• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

another new to overclocking

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.
ok i got i installed Coolermaster 212 Evo and also a new 120cm front fan... most temps in HWMonitor on idle seem lower but i damaged my graphics card in the process.

clearly the problems are comming from my ATI Radeon 4850.

unfortunatly i got a bit carried away when i dissassembled the PC and sprayed the card/fan with an air cannister and im pretty sure ive damaged the fan when i focused on it (it made a noise and some blades were misaligned/stuck i touched them to try let them go to natural position).

any tips for repair or do i get a new video card?
i done some google'ing and apparently upto 100C's is when it gets dangerous and it can withstand even 120C before it goes off automatically, i just hope it doesnt damage my board. i might get a new card instead of try any repair.

ANYWAY, for overclockers purprose.. i will need post another screenshot because this 1 i posted has BIOS with C1E on and also AMD QuietandCool on.
 

Attachments

  • nowwtf.jpg
    nowwtf.jpg
    125.3 KB · Views: 116
Last edited:
Whenever, I repeat, whenever you clean pc components that are cooled with a fan with compressed air your should always, I repeat, always prevent the fan from turning. The compressed air will make the fans turn at super fast speed and damage the fan bearings. PC Maintenance 101.
 
im a bit of a rookie i know i'd heard that before but i got carried away.. was quite fun taking my pc apart and i there are so many precautions e.g, static wristbands that can take the joy out of this learning process.. maybe it was just the evil side of me wanting to get rid of that old 4850!

ANYHOW, i didnt mean to sidetrack my own post so much but the good news is i ordered a new MSI Raedon 6850 from www.overclockers.co.uk.. i wont get it till Friday (2 days of no gaming very hard for me!!) but once i do get it i will be back here to resume were i started, lol! :)

btw i am actually very happy with my new cooler and red LED front fan i installed today myself, quite chuffed there has been a much improved air flow :)
 
well, theres good news and bad news lol... a picture tells a 1000 words, this is on idle...
 

Attachments

  • goodnewsorbadnews.jpg
    goodnewsorbadnews.jpg
    28.8 KB · Views: 111
You could almost fry an egg on that GPU!

lol yea as soon as i seen that i turned the PC off and took the card out, went without a video card for 2 days..

today my 6850 arrived and i have it installed now, will use the system to ensure stability and wait for a temperature increase (cold climate atm) before i perform the blend test with p95 for 20mins.
 
It's a good thing you bought a new CPU cooler, the stock cooler for the Athlon II x4s is absolute trash, one of the most pitiful coolers I've ever seen hahaha. That CM cooler should put you in a good place to start overclocking.

Just to warn you, don't aim too high, best I was able to get out of a similar cpu with a $60/40gbp cooler was 3.6GHz stable. While every setup is different, 3.8GHz, if it's possible on your setup, will take a LOT of careful tweaking. Either way though, Trents is the guy to talk to with this stuff, he will get you up and running with a solid overclock in no time.
 
Yeah, 3.6 is about the ceiling. I have a AII 630 X4 and that's all I could get out of it with top of the line air and even then it took 1.485 vcore. I ran it everyday at 3.5 on 1.412 vcore. That last .1 ghz took a lot more vcore.
 
please do not only disable a software oc genie please remove the whole program, i had one i thought i had disabled but it still ran in the background and drove me fruit bats for a week or more.
it was easytune5 that came with my biostar board.
 
Yeah, 3.6 is about the ceiling. I have a AII 630 X4 and that's all I could get out of it with top of the line air and even then it took 1.485 vcore. I ran it everyday at 3.5 on 1.412 vcore. That last .1 ghz took a lot more vcore.

cool cos me having 2.9 atm i would be ok with 3.2 and chuffed at 3.4, for me stability is the most important factor.. 3.5 would be beast if no issues!
my PC is on at home at least 12hours a day every day there have been times its 24hrs left on overnight, probably its 16hrs/day on average.

anyway, back to where i started with some new hardware i performed a P95 stress test for over 20mins in a hot environment that has me sweating wearing only boxer shorts!

This is the HW Monitor screenshot at over 20mins of P95 testing:
(i compared to my previous results before the cooler and new front fan were installed its quite plain to see, i also upgraded video card from 4850 to 6850)
 

Attachments

  • newp95.jpg
    newp95.jpg
    120.7 KB · Views: 101
wonderful case, i have my 955 in a cooerlermaster 335 elite nvidia edition.
what was causing the other gpu to overheat so badly?
 
wonderful case, i have my 955 in a cooerlermaster 335 elite nvidia edition.
what was causing the other gpu to overheat so badly?

yea man case is awesome, looks quite sleek too..

the fan on the gpu was damaged during maintenance so thats what messed it up.
 
Yeah, 3.6 is about the ceiling. I have a AII 630 X4 and that's all I could get out of it with top of the line air and even then it took 1.485 vcore. I ran it everyday at 3.5 on 1.412 vcore. That last .1 ghz took a lot more vcore.

plz can u help me where should i start to get this going?

i dont want to just play with it cos the guys i play mw3 with online say i could seriously damage my pc.
 
There we go, those temps are looking good. While it is true OC'ing a computer can lead to damage, as long as you do it carefully, follow our guidance, and have a respectable PSU(which you do), the odds of damage are pretty low.

trents is the better guy to walk you through this but I can try and get you started.

First off, you understand how to get into and navigate your bios right? I'm not trying to be condescending or anything if you do, just making sure. What you are essentially going to be doing is changing your FSB(may also be called CPU Frequency or HT reference clock) in the bios, which is what increases overall speed. So what you need to do is increase it by intervals of 5, test it, then repeat until you crash or fail a 20 minute prime95 test.

BUT WAIT THERE'S MORE!

As the FSB increases, it also increases your HT link and RAM speeds, so these need to be adjusted so they don't leave stable ranges as you increase fsb.

My suggestion is for now, make sure windows is running high performance power settings, all cool`n`quiet or c1e type features are disabled in your bios, then bump the fsb up by 10, prime95 for 20 mins, and come back to us with screens of the results in CPU-z, HWMonitor, and Prime95.

As a side note, it shouldn't be an issue with the new cooler, but if your core temps go above 65C turn off prime95, and bring us the results of that abbreviated test.
 
ok thanks man.. yea i hope if i do it right then it should be ok, is it worst case just my cpu that can get damaged or can it stretch out onto my mobo as well?

i do understand how to get into and navigate bios, im not expert at it all by any means tho i am now slightly familar with the terms u used.

i dont know much about RAM speeds and how i should adjust them, i will have a look at them on BIOS and also the HT link. i will disable cool n quiet and c1e again.

anyway, i will do as u said for now and run the p95 test for 20mins with fsb up by 10 and post the screenshots a bit laters. cheers.
 
Ok so I up'd the FSB by 10 in BIOS and also disabled both C1E and Cool&Quiet.

Here is how my CPU-Z is looking:
ZQCcP.png
Ysslk.png
0iJKn.png
YJrf8.png

I then ran Prime 95 for 20mins while I was using my PC for other programs, here is the screenshot after just over 20mins for HW Monitor also.

nFgmc.png
nBaQs.png

I got a hardware error in 1 test so I stopped using other programs and left PC idle for another 20 mins while I ran the Prime 95 stress test again, here are the screenshots.

5Zmm3.png
5wDuB.png

What do they results mean and where can I go from here?
 
Ok that's looking good, though your multiplier should be on 15, not 14.5.

Also, don't use your computer for anything besides just prime95 when you are running it. The program is specifically designed to max your processors all by itself, so trying to run it and other things is just begging for errors or a BSOD.

Now, from HWMonitor we can see your temps are maxing at under 55C so that's good, your RAM seems ok since it's DDR3 1600 I believe. Your HT Link however is a tad high, so we will need to bring it down some before we continue.

So what you wanna do now is find the two settings that control RAM and HT Link, once you do, turn them both down one setting. Then, turn your FSB up by intervals of 5 at a time, testing for 20 minutes after each jump. What you wanna do is keep going until you get an error in prime95, your core temps go over 55C, or you get a BSOD. When that happens, go down a setting to your last successful setting, run like a 10 to 15 prime and HWMonitor, and post the results here again.
 
You have about 5-8c worth of temp headroom to add more CPU voltage if needed. But like WoddenK said, first put your multiplier back to the stock number first and then do a Prime95 blend test for 20 minutes to check for stability at the higher speed. Don't add more vcore until you encounter instability. Also, go ahead now and increase you CPUNB voltage to 1.225 . Watch your memory speed. It's creeping up there. The integrated memory controller for that CPU is rated at 1333 but will handle about 1500 mhz. At some point you may need to lower the memory divider from the 1333 speed to the 1066 speed to make more headroom.
 
Back