Jump to content

Hey all.

I've managed to get my 9900K stable @ 5200MHz. I've thrown everything at it like Cinebench R15 and R20, XTU, AIDA64, CPU-Z Bench, 3DMark, and hours of gameplay on various games without a single blue screen. My voltages and LLC are as follows...

Adaptive Mode
Additional turbo vcore @ 1.150v
Offset @ 0.210
= 1.360 vcore
LLC6

I'm using an ASUS Maximus XI Hero Z390 board.

When I run lets say Cinebench with Core Temp running besides it to monitor vcore my vcore will overshoot to 1.50v!! Is this safe? Should I be concerned with this? Or are these readings inaccurate? Cheers guys!

Link to comment
https://linustechtips.com/topic/1127404-is-my-9900k-overvolting/
Share on other sites

Link to post
Share on other sites

2 hours ago, evo4ever said:

Hey all.

I've managed to get my 9900K stable @ 5200MHz. I've thrown everything at it like Cinebench R15 and R20, XTU, AIDA64, CPU-Z Bench, 3DMark, and hours of gameplay on various games without a single blue screen. My voltages and LLC are as follows...

Adaptive Mode
Additional turbo vcore @ 1.150v
Offset @ 0.210
= 1.360 vcore
LLC6

I'm using an ASUS Maximus XI Hero Z390 board.

When I run lets say Cinebench with Core Temp running besides it to monitor vcore my vcore will overshoot to 1.50v!! Is this safe? Should I be concerned with this? Or are these readings inaccurate? Cheers guys!

Offset +0.200v?  Yikes.  Yeah.  1.50v sounds right.

You should read up on how additional turbo vcore works.

1.150v does not work.  I don't have an Asus board but I can tell you right now that this value must not be set below the VID or it won't work.

Looks like it's setting 1.30v + 200mv.  Check the VID.  And reduce that offset before you degrade that chip. 

And how the hell did you cool that thing at 1.50v?

Link to post
Share on other sites

1 hour ago, Falkentyne said:

Offset +0.200v?  Yikes.  Yeah.  1.50v sounds right.

You should read up on how additional turbo vcore works.

1.150v does not work.  I don't have an Asus board but I can tell you right now that this value must not be set below the VID or it won't work.

Looks like it's setting 1.30v + 200mv.  Check the VID.  And reduce that offset before you degrade that chip. 

And how the hell did you cool that thing at 1.50v?

Yeah I've just been doing some research on the 9900K's stock volts and VID's. I've changed my settings to:

 

Adaptive Mode
Additional turbo vcore @ 1.250v
Offset @ 0.110
= 1.360 vcore
LLC6

 

Would say those settings are about right? If you have any suggestions for better settings im all ears!

 

I'm using a Corsair Hydro Series H115i Platinum RGB with Thermal Grizzly Conductonaut Liquid Metal Paste. Also the ambient temps in my room are very low.. around 10c. Now that its winter it'll get even colder, ive have to start wearing a coat soon lol.

 

Idle temps vary between 23c - 27c across 8 cores.

Gaming temps vary between 55c-68c across 8 cores.

Stress temps vary between 80c-95c (sometimes 100c) across 8 cores.

 

"Stress Tests" and the temps they create don't concern me as I'm purely a Gamer. It's the gaming temps I take notice to which imo seem reasonable!

 

EDIT: So is it safe for my CPU overvolt to 1.5v on the odd occasion?

Link to post
Share on other sites

3 hours ago, evo4ever said:

with Core Temp running besides it to monitor vcore

But Core Temp does not report vcore.  It only reports the VID voltage.  Use HWiNFO or CPU-Z to see how much voltage your CPU is really getting when stress testing.

Link to post
Share on other sites

1 hour ago, evo4ever said:

Ok I installed HWiNFO64 and I done a quick CPU-Z bench and HWiNFO64 is reporting a maximum vcore of 1.40-1.41v. This is much better right? I wonder why CoreTemp was reporting 1.5v? Sorry guys im a little confused here lol..

Coretemp is a pretty old program.  I see it's been updated, and it has its uses, but it doesn't even show CPU Vcore.  It shows VID, which can't be used as VCORE unless you are using 'Auto voltage and certain values for "AC Loadilne", "DC Loadline" must match "VRM Loadline" (loadline calibration), which is too confusing to average users (Loadline Calibration is a mOhms rating as shown here: https://www.overclock.net/forum/6-intel-motherboards/1638955-z370-z390-vrm-discussion-thread-398.html#post27860326 ), and even matching "DC Loadline" to your VRM loadline won't necessarily make VID sync to your vcore.


Plus Coretemp tries to install adware.  Yikes.  Avoid it.

 

Coretemp is only useful for temps.  Nothing more.  And with HWinfo64's ability to 'Hook' into Rivatuner Statistics Server so you can see your temps while playing games, why do you need Coretemp?

 

https://www.hwinfo.com/add-ons/

 

Your offset is still too high.

Try Additional Turbo Voltage: 1.35v and remove the offset completely and see if you're stable.

If you're not, increase offset by 0.01v (10mv) at a time from 0.

Link to post
Share on other sites

7 hours ago, Falkentyne said:

Coretemp is a pretty old program.  I see it's been updated, and it has its uses, but it doesn't even show CPU Vcore.  It shows VID, which can't be used as VCORE unless you are using 'Auto voltage and certain values for "AC Loadilne", "DC Loadline" must match "VRM Loadline" (loadline calibration), which is too confusing to average users (Loadline Calibration is a mOhms rating as shown here: https://www.overclock.net/forum/6-intel-motherboards/1638955-z370-z390-vrm-discussion-thread-398.html#post27860326 ), and even matching "DC Loadline" to your VRM loadline won't necessarily make VID sync to your vcore.


Plus Coretemp tries to install adware.  Yikes.  Avoid it.

 

Coretemp is only useful for temps.  Nothing more.  And with HWinfo64's ability to 'Hook' into Rivatuner Statistics Server so you can see your temps while playing games, why do you need Coretemp?

 

https://www.hwinfo.com/add-ons/

 

Your offset is still too high.

Try Additional Turbo Voltage: 1.35v and remove the offset completely and see if you're stable.

If you're not, increase offset by 0.01v (10mv) at a time from 0.

Ok I've just tried your suggested settings and I got a blue screen just as windows was loading. I tried applying some offset (+0.010) like u said and that blue screened as well!

 

So I reverted back to my settings (1.250, 0.110) and no blue screens!

 

Whats goin on here? Do higher offsets give better stability?

Link to post
Share on other sites

I'm not sure what your 'default' VID is.

Offsets are based on your base VID.  The "AC Loadline" setting influences your base VID also, but the AC LL setting (I think "SVID behavior" will set this to a certain value as well) tweaking is not for anyone but very advanced users.

 

However I assumed that 1.35v for additional turbo voltage would give the CPU at least 1.35v.

I'm surprised that it crashed.

 

I guess without an Asus board, I can't give you valid suggestions.  Maybe you can try 1.35v additional turbo voltage and +50mv ?

 

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×