[SOLVED] Incorrect Voltage Displayed

one-called-kane

Honorable
Oct 8, 2013
29
0
10,530
0
So in the process of setting up my Overclock the voltages that I am setting do not match the voltages being reached during my stress tests. For example when I set the voltage to 1.2 the the voltages will spike to 1.4 during testing, 1.3 reaches 1.5, etc.

At first I thought the Intel XTU utility I was using was the problem, but it happens when I make the changes in the BIOS too. I can change the voltage to any number and it's like the system takes it as a suggestion and then does whatever it pleases. Right now I'm running at 4.5GHz with what it says is 1.20V but is actually 1.41V according to HWiNFO. Even the idle voltage is ~1.35.

My motherboard is an AsRock Z270 Extreme4 and the CPU is an i7 6700k. What I have right now is technically stable at least thermally, but the voltage makes me nervous. I'd be a lot more comfortable if it would just do as I tell it.
 

Darkbreeze

Titan
Moderator
Something is wrong. Voltages should go DOWN during stress testing, due to vdroop, not up.

Are you sure it's the CORE voltage you are changing? Do you have the MOST recent BIOS version installed? I think I'd reset the bios to the defaults by doing a hard reset, or if there is an update, then update, and then try again.

Power off the unit, switch the PSU off and unplug the PSU cord from either the wall or the power supply.

Remove the motherboard CMOS battery for five minutes. During that five minutes, press the power button on the case for 30 seconds. After the five minutes is up, reinstall the CMOS battery making sure to insert it with the correct side up just as it came out.

Now, plug the power supply cable back in, switch the PSU back on and power up the system. It should display the POST screen and the options to enter CMOS/BIOS setup. Enter the bios setup program and reconfigure the boot settings for either the Windows boot manager or for legacy systems, the drive your OS is installed on if necessary.

Save settings and exit. If the system will POST and boot then you can move forward from there including going back into the bios and configuring any other custom settings you may need to configure such as Memory XMP profile settings, custom fan profile settings or other specific settings you may have previously had configured that were wiped out by resetting the CMOS.
 

Darkbreeze

Titan
Moderator
The other possibility is that you have the LLC (Load line calibration) set way too high. Rather than countering vdroop, if it is set too high, it can actually cause voltage to increase beyond the defined setting during high load conditions. Make sure your LLC setting isn't more than about 4 or 5 on a setting scale of 10, if your board has that setting, or set it to standard or high, rather than extreme or very high. See what settings it has and play around with that a bit. I think that could be the issue. It's the only thing I can think of that might cause that.
 

one-called-kane

Honorable
Oct 8, 2013
29
0
10,530
0
So I cleared the CMOS and updated to 2.70, then changed LLC to level 4 which is the mildest it will go.

In the BIOS the voltage is set to 1.3 and HWiNFO is showing the maximum usage during x264 to be 1.23. I have never had a CPU be so reluctant to do as it's told.

Is there any chance HWiNFO's VID values are incorrect?
 
Last edited:

Darkbreeze

Titan
Moderator
Unlikely. If it was HWmonitor, Open hardware monitor, CPU-Z or Speccy, then I'd say yeah, it's probably wrong. HWinfo and Core Temp are generally pretty spot on, but nothing is perfect so anything is possible.

Before, you were EXCEEDING the voltage that was specified. Now it seems you are below it, which is correct, and normal, when under a load. All systems have vdroop. The hot ticket is getting to a place where you find a balance between vdroop (Which is normal, and drops voltage under a load) and stability, which generally requires that the voltage not drop TOO FAR when under a load. Try bumping the LLC to 5 if you are having trouble with stability. If you are not having stability issues, then 1.23v under a load with a configured 1.3v setting, is probably not too bad.

Also, be sure you are looking at vcore and not VID. They are different.

VID is the voltage that the CPU requests from voltage regulator and which it "thinks" it is getting. Vcore on the other hand is the real monitored CPU voltage value that the CPU is getting. If the mainboard implements additional Vcore "offsetting" then the CPU doesn't know about this and so the difference between VID and Vcore can be explained.
 

one-called-kane

Honorable
Oct 8, 2013
29
0
10,530
0
Yeah, I was looking at VID, Vcore is much more stable. Now I'm having a different issue though. Now the Vcore shows that it's hovering around 1.21 during stress, but I get a crash even though the voltage is set to 1.3.

It isn't crashing due to temp so I'm assuming it's voltage but it doesn't seem to be using
as much as its able to. Do I keep cranking up the voltage? I set the LLC back to auto and it doesn't seem to be helping.
 

Darkbreeze

Titan
Moderator
LLC should probably not be on auto. I'd leave it on 4 or 5 personally. That's where my 6700k lives.

Bump the core voltage a bit. For 4.5Ghz, when I was running at that speed, I had mine set to 1.335v on both my Hero VIII and my Gaming 5 Z170 boards. You have to play around a bit to find the stable voltage but IIRC anything at 1.3v or less would not be stable at 4.5Ghz.

Below are screenshots of my BIOS settings if you'd like to take a look. Might be something there that's helpful, BUT, keep in mind that on every board and with every CPU, it will be somewhat different. Even different BIOS versions on the same board may require different configurations and settings. Every time a new BIOS has been released if I upgraded I'd have to do it all over again.

Here's my quick and dirty validation process.

Quick and dirty overview of overclocking/stability validation procedure.

Set CPU multiplier and voltage at desired settings in BIOS. Do not use presets or automatic utilities. These will overcompensate on core and other voltages. It is much better to configure most core settings manually, and leave anything left over on auto until a later point in time if wish to come back and tweak settings such as cache (Uncore) frequency, System agent voltage, VCCIO (Internal memory controller) and memory speeds or timings (RAM) AFTER the CPU overclock is fully stable.

Save bios settings (As a new BIOS profile if your bios supports multiple profiles) and exit bios.

Boot into the Windows desktop environment. Download and install Prime95 version 26.6.

Download and install either HWinfo or CoreTemp.

Open HWinfo and run "Sensors only" or open CoreTemp.

Run Prime95 (ONLY version 26.6) and choose the "Small FFT test option". Run this for 15 minutes while monitoring your core/package temperatures to verify that you do not exceed the thermal specifications of your CPU.

(This should be considered to be 80°C for most generations of Intel processor and for current Ryzen CPUs. For older AMD FX and Phenom series, you should use a thermal monitor that has options for "Distance to TJmax" and you want to NOT see distance to TJmax drop below 10°C distance to TJmax. Anything that is MORE than 10°C distance to TJmax is within the allowed thermal envelope.)

If your CPU passes the thermal compliance test, move on to stability.

Download and install Realbench. Run Realbench and choose the Stress test option. Choose a value from the available memory (RAM) options that is equal to approximately half of your installed memory capacity. If you have 16GB, choose 8GB. If you have 8GB, choose 4GB, etc. Click start and allow the stability test to run for 8 hours. Do not plan to use the system for ANYTHING else while it is running. It will run realistic AVX and handbrake workloads and if it passes 8 hours of testing it is probably about as stable as you can reasonably expect.

If you wish to check stability further you can run 12-24 hours of Prime95 Blend mode or Small FFT.

You do not need to simultaneously run HWinfo or CoreTemp while running Realbench as you should have already performed the thermal compliance test PLUS Realbench will show current CPU temperatures while it is running.

If you run the additional stability test using Prime95 Blend/Small FFT modes for 12-24 hours, you will WANT to also run HWinfo alongside it. Monitor HWinfo periodically to verify that no cores/threads are showing less than 100% usage. If it is, then that worker has errored out and the test should be stopped.

If you find there are errors on ANY of the stability tests including Realbench or Prime95, or any other stress testing utility, you need to make a change in the bios. This could be either dropping the multiplier to a lower factor or increasing the voltage while leaving the multiplier the same. If you change voltage or multiplier at ANY time, you need to start over again at the beginning and verify thermal compliance again.

A more in depth but general guide that is still intended for beginners or those who have had a small amount of experience overclocking can be found here:


*CPU overclocking guide for beginners



And the BIOS screenshots.



















 

one-called-kane

Honorable
Oct 8, 2013
29
0
10,530
0
I'll check again when I get home from work and see if your BIOS pics can shed some light for me. The issue I've been having now is the CPU absolutely refusing to be stable at anything over 4.4 GHz. For that it runs stable at ~1.33V, but increasing the multiplier is no bueno. I've tried stepping the voltage all the way up to 1.39 with no luck. It fails OOCT within ~5mins every time.
 

Darkbreeze

Titan
Moderator
Perhaps, and this is just a possibility not an assurance by any means, it's not the CPU causing the instability?

What is your memory configuration? Are all memory modules from a matched set?

What speed is the memory set at?

Is it running at the XMP profile or the default JEDEC profile currently, or have the timings been manually adjusted at all? Of note is the fact that until you 100% confirm the stability of your CPU overclock, you should not even set the memory at the advertised XMP profile. They should be at the stock JEDEC 2133mhz profile so that any problems you encounter will be identifiable as being related to the CPU overclock and no other factors. After validating stability, THEN you can set the memory to the XMP profile OR custom settings of your choosing.

Also, I would recommend that your forego using OCCT for testing stability. For now at least. I don't use it at all.

Try Realbench, as per the outline in my earlier post and guide. If you use Prime, be sure to use a non-AVX version like version 26.6 for thermal compliance. Other versions can be used too, by removing the AVX factor. AVX version WITH AVX can be used for stability testing, but you are going to get extremely high thermals.
 

one-called-kane

Honorable
Oct 8, 2013
29
0
10,530
0
Perhaps, and this is just a possibility not an assurance by any means, it's not the CPU causing the instability?

What is your memory configuration? Are all memory modules from a matched set?

What speed is the memory set at?

Is it running at the XMP profile or the default JEDEC profile currently, or have the timings been manually adjusted at all? Of note is the fact that until you 100% confirm the stability of your CPU overclock, you should not even set the memory at the advertised XMP profile. They should be at the stock JEDEC 2133mhz profile so that any problems you encounter will be identifiable as being related to the CPU overclock and no other factors. After validating stability, THEN you can set the memory to the XMP profile OR custom settings of your choosing.

Also, I would recommend that your forego using OCCT for testing stability. For now at least. I don't use it at all.

Try Realbench, as per the outline in my earlier post and guide. If you use Prime, be sure to use a non-AVX version like version 26.6 for thermal compliance. Other versions can be used too, by removing the AVX factor. AVX version WITH AVX can be used for stability testing, but you are going to get extremely high thermals.
Memory is 2 sticks, 16GB GSkill DDR4. This is exactly what I got from Newegg maybe ~18 months ago:
https://www.newegg.com/g-skill-16gb-288-pin-ddr4-sdram/p/N82E16820231888?Item=N82E16820231888
In the correct slots per the mobo's manual (2 & 4). Should be all stock, I have done nothing to tweak it at all. I started the OC from the Default settings in the BIOS, no XMP profile or any kind of modification. I'm trying to tweak as few things as possible at once to (I thought) make issues easier to diagnose. Haven't had any other issues that would point to memory issues that I have noticed, but I suppose memtest should clear that up for certain. Bad RAM would be annoying, but far less expensive than the alternatives I guess.

I won't be able to try any of what you shared in that link above until tonight unfortunately, but I'm hoping Realbench gives me better results than the other ones I've tried so far. What is your opinion of the "mainstream" stress testers? I've tried x264, XTU, CPU-Z and OCCT and the only ones I've had run a 4.5GHz OC for any length of time have been CPU-Z and XTU, but those seem to be considered weaker than the other options and after I failed the other two I stopped using them to test altogether.
 

Darkbreeze

Titan
Moderator
Make sure there are no newer BIOS updates than what you have installed. If there are, I'd update to the last version that was available BEFORE the Specter and Meltdown microcode patches in January of 2018. If you already have a version newer than January 2018, and there are even newer versions than what you have installed, then I'd go ahead and install the latest version that is available, if any.

Realbench is the best stress utility aside from Prime, that I've found. For Prime, especially for memory testing but for anything really, I like Blend or a custom configuration of the Blend mode tests which can be found in my memory guide here:


But mainly I use Realbench (Recently some users have found they are having errors due to Nvidia cards, I haven't seen this but worth keeping in mind), Prime95 and the x264 variants.
 

one-called-kane

Honorable
Oct 8, 2013
29
0
10,530
0
One thing of note I forgot to mention, the BCLK for my CPU seems to default in at 100.9 rather than the typical 100.0 flat. Is this intentional or should I manually back it off to 100.0? If nothing else it sets off my OCD when the numbers aren't all nice and even.
 

one-called-kane

Honorable
Oct 8, 2013
29
0
10,530
0
Run Prime95 (ONLY version 26.6) and choose the "Small FFT test option". Run this for 15 minutes while monitoring your core/package temperatures to verify that you do not exceed the thermal specifications of your CPU.
Silly problem, the link here: http://windows-downloads-center.blogspot.com/2011/04/prime95-266.html for v26.6 64 bit is not valid, and for the life of me I can't seem to find a valid link for anything but the newest version of Prime. Is there another mirror somewhere where I can download the old one?
 

Gmoney06ss

Reputable
Jul 3, 2015
698
103
5,190
15
My apologies. That is where I initially downloaded from and had bookmarked. Updating post now to avoid confusion.

Also just want to extend a thank you to darkbreeze, following your oc guide and memory troubleshooting guide, I finally achieved a stable oc on my 6600k!!
 

one-called-kane

Honorable
Oct 8, 2013
29
0
10,530
0
Well I didn't have much time to play with it last night, but long story short I'm pretty sure I lost the lotto (again) with this chip, so I'm now 2/2. My 4.4 GHz test passed thermals and stress but it looks like 4.5 may as well be on the moon. I was able to get I think ~1.37V to pass the 15min thermal check on Prime but it failed the Realbench test.

I'll probably check to see academically what it's actually going to take to run 4.5 but its probably in the ~1.4 range so a moot point anyways, since even at 4.4 I was hitting 90C in Realbench. Skylake has been pretty underwhelming.
 

Darkbreeze

Titan
Moderator
It could simply be your motherboard as well. Not a bad board, but definitely not a top notch overclocking board either.

On my Gigabyte Z170 gaming 5 the best I could get was 4.5Ghz. On my Hero VIII I can easily do 4.8Ghz, although I've dropped it 200mhz since it's a daily driver and I wish to prolong it's useful lifespan.
 

one-called-kane

Honorable
Oct 8, 2013
29
0
10,530
0
The real kicker is I cancelled my first order for a z170 Gaming 7 for the one I have now because the Gigabyte was only Like New and I wasn't sure I trusted it. I had hoped the newer hardware would make up the difference.
 

Darkbreeze

Titan
Moderator
Newer never specifically means "better". There is just as much "new" garbage and low quality product as there is "old" garbage and low quality product. And I'm not saying that board is either of those, but as I said, it's also not a model I'd typically look to as being an exemplary overclocker either. The Gaming 7 likely would have been the far better choice, all things considered.

I'd definitely double check all the BIOS settings though. Even if you can't use exactly what I have set it may give you an idea on a few things.
 

one-called-kane

Honorable
Oct 8, 2013
29
0
10,530
0
Yeah, lesson learned there. I'm definitely not done trying, it just takes so long to do each test and there are only so many hours after work, haha.

I'll cross check your BIOS values with mine again, although I think I've got about as much translated into mine as I could. Disabling Intel's Turbo in my BIOS resets the core multiplier to 'Auto', so I can't fiddle with that unfortunately.

I'm assuming Realbench spits out a dump file, but I didn't have a chance to look for it this morning before work. Is there anything specific in it I should be looking for regarding diagnosing errors or is it just going to be generic crash info?
 

ASK THE COMMUNITY

TRENDING THREADS