The good, the bad, and the OC!

[Mag10]

So I have been toying around with my 8320 once again and seeing what I would be able to get as far as voltages, temps, and of course frequencies. I managed to come up with the following screenshot:


Now what I am curious about is the voltages that are being read by the software. where HWINFO says its about 1.225V meanwhile CPU-Z states 1.284-1.296V. From what I have gathered the voltage is extremely good to at least very good for this kind of OC. I have a feeling that with the right cooling this CPU could actually end up somewhere between 4.5-5.0 or better? Would anyone have any thoughts on this? Also I have been tempted to adjust the bus speeds but just haven't gotten there yet. I think the max was like 210 or 220MHz when I did a baseline test for the MB only.

Any convo on this is welcomed, and any additional tips as well. It just seems like I have a very good chip for OCing, or am I mistaken?

 

[Mr.Scott]

Will it pass 2 hrs prime with those settings?

 

[Mag10]

yes completely stable

 

[Mr.Scott]

Nice.
My 8320 will do 4.5 @ 1.3v.
On high end water I can bench over 5.3 @ 1.55v

 

[Mag10]

Yeah, that's a very thirsty chip you have there! I still need to do some tweaking with my cooling also, my 140XL can keep up with this CPU thus far with default fan speeds. It idles anywhere between 22 and 28ish with the load never breaking about 46.

But back to my voltage Q which voltage is more reliable in this case? I am simply not quite as familiar with HWINFO

 

[Mr.Scott]

It's an early 8320. I can do 5 gig rock stable @ 1.475v.
As far as reading voltage, I don't use HWINFO, only here for comparisons.
I use a combo of the bios, CPUz, and a DMM, depending on what I'm doing.

 

[bartx]

Notice that HWINFO shows two voltages - VID and Vcore, which is equal to CPU-Z. So I guess that VID is based on bios setting, but Vcore means real voltage. If you want to be sure check with DMM.

 

[Mag10]

Ah ok, well either way I wonder how much more I can get out of it at that voltage. If i could push it to 4.7 to 4.8 would be great!

 

[Johan45]

Notice that HWINFO shows two voltages - VID and Vcore, which is equal to CPU-Z. So I guess that VID is based on bios setting, but Vcore means real voltage. If you want to be sure check with DMM.

VID(voltage identification) Is factory set(programmed) within the CPU. There will be multiple VIDs that correspond to certain PStates used for downclocking(energy savings)
Here's an example of an 8350 with VIDs for corresponding PStates

P-State FID 0x1A - VID 0x0A - IDD 13 (21.00x - 1.425 V)
P-State FID 0x19 - VID 0x0B - IDD 13 (20.50x - 1.412 V)
P-State FID 0x18 - VID 0x11 - IDD 12 (20.00x - 1.337 V)
P-State FID 0x12 - VID 0x19 - IDD 10 (17.00x - 1.237 V)
P-State FID 0xC - VID 0x21 - IDD 8 (14.00x - 1.137 V)
P-State FID 0x5 - VID 0x2B - IDD 5 (10.50x - 1.012 V)
P-State FID 0x10C - VID 0x35 - IDD 4 (7.00x - 0.887 V)

 

[Mag10]

Thats good to know, I appreciate the information

 

[Johan45]

You can get your Pstates by using CPUz. The PC has to be at defaults in bios, open CPUz go to about tab and export to a .txt file. There's a lot of info in there so you need to go through it a bit but you'll find what I listed in the post above for your CPU.

 

[Mag10]

Well, I managed to push it to 4.72GHz unfortunately it doesn't seem it was completely stable, as I got a BSOD. It would boot into windows, but the temps were going nuts bouncing between 38-50*C while doing basic tasks. The good news is I did not need any additional voltage, I also found out how the CPU handle 1.4V @ temp in the process of tinkering around. I would definitely say I need a better cooling solution to push the CPU much further, the 140mm AIO does give me a small margin better than say a 120 in terms of dissipation but once I start cranking the voltage/freq past 1.3/4.5 it does not handle it well and turned into a jet just trying to keep up. I do have the Fans setup to run on PWM and it of course throttles the fans but I don't think I have quite gotten the sweet spot for cooling and silence. I have a strong feeling that I could possibly hit 5.0 range with this CPU given better cooling at around 1.35-1.4V. I have always intended to put it under custom water(and my AIO was only $60 at the time), but that will still have to wait.

 

[Johan45]

For 5.0 water is a must and I think you'll find you'll be closer to the 1.5v range. I have had one chip that I thought was quite exceptional that way but I also use chilled water in my loop so I have a bit of an advantage. Believe me I have tested quite a few and this one was an exception.

 

[Mag10]

For 5.0 water is a must and I think you'll find you'll be closer to the 1.5v range. I have had one chip that I thought was quite exceptional that way but I also use chilled water in my loop so I have a bit of an advantage. Believe me I have tested quite a few and this one was an exception.

View attachment 167839

well with having a clock of 4.7 and just shy of 1.3v you really think pushing to 5.0 requires and additional 0.2v? even though my clocks weren't 100% stable it does show great promise, unless the voltage scaling completely goes out the window after 4.6-4.7GHz. I may just manually crank the AIO all the way up to 100% in order to attain reasonable temps but just the gut reaction I have gotten from briefly bumping over 4.5 seems to show that the chip is very reasonable and isn't extremely power hungry from what I can tell.

 

[Johan45]

You'll be surprised how hot and thirsty these FX can get in a hurry. You might get lucky but the odds aren't in your favour.