Please help me get my 4400 to FX62 Speeds :-)

squiffy said:
Mines the "Deluxe" version, so will look different. btw I left all of the non cpu related over voltage to "auto" after a reboot it changed them automatically, so I presume it requires a higher voltage if overclocking CPU. I think when I set to "disable" all over voltage when overclocking it didn't post..guess try again, no point over volting if it's not required. Do need increasing vcore on CPU though after 2.5ghz.
Click to expand...

So your card isn't the A8R-MVP as in your sig, it is the A8R32-MVP, kk all is clear now
 
squiffy said:
oops :o damn long model names. sig now changed.
Click to expand...

Yeah i know, reckon they compete to have the longest model name? Question about your PSU, how do you find the liberty module sytem, i find it messy in my case doesn't seem to be enough room for the cabling, i have a thermaltake soprano case but the cabling is a bit of a mess. Any tips for tidying it up? Would have PM u but it wont let me
 
hmm I don't have that case, but the PSU cables should be long enough. Have you pushed aside and tie wrapped the ATX, IDE power, IDE data cables together? Try and avoid cables bundling over mobo and other components, don't use flat IDE cables. 4 pin mobo power should trail just above mobo. Wrap the mobo power cable to one side. I've linked pics in another thread http://forums.overclockers.co.uk/showthread.php?t=17587930 having a bigger case helps can be a bit more messy and still have plenty of airflow :p
 
I think for ati cards PCI Freq is best set at 105, the similarity in systems is uncanny squiffy lol... but for some reason or other my cpu just doesnt like 11*245. May try the folding for a bit and attempt again after that.

Have you also tried upping the mem divider? Surely you can oc the memory a little. having it at 183 would be nice ;)
 
Last edited:
Tenzen said:
I sincerely doubt any 4400 will beat a fx60 at super pi. The fx simply have better quality chips.
Click to expand...
They are exactly the same chip. A 4400+ at 2.6 with an identical board and memory will give exactly the same scores as an fx 60.
The only difference between an fx 60 and a 4400+ is the fx 60 can do 2.6 at stock 1.4v but a 4400+ will require between 1.45 and and 1.55 and therefore produce more heat and have a slightly shorter lifetime.

Someone asked if watercooling was the reason they don't require as many volts as someone else to get a certain speed. It is not due to watercooling, its due to the characteristic of the particular chip. Watercooling will only give you higher clocks if your limited by your temperatures, and most people seem to be limited by the volts they are prepared to put through the chip.

Someone else said they were changing the affinity of folding to run it on each core. You can run multiple instances of folding so your running one on each core. Read the guide in the dc forum to find out how.

The pci-e frequency should be locked at 100.

Giving the memory more volts only helps if your overclocking the memory and its failing memtest at a certain speed.

Not all X2/Opterons can hit 3ghz. Many can only reach 2.8 or 2.6, there is no guarantee its pure luck which stepping you get.
 
Last edited:
glen8 said:
Hi guys,

I am very new to overclocking but would like to bump my CPU up a bit to save time encoding video.

I have an AMD X2 4400+, Ninja HSF, Dfi CFX3200 Mobo, 2gb (2x1gb) Corsair XMS3500 Pro XL

I had a go playing with the bios lastnight but whatever I tried the pc refused to boot :mad:

It would be ideal if I could match FX speeds but not sure if i'm asking a bit much?

I have searched numerous forums including this one and also read some guides but nothing seems to match the settings I have in the bios. Most guides are too old now to be of any use

Some pointers would be brilliant please :-)
Click to expand...


ill be able to help you out with all this

my msn is [email protected]

add me and ill help you out with this
 
Ahh I am glad I'm not the only one who can't seem to push their 4400+ overclock. My 4400+ is struggling to get 11x235 stable! Pumped up the voltage to 1.4 to see if it passes 2 hours on Prime95.

What's the absolute highest voltage you can push through these chips? I have set myself a limit of 1.5v... this safe for air cooling? My PWM IC chips seem to be getting a little toasty at about 50C.

SiriusB
 
SiriusB said:
Ahh I am glad I'm not the only one who can't seem to push their 4400+ overclock. My 4400+ is struggling to get 11x235 stable! Pumped up the voltage to 1.4 to see if it passes 2 hours on Prime95.

What's the absolute highest voltage you can push through these chips? I have set myself a limit of 1.5v... this safe for air cooling? My PWM IC chips seem to be getting a little toasty at about 50C.

SiriusB
Click to expand...

pm me your specs
 
SiriusB said:
Ahh I am glad I'm not the only one who can't seem to push their 4400+ overclock. My 4400+ is struggling to get 11x235 stable! Pumped up the voltage to 1.4 to see if it passes 2 hours on Prime95.

What's the absolute highest voltage you can push through these chips? I have set myself a limit of 1.5v... this safe for air cooling? My PWM IC chips seem to be getting a little toasty at about 50C.

SiriusB
Click to expand...

Most seem to agree that 1.6 is the absolute highest anyone should go on air. My DC opty went this high for 12hr dual prime (100% load) and suffered no problems. What you need to bear in mind is that the higher the voltage the lower the reported lifespan will be. So running 1.6vcore through your CPU 24/7 will shorten its lifespan, as will any voltage increase from spec, how much nobody seems to know exactly but I would be confident in saying that it will last until you come to upgrade again. Voltage increase will cause electron migration which in turn degrades the cpu's performance as the electrons form new paths afaik.

With voltage increase comes temperature increase and this is what should concern primarily. I wouldn't want my CPU running over 55c for any length of time and for opterons (maybe x2 too), keeping the temps below 45c seem to yield better results.

Make sure you have some active cooling over the mosfets as this will reduce there temp dramatically. PWMIC temp isn't too much to worry about as these are rated to apx 125c although I would start to worry once they go past 60c. My stability testing was effected by pwmic temp (they went to 82c!) before I actively cooled them. Once they reach a certain temp I reckon the voltage they regulate becomes unstable which in turn destabilises CPU performance, it did in my case anyway.

If CPU temps are not a problem (<60c load) then increase vcore a little. Make sure your HT is 3x and maybe consider putting your memory on a divider/slacken timings, its difficult to tell without looking at the settings you have already.

Hope this helps.
 
Well last night I wacked up the voltage on the CPU to 1.5v and I ran Prime95 for about 6-7 hours with no errors on 240x11.

My PWM IC hovers at around 59/60C so not too bad. I will probably get a nice quiet 80mm fan for the side panel on my CM Stacker 810 - it is almost exactly above the PWM IC area so should cool it a treat. Some tidying up of the cables should help too :D

Load temps for the CPU didn't go higher than 52C and my room was quite cool. Reluctant to try for a higher OC because of the warm weather.

At the moment I am OCing my memory to see what it can do. Memory is going through a longer stability test currently at 220MHz with unchanged timings of 2.5-3-2-5. Voltage is at 2.7v. Memory wouldn't budge past this even with 2.8v and I'd rather not slacken the timings if I can help it.

My spec:

AMD Athlon 64 X2 4400+ - Air cooled with AC Freezer 64 Pro
OCZ 2x 1GB Platinum PC3200 Dual Channel Kit
DFI LanParty UT Ultra-D nF4
CM Stacker 810 - 1 x 120mm Intake, 2 x 120mm Exhaust, 1 x 40mm Blowhole

SiriusB
 
Okay I either have the worst 4400+ EVER or I have missed something!

It passed over 6 hours of Prime95 at 240x11 with the RAM on a low divider to take it out the picture.

Today I have been testing my RAM and the best I have got out of it is 210 with the current timings. Not all that great to be honest.

So, deciding not to bother OCing the RAM and keeping it as close to 200MHz as possible I go ahead and see what happens when I run my system at 240x11 with the RAM on a 9/10 divider...

No boot.

So I try lowering to 235 and again.. no boot...

So I let the BIOS set the divider by telling it to automatically limit my RAM to SPD... so I assume its holding it at 200MHz... go to boot.... and nada.

In fact, at the moment I am running at 210x11 and to be quite honest its pathetic! I got this 100MHz OC by accident weeks ago :(

Going to try 220x11 but 2400 OC is a bit dissapointing considering I got 2600 when RAM was underclocked. *glares at IMC*

Anyone shed any light at all? I have been following the 939 OC guide on Short Media. http://www.short-media.com/review.php?r=300

EDIT: Gah, what a fool I am! The Auto RAM limit from SPD is basically another way of saying 1/01 it seems on the DFI boards :p

Currently stress testing 2500MHz [250x10]. Approx 2 hours so far with no errors. If this lasts until I get up I am gonna go for 2600MHz.

SiriusB
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom