Overclock.net › Forums › Components › Memory › Exceleram 4GB (2x2GB) 1600MHz DDR3 E30110A, a little problem.
New Posts  All Forums:Forum Nav:

Exceleram 4GB (2x2GB) 1600MHz DDR3 E30110A, a little problem.

post #1 of 6
Thread Starter 
Got this memory long time ago, it's working on 1.5v 9-9-9-24 timings, 1333MHz. Since i started doing a little bit of overclocking to my 955BE i thought why not give it a try and play a little with my memory.

Now, this memory is labeled as CL7 and if i remember correctly on the box it said 7-10-10-24 timings but whenever i set this timings OS refuse to boot. Also i set it to 1600MHz, 7-10-10-24 timings and 1.660v because in my bios i cannot set it to 1.65 only the 1.64 or 1.66. These timings and voltage was suggested by CPU-Z timings table. The problem is i cannot find this memory on my manufacturer's website. It's like it doesn't exist.

Here is the link to the products page: Exceleram

Also i can only find this exact memory kit on my local website ITSvet and in the comments there is one guy claiming he is running this memory with 7 10 10 24 32 1 timings 1600MHz on the exact same motherboard i have.

I am really confused.
post #2 of 6
Try 7-9-9-30 @ 1.66v. Without knowing the specified latencies you'll have to try different combos to see what works for you. It doesn't matter if you use 1.64v or 1.66v as the 1.65v is just a nominal voltage used for OC'ing RAM. You might be able to e-mail your RAM maker for the proper specs?

The links below may help. Be advised that independent testing by three reputable PC industry sources has shown that on AMD and Intel desktop PCs there is no tangible real world performance gains with increased frequency or lowered latencies once you get to ~1333 MHz. RAM frequency because the RAM isn't a boottleneck. In addition as RAM frequency increases one clock cycle becomes less in real time so latencies tend to cancel out.

Llano APUs are the exception as the GPU gets to use the higher RAM frequency for improved performance.


Edited by AMD4ME - 3/15/12 at 12:36pm
post #3 of 6
Thread Starter 
Thank you very much, will try and play a little more with it but it's good to know that if i don't succeed i wont be missing too much.
post #4 of 6
^^^ Glad to help. thumb.gif
post #5 of 6

Watch out for the secondary timings.  Try 2T over 1T.  Make sure the tRC is est to 32.  Watch out for the tRFC - keeping that value low with higher RAM speeds can cause instability sometimes.  Raise it if you find instability - that raise is not going to nearly make as much of a difference as changing the main timings.


Only one of the sources in the above link even bothers to mention an AMD system (it is immoral to be using Intel memory bandwidth results to be comparing to AMD bandwidth results because of differing patterns in IMC design) and the results may be outdated as they make use of older CPU revisions and do not take into account a CPU-NB or CPU overclock, both of which may increase throughput into the memory and amplify the difference, which may or may not be tangible*

* "tangible" is not definable because the perceptions of the end-user may be different depending on what he or she does with this computer.  However I stress that there are differences, even if marginal, that can be noticeable - many of them are in the due processes in things such as video editing and cannot be measured with benchmarks.

Edited by xd_1771 - 3/15/12 at 1:38pm
post #6 of 6
Some systems slow considerably with 2T so 1T is preferred if possible.

A tangible change in system performance in real applications vs. synthetic benches = perceptible, real, significant system gains. If you can't see it or feel it, then it really doesn't matter though some folks would argue just for the sake of arguing. That is why all three reputable independent PC industry sources confirmed that there were "no significant" or "insignificant" (their words), real world application performance gains from increasing the RAM frequency or reducing the latencies above ~1333 MHz.

People are free to run their own tests and see what works for them. The results from Tom' Hardware, AnandTech and X-bit Labs mirrors my testing and that of many other people with both AMD and Intel systems. This is no surprise because DDR3 RAM @ ~1333 MHz. is not a bottleneck for current desktop PC systems.
Edited by AMD4ME - 3/15/12 at 7:53pm
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Memory
Overclock.net › Forums › Components › Memory › Exceleram 4GB (2x2GB) 1600MHz DDR3 E30110A, a little problem.