heres something i came across in the mushkin forum. posted by lorven (site admin):
====start====
...we never recommend running the Active to Precharge Delay at 5T. Some of our testing has indicated that running this setting at 5T for extended periods of time may eventually lead to hard drive corruption, regardless of the memory type, so we always recommend running it at 6T. We feel that the miniscule difference in performance isn't worth the risk of hard drive problems.
====end====
thought that might be of interest to some of us here running at 5-2-2-2 timings.
__________________
NEW PUTER!!
SL-75FRN2-L, XP1700 dlt3c juihb, slk800 - smart fan2,
mushkin pc3200 222, Ti4800SE 128mb
++++++++++++++++++++++++++
AMD K6-2+ 450 underclocked to 400mHz (66 x 6)
DFI P5BV3 baby AT
Crucial 2x128M PC133 @ 66mHz (Whew!!)
Inno3d GeForce 2 MX400 64M Twinview
Creative Live 5.1 oem
Creative DXR3 dvd card (full screen dvd.... yeay!!)
You know what, I've read that somewhere else, and although I haven't had any problems running 5-2-2-2, I'm going to bump it up one just in case. Thanks for the tip.
I loosen my ram timings to help ensure stability. Anyway, in my systems I do not notice any performance gain except slight gains when benching. The performance gain is not enough to notice in gaming (at least in the games I have tried).
Originally posted by screaminglemon heres something i came across in the mushkin forum. posted by lorven (site admin):
====start====
...we never recommend running the Active to Precharge Delay at 5T. Some of our testing has indicated that running this setting at 5T for extended periods of time may eventually lead to hard drive corruption, regardless of the memory type, so we always recommend running it at 6T. We feel that the miniscule difference in performance isn't worth the risk of hard drive problems.
====end====
thought that might be of interest to some of us here running at 5-2-2-2 timings.
Nice info to know however I consider that I have been running 5,2,2,2 for a long period of time both on an 8RDA+ and recently on my 75FRN2-L and I have not had a problem. Using OCZ PC3500 EL DDR
you know, if you raise it higher like more than 6, the mem performance will be little bit better. I could do 11-2-2-2 but I set it at 9-2-2-2. it depends on memory too of course
so it looks like hdd corruption would only happen if the machine is shut down and the corrupted memory content is written back to the hard disk drive before the content was fixed. sounds to me like it would only happen once in a blue moon like during a power outage or something.
but mushkin is also claiming that 6-2-2-2 would be faster than 5-2-2-2. has anyone noticed this in their benchies?