Jump to content

Lower Frequency = Lower Timings?


cirro

Recommended Posts

im going to pick up some pc8500 ram and i wonder if i unedclock it by 200mhz at least back down to 800 if i can get some nasty timings or am i just being too hopefull

Share this post


Link to post
Share on other sites

some RAM can do 3-3-3-6-1T @ 400 MHz... you most probably won't do it on a reference 780i board though... so you'll probably get like 4-3-3-8-2T and gain 10 points in 3DM06 lol

 

running 8x450=3600 with ram Sync'd at 450, 4-4-4-12-2T would be better than how you have it now...

Share this post


Link to post
Share on other sites

Memory Testings

 

Evga 780i p04 Ultra x3 550w PSU

Q6600 @ 3.6Ghz 1600QDR 1.5875v

2048 Pc8500 ram Sync'd @ 800 Mhz

Stock RAM speeds 1066Mhz 5-5-5-15 2T

Tests Ran @ Spec. Votage of 2.2v

Secondary timings left on auto, except Forced 7.8us

 

Score Based On Using 3dmark06 CPU test

 

* what i originally thought would be best

** Suprised it Ran

----------------

Original EEP #2 1066Mhz 5-5-5-15 2T

CPU:

 

----------------

 

**5-5-5-15 2T (Stock Timing for it @ 1066mhz)

CPU: 5491

 

5-5-5-15 1T

CPU: 5517

 

*4-4-4-12 2T

CPU: 5521

 

**4-4-4-12 1T

CPU: 5554

 

4-4-4-10 2T

CPU: 5531

(hahaha theres your 10 point increase :P LOL)

 

4-4-4-10 1T (So Far lowest 1T stable)

CPU: 5567

 

4-3-3-10 2T

CPU: 5561

 

4-3-3-9 2T (Beat by 4-3-3-10)

CPU: 5548

 

**4-3-4-10 1t (not fully stable maybe lowest)

CPU:5586

 

4-4-4-9 1T

CPU: 5574.

 

4-4-4-8 1T

CPU: 5604 (surpised)

 

4-3-4-9 1T

CPU: 5566

 

 

Currently Running Prime95 on some of these timing to check stability.

so far 4-4-4-9 1T is running its test

Share this post


Link to post
Share on other sites

hmmm, that's something I wouldn't have expected...

 

what are the values on the memory page in the BIOS? (even though they say Auto, the value used should appear, or alternatively, download and run MemSet to see the values, btw, you can use MemSet to change timings without rebooting ;))

 

also, with ram timings, it's common for a reduction of tRP to be more stable than a reduction of tRCD when both are equal

 

e.g. 4-4-3-10-1T is much more likely to be stable than 4-3-4-10-1T

Share this post


Link to post
Share on other sites

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
×
×
  • Create New...