Unstable system

[morlud]

Hi there

 

I have some insability problems with my dfi lp ut sli-dr expert mb (i think.... or maybe it's my ram)

I regulary have bsod with "irql_not_less_or_equal" or something else sometimes and .... reboot

When i test my g.skill ram with memtest, there are some errors.

When i test one at a time, one have errors and the other not, but when i boot my computer with the good one, i still have bsod.

 

The most curious is when i run memtest with different ram, there is no error and my computer is stable.

And when i run memtest with my g.skill ram on a different computer, there is no error too.

 

I have my g.skill ram since 2 weeks and it worked perfectly on my last mb (an asus p5gdc-v deluxe with a p4) since a week ago when i change my asus mb for the dfi expert mb.

I put my g.skill ram on the dfi mb and problems started

 

I read on this forum that there is problems with some tccd and the dfi expert, is that right ? it could be my problem ?

 

For information, i don't o/c.

Thx for any help, and sorry if my english isn't perfect (i'm french )

[boatasiaus]

How many volts are you giving the RAM? Spec allows for 2.7-2.9. Maybe try increasing DRAM Voltage and rerunning memtest

[UncleDavid218]

If you're not giving them enough voltage you're not going to be stable. Try 2.8v.

[DarkjeThe2nd]

I'm having some problems with my g.skill tccd's too, but I seem to be making some progress in getting them steady.

 

Memtest passes at a lot of settings, but mprime from tMods excellent boot cd often fails within a minute if the settings are not okay.

 

I'm quite sure you problem is also setting related and your current settings will fail mprime.

If so try to loosen all timings as far as you can and see if that will let you memtest and mprime for at least an hour or so... after that you can almost certainly reach windows, to do further stability testing there. If that all works out you can try tweaking the timings back to normal values

 

I've read a couple of times that drive strength/data drive strength are the key to the expert, for me ds=7/dds=2 seems to work.

[St0ffel]

I'm having some problems with my g.skill tccd's too, but I seem to be making some progress in getting them steady.

 

Memtest passes at a lot of settings, but mprime from tMods excellent boot cd often fails within a minute if the settings are not okay.

 

I'm quite sure you problem is also setting related and your current settings will fail mprime.

If so try to loosen all timings as far as you can and see if that will let you memtest and mprime for at least an hour or so... after that you can almost certainly reach windows, to do further stability testing there. If that all works out you can try tweaking the timings back to normal values

 

I've read a couple of times that drive strength/data drive strength are the key to the expert, for me ds=7/dds=2 seems to work.

 

Try the OCZ Tony Beta Bios, solved the problems for me. Had the same things, Memtest and Goldmemory stable, but Prime failed. With the new bios im stable (8h blend) 265 - 1/1 - 2.5-4-3-6-1t - 2.9v.

[morlud]

I didn't say it in my first post but i already tried 2 days ago to increase the voltage from 2.66 (default setting) to 2.71, but it's the same.

Some hours ago, i tried the opposite and i have set the voltage of the ram to 2.4, and run memtest.

Curiously, with my 2 rams together, 20 pass of the test 7 (i always have errors on it and only on it) and no error.

then 5 pass of all tests and no error too.

I'm running my computer in dual channel and i hope it will be stable for next days :angel:

I'll holla at y'all for the results if someone is interested

[DarkjeThe2nd]

Try the OCZ Tony Beta Bios, solved the problems for me. Had the same things, Memtest and Goldmemory stable, but Prime failed. With the new bios im stable (8h blend) 265 - 1/1 - 2.5-4-3-6-1t - 2.9v.

 

thanks, that's working fine indeed thnx St0ffel...

 

now primestable for 3h @ 285x10, 1:1, c3448,1t, 2.91v...

will post complete prime run etc. in my topic about this later on, but it seems now it was a bios problem ...