Tight Timings: Risks and Benefits

J Ross

Distinguished
Jul 13, 2011
5
0
18,510
Yo,

I recently assembled my first computer and I have now started dabbling with overclocking. The memory I am using is PNY XLR8 2 x 2GB DDR3 . This memory is stock rated at 8-8-8-24-2T DDR3 1600 @ 1.65v. However, it defaults to DDR3 1333 and when I change it in BIOS to 1600 I have stability issues; prime95 cannot complete even one test.

As I now understand it, the Phenom II x4 955 BE that I am using is simply not designed to go above DDR3 1333 (is this correct?) so I am not too concerned with getting up to DDR3 1600. That being said, since I am now underclocking it to DDR3 1333, I figured I could tighten the timings.

I currently have it set to 7-7-7-18-1T @ stock 1.65v. It seems stable, I ran prime95 for about an hour and a half with no errors and have not noticed any problems in day-to-day use, but I am wondering if these timings are efficient and/or deleterious to the system lifespan. I understand that raising the voltage to RAM can cause damage, but I don't know how changing timings affects it.

Basically, I'm just wondering if these are good timings, and if not, what would be better?

Thanks,

J

P.S. My CPU is a Phenom II x4 955 BE overclocked to 3.6 GHz @ stock 1.35v and my motherboard is an Asrock 870 Extreme3
P.P.S. Here is the CPUID for my RAM
RAM.png
 

J Ross

Distinguished
Jul 13, 2011
5
0
18,510
I'm not sure I understand why I'd want to do that. Is it that my voltage is too high or that some timings work better together? I'm guessing the former, so I decided to try just lowering the voltage as far as I could with the 7-7-718-1T timings. I got down to 1.575v stable. The reason I didn't do this before was because the RAM is marketed as 1.65v RAM, so I though the stock voltage would be best. What do you think?

J
 


Just a note to you, If you're trying to weed out memory issues using Prime95, make sure you're running the Blend Test.