overclocking an fx-8320 on asrock 970 pro3 r2.0

Page 3 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.

Osoclocker69

Honorable
Jan 12, 2014
138
0
10,710
hello fellow tom's hardware people.
i was wondering if it will be a good idea to overclock my amd fx-8320? will it void my warranty if i do overclock? here are some of my specs:
> mobo: asrock 970 pro3 r2.0
> cpu: amd fx-8320
> cooler: deepcool gammaxx 300
> gpu: msi twinfrozr iii 7950 3gb
> psu: corsair cx 600w 80+

if you guys recommend or say it would be ok to overclock my cpu at least by 5% and what cpu fan speeds should i use? and what temps would i be getting or be aiming to get?
 
It's completely safe to overclock your 8320 to 4Ghz. Overclocking it to 4Ghz practically makes this cpu 8350 which is supported by this motherboard. I have the same mb and I'm running my 8320 on 4Ghz. It runs on 4.2Ghz too but I'm still using the stock cooler and it gets very hot so if you want to do more than 4ghz, let's say 4.2, I suggest getting a new cpu cooler. Also I wouldn't recommend going above 4.2ghz on this mobo.
 
Update: I stand corrected by myself after a bit of research. Apparently, some of these cooling fans DO push, rather than pull. Having been in the automotive industry for 25 years and working on every kind of cooling system out there including reverse flow and air cooled, I think it's a poor decision to design this way. If pushing air through heat fins was efficient every automotive manufacturer would put the fans in front of the radiator rather than behind the radiator. These systems are no different so I don't understand their reasoning for this, but some of them are in fact that way so my apologies to anybody I may have contradicted.
 


So you are running an FX-8320 at 4.0Ghz, on an UN-MODDED ASRock Pro3 970 R2.0 M/B and you have no issues at all? Man I should then try overclocking mine also, but slowly.

BTW Guys I'm Osoclocker, the creater of this thread, i forgot my password so I made this new account. Add me on origin if you want to play some Bf4 with me :) Mr_Venbeer

EDIT: Sorry for the late reply 🙁
 
Welcome back. Yeah, although I"m using a different board then you guys I've currently got mine at 4.2 stable WITHOUT touching voltage at all, multiplier only, Prime95 16hr small fft and 16hr blend without issue. I am using an ASUS Sabertooth though but you might want to try your oc with just the multiplier if you have fears of the boards capability.
 


What voltage are you running? And what do you mean by try our oc with just the multiplier, what's the difference?
 
Stock voltage, multiplier only on the overclock. In other words the voltage is untouched and the multiplier is at 21 giving me a 4213.3mhz overclock, up from the 3500mhz stock clock. Since there is no voltage increase the amount of additional heat created is drastically reduced but don't expect to be able to go past 4.2 without increasing voltage or it will likely become unstable, as only a few rare chips I've read about have had any success with that.(And of course is unverified.) If you click on my cpu-z banner it will pull up my verified settings for you to take a look at.
 


Sorry but what is the exact stock voltage? Isn't it something like 1.36v?
 
A good way to figure out your overclock is to use AMD overdrive and play with the settings in there until you find your best stable overclock and then note those settings and use them in the BIOS. A few purists might disagree with that, but there's always somebody that's gonna disagree with everything. I know plenty of people who have found their stable overclock this way and then just uninstalled overdrive after making the switch in the BIOS. Using overdrive as a permanent solution probably, well, sucks. It's too quirky and likes to reset things after sleep and reboots even if you save the settings as the default ones and set it to reload every boot.
 


On my cpu-z it says on the multiplier (7 - 20), and yours says (7-21). Does that mean I can only overclock using multiplier at a max of 20? I'm currently running it at 18.5
 
Right now, I have my FX-8320 @ 4 Ghz (200*20) under 1.225-1.235 V running stable with 970 Pro3 r2.0...
In other words you can both overclock and undervoltage your FX-8320!
A futher research can be found over here...(Make sure you have checked the 3rd page)

http://www.overclock.net/t/1409067/undervolting-8350fx-without-losing-performance/http://www.overclock.net/t/1409067/undervolting-8350fx-without-losing-performance/
 


Have you got any fans or heatsinks or any cooling on the vrm's of the motherboard?
 
I don't know about the creeper, but I don't, and I have a tower cooler so not much of the cpu cooler airflow is going to help them out either. That being said, the Sabertooth board power design is superior to most boards and my motherboard temps are staying plenty cool simply with the case cooling. I have spot checked with an IR gun and nothing was ever above 32c. In fact, under full load the temps on the VRM's actually went down due to the cpu and case fans increased speeds from the cpu temp response.
 


If I buy a 120mm fan and put it on my case side panel, will it help keep the VRM's cool and help overclock? I have a CM Force 500 Case (it is bad i know)
 
Yes, it will help keep the board components cool and likely keep the graphics card cooler as well. Oh crap, you need to ditch that case. With a top mounted PSU you're never going to keep things cool. You can't have a rear top or actual top mounted case fan with that case and that's a problem for heat, both for the board and the PSU itself. I'd forget about another fan for now and get a different case with a bottom mounted PSU location. Any stupid case that's wide enough to accommodate your cpu cooler, has at least some cable management features and has locations for front, rear and top case fans should be ok.

It doesn't need to be a 100 dollar or up case. Back to the question though, even with that case it would probably help to plant a fan there if you're going to be using it for a while.
 


I do want to get rid of this case, unfortunately I won't be able to for a while.
 
Well, sometimes you have to make due with what you've got. I see that there are mounting locations for two intake fans on the right side so I'm assuming there is no front fan location. Do you have fans in both of those holes for intake? If not, I'd add an extra fan there, doesn't have to be PWM, a 3pin fan would be fine, which you can get for like ten bucks, as well as one in the left hand case panel for intake.

That would be about a twenty to twenty five dollar investment that would leave you with three intake fans, one exhaust case fan and the power supply exhausting to some degree as well which would probably drop your internal case temp by about 10-15c over a single intake and a single exhaust which is probably what you've got now huh?
 


I have 2 front intake fans, 120mm i think. A 120mm exhaust at the back, and that's it. The front case panel has barely any holes where the air can come in through.
 
Yeah, here's the thing. You really don't want any air coming in OR out except through the fans. Otherwise you defeat your pressure assignment regardless of whether you intend for positive or negative pressure. If you intend positive pressure, which is gained by having more intake than exhaust and the main feature of which is keeping dust inside the case to a minimum, you defeat it by allowing air pressure to escape through holes in the case.

In this scenario you only want air leaving the case through your designated exhaust fan so it will take all the heated air with it and leave the highly charged cooler air that was brought in by the intakes in the case until it is warmed and rises to where the exhaust fan is. With holes in the case you end up pushing some of your cool air that you brought in with the intake fans out of the case before it has had the chance to exchange heat from the components to the air.

If a negative pressure setup is your intention, (And IF the case is designed well for this purpose, it's more efficient at removing heat. But you had better either have filters over the intakes or regularly clean the dust from inside the case.) it is defeated once again by drawing air through those same holes, this time inwards rather than outwards. This is a problem because the effect of vacuum created by negative pressure cannot be well demonstrated with holes in the case. Imagine a vacuum cleaner hose. It has good suction at the end and will draw in air at a specific rate, which is then blown through the unit and replaced by the surrounding air which repeats the cycle. If you puncture several pencil sized holes in the vacuum line you will feel a noticeable difference in the amount of air being drawn through the hose and it will decrease the amount of air it is exchanging by the vacuum process.

If that loss of exchange considered hot air as the focus you would now effectively have reduced how much of it you are moving from where it is to where you want it to be, which in this case is OUT of the case. Bottom line is, the more sealed your case is with either configuration, the more efficiently it will perform the task. Some people will say, "but open bench cases run cooler than either of those designs" so just leaving the side panel off should be better. Nope. Because an open air bench testing style case has ALL components exposed to the ambient air with no chance of there being hot areas or pockets. With a case that still has five of it's six sides in place there is no negative nor positive pressure possible to either force or vacuum existing heat from the case so there will be the potential for hot pockets or "eddies" similar to what you see in parts of a river or stream where the water just swirls in circles while the rest of the tributary flows away to wherever they flow away to. Plus, we all know that to some degree at least, moving air is cooler than the ambient air that's still. I don't particularly understand this component since science tells us that all moving molecules create heat, but it works, so I'm going with it.

So in all reality, your case would probably work better if you sealed the small holes I see in the back beside the I/O shield and any that are in the front that are not for fans. Duct tape would probably work fine for this. You can probably even find black duct tape so it doesn't stand out.
 
Both. Seal off those small holes so the system is closed, or at least more closed. Few systems are really anywhere close to actually closed, but it's close enough. It doesn't need to be 100% sealed. 95% is probably fine. I would then add the other two fans using something like this or whatever you like. Adding fans is also cost effective because later on when you move to a better case these can be reused so you don't have to add fans to the cost of the case at that time.

http://www.newegg.com/Product/Product.aspx?Item=N82E16835553002