factors limiting my overclock.

footman

footman

footman

Soldato
Joined
21 Oct 2002
Posts
4,294
Ive swapped the opty 146 for a 165 CCBBE 0610DPMW), everything else is the same.
After several days messing about Ive got it (on air) to 9x295=2.655GHz.
The memory is running at 241.4 (166MHz) 3 3 3 7 1T
voltage is 1.5v temp is 42 degrees. raising the voltage is still no go.
HT linkspeed 1GHz (ie x5)
It wont load Windows above 295 and at 295 Prime 95 runs merrily when not testing much RAM at a temp of 47.
When testing RAM Prim495 falls over within seconds.

So, does this mean that the mobo max's out at 295 or the RAM is pushed too far? (tomorrow Im getting 2g of Corsair 4400)

If I put the 165 on water (all the kit arrived today..scared!) will the mobo limit anything above 295?

thanks for looking
 
if you lower your processor multiplier and set divider on memory, you should be able to get a decent indication of exactly what sort of clock your motherboard is capable of, remove the processor as the limiting factor ;) i did that with 3200+ venice and got to 325Mhz on my MSI platinum before ram became a limiting factor i feel :eek:

Edit: lol forgot to try answer your question, if thats the limit of your motherboard then watercooling proberbly won't help, unless your watercooling voltage regulators and northbridge, might get bit more out of it that way
 
I lowered the cpu multi to 8 and got up to 8x298=2384 with the memory at 238.4. If I lower the memory speed to 133 it wont boot.
I then tried to restart but the pc wont even post so I guess its reset the cmos time!
So as the RAM was running lower than when at the stable overclock can I assume that the RAM is not the limiting factor here, but the mobo max's out at 297?
Would this mobo,DFI LanParty UT CFX3200-DR RD580, known to go somewhat higher than 300 be the answer?
 
Some RAM doesn't like certian dividers.. mine doesn't like 266 (133) at all.. i run it at 333 (166) and accept putting 3.2v through it to keep it happy...
 
i doubt it, i think you've just reached the limit of your CPUs actual potential, in which case getting a better clocking board wouldn't help at all
 
Gashman said:
i doubt it, i think you've just reached the limit of your CPUs actual potential, in which case getting a better clocking board wouldn't help at all
Click to expand...
I know I should just go away and shut up but the "math doesnt add up" (apologies to Saving Ryans Privates)

The CPU speed when the PC failed to post was 2.376 (8x297) and as Id got it Prime stable at 2.655 when using the stock multi of x9 then it couldnt have been the cpu that was maxxed out.
 
iraiguana said:
did u lower the htt link speed to below eight ? as around the 300mhz barrier it needs to be 600 (3x). and u could lower it further still.
Click to expand...
ahh....If I recall it was set at 1GHz (x5). I'll give it a try later, cheers.
 
iraiguana said:
did u lower the htt link speed to below eight ? as around the 300mhz barrier it needs to be 600 (3x). and u could lower it further still.
Click to expand...
Agreed you need to keep the htt below 1Ghz
 
jmhduck said:
Agreed you need to keep the htt below 1Ghz
Click to expand...
I left evrything on auto to see what would happen and its now set to 400MHz. Its prime stable (24 hrs) at 295 but everything goes weird at 296!

Im getting the DFI crossfire board to see if that can take the HTT higher.
 
footman said:
I left evrything on auto to see what would happen and its now set to 400MHz. Its prime stable (24 hrs) at 295 but everything goes weird at 296!

Im getting the DFI crossfire board to see if that can take the HTT higher.
Click to expand...
Yep DFI should be better, was about to suggest sli-d but not with those 2 beastie gfx's :D
 
You must log in or register to reply here.
Back
Top Bottom