When OCing, go Loose or Tight (RAM)

C

Carnival Forces

Supreme [H]ardness
Joined
Feb 9, 2003
Messages
4,297
i think i saw this question asked before, but i never remember it getting a satisfactory answer; and, the search is down =\

so, when OCing, is it better to set RAM at the tightest possible settings and then go for broke?

or to set it for the loosest settings, then go for broke, and then tighten up as much as possible?

i was thinking to "go loose" would be better, since you can be pretty sure that your RAM is not the bottleneck (outside of the contingency that it's not good performing RAM or something, just timing wise).
 
I would think to start with loose timings so as to uncomplicate the o/c. Then tighten up when you are sure the o/c is going to work...
 
YOu run tight timings on my memory, I'll give you ten bucks.

I'm only using Pc2700 at 192fsb right now(2.3ghz)
 
I would oc in parts

Remember in Biology you must set up an expirment with least amount of vairibles and the ram is the easest thing the get rid of

Here is cloud's guide to ocing :p

set ram divider to 3:2 and put timings at ram recommended settings
But boost voltage to 2.9V

Lock PCI/AGP bus

Up cpu voltages .1-.2 v's

Oc to highest max stable oc(use prime95 or another program to crunch proc)

Once you get to your max stable you should know your ram speeds. EG: PC3200=ddr400* we will use this ram in examples
good-great ram should o/c at least 20 mhz without hesitation
Using a calc or brain power find out the fastest possilbe ram divider you think your ram can handle
EG:
p4 2.4c with 260mhz fsb=1040 effective and ram would need a 16mhz oc to run 5:4. So assuming its corsair xms pc3200, with 2.9v's that should be perfect. 1:1 in unachievable with this ram and 3:2 is performance loss

Hell with 2.9v you should get that ram to run 1:1 with 16mhz oc at 2-2-2-5

now if your ram oc is 30 mhz but going to 5:4 would make your ram declock, try to boot 1:1 with 2.9v's, becuase it might do it
 
You must log in or register to reply here.
Back
Top