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

fx 8350 and overclock? or fx 9370?

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.
Is it me but do his temps and core voltage seem very high for 4.7?(at the top of the page with all that plumbing) mine @ 4.5 below. (CPUID while stress testing)
It's 4.8 not 4.7 and every chip is different. These Fx's get how when you really start putting your foot on them. You also do not know what the ambient temps are. I just want to give you an idea of the differences between chips. I have a 4770k and so does Johan. With mine running at a lower voltage and the same clock speed his runs 10c less then mine in temps using a closed loop thermal take 2.0 and I'm running a custom loop with a 360 rad. With the ambient temps being the same.
 
yea, his vcore is high, he should work it down, BUT, some of us pick a vcore we never will go over and then clock to that vcore, a fine way to do things in my book.
 
Ya some are some aren't. I had my piggy of an FX at nearly 1.6v to get 4.9. Here's a SS, this is a thread I started a while back Title Not all FX are the same. http://www.overclockers.com/forums/showpost.php?p=7484325&postcount=1

I could probably go to the AMD mobo section and find my thread over there where I was testing an FX-6300 and where I stated that I did not like how the FX-6300 I was using for test was comparing to my FX-8350 which I pretty much knew was good.

However I just reached up on the desk and got my cpu voltage log and here is what I wrote that statement from months ago when using FX-6300. From 4.1Ghz to 4.5Ghz the FX-6300 was just 0.01 Vcore less than my FX-8350 but at 4.6Ghz the FX-6300 was needing 0.02 more Vcore than my FX-8350 and when the FX-6300 reached 4.9Ghz speed the Vcore was now 1.54Volts or some 0.05Volts greater than my FX-8350. Knowing that four more FX-6300's bought at the same time clocked about the same, caused me to think if not say, that I had little need of an FX-6300.

And remember I said I "thought" if not said, that I did not think I would get an FX-6300 based on my little bit of testing. I likely did not say such because there is such a wide variance possible in cpus and systems that blanket statements are often mis-leading.

I will say I bought a used FX-8350 with the price right, rather than get a new FX-6300, but again, there can be great differences in a piece of cpu silicon. And of course one can see if cpu voltage can be lowered by testing, but sometimes the cpu is just a voltage pig once a certain speed level is reached.
RGone...
 
Is it me but do his temps and core voltage seem very high for 4.7?(at the top of the page with all that plumbing) mine @ 4.5 below. (CPUID while stress testing)

You would be amazed how much juice that fx processor wants when pushing towards 5 ghz. And mine is clocked at 4.8 ghz. This also has alot to do with how hot both my socket and VRM heatsink was getting because all those silver fans you see only move 31-ish CFM. VRMs lost efficiency when they're hot. I recently upgraded to much faster fans hooked up to my fan controller (103cfm each) and temps don't get past 45C under 100% load. And idk why core voltage is showing 1.5 something. Bios displays 1.3875.
 
And idk why core voltage is showing 1.5 something. Bios displays 1.3875.

tHAT'S SOMETHING ELSE. tHAT ss ON THE PREVIOUS PAGE SHOW 1.488- 1.554V THAT'S A HUGE JUMP AND YOU SAY YOU'RE SET A 1.38V?? wHAT ARE YOU DOING WITH llc ?

Sorry caps lock, trying to eat
 
tHAT'S SOMETHING ELSE. tHAT ss ON THE PREVIOUS PAGE SHOW 1.488- 1.554V THAT'S A HUGE JUMP AND YOU SAY YOU'RE SET A 1.38V?? wHAT ARE YOU DOING WITH llc ?
:rofl: need I say more.
 
tHAT'S SOMETHING ELSE. tHAT ss ON THE PREVIOUS PAGE SHOW 1.488- 1.554V THAT'S A HUGE JUMP AND YOU SAY YOU'RE SET A 1.38V?? wHAT ARE YOU DOING WITH llc ?

Sorry caps lock, trying to eat

:rofl: need I say more.

I assume you're asking what my load line calibration settings are? I believe it is turned off if I am not mistaken. I turned it off along with CIE6, power saver, cool n quiet, etc. Either that or it is set to auto, which in turn is turned off automatically when cpu OC mode is turned to manual. Why? LLC enabled or disabled would cause this voltage jump appearance in cpuz?

I would check right now, but I'm currently in class listening to my professor babble on about if else statements in c++.
 
I would try and control that voltage swing a bit. That's huge really. If I remember correctly the LLC on Asrock is kind of backwards but from the look of it it must be on auto. I would turn it on and set it to some middle ground. You'll have to up your CPU voltage to compenste and keep you at the 1.55v mark. Typically we try to find a setting that will keep the idle and load voltages fairly close to each other. so it will take a bit of tinkering on your part.
 
I would try and control that voltage swing a bit. That's huge really. If I remember correctly the LLC on Asrock is kind of backwards but from the look of it it must be on auto. I would turn it on and set it to some middle ground. You'll have to up your CPU voltage to compenste and keep you at the 1.55v mark. Typically we try to find a setting that will keep the idle and load voltages fairly close to each other. so it will take a bit of tinkering on your part.

hmmm. Might have to just start the whole overclock over again. Might just come here and ask for help from the get-go. I just upped the voltage and multiplier on this current clock. And how do you mean backwards on asrock boards? meaning if set to auto it does the opposite of other brand boards? I was getting mixed messages from everywhere about whether llc was something to be left on, off, or on auto for overclocking. Seems everyone had a different opinion.

Edit: well after snooping around some about special cases of LLC, Mine must have been left on auto because it is doing something called V-boosting. common in Asrock boards it seems. It is way over-compensating for any possible voltage droop. I'll make sure to fix this issue when I have the chance. I may not particularly care about this board or chip (both are under warranty and I plan on buying top shelf stuff if I burn something out) but I do wanna milk at least another year out of them.
 
I'm fairly certain that everyone here in the AMD section uses it if their board is equipped. The backwards part was referring to it's implementation. On my board a lower setting gives less voltage bump where a higher one gives more. From what I have seen of the Asrock the higher you set the LLC the less bump in voltage you get.
 
I'm fairly certain that everyone here in the AMD section uses it if their board is equipped. The backwards part was referring to it's implementation. On my board a lower setting gives less voltage bump where a higher one gives more. From what I have seen of the Asrock the higher you set the LLC the less bump in voltage you get.

I see. Alright. Like I said, I'll fix it asap. thanks for the input.
 
I'm fairly certain that everyone here in the AMD section uses it if their board is equipped. The backwards part was referring to it's implementation. On my board a lower setting gives less voltage bump where a higher one gives more. From what I have seen of the Asrock the higher you set the LLC the less bump in voltage you get.

On my Asrock Extreme 4 Z77 the LLC is 1 for the highest and 5 for the lowest, I believe it's the same on the AMD boards. You should try 3 and run prime to see if the Cpu V droops if it does up it to 2. The idea is to not have the LLC overshoot the set voltage by too much.
 
On my Asrock Extreme 4 Z77 the LLC is 1 for the highest and 5 for the lowest, I believe it's the same on the AMD boards. You should try 3 and run prime to see if the Cpu V droops if it does up it to 2. The idea is to not have the LLC overshoot the set voltage by too much.


got it. I can almost guarantee it is currently set to auto, and the board is WAY over-compensating. Now I am itching to get the hell outta this lecture.

Edit: turns out llc was disabled, not set to auto. The cpu has illegal sumout errors when using from lowest to highest voltage fluctuations. Fails at 100%, 75%, and 50%. Currently testing at 25%. These are the only options. Other than these 4, there is auto amd disabled. Im really hoping it passes 25% llc becauss my max voltage doesnt go past 1.472 now compared to 1.552 before. Temps dropped severely as well.
 
Last edited:
LLC changes.PNG

Alrighty gentlemen, Here it is. Passed 25 min of P95 blend testing. Voltage never went near the 1.5ish volts compared to before. Temps too. Thanks for the help. Does everything look to be good?
 
Do you still have the power saving features enabled? Cool and Quiet C1e etc...? The Cpu V is still jumping around more then I would like but that may be the best you can do with it.
 
Do you still have the power saving features enabled? Cool and Quiet C1e etc...? The Cpu V is still jumping around more then I would like but that may be the best you can do with it.

Nope. All other power saving stuff is turned off. And yea, this seems like the best i can do. Like i said, everything has illegal sumout errors beyond 25%. Not too bad though. Could be worse and still peaking at 1.552
 
Nope. All other power saving stuff is turned off. And yea, this seems like the best i can do. Like i said, everything has illegal sumout errors beyond 25%. Not too bad though. Could be worse and still peaking at 1.552
This may be because you have the set voltage too low. When you set the LLc lower open HWmonitor and run prime again. If it isn't dropping or overshooting the set voltage as much as it is when set at 25% then you may be better off. If you look at the last SS of HWmonitor you have a low of 1.384 and a max of 1.496 that is still a pretty big fluctuation. I would rather see it held in a tighter range. What's probably happening is with the LLC set where it is, the overshoot on the voltage is what is keeping it stable. This could cause a couple of issues, you may find the PC crashes periodically because when it's not under load it may not be getting the correct voltage. You may also find that your temps may get better with the LLC not overshooting by as much and the "set" voltage a bit higher. If I'm not making sense let me know.
 
Back