• Happy holidays, folks! Thanks to each and every one of you for being part of the Tom's Hardware community!

AMD 8320 OC Question

LouieSmith66

Honorable
Feb 10, 2015
167
0
10,690
Basically I was able to hit 5.1Ghz at 1.48 but then I realise I had a little to much thermal paste and I removed the chip to clean off and put less on and the temps were fine but when I try 4.7 ,8, 9 and 5.0ghz my PC freezes even though before it was stable before I removed the chip ? I don't know what's happening and I'm getting really annoyed because my chip can hit over 5Ghz


Anyone had this problem ? I would really appreciate it if you wishful guys could help me out its driving me crazy

Specs:
AMD 8320 was at 5.0ghz
GTX 770 3way SLI
1200w PSU
16GB RAM 1333mhz
Asus Crosshairs Formula V
Corsair H100i
 
What do you mean by "stable"? Did you run Prime95 version 26.6 on small FFT's? If so, for how long. Stability is determined by a 24 hr run of Prime95 26.6 small FFT steady state testing. It's also the best way to determine thermal compliance.

I don't see a CPU cooler listed. What CPU cooler are you using? I can't say you're wrong, but I've seen very few 8320 chips that could be stabilized beyond 4.7Ghz, and those needed very extreme cooling. 4.5-4.7Ghz is pretty well the norm across the board with a few that act quirky even there.

It's possible you got a poor mount the second time or botched the paste job on the most recent attempt. How much paste did you use, did you fully clean both the CPU and the heatsink base and what cooler are you using? Considering the board, 5Ghz might be possible, but in light of the fact that the 9370 and 9590, which are the same processor, just much higher binned chips, have major heat issues without very high end cooling, I'm interested in what you were "stable" with.
 
And I prime 95 it for a day it was fine no workers failed it was all stable but when I put the chip back onto the board with less thermal paste to get the temps lower it wouldn't like it at 5ghz it would freeze in cinebench
 
Yeah it's alright ive oced it and it made no difference and I left it at auto before and everything was fine .... So I did a cmos reset and did the same setting as last time and still crashes but it didn't the first time ???
 
I didn't ask if you "thought" your PSU was fine. Everybody "thinks" that their PSU is fine until it's proven to not be. I simply asked what the brand and model number were. If you don't want to provide the information necessary to help eliminate potential issues, of which PSU quality when overclocking is definitely one, I can't help you.
 
Ok, the AX is a good product, for Corsair. Heh. However, 3 way SLI on the GTX 770 calls for 1100w. Your CPU is a 125w chip without ANY overclocking. Overclocking is going to take that up to more like 225w. Clearly your beyond the capacity of the unit. If your graphics cards are also overclocked, you're even more so. I'd pull two of those cards out, work out your stability issues, then see about reinstalling them. If the instability returns with three cards installed, you may need a bigger PSU. Personally, I don't think there is much to gain from a 5Ghz overclock. The FX-9590 does not show any gaming benefits over any FX 8 core chip at 4.5Ghz except in a very small handful of titles. And even then, it's not much. I'd take your chip down to 4.6 if you can get it stable there, and call it a day.

It's also possible you may have severely taxed the PSU with your previous efforts which is where the stability issue is coming from now. A PSU tester doesn't tell you the whole story. Only part of it.
 
5Ghz might have been running stable, or "somewhat" stable, but may have just been too much for the hardware. There are still a lot of things to be considered here. I'm going to ask another member to chime in on this. I'm fairly well experienced with overclocking but am still by far not an authority in this area.

I think I'd start by resetting the bios to defaults and seeing what you're able to take the chip to in small increments before any problems crop up.
 
Are we sure temps are NOT the issue here? Are you monitoring with Core Temp? What are your temps peaking at? As much as you might not want to hear it, being a proponent of Corsair products, the H series coolers are known to be somewhat problematic and underperformers, so I'd think a serious interest in what the temps are doing at the time it freezes would be of great interest to your success or lack thereof.
 
MemTest86+ on your RAM? Seems any time that really strange oddities show up with overclocking, RAM or MB DIMM slots seem to be a culprit. Well, that and MB power management can go on the fritz. Your PSU might be OK, but it might also be close to being beyond load.

Here's what I would do: reset your MB to everything AUTO/default. Use only 1 GPU. Test your RAM with MemTest86+...you'll know within a few minutes if you have a bad stick. If you do one at a time, it will easily identify which is bad. Also MB DIMM slots can go bad, so testing that is also a good idea.

The one thing you said you did was pull the CPU cooler and fix the TIM. Assuming that you have good contact and you are reporting decent temps, have you tried loading the CPU at stock settings to ensure it isn't something else? Doing load tests to monitor temps and performance at stock speeds on everything is the only way to benchmark your hardware when you do overclock it. Otherwise, you don't have a reference for any values.