Is my card dying? (Sudden FPS drops in-game affecting whole PC)

[bluewaves]

First off, specs: http://i.imgur.com/XpIn2sc.png
^ CPU/GPU-Z stats are just after booting up my PC.

So, recently I've been playing Witcher 3. Every time I started up the game, it crashed about 5 minutes into it so I had to restart the PC then it worked just fine for as long as I played (this became a usual routine). It started to get annoying so I decided to do something about it, I saw somewhere that decreasing core clock (using Afterburner) can help the game run a lot smoother, so I did (to -35Mhz) and by the gods did it work (I was even able to run the game in High settings with smooth FPS (used to play in Medium before). Maintained temperature in the CPU was around 70-80C and the GPU ran on about 65-70C (when running the game, of-course). However, after a week, I began to suddenly experience huge FPS drops just after starting the game so I put back the core clock to 0 (where it was). This did not solve it, the FPS drops got worse. The moment I start up the game, I get 5 FPS at most and it even affects my whole PC (hiccups, lag and what not) when I minimize it to try to close it. I thought there was a problem with voltage (keep in mind I haven't got the slightest clue about overclocking, yet alone voltages). I went into BIOS to try to find some answers (I have OC Tweaker in my MB (Asrock Z77 Extreme4)). I saw that I had Advanced Turbo, CPU/GPU OC and all that Auto stuff enabled by default (which is weird because even if the OC was set to 4.something GHz, CPU-Z was reporting 3.4 Ghz which is my default. I paid no mind to and I proceeded to changing PLL voltage from around 1.83 (AUTO) to 1.84 (without knowing what I was doing), saved changes & exit to boot up and VOILA I kept getting BSOD everytime I tried to start my PC. So I went back to BIOS, changed PLL voltage to Auto and nope, still getting BSOD.. I went back once more, disabled Advanced Turbo and all that and it booted up perfectly fine. Now I'm back to where I was. Some other stuff I tried:

- Setting Power Options to High Performance and tried every power mode on NVIDIA control panel (Optimal was default, I tried Adaptive and Maximum).
- Upgraded NVIDIA driver to the latest version (didn't work), downgraded to 347.88 (didn't work), upgraded to 365 (didn't work), upgraded a little higher to 372 (didn't work). I also used DD-U everytime when uninstalling the driver.
- Tried running the game on the lowest of the low settings.
- After all that BIOS adventure, I tried overclocking using Afterburner, set power limit 132 (which is apparently the maximum for my card), set core clock to around +70 MHz. This actually helped a little but nah the FPS was still incredibly low (although it stopped affecting my PC). Temperature in the CPU & GPU was for some reason at around 60C which is weird even after overclocking (as I said above, when I underclocked it was 70-80C for the CPU and around 70C for the GPU). Basically both my GPU & CPU stopped working as hard, became lazy.

I have now set everything back to normal, still experiencing this 5 FPS (that is also affecting my PC). Is my CPU/GPU dying because I underclocked it before? Are they just throttling? What is going on here? I'll say this again, I have no idea how to overclock and all that voltage stuff. So please be gentle.

EDIT:

If it is of any relevance, my PSU is: Corsair 600W Gaming Series

EDIT 2:

Wow, okay. I just ran Heaven Benchmark 4.0 (with nothing overclocked, just to see the FPS on High setting), was getting around 15 then after closing it, my screen suddenly started going black (on and off) and simultaneously started glitching. I just did a hard shutdown. That was scary.

 

[aquielisunari]

First off I would perform a CMOS reset. Tinkering with voltages and all that may have contributed to this debacle. You will have to go into your BIOS and set your boot options and maybe the time as well as any other personal changes you may have made.

Second I would buy new RAM. Not just one more stick. Purchase a dual channel RAM kit. The fluctuating RAM speed suggests a fluctuating CPU frequency. Is that due to heat(thermal throttling) or one of Intel's features such as Speedstep? My RAM speed is constantly at 1866MHz. Odd fluctuation. Can you say fluctuation 5 times fast?

Click on your Start button and type mdsched and then click enter. Type y and press enter. Type exit and press enter. Restart your computer. That will test your memory.

I'm betting on RAM or an unstable GFX card. In Nvidia control Panel click help and then Debug mode and try gaming again. 15FPS strongly suggests the integrated GPU but your monitor isn't plugged into your motherboard.

I can no longer access my BIOS after switching default graphics adapter from "PCI Express" to "Onboard" as per the other guy's suggestion. My PC boots up but the screen is black while the sound and stuff is on. After I do a hard shutdown it skips BIOS and boots up to login page, when I restart, the black screen is back. I tried resetting the BIOS using a "Reset" button on the motherboard, it restarted the PC but there were no changes. I also cannot seem to find a CMOS battery to do it manually (I have an Asrock Z77 Extreme4).

Did you move your monitor's cable from the GFX card over to your motherboard's GFX connection after that change? Did you remove the GFX card?

Your CMOS battery is hiding beneath your GFX card.

Ah yes, forgot to move the cables. Anyhow, I was able to reset the MB by moving the CMOS jumper. Also I ran the memory diagnostics, no issues detected. Perhaps the reason why the RAM clock was set to 333 was probably because I was tinkering with Afterburner and forgot to revert to default settings. Also, looks like I'm already in debugging mode: http://i.imgur.com/CzlcT3J.png

Odd. What's your core clock at in your game? You can set MSI's OSD to monitor and display that info.

I'm getting 1.7 FPS while the core clock is at 1097 MHz, the game is crashing 10 seconds in (black screen, I have to do hard shutdown).

@Tolis_GR

No I did not touch any voltage option in Afterburner, also I included both CPU-Z, GPU-Z and Speccy specs in my original post.

What does your event viewer say for the time of the crash? Leaning towards bad card(your original thought). Still, I'd go to 16GB RAM.

 

[bluewaves]

First off I would perform a CMOS reset. Tinkering with voltages and all that may have contributed to this debacle. You will have to go into your BIOS and set your boot options and maybe the time as well as any other personal changes you may have made.

Second I would buy new RAM. Not just one more stick. Purchase a dual channel RAM kit. The fluctuating RAM speed suggests a fluctuating CPU frequency. Is that due to heat(thermal throttling) or one of Intel's features such as Speedstep? My RAM speed is constantly at 1866MHz. Odd fluctuation. Can you say fluctuation 5 times fast?

Click on your Start button and type mdsched and then click enter. Type y and press enter. Type exit and press enter. Restart your computer. That will test your memory.

I'm betting on RAM or an unstable GFX card. In Nvidia control Panel click help and then Debug mode and try gaming again. 15FPS strongly suggests the integrated GPU but your monitor isn't plugged into your motherboard.

I can no longer access my BIOS after switching default graphics adapter from "PCI Express" to "Onboard" as per the other guy's suggestion. My PC boots up but the screen is black while the sound and stuff is on. After I do a hard shutdown it skips BIOS and boots up to login page, when I restart, the black screen is back. I tried resetting the BIOS using a "Reset" button on the motherboard, it restarted the PC but there were no changes. I also cannot seem to find a CMOS battery to do it manually (I have an Asrock Z77 Extreme4).

Did you move your monitor's cable from the GFX card over to your motherboard's GFX connection after that change? Did you remove the GFX card?

Your CMOS battery is hiding beneath your GFX card.

Ah yes, forgot to move the cables. Anyhow, I was able to reset the MB by moving the CMOS jumper. Also I ran the memory diagnostics, no issues detected. Perhaps the reason why the RAM clock was set to 333 was probably because I was tinkering with Afterburner and forgot to revert to default settings. Also, looks like I'm already in debugging mode: http://i.imgur.com/CzlcT3J.png

Odd. What's your core clock at in your game? You can set MSI's OSD to monitor and display that info.

I'm getting 1.7 FPS while the core clock is at 1097 MHz, the game is crashing 10 seconds in (black screen, I have to do hard shutdown).

@Tolis_GR

No I did not touch any voltage option in Afterburner, also I included both CPU-Z, GPU-Z and Speccy specs in my original post.

What does your event viewer say for the time of the crash? Leaning towards bad card(your original thought). Still, I'd go to 16GB RAM.

The description for Event ID 14 from source nvlddmkm cannot be found. Either the component that raises this event is not installed on your local computer or the installation is corrupted. You can install or repair the component on the local computer.

If the event originated on another computer, the display information had to be saved with the event.

The following information was included with the event:

\Device\Video3
1701(1de0) 00000000 00000000

the message resource is present but the message is not found in the string/message table

Also this warning:

Display driver nvlddmkm stopped responding and has successfully recovered.

I will try adding additional RAM sticks.

 

[Tolis_GR]

First off I would perform a CMOS reset. Tinkering with voltages and all that may have contributed to this debacle. You will have to go into your BIOS and set your boot options and maybe the time as well as any other personal changes you may have made.

Second I would buy new RAM. Not just one more stick. Purchase a dual channel RAM kit. The fluctuating RAM speed suggests a fluctuating CPU frequency. Is that due to heat(thermal throttling) or one of Intel's features such as Speedstep? My RAM speed is constantly at 1866MHz. Odd fluctuation. Can you say fluctuation 5 times fast?

Click on your Start button and type mdsched and then click enter. Type y and press enter. Type exit and press enter. Restart your computer. That will test your memory.

I'm betting on RAM or an unstable GFX card. In Nvidia control Panel click help and then Debug mode and try gaming again. 15FPS strongly suggests the integrated GPU but your monitor isn't plugged into your motherboard.

I can no longer access my BIOS after switching default graphics adapter from "PCI Express" to "Onboard" as per the other guy's suggestion. My PC boots up but the screen is black while the sound and stuff is on. After I do a hard shutdown it skips BIOS and boots up to login page, when I restart, the black screen is back. I tried resetting the BIOS using a "Reset" button on the motherboard, it restarted the PC but there were no changes. I also cannot seem to find a CMOS battery to do it manually (I have an Asrock Z77 Extreme4).

Did you move your monitor's cable from the GFX card over to your motherboard's GFX connection after that change? Did you remove the GFX card?

Your CMOS battery is hiding beneath your GFX card.

Ah yes, forgot to move the cables. Anyhow, I was able to reset the MB by moving the CMOS jumper. Also I ran the memory diagnostics, no issues detected. Perhaps the reason why the RAM clock was set to 333 was probably because I was tinkering with Afterburner and forgot to revert to default settings. Also, looks like I'm already in debugging mode: http://i.imgur.com/CzlcT3J.png

Odd. What's your core clock at in your game? You can set MSI's OSD to monitor and display that info.

I'm getting 1.7 FPS while the core clock is at 1097 MHz, the game is crashing 10 seconds in (black screen, I have to do hard shutdown).

@Tolis_GR

No I did not touch any voltage option in Afterburner, also I included both CPU-Z, GPU-Z and Speccy specs in my original post.

Some options left: Blow some air to the PCI slot to make sure it is not a matter of dust. Reseat your card gently and properly. Do the same for your RAM.
How old is your PSU? You have a power hungry card if i am not mistaken. Aged psus will drop their performance. Is yours 80+ certified or something? You can also check you 6+2 pin connector if it properly seated

 

[aquielisunari]

First off I would perform a CMOS reset. Tinkering with voltages and all that may have contributed to this debacle. You will have to go into your BIOS and set your boot options and maybe the time as well as any other personal changes you may have made.

Second I would buy new RAM. Not just one more stick. Purchase a dual channel RAM kit. The fluctuating RAM speed suggests a fluctuating CPU frequency. Is that due to heat(thermal throttling) or one of Intel's features such as Speedstep? My RAM speed is constantly at 1866MHz. Odd fluctuation. Can you say fluctuation 5 times fast?

Click on your Start button and type mdsched and then click enter. Type y and press enter. Type exit and press enter. Restart your computer. That will test your memory.

I'm betting on RAM or an unstable GFX card. In Nvidia control Panel click help and then Debug mode and try gaming again. 15FPS strongly suggests the integrated GPU but your monitor isn't plugged into your motherboard.

I can no longer access my BIOS after switching default graphics adapter from "PCI Express" to "Onboard" as per the other guy's suggestion. My PC boots up but the screen is black while the sound and stuff is on. After I do a hard shutdown it skips BIOS and boots up to login page, when I restart, the black screen is back. I tried resetting the BIOS using a "Reset" button on the motherboard, it restarted the PC but there were no changes. I also cannot seem to find a CMOS battery to do it manually (I have an Asrock Z77 Extreme4).

Did you move your monitor's cable from the GFX card over to your motherboard's GFX connection after that change? Did you remove the GFX card?

Your CMOS battery is hiding beneath your GFX card.

Ah yes, forgot to move the cables. Anyhow, I was able to reset the MB by moving the CMOS jumper. Also I ran the memory diagnostics, no issues detected. Perhaps the reason why the RAM clock was set to 333 was probably because I was tinkering with Afterburner and forgot to revert to default settings. Also, looks like I'm already in debugging mode: http://i.imgur.com/CzlcT3J.png

Odd. What's your core clock at in your game? You can set MSI's OSD to monitor and display that info.

I'm getting 1.7 FPS while the core clock is at 1097 MHz, the game is crashing 10 seconds in (black screen, I have to do hard shutdown).

@Tolis_GR

No I did not touch any voltage option in Afterburner, also I included both CPU-Z, GPU-Z and Speccy specs in my original post.

What does your event viewer say for the time of the crash? Leaning towards bad card(your original thought). Still, I'd go to 16GB RAM.

The description for Event ID 14 from source nvlddmkm cannot be found. Either the component that raises this event is not installed on your local computer or the installation is corrupted. You can install or repair the component on the local computer.

If the event originated on another computer, the display information had to be saved with the event.

The following information was included with the event:

\Device\Video3
1701(1de0) 00000000 00000000

the message resource is present but the message is not found in the string/message table

Also this warning:

Display driver nvlddmkm stopped responding and has successfully recovered.

I will try adding additional RAM sticks.

Another nail in the coffin for that 680. http://www.instructables.com/id/How-to-Fix-the-Nvlddmkm-Error/ has some possible fixes but it seems you have tried many TS steps before you got here, tossing the Hail Mary.

 

[bluewaves]

Another nail in the coffin for that 680. http://www.instructables.com/id/How-to-Fix-the-Nvlddmkm-Error/ has some possible fixes but it seems you have tried many TS steps before you got here, tossing the Hail Mary.

Some options left: Blow some air to the PCI slot to make sure it is not a matter of dust. Reseat your card gently and properly. Do the same for your RAM.
How old is your PSU? You have a power hungry card if i am not mistaken. Aged psus will drop their performance. Is yours 80+ certified or something? You can also check you 6+2 pin connector if it properly seated

An update to this thread:

I picked everything apart and put them all back with better cable management. This somehow magically fixed the issue. I still have no idea what the source of the problem was and how and why it got fixed. I won't complain. Anyhow, I just decided to take the safer route and ordered all new parts (CPU, GPU, MB). About time I upgraded this sucker. Thanks for your guys' help.

 

[Tolis_GR]

Another nail in the coffin for that 680. http://www.instructables.com/id/How-to-Fix-the-Nvlddmkm-Error/ has some possible fixes but it seems you have tried many TS steps before you got here, tossing the Hail Mary.

Some options left: Blow some air to the PCI slot to make sure it is not a matter of dust. Reseat your card gently and properly. Do the same for your RAM.
How old is your PSU? You have a power hungry card if i am not mistaken. Aged psus will drop their performance. Is yours 80+ certified or something? You can also check you 6+2 pin connector if it properly seated

An update to this thread:

I picked everything apart and put them all back with better cable management. This somehow magically fixed the issue. I still have no idea what the source of the problem was and how and why it got fixed. I won't complain. Anyhow, I just decided to take the safer route and ordered all new parts (CPU, GPU, MB). About time I upgraded this sucker. Thanks for your guys' help.

Keep in mind for future problems that reseating and cleaning your slots and connections is crucial. Many problems appear due to bad connectivity when the continuous change of operating temperature involves contraction and expansion