Question Need help to overclock i7-9700k

[katulen]

Hey guys.

I am currently attempting to overclock my i7-9700k.

Here is my current specs:
CPU: Intel Core i7 9700K / 3.6 GHz
CPU Cooler: NZXT Kraken X72
MOBO: Asus Rog Strix Z390-F GAMING
GPU: Asus Rog Strix RTX2070 O8G GAMING 8GB GDDR6
RAM: Corsair Vengeance RGB 16GB: 2x8GB DDR4 3000MHz DIMM
SSD: Samsung 970 EVO Plus MZ-V7S500BW 500GB - PCI Express 3.0 x4 (NVMe) - M.2 Card
HDD: Toshiba P300 3.5´´ 2TB - 7200rpm 64MB SATA-600
PSU: Corsair RM750x
CASE: NZXT H series H700i Window Black

Attempt 4.9GHz:
Multiplier: 49
AVX Offset: -3
XMP: XMP 1 Enabled
MCE: Disabled
SVID Support: Disabled
BLCK Aware Adaptive Voltage: Disabled
CPU Core/Cache Voltage: 1.350

• Cinebench r15 - Passed.
• Realbench - Passed.
• Prime95 - Passed.
• Highest temps was 85c.

Attempt 5.0GHz:
Multiplier: 50
AVX Offset: -3
XMP: XMP 1 Enabled
MCE: Disabled
SVID Support: Disabled
BLCK Aware Adaptive Voltage: Disabled
CPU Core/Cache Voltage: 1.360v
CPU Load-line Calibration: Level 6
CPU Current Capability: 170%
Long Duration Package Power Limit: 4095 (max)
Short Duration Package Power Limit: 4095 (max)
CPU Core/Cache Current Limit: 255.75 (max)

• Cinebench r15 - Passed.
• Realbench - Passed.
• Prime95 - BSOD after 2 minutes.

Failed.

2nd attempt 5.0GHz:
Multiplier: 50
AVX Offset: -3
XMP: XMP 1 Enabled
MCE: Disabled
SVID Support: Disabled
BLCK Aware Adaptive Voltage: Disabled
CPU Core/Cache Voltage: 1.360v
CPU Load-line Calibration: Level 5 (Changed)
CPU Current Capability: 170%
Long Duration Package Power Limit: 4095 (max)
Short Duration Package Power Limit: 4095 (max)
CPU Core/Cache Current Limit: 255.75 (max)
Changes: Only changed LLC from Level 6 to Level 5.

• Cinebench r15 - Passed.
• Realbench - Passed.
• Prime95 - Worker stopped after 3 minutes, no BSOD.
• XTU - Passed.

If you need some Min/Max/Avg screenshots of temps/voltages let me know, apparently cant paste them in this thread.

So a few questions here.
I've heard/read that if a worker stops in Prime95, it means your overclocking is unstable. Is it worth trying to let it finish and see what happens or pointless ?

Also i noticed when using XTU, that every 10 sec, 1 or more cores drop to 4.7GHz for a split second and go back to 5.0GHz. Is that due to AVX instructions coming into play or is it my cpu throttling? And how to fix it if needed.

Temps are fine, hitting max 82c on 1 core, rest high 70's.
VCORE is at 1.360v and droops to 1.332v and never go over 1.360v.

Update on my latest 5.0GHz attempt:

I just tryid following:
1.375v vcore fixed, LLC 4. Vdroop to 1.305v and BSOD in Prime 95.
1.375v vcore fixed, LLC 5. Vdroop to 1.350v and BSOD in Prime 95.
1.375v vcore fixed, LLC 7. Vboost to 1.421v and BSOD in Cinebench r15. (No vdroop - Probably overshooted the voltages with Level 7 i assume)

I tryid LLC 6 together with 1.360v without success.

Im trying with LLC 6 now with a higher vcore voltage.
Settings:
Multiplier: 50
AVX Offset: -3
XMP: XMP 1 Enabled
MCE: Disabled
SVID Support: Disabled
BLCK Aware Adaptive Voltage: Disabled
CPU Core/Cache Voltage: 1.380v
CPU Load-line Calibration: Level 6
I removed all the other stuff i added earlier. (See above).

Did OCCT Linpack Non-AVX.
1.380v, vboost to 1.403v. No vdroop.
Max temp 94c.
Distance to TJMax 6c.
After 13min of OCCT i got BSOD.

I dont know what to try next tbh.
Is there ANYTHING I am missing or something i need to dial/tweak for 5.0ghz to happen? Or have i done everything correctly but my chip just won't hit 5.0GHz due to silicon lottery ?

 

[rigg42]

Start with them both at stock and see what happens when you run IBT. It should have the stock values greyed out next to the columns. If it freezes or the window pops up increase to 1.1 then 1.2 if needed.

 

[katulen]

Start with them both at stock and see what happens when you run IBT. It should have the stock values greyed out next to the columns. If it freezes or the window pops up increase to 1.1 then 1.2 if needed.

Ehm. There is a blackbox next to it, saying 1.312v and vccsa 1.216v . Is that it?
When i hover over vccio it says In the buttom of bios min voltages xxx / max xxx / standard 0.950v.. or is that it?

Dont i also gotta keep In mind that my system is oc, and default voltages prolly just gonna make it crash?

 

[rigg42]

Start with standard. See what happens. You want to get these voltages as low as possible or you will degrade your memory controller on the CPU. You shouldn't need much more or anything above stock for a 3000mhz kit.

 

[rigg42]

Start with standard. See what happens. You want to get these voltages as low as possible or you will degrade your memory controller on the CPU. You shouldn't need much more or anything above stock for a 3000mhz kit.

Save your profile in case it wont post and we have to clear cmos.

 

[katulen]

Save your profile in case it wont post and we have to clear cmos.

Yeah i did just save when i was in BIOS just now.

I got it to pass IBT at these values:
VCCIO 1.250v
VCCSA 1.150v

What do you think?

 

[rigg42]

I think you have a terrible memory controller. That voltage should be safe though.

 

[katulen]

I think you have a terrible memory controller. That voltage should be safe though.

Hahaha. Probably is. But it failed in IBT 5/10 with lower than the current values listed above 😀.
I did move both of them at once. Perhaps work on 1 at a time is better way, as usually.

EDIT: When browsing the web for peoples VCCIO/VCCSA, they have set them higher than mine, but lower than default. By default i mean asus' default.
Article: https://rog.asus.com/forum/showthre...m-Agent-What-do-you-guys-set-your-voltages-at

What you reckon next step is ?

 

[rigg42]

Hahaha. Probably is. But it failed in IBT 5/10 with lower than the current values listed above 😀.
I did move both of them at once. Perhaps work on 1 at a time is better way, as usually.

What you reckon next step is ?

Yeah 1 at a time would be better. I probably should have mentioned that. Play some games until you are ready to go to bed. The more cpu intensive the better. See how your cpu temps look while gaming. Before you go to bed start an 8 hour realbench stress test with 8 gigs of memory. Turn the sleep settings off in windows first.

 

[katulen]

Yeah 1 at a time would be better. I probably should have mentioned that. Play some games until you are ready to go to bed. The more cpu intensive the better. See how your cpu temps look while gaming. Before you go to bed start an 8 hour realbench stress test with 8 gigs of memory. Turn the sleep settings off in windows first.

Read my edit in previous post.
But i can always mess with that later, if, when i get this current OC stable.

Ok.
If i run 8 hour realbench stress test when sleeping, i can't monitor tempatures. Should i monitor the temps 10,20,30 mins in or doesnt it matter? And if shit goes down, will be pc just shut down to protect it self or what?

 

[rigg42]

Read my edit in previous post.
But i can always mess with that later, if, when i get this current OC stable.

Ok.
If i run 8 hour realbench stress test when sleeping, i can't monitor tempatures. Should i monitor the temps 10,20,30 mins in or doesnt it matter? And if shit goes down, will be pc just shut down to protect it self or what?

The temps should level out after 20-30 mins. If you are not comfortable with leaving it unattended for 8 hours just do it during the day and check in on temps from time to time. At least then you'll know if it crashes. If the cooler takes a dump its possible it could leak before the machine shuts itself down. I've never really been that worried about it though.

 

[katulen]

Should i put any high fan curve or just have it as default during this 8hour test ?

Also,
i disabled windows sleep thingie to never. But there is also something called Hypernate. Disable that as well, or it maybe doesnt take effect since sleep is disabled?

At least then you'll know if it crashes

If the stresstest failed i assume it will write so in Realbench right ? And if i wake up tomorrow and my PC is shut off or its on desktop with nothing opened, i can assume it failed at some point?

 

[rigg42]

Should i put any high fan curve or just have it as default during this 8hour test ?

Also,
i disabled windows sleep thingie to never. But there is also something called Hypernate. Disable that as well, or it maybe doesnt take effect since sleep is disabled?

Disable that too. The fans are up to you. I'd run them as high as you can stand. Set them how you plan to use them or you aren't really doing a valid test.

 

[rigg42]

Be prepared for it to crash or stop the test if it detects instability. You may need to add more voltage.

 

[katulen]

Disable that too. The fans are up to you. I'd run them as high as you can stand. Set them how you plan to use them or you aren't really doing a valid test.

Well, as you said yourself earlier, these stresstest is way more demanding than the games i would be playing. Therefor i would imagine i can assume lower temperatures at real world applications ?

And yeah, probably will crash. I guess ill see tomorrow morning 😀.

 

[katulen]

Be prepared for it to crash or stop the test if it detects instability. You may need to add more voltage.

There we go.
It passed Realbench 8 hours, with no instabilities shown.
Max temp was 84c, avg around the 79c mark. 0 WHEA errors shown in HWiNFO.

So i assume this 5.0GHz profile is stable. Thanks alot for the help, much appreciated !

Question is now,
how do i get this to work with adaptive voltages, so i can have lower voltages when idling ?

So if I want to change from fixed voltages to adaptive voltages to get the benefit of that, how do I get around that correctly ? Do I just put in 1.385v as Adaptive Voltage or do I also add some sort of +/- offset as well?
Do i change IA DC load lines set to 0.01 ?

 

[rigg42]

There we go.
It passed Realbench 8 hours, with no instabilities shown.
Max temp was 84c, avg around the 79c mark. 0 WHEA errors shown in HWiNFO.

So i assume this 5.0GHz profile is stable. Thanks alot for the help, much appreciated !

Question is now,
how do i get this to work with adaptive voltages, so i can have lower voltages when idling ?

No problem. Happy to help. The voltage is mostly irrelevant when idling since you are pulling little to no current. I suggest you leave well enough alone and be happy to join the 5ghz club!

 

[katulen]

No problem. Happy to help.

You were awesome rigg, so again thanks so much for guiding me through this

The voltage is mostly irrelevant when idling since you are pulling little to no current. I suggest you leave well enough alone and be happy to join the 5ghz club!

Hmm not sure i understand that? Looking at CPU-Z & HWiNFO i can see it sits on 1.376v-1.385v at all times during idle. So what do you mean by "pulling little to no current" ?
Isn't it more "healthy" for my CPUs longevity to get less voltages over time, by using adaptive. So when it clocks 800mhz it only gets 0.9XXv ?

Also my cpu is at 35-40c on idling. I remember at stock it was sitting in the 28-32c on idling.

 

[rigg42]

I'm home now so I can play around with my setup. Out of curiosity I stuck my motherboard in offset voltage mode and set the minimum .005 negative offset. My motherboard is delivering way less voltage than it took to get stable with a fixed voltage and is seemingly stable right now. I'll have to do more stability testing but you may want to try this and see what it does in your setup. This is bizarre.

 

[katulen]

I'm home now so I can play around with my setup. Out of curiosity I stuck my motherboard in offset voltage mode and set the minimum .005 negative offset. My motherboard is delivering way less voltage than it took to get stable with a fixed voltage and is seemingly stable right now. I'll have to do more stability testing but you may want to try this and see what it does in your setup. This is bizarre.

Well that is interesting.

From what i understand with the information i've gathered upon the subject is.

You start with fixed voltages. Find your minimum voltages your stable at. Then you switch to adaptive voltages, with the last known stable voltage. Then you change SVID to Best Case Scenario, which will automaticly set AC/DC loadline to 01.
You load up windows, put load on the CPU and see where you're vcore voltages is. If you are higher than what you set in BIOS, you set a negativ offset to match the voltages you want. If you're lower than what you set in BIOS, you set a positive offset to match what you want.

I don't know if you can confirm the above is correct or not. But i assume this is what im gonna try.

Also, please reply to my previous message about voltages and little to no current etc.

 

[rigg42]

If that is what you want to achieve that seems like a solid plan. I'm really not sure what is going on with my setup right now. The voltage is 500mv lower according to HWinfo yet my temps are identical. I'm confused.

As to the current thing..... So when your CPU is idle it is drawing very little current compared to at load. This is why you need load line calibration. The voltage droops under load because the CPU is suddenly drawing a lot of current when its working hard. LLC ups the voltage to compensate. This is why v-boost is dangerous. If the voltage is increasing under load when it shifts back to idle the voltage compensation doesn't stop instantaneously. So when it shifts back to idle and the CPU isn't drawing a bunch of current anymore you get a short spike of voltage while the motherboards LLC is still pumping extra voltage to the CPU. If you use too much LLC this could potentially cause a spike that is dangerous to the CPU. This is why its a good rule of thumb to never use a level of LLC that increases voltage under load. The v droop offers some headroom for this voltage spike.

Since the CPU isn't pulling any significant current at idle you wont damage or degrade the CPU. You'll just use a bit more electricity than you normally would.

 

[katulen]

The voltage is 500mv lower according to HWinfo yet my temps are identical. I'm confused.

WHAT? What vcore are you at under load ? :O
Please update me on this, cause this is super interesting.

Since the CPU isn't pulling any significant current at idle you wont damage or degrade the CPU. You'll just use a bit more electricity than you normally would.

Thanks for answering.
I think where i dont understand this, is the difference between "voltage" and "current". So let me see if I get this right.
At this moment my CPU is getting 1.376v-1.385v while idling, but little to no "current". But under load, it is pulling the same voltage (1.350v to be precise - because of vdroop) but with "current". Therefor using more electricity, producing more heat, but at idle with same voltage, it pulls little to no current, therefor little electricity and producing less heat. Correct? 😀 Sorry im dumb on this subject, hahaha.

 

[rigg42]

Google Ohm's law

 

[katulen]

Google Ohm's law

Damn i missed out in school, ROFL.

Anyway, what is your current vcore under load?

 

[rigg42]

WHAT? What vcore are you at under load ? :O
Please update me on this, cause this is super interesting.

1.3 is what it says at load. I think the auto voltage is reporting incorrectly. Temps were worse than with a 1.35 fixed. I'm dialing down to a 5.0 overclock right now. 1.265v yeilds max 80c with Prime small fft. I think I had it stable at 1.25 previously but I'm getting ready to sell this rig and want to make sure it's as stable as possible for the next guy. These chips are so hot it's ridiculous. 80c at 1.265v (1.25 actual) with a 360 AIO in a well ventilated case is silly.

 

[rigg42]

Damn i missed out in school, ROFL.

Anyway, what is your current vcore under load?
And also, you didnt pick adaptive mode but offset mode, correct?

Yes offset mode.