Zotac 1060 running capPed at 139mhz core clock

[yesyamaha]

Well I am having a spate of problems with my PC recently and the latest is that my 1060 3gb card is running running capped at 139 mhz even while gaming. (see it in afterburner, firestorm, gpu-z) I had the same issue before and you can read about it here if you want. I solved it by uninstalling the via hd audio (onboard audio) drivers and disabling the nvidia hd audio drivers in device manager. I have my Headset installed so I still have audio so I was okay with running the system like that.

Win10 Pro
I5 4550
zotac 1060 3gb
8gb ddr3 ram
psu: 500w Artis gold (its not as reputable as say corsair or cooler master but it does the job)


As I said, its again running capped and gpu-z shows the reason as pwr limit same as before.

Now I was having an issue till yesterday where my seagate baracuda 2 gb drive was running constantly at 100% in task manager and I solved that by changing the sata cable AND using a different power cable for my hd1 a samsung ssd and my hd2 the seagate baracuda drive.

I also disabled windows search and another windows service which helps with faster starts but I think its changing the power cable which solved it.

I did that yesterday and after both my hd's were running normally again, played for like 3 hours at least with my 1060 also running normally.

However when I booted my computer again today and launched PUBG, I was again getting low fps same as before and when I went into gpu-z it said perfcap -> pwr sure enough.

I am not sure if this caused by the psu supplying limited power or if its because of some other reason and buying a new psu just so I can test if this is the issue doesn't sound like a smart thing to do... can anyone please advice me on this...


p.s when I got my 1060 running normally, gpu-z was still giving vrel as a perfcap reason a lot of the time while gaming. I think Vrel means that it wasn't getting reliably constant voltage.

p.p.s I apologize for any mistakes in my english I am not a native speaker

 

[Nuckles56]

If you could provide the system specs that you have, that would also be quite helpful, especially what power supply you are using

 

[yesyamaha]

If you could provide the system specs that you have, that would also be quite helpful, especially what power supply you are using

oops sorry

Win10 Pro
I5 4550
zotac 1060 3gb
8gb ddr3 ram
psu: 500w Artis gold (its not as reputable as say corsair or cooler master but it does the job)

Also updated the original post with the specs.

Btw this psu is now at least 3 years old and I have used it with a geforce 670 previously and the 1060 for about a year...

 

[Vayra86]

1. Update to latest drivers, there were issues with 1060's drivers recently. Just so we can rule it out
2. DO NOT Flash a bios on these cards (did you, by chance?)
3. VRel is normal to see and does not point at power issues. 139mhz under load is not normal however

 

[yesyamaha]

1. Update to latest drivers, there were issues with 1060's drivers recently. Just so we can rule it out
2. DO NOT Flash a bios on these cards (did you, by chance?)
3. VRel is normal to see and does not point at power issues. 139mhz under load is not normal however

1. well I already have the latest driver version 397.64. I installed it when I first started having this issue and I installed it after using ddu in safe mode.
2. Yes I did flash a new bios I found on techpowerup.com site but it has the same version as the one I had originally. (btw maybe this is the reason the card was running boosted at 1898 mhz instead of 1506 when this perfomance cap issue was resolved the first time?)

edit:- I reinstalled the drivers just to be sure and sadly it didnt help

 

[Vya Domus]

(btw maybe this is the reason the card was running boosted at 1898 mhz instead of 1506 when this perfomance cap issue was resolved the first time?)

It's supposed to run at around 1900mhz out of the box without touching anything. You flashed it because you thought it's supposed to run at 1506 ? I don't understand.

 

[yesyamaha]

It's supposed to run at around 1900mhz out of the box without touching anything. You flashed it because you thought it's supposed to run at 1506 ? I don't understand.

no I flashed it when I was troubleshooting the exact same issue I have now: 1060 running capped at 139 mhz (gpu-z showed perfcap reason as pwr and gpu shark showed nvidia limiting policy and power limit reached status)

What I was saying in the last post is that when I did get the card to run normally after uninstalling my via hd audio drivers, I noticed it was running at 1898 mhz instead of 1506 mhz which I thought was the default clock for zotac 1060 3gb card...

 

[Vayra86]

no I flashed it when I was troubleshooting the exact same issue I have now: 1060 running capped at 139 mhz (gpu-z showed perfcap reason as pwr and gpu shark showed nvidia limiting policy and power limit reached status)

What I was saying in the last post is that when I did get the card to run normally after uninstalling my via hd audio drivers, I noticed it was running at 1898 mhz instead of 1506 mhz which I thought was the default clock for zotac 1060 3gb card...

Pascal cards will boost beyond the advertised clocks, its very normal behaviour. They keep boosting until they hit a thermal or voltage ceiling.

As for the BIOS flash: are there substantial pieces of evidence that detailed your problem and a specific BIOS update to fix this problem? Otherwise flashing BIOS on a GPU is a big no no and I would personally look at reverting that by flashing the official BIOS back on it. Not in the least because of warranty and narrowing down the cause. After you've done that clean install the latest driver and run a few benches at stock settings, see how it goes. If it doesn't boost proper, RMA. Before you waste that opportunity...

 

[yesyamaha]

but thats the thing I don't want to rma because its already a replacement (long story ) card from zotac. If I wait ten days and the same shit starts happening with the new card what will I do. Besides it ran fine for the first 2 days before this perfacp bullshit started happening and then again for like 3 days till today after I don't know what I did (it probably wasn't me disabling via hd audio drivers as I have been saying because its not helping now)

edit:- I will look into replacing the original bios (I know I am an idiot but I lost the rom backup. Maybe zotac will giive me the bios if I ask them) and then take the PC to a decent shop because its now beyond me

 

[yesyamaha]

I cant take my PC/card anywhere today cuz its a local holiday but guess what I was playing around with services in windows and changed some services to start automatically and some to disabled and now the card is unlocked again. I am thinking this what helped me when I got this resolved before as well not uninstalling via audio. I had done a system restore to try to fix this other issue with my hard disks I mentioned in my op and thats probably why the card became locked again.

BUT even tho the card goes up to 1800's mhz in clock now its still saying perfcap -> pwr in gpu-z most of the time (or idle even while stress testing) unlike the last time I had it unlocked when it used to say idle when not in use and vrel when being used for gaming.

As for my orgi bios, zotac wrote back to me asking for my s/n no so they can give me the right bios but I am a bit confused should I really RMA this card when its clearly a Windows issue. The new updates are what TOTALLY MESSED MY SYSTEM UP!! SCREW MS!!!

edit: please ignore that last bit about pwr pefrcap. Thats because of the power limitation in the bios I have flashed. I am guessing as soon as I have the correct bios from Zotac flashed, this problem will also go away

 

[Vayra86]

Its weird though, Windows services that destroy card performance. Would be interesting to figure out which services those are and why the problem actually appears, so you know what you're really troubleshooting here. It sure as hell ain't the card, that much is clear now... so no, I would not RMA this with current knowledge.

 

[Vario]

Are you using GeForce Experience? Maybe try installing the drivers alone, should be an option when you do the install. I don't know if this will work but its worth a shot.

 

[thebluebumblebee]

Even though I don't put much faith in software voltage reading, I would like to see a screen shot from HWinfo.

Your PSU makes me nervous. I can find only limited information about it and even the manufacturer hides the label. The gold in the name refers to its color, not it's rating. The manufacturer mentions "70% High efficiency protection" but a listing on Flipkart says 90% efficiency, which I really doubt. Flipkart also says: Maximum Current: Current at DC Output of +12 V: 24 A which is only 288 watts, which indicates that this PSU is of a very old design - before 2003. A modern PSU will output almost all of its rated capacity in the 12V rail.

Btw this psu is now at least 3 years old

I would not be surprised if it is coming to the end of its life span.

I know that decent PSU's are hideously expensive in your part of the world, but I would suggest replacing that as soon as possible.

 

[newtekie1]

I cant take my PC/card anywhere today cuz its a local holiday but guess what I was playing around with services in windows and changed some services to start automatically and some to disabled and now the card is unlocked again. I am thinking this what helped me when I got this resolved before as well not uninstalling via audio. I had done a system restore to try to fix this other issue with my hard disks I mentioned in my op and thats probably why the card became locked again.

BUT even tho the card goes up to 1800's mhz in clock now its still saying perfcap -> pwr in gpu-z most of the time (or idle even while stress testing) unlike the last time I had it unlocked when it used to say idle when not in use and vrel when being used for gaming.

As for my orgi bios, zotac wrote back to me asking for my s/n no so they can give me the right bios but I am a bit confused should I really RMA this card when its clearly a Windows issue. The new updates are what TOTALLY MESSED MY SYSTEM UP!! SCREW MS!!!

edit: please ignore that last bit about pwr pefrcap. Thats because of the power limitation in the bios I have flashed. I am guessing as soon as I have the correct bios from Zotac flashed, this problem will also go away

I'd be interested in what services caused this. Are you sure they were even MS services? Lots of other applications will add services to Windows.

I'd trouble shoot some more, reversing the changes you made one by one, until you identify what service is actually causing the problem.

 

[eidairaman1]

Reset the nvidia control panel to defaults, if you are running a oc tool from zotac or msi etc , reset it to defaults in advanced settings too.

 

[yesyamaha]

Its weird though, Windows services that destroy card performance. Would be interesting to figure out which services those are and why the problem actually appears, so you know what you're really troubleshooting here. It sure as hell ain't the card, that much is clear now... so no, I would not RMA this with current knowledge.

Ive not had the time to go and look into the services I stopped otherwise I would try and give the exact list here. But at least a couple of services that I started said that they might cause system instability if turned off but were for some reason set to start manually! The same or other services I enabled also had to do with installing/discovering new hardware correctly. (I did install a new mouse logitech g502 on the eve of this perfcap issue)

Even though I don't put much faith in software voltage reading, I would like to see a screen shot from HWinfo.

Your PSU makes me nervous. I can find only limited information about it and even the manufacturer hides the label. The gold in the name refers to its color, not it's rating. The manufacturer mentions "70% High efficiency protection" but a listing on Flipkart says 90% efficiency, which I really doubt. Flipkart also says: Maximum Current: Current at DC Output of +12 V: 24 A which is only 288 watts, which indicates that this PSU is of a very old design - before 2003. A modern PSU will output almost all of its rated capacity in the 12V rail.

I would not be surprised if it is coming to the end of its life span.

I know that decent PSU's are hideously expensive in your part of the world, but I would suggest replacing that as soon as possible.

Fortunately I do have enough money saved for a new psu. I can afford to get a Corsair VS450 or maybe even a VS 550 which are entry level psus but they provide reliable continous voltage and more than what my PC needs apparently. I just wanted to be sure I need to change my psu because Amazon/Flipkart etc have a no returns policy for Power Supplies here so I can't return them saying "no longer needed" ...

Also, I do plan to troubleshoot this further when I get the time, here's a couple of screenshots of HWinfo :- https://www.dropbox.com/s/59o5r0chdpiwx5e/Screenshot 2018-05-19 10.58.26.png?dl=0 (copy pasta so the system is busy)

and another https://www.dropbox.com/s/37pq1635y6v9bpp/Screenshot 2018-05-19 11.23.29.png?dl=0

You know why that vmx thing is the red in both those screens?

Are you using GeForce Experience? Maybe try installing the drivers alone, should be an option when you do the install. I don't know if this will work but its worth a shot.

I accidentally did that a couple of times when I was troubleshooting this didn't help any sadly

Reset the nvidia control panel to defaults, if you are running a oc tool from zotac or msi etc , reset it to defaults in advanced settings too.

done both of those things sooo many times. I don't think its related to settings in Nvidia anymore. Its really my windows and/or my hardware

 

[newtekie1]

but were for some reason set to start manually!

Start manually is not the same as disabled or turned off. All setting a service to start manually means is the service won't start when the computer is booted, but the moment something runs that requires that service, it will start.

If you want a good list of the default service settings for Windows 10 check here: http://www.blackviper.com/service-configurations/black-vipers-windows-10-service-configurations/

It also gives a good list of a "safe" configuration of services that you can disable without really worrying about problems.

 

[thebluebumblebee]

You know why that vmx thing is the red in both those screens?

It means that Intel virtual machine extension is disabled, most likely in your BIOS. Unless you're trying to rum VM's, there's no need to worry about it.

Thanks for the screen shots, but would you do one with the Sensor window like this showing the voltages:

 

[yesyamaha]

It means that Intel virtual machine extension is disabled, most likely in your BIOS. Unless you're trying to rum VM's, there's no need to worry about it.

Thanks for the screen shots, but would you do one with the Sensor window like this showing the voltages:

Hey thank you for helping me with this infact!

Here's the screen https://www.dropbox.com/s/agopylo006qeu1k/Screenshot 2018-05-20 10.36.47.png?dl=0

and here's one showing gpu voltages https://www.dropbox.com/s/9nwfiv4v297y9vu/Screenshot 2018-05-20 10.40.08.png?dl=0

edit: btw the Hard Drive problem I mention in the original post, my HD2 goes to 100% disk usage in task manager and stays there for upto 20 secs at a time (I think opening the system and using seperate power cable for my ssd and my hd2 solved it first time. Also attached it to different sata cable) It completely halts my system but only when I am running stuff installed on D: and E:

In case you or anyone has any insight on this it would be welcome. It happening both the first time when this GPU performance cap issue was resolved and again when I resolved it this time is giving me doubts this too could be related to low power from my psu?

Damn!! If I could only be a 100% I need a new PSU lol

And I am only asking about the HD problem because I think it could be related

p.s I have tried various solutions for it I found on google already and windows gui chkdisk doesn't report any errors with the drive.

 

[eidairaman1]

Ive not had the time to go and look into the services I stopped otherwise I would try and give the exact list here. But at least a couple of services that I started said that they might cause system instability if turned off but were for some reason set to start manually! The same or other services I enabled also had to do with installing/discovering new hardware correctly. (I did install a new mouse logitech g502 on the eve of this perfcap issue)



Fortunately I do have enough money saved for a new psu. I can afford to get a Corsair VS450 or maybe even a VS 550 which are entry level psus but they provide reliable continous voltage and more than what my PC needs apparently. I just wanted to be sure I need to change my psu because Amazon/Flipkart etc have a no returns policy for Power Supplies here so I can't return them saying "no longer needed" ...

Also, I do plan to troubleshoot this further when I get the time, here's a couple of screenshots of HWinfo :- https://www.dropbox.com/s/59o5r0chdpiwx5e/Screenshot 2018-05-19 10.58.26.png?dl=0 (copy pasta so the system is busy)

and another https://www.dropbox.com/s/37pq1635y6v9bpp/Screenshot 2018-05-19 11.23.29.png?dl=0

You know why that vmx thing is the red in both those screens?


I accidentally did that a couple of times when I was troubleshooting this didn't help any sadly


done both of those things sooo many times. I don't think its related to settings in Nvidia anymore. Its really my windows and/or my hardware

https://www.hardwareinsights.com/corsair-vs450-review/4/

 

[thebluebumblebee]

Hey thank you for helping me with this infact!

Here's the screen https://www.dropbox.com/s/agopylo006qeu1k/Screenshot 2018-05-20 10.36.47.png?dl=0

and here's one showing gpu voltages https://www.dropbox.com/s/9nwfiv4v297y9vu/Screenshot 2018-05-20 10.40.08.png?dl=0

edit: btw the Hard Drive problem I mention in the original post, my HD2 goes to 100% disk usage in task manager and stays there for upto 20 secs at a time (I think opening the system and using seperate power cable for my ssd and my hd2 solved it first time. Also attached it to different sata cable) It completely halts my system but only when I am running stuff installed on D: and E:

In case you or anyone has any insight on this it would be welcome. It happening both the first time when this GPU performance cap issue was resolved and again when I resolved it this time is giving me doubts this too could be related to low power from my psu?

Damn!! If I could only be a 100% I need a new PSU lol

And I am only asking about the HD problem because I think it could be related

p.s I have tried various solutions for it I found on google already and windows gui chkdisk doesn't report any errors with the drive.

These last screen shots show your system running at full load on the GPU and the CPU??? You did not show me the motherboard voltages that I am looking for. Your GPU is running at full speed however. Both your CPU and GPU are running hot. What case do you have? How many fans do you have in the case and how are they oriented?

 

[yesyamaha]

These last screen shots show your system running at full load on the GPU and the CPU??? You did not show me the motherboard voltages that I am looking for. Your GPU is running at full speed however. Both your CPU and GPU are running hot. What case do you have? How many fans do you have in the case and how are they oriented?

I had PUBG running in background thats why its on full load. I didnt know you wanted to see motherboard stats sorry. Here they are https://www.dropbox.com/s/9ib0z23xbo01v0l/Screenshot 2018-05-20 20.45.58.png?dl=0 (busy now as well)

And yes I have a very basic cooler master case with just one big fan so yes my system runs hot. Normally I boost the cpu fan in bios which keeps all cores in green, gpu also remains below 50 at idle. I do plan to upgrade cpu/mobo in a couple of months so I am making do for now.

edit:- and please ignore my question about hard drive. I just had errors in file system on it. Chkdsk seem to have fixed it.

 

[Vayra86]

That GPU at 89 C with fans at nearly full throttle is not a good sign. If you applied extra voltage or higher power target I would suggest you take it back to stock. If you didnt, look at improving case airflow. Whats your ambient temp in the room?

Seeing those temps, it may also be related to your GPU downclocking. When you exceed temp limits like this and the driver cant push the temp back under 90 C by lowering voltage you will see heavy clock reductions. Perhaps youve even exceeded so far that it locked the card into lowest power state to avoid death.

 

[thebluebumblebee]

the 82°C cutoff beyond which Boost 3.0 will reduce clocks

You're losing performance due to temperatures. I'm assuming you've cleaned your system? If so, the first thing you need to address is airflow in your case.

 

[Final_Fighter]

can we get a picture of the inside of your case? also i recommend flashing the bios back to the original. there is very little to gain from flashing the bios on a pascal gpu unless you know what you are looking for. start from there and do a complete uninstall of the gpu drivers and revert back to version 368.81 and see if it persists.