AMD FX-8320 3.5GHz ( Eight-Core )

Wolfy_

Commendable
Feb 17, 2016
25
0
1,530
I've looked all over the place and at different guides, but nothing that helps me to the end.

=====================Components=======================

Motherboard: Gigabyte GA-970-Gaming SLI
CPU: AMD FX-8320 3.5GHz( Eight-Core)
GPU: AMD Radeon R7 360 2GB 16x GDDR5
Cooling: Coolermaster Storm Devastator
Cooling: ASETEK 550LC 120MM Watercooler
RAM: ADATA 4GB DDR3 2-4GB

Any ideas on overclocking?
 
Solution
You don't seem real comfortable with this stuff, but from memory since I don't have my amd anymore to oc in bios (the prefered method) my setup was

multiplier = 21/4200mhz
turbo boost = disabled
xmp mem profile = on
voltage +.28v @ 4200 mhz

For me I'd usually disable all the power saving features except for "cool & quite". I found raising the fsb (the "200" setting) caused more trouble than it's worth, and anything above 4300mhz to be unstable no matter how much voltage you threw at it, but every chip is different so your results may vary..
My max stable oc for that chip was 4.2ghz with +.28v, 4300 was stable also, but for some reason an odd number for my oc just didn't seem right, and 4.4ghz was inevitably unstable no matter how much V I threw at it. Also multiplier only oc'ing was much more stable than adding fsb, even upping it 205/210 on the fsb caused more trouble than it was worth, at least 4 me..
 


Do I have any at all room to overclock or is it simply it can't spare the voltage, or is it just wouldn't be recommended?

If I could squeeze it a little though, how would I do so and what would be the warnings if even?
 


I'm just trying to get more performance per core and this seems to be the only option other than upgrading my CPU.
 
"EVGA 600 WATT 80 Plus Power Supply"

I don't know why everyone is so squeamish, you're at 225w tdp for your cpu/gpu so its got plenty of wattage, and I've personally rocked that psu with a fx-8320 @ 4.2gz and a moderate oc on a gtx 760, and as far as headroom/durability it is taking in the teeth from my new build and running with no stability issues at around 545w (according to the psu calc) while I wait for my 750w to show up..
 
Overhead wattage isn't the only part of not recommending overclocking. There's the quality of the PSU to look at as well. If it isn't incredibly stable/doesn't give out constant power/doesn't necessarily have the bells-and-whistles safety precautions on it, it's not recommended to overclock with. Wattage isn't the only determining factor in overclocking potential on a PSU.
 


It's not the maximum output that has people squeamish, it's the quality. Voltage regulation, filtering, etc. become even more important when overclocking a CPU and it's not recommended on a rather middling PSU.
 
It may not be the best, but it's not garbage either, it's probably your best entry level value in psu's..

https://www.techpowerup.com/reviews/EVGA/600B/11.html

Pros:
Affordable
Delivered full power at 46°C
Great ripple suppression (much better than the 500B I tested)
Voltage regulation at 5V
Quieter operation compared to the 500B
Low inrush current
Compact dimensions

Cons:
Should have four PCIe connectors and some more peripheral ones would be very welcome
Not Haswell ready (according to Intel's testing procedure)
Low hold-up time

Overall rating 8.3

Take a quick look at the 600B and it will look similar to its smaller brother, the 500B, since both are based on the same platform/design, have exactly the same connectors, and only 100 W of power as the difference, at least on paper. However, the 600B unit performed much better than the 500B I reviewed since it registered excellent ripple suppression, which is contrary to the mediocrity of its smaller brother, and a quieter overall operation. Voltage regulation and hold-up time of both units was almost the same, but the 600B exhibited much better ripple suppression, which was enough to give it a significantly higher score than the 500B. Some components apparently changed to bring these improvements about, though the 600B still wasn't able to pass my Haswell test based on Intel's leaked instructions. Here I should note that the latter are admittedly based on an extreme scenario, and EVGA informed me that they actually tested the 500B and 600B with Haswell units, which had them perform well in all instances. Yet I still have to use Intel's methodology, even if it looks totally out of this world, and, yes, its specifications are impossible to meet for group-regulated PSUs, which includes these EVGA Bronze units.
 
Just because it's of good entry level value doesn't necessarily mean that it's cut out for overclocking. As you said, it's absolutely doable. I got my FX6100 to 4.4GHz stable on a CM GX 750. Also not a great PSU. It's certainly doable--just know going in that you are taking a risk.
 
This mobo/cpu can be had for $225, so the real question is the best use of your $$ the $100 or so bucks for a good, well probably a decent mid-tier psu, for that price in order to protect your $225? I say Keep the psu and take the $100 and get more ram or a samsung evo and the roof, the roof, is on fire.. let mother'fer burn..
 
"Or to save up for a better Intel/Zen platform. Yeah."

Yeah, THEN you need a psu upgrade, but in the meantime I would run that amd until the caps glow red knowing my downside of blowing it out was being forced to upgrade.. cause sorry the am3 platform is well deader than dead...
 
And bash the evga all you want, I know it can rock the redonkulous and keep on ticking..

image.png


Capture.png
 


That is the same psu he has, and the opinion seems to be that it is not adequate for oc'ing , I disagree
It was running pretty much his exact setup
image.png

and is running this setup as I type..
image.png
 


I definitely am taking your cautions and the other responses into consideration, but I still am only left with the option to not overclock as I do not know what to edit in the Gigabyte Bios, I've checked around and it's usually an ASUS which does not help me.

I am still being cautious with this, but I'd like to see what it could do if I did decide to stick with the possible harming overclock.