Intermittent blue screens and elusive memory faults

[simonstandish]

I have a one-month old system running Vista Home Premium, based on an Intel Core Duo E6750 processor, a Gigabyte P35-DS3R motherboard, and 4 Gb of Corsair PC6400 memory. In the last couple of weeks I have started to experience the occasional blue screen crash when doing something demanding such as video encoding. If I run the Vista memory diagnostic tool immediately after such a crash, it usually tells me that there is a memory fault. I then go through the process of trying combinations of memory sticks in combinantions of slots without being able to reproduce an error. The system then stays stable for a few days until the problem rears its head again and I repeat the same futile attempts at trying to identify the rogue stick or slot.

It's now driving me mad! I'd love to know if there is a better way of homing in on the problem, or if I am just going to have to accept that I need to live with this intermittent problem until something more conclusive happens.

Simon Standish

 

[cb62fcni]

I have a one-month old system running Vista Home Premium, based on an Intel Core Duo E6750 processor, a Gigabyte P35-DS3R motherboard, and 4 Gb of Corsair PC6400 memory. In the last couple of weeks I have started to experience the occasional blue screen crash when doing something demanding such as video encoding. If I run the Vista memory diagnostic tool immediately after such a crash, it usually tells me that there is a memory fault. I then go through the process of trying combinations of memory sticks in combinantions of slots without being able to reproduce an error. The system then stays stable for a few days until the problem rears its head again and I repeat the same futile attempts at trying to identify the rogue stick or slot.

It's now driving me mad! I'd love to know if there is a better way of homing in on the problem, or if I am just going to have to accept that I need to live with this intermittent problem until something more conclusive happens.

Simon Standish

Do you have 4 X 1GB sticks, or 2 X 2 GB sticks? What is the recommended voltage for your RAM. If you're unsure, check the specs on Newegg. Go into bios and ensure you have enough voltage provided. Make sure timings are set properly as well. If this stuff is good, the problem could be heat or power supply related. Is there anything that ties the crashes together? I.e., does it tend to happen at night when you've turned the heat in your home up? Download speedfan and check your temps. Use a program such as orthos or prime95 to stress your rig and watch your temps as it does. Make sure you have proper airflow over your RAM modules. Check the temp of your northbridge, there have been many instances where motherboards have poor quality control over the thermal interface between the northbridge heatsink and the actual chip. In operation, this heat sink should be rather warm. If it is cool, heat from the chip may not being transferred well. If it is blazing hot, the airflow is insufficient. An overheating NB may present as a memory error, as it is where your memory controller resides. If there are not heat issues, the power supply could be suspect. What brand/model are you using? Do you have a spare that you could swap in?

Anyway, this is a ton of info, but without more information about your rig and problem I felt I should cover the gamut of likely issues. Hope it helps, let us know what you find.

 

[simonstandish]

Many thanks for the extensive advice. I'm pretty sure it isn't a temperature problem as, even during video enncoding, the processor neverises above the low forties, and the motherboard never exceeds the high forties. I will look into the other steps you recommend though.

Simon Standish

 

[simonstandish]

Do you have 4 X 1GB sticks, or 2 X 2 GB sticks? What is the recommended voltage for your RAM. If you're unsure, check the specs on Newegg. Go into bios and ensure you have enough voltage provided. Make sure timings are set properly as well.

I have 4 x 1Gb sticks of Corsair XMS2 6400 RAM. The timings on the label are 5-5-5-12. I checked the BIOS where the default settings were 5-5-5-18, so I changed them in line with the RAM. So far no more blue screens, so I'm hoping this has fixed it.

Simon Standish

 

[Xtreeme]

lol odd cause your just tightened the timings (faster).

ITs not the cas or otehr typical timings I bet. Its the RC or RFC maybe even RRD or WR RTP etc.

Is command rate on 2 T that has to be set 2T for stablilty. Run everest click motherboard then spd. Right down mem timings and set those. Then run memtest86 if it goes 2 passes without error you should be ok.

I had same issue on Tforce550 and it was the other timings.