled) - FSB: 333MHz - Type: DDR2
Settings: RAM: 400MHz (DDR800) / CAS: 5-5-5-18 / Dual Channel

Need help desperately!

[Yitzchak909]

Update BIOS. This fixes many types problems with 4GB of RAM. Most important if you use vista.

Dram voltage: 2.1V
NB Core Voltage: 1.4v -> important and this increases slightly for 4gb.

TCL 4 (CAS Latency 4) -> timings for the OCZ DDR2 2GB Platinum XTC Dual Ch.
tRCD 5
TRP 4
tras 15
Command Per Clock (CMD): 2 clock (2T)

Advance Memory Settings
tRRD: AUTO
TRC: AUTO
tWR: AUTO
tWTR: AUTO
TREF: AUTO
Async Latency: AUTO

By proper voltage and timings in the bios, only 0 errors are ok (in all 9 tests), everything else is defective memory.

[Segvoia]

Downloaded last BIOS. Generally I use the manufacturer's website for BIOS update. However can I use other BIOS's also?

The only place I can ask for voltage's etc are under "MB Intelligent Tweak (MIT)". I have the following options;

Robust Graphics Boosters (Auto/ Fast/ Turbo)
CPU Clock Ratio (number from 6-9)
Fine CPU Clock Ratio (0.0/ 0.5)
CPU Frequency (info about the current GHz, 3.00 (333x9))
CPU Host Clock Control (Disabled/ Enabled)
CPU Host Frequency (info about the current Hz, 333)
PCI Express frequency in MHz (Auto/90-150)
Enhance Performance (Standard / Turbo / Extreme)
System Memory multiplier SPD (Auto/1.60/2.00/2.40/3.20)
Memory Frequence in MHz (info about the current Hz, 800)

Voltage Control System (Auto / Manual)
DDR2 Over Voltage Control (Normal / 0.1 V / 0.2 V / 0.3 V / 0.4 V)
FSB Over Voltage Control (Normal / 0.1 V / 0.2 V / 0.3 V)
(G) MCH Over Voltage Control (Normal / 0.1 V / 0.2 V / 0.3 V)
CPU Voltage Control (Normal / Long list from 0.5V to 1.6V)

You said to Dram Voltage to 2.1V, ie "DDR2 Over Voltage Control" in my case. If the standard is 1.8, so I have to put it to +0.3 V, right?

You say Core Voltage ... what do I have set for the options I have available?

And I can not find any place where I can ask for timings on RAM ... they are hidden away in a "secret" menu, or does not support my motherboard to ask for this?

[Segvoia]

4-4-4-12
NB Voltage 0.3 V
DDR2 Voltage 0.3 V

1 run pass with memtest86 - 0 errors.

Then again continuous problems with my computer.

Dual Channel has been running all the way, and it has been unstable. Last attempt was 2x1GB, in Single Channel. Ran manually [email protected] / 2.0V, [email protected] / 1.4V. Did stable, it seems that, but then came BSOD's on the number and a row - and plenty of errors in memtest after reboot.

It seems like it works for a while, but then "build" that up until the last drop that gets it to run over. I have repeatedly used the same settings, and very few errors in memtest, but when after a while intersects with the BSOD's, and I rest species to run memtest, can I get hundreds of errors - although the settings are the same.

I'm running now 2x1GB in Single Channel @ auto in the BIOS. Then we will see how it goes.

[pddaugusto]

Put the excitement of memory to 1.9V and 2.0V but 1.9V to give.
Then set the timings to 5-5-5-15, it will say:
5 at TCL
5 tRCD
5 on the TRP
15 on tras

[baylock]

Hi, time to resurrect an old thread i suppose but here goes.

I seem to be suffering the same problem with an AMD system:

GA-MA790X-UD4P
Phenom 2 940 BE Stock cooler
4Gig Corsair Dominator CM2X2048-8500C5D 1066 2.1v
4x SATA HDD 7200RPM
PALiT ATI Radeon HD4870 1Gb
Corsair VX550 Power Supply
Vista X64 SP2
---------------------------------------
The closest have been able to get it running relatively stable by forcing the ram to 800Mhz at 1.8v, but it still crashes after about 24hours (eather BSOD, or random programs become unstable) or so, and also throws up errors in IntelBurnTest once it has been running for a while.
If i run memtest immediately after a BSOD event i git a lovely xmass tree effect rather quickly, but if i shut down completely for about 5 min then the errors go away.

was a solution reached for yours, or is there any further advise??
thanx for the asist

[PeterAGASSI]

I am listing you some suggestion of why a system crashed frequently. This can help you out to sort your problem. The main reason behind this first a registry error, hardware error and then software error. If you windows registry is corrupted by a virus or any other malicious software then often you windows will crash. The effect starts from slow pc to crashed. The solution for this is download a registry cleaner tool and run it. Or even you can try system restore. Then comes hardware error. This happens when a hardware is bad or over heated. In case of RAM if it is not compatible with motherboard or heating up the system will crash. Try to change the slot or better replace it. Second thing check that your cpu is not getting hot. If yes then it can due to inadequate power supply, improper cpu fan and absence of heatsink. Even if your hard disk to rotating properly or heating up the same problem come out. Third comes the software error. Some software are incompatible with windows. There corrupt system files make windows crash frequently.