Question Massive Degredation in PC Performance

Spielen1

Honorable
Jun 28, 2013
11
0
10,510
Hi everyone,

I've been using the same setup for years now and been enjoying 1440p gaming on High to Ultra settings with very consistently high framerates, but recently the performance in all my games has massively degraded and now I can't get a stable framerate in anything I used to play. The problem started about a week into playing Destiny 2 where I would experience huge frame variance and lag, and sometimes my PC would hard reboot with the kernel power event 41 showing in the event viewer. I was playing in the first week with a framerate cap of 120 and would rarely if ever dip below that, and now if I cap at 90 I can barely manage that, and will regularly tank down to 50-55fps.

I initially thought this was an issue with Destiny 2's performance so I made a thread about it on the forums there, but since checking with other games my performance has degraded massively and I can't maintain a stable framerate at 1440p with anything I used to play with ease at the same settings. Battlefield 5 now has massive frame variance and where I used to play with 90+ fps on high/ultra at 1440p I now go below 40 and in a 64 player lobby my framerate is all over the place. Hitman 2 lags horribly too when I used to have every setting maxed out and play at 80+ fps. I've kept logs in HWinfo64 and there's no signs of overheating in my GPU or CPU, and I've taken many steps to attempt to address the issues, including completely reinstalling my OS, formatting all drives, reapplying thermal paste and reseating the cooler tower. and more. My hardware has passed every test I've thrown at it so far so I'm out of ideas on what to do next.

Any ideas on what could be the culprit here? I can provide logs or further data on request. My DxDiag

My specs:
ASUS Z270E Motherboard
i7 7700k @ 4.2Ghz
ASUS ROG STRIX GTX 1080
16GB Corsair Vengeance DDR4 RAM
256GB M.2 SSD
Corsair 750W PSU
 
" sometimes my PC would hard reboot with the kernel power event 41 showing in the event viewer".

"I've been using the same setup for years now"

How old is that Cosair 750W PSU?

It may be reaching its' EOL (End of Life).

Two things you can do:

1) Also look in Reliability History for error codes and warnings that correspond with the degradation incidents.

2) Test the PSU.

https://www.lifewire.com/how-to-manually-test-a-power-supply-with-a-multimeter-2626158

Not a full test (not under load) but if any voltages are going out of tolerance then that in itself indicates a problematic PSU.
 
" sometimes my PC would hard reboot with the kernel power event 41 showing in the event viewer".

"I've been using the same setup for years now"

How old is that Cosair 750W PSU?

It may be reaching its' EOL (End of Life).

Two things you can do:

1) Also look in Reliability History for error codes and warnings that correspond with the degradation incidents.

2) Test the PSU.

https://www.lifewire.com/how-to-manually-test-a-power-supply-with-a-multimeter-2626158

Not a full test (not under load) but if any voltages are going out of tolerance then that in itself indicates a problematic PSU.

So when I did my newest build, I replaced everything except the case and the PSU, so the PSU is the oldest hardware in my build by quite some margin. I'll have to double check but it's at least 6 years old now so it's quite possible the EOL has been reached.

I've checked the reliability history but unfortunately I did a full reinstall of Windows a couple of weeks ago to rule out some potential causes and because I was long overdue a refresh (this is before I realized my performance was reduced in ALL games, not just Destiny) so my history only goes back a couple weeks. As I expected there's a lot of "Windows was not properly shut down" due to the amount of crashes I've experienced recently.

As for the Power Supply test, I'll do that at my earliest convienience. I hadn't got round to it yet because my performance outside of gaming has been fine so I assumed something as critical as the PSU would have affected me in other ways than the occasional shutdown, but after seeing how my overall FPS has tanked in all games I can't see many other options at this point. Maybe it's still in warranty...

Anyway thanks a lot for the reply and suggestion! I'll get back to you with my findings from the test.
 
Update on the situation: I've recently purchased a new Corsair HX750 and installed it on Monday. I managed to test Destiny 2 with 0 crashes the entire next day which is encouraging. However because the crashing has been extremely sporadic I'm not willing to call that issue completely fixed, but considering how old my previous PSU was it seems likely that was the issue, so thanks for the suggestion @Ralston18 !

Unfortunately my framerates haven't really improved at all. I assumed the crashing and low fps were related and caused by the same issue because they started around the same time, but I've tested Hitman 2, Battlefield 5 and Destiny 2 and all are way worse performing than they were a month ago, so now I'm thinking the low fps issue is coming from somewhere else.

I'm almost ready to rule out software being the issue because since the problem started I've reinstalled my OS + drivers and I basically have nothing on this PC except the above games I've mentioned and monitoring software. As for hardware, I've run many tests on pretty much every part in my PC but as I didn't do this extensively before the problem began and I didn't save any logs I have nothing to compare my results to. Pretty much the only thing I can rule out (other than the PSU) is overheating being a factor, as my setup runs very cool even under heavy load.

Any further ideas would be greatly appreciated!
 
There should be some accumulated error code history in Reliability History and Event Viewer. Go back and take another look at both. No need to immediately react to any given issues that are found. Just do some additional research and determine if there is some meaningful or relevant issue with respect to your build.

No harm in opening the case and doublechecking your build - something may have worked loose - just enough to make some connectivity intermittent.

Another path to take is to use Task Manager, Resource Monitor, and Performance Explorer to observe your computer's behavior and performance.

Use only one of three at a time to observe while; 1) system is just sitting there idling, 2) while you do light work, browsing, backups, av scans, etc., and 3) while gaming.

You may spot some bottleneck that develops and the process causing that bottleneck.

Take your time, be methodical, keep notes.
 
I've been using HWinfo64 to log data during gaming but I haven't been able to recognise any issue from the data I've captured. I've got logs from an extended play session of Destiny 2 and managed to capture a hard reboot in the logs but nothing jumped out at me as strange. I've checked over the internals many times and made sure to reconnect everything properly when installing my new PSU so the connections shouldn't be causing an issue.

As far as Event Viewer and the Reliability History goes, it still only reports the "Windows did not properly shut down" and kernel power event 41 associated with my reboot problem. I'll use the utilities you suggested for some further troubleshooting and testing. Mind if I upload a HWinfo64 log in a while to see if anyone can spot anything wrong that I couldn't? Cheers
 
To clarify. Have you experienced any random restarts since the psu replacement? Since the operating system reinstallation did you make sure 100% that your windows version is operating on the latest build? This would absolutely play a role in performance degradation. Perhaps your OS wasn't fully updated both pre and post clean installation. Next thing to check is load & thermals for both cpu and gpu if you're still experiencing degraded performance. Don't use task manager for load because it's inaccurate. On screen display via MSI Afterburner is my recommendation.
 
Last edited:
To clarify. Have you experienced any random restarts since the psu replacement? Next thing to check is load & thermals for both cpu and gpu if you're still experiencing degraded performance. Don't use task manager for load because it's inaccurate. On screen display via MSI Afterburner is my recommendation. Since the operating system reinstallation did you make sure 100% that your windows version is operating on the latest build? This would absolutely play a role in performance degradation. Perhaps your OS wasn't fully updated both pre and post clean installation.
Since installing my new PSU on Monday evening I haven't experienced any further random restarts, but considering that I've only tested it for a day I'm going to wait a while longer before I call that issue fixed.

I don't think my PC is experiencing any thermal or overheating issues as I've been keeping an eye on HWmonitor and both the GPU and CPU stay within reasonable boundaries. As for load, I'm going to take your advice on MSI Afterburner's on screen utility and do some testing in Hitman 2 with it soon.

I was keeping an eye on Windows updates before and after installation and I think I've been up to date (version 1903) throughout the process. I'm definitely on the right version now at least.
 
Did you plug one psu cable from the PSU to the graphics card? Using one dedicated PSU cable vs two can have a significant impact on performance. Two dedicated PSU cables going from the PSU to the GPU are recommended. Also now that you have a fresh installation of windows, make sure your power plan is set to high performance.
I've got both of the PCIe sockets on my card going to the PSU. Both use the same cable, but that's the setup I was using before the performance drop occured so I don't think that's the issue. I've also tried to move the card to the other slot on my motherboard but I still have the same results.

On a side note, my GPU has recently started making noise it hasn't made much before. Sort of like coil whine but a bit more "grindy"?? Kinda hard to describe a noise lol. I thought that the noise was being created by the combination of the GPU and my faulty PSU but after replacing the PSU the noises persist. Nothing looks out of the ordinary to me from monitoring usage, ram, clocks etc. I think I'm going to check the warranty status of my GPU and see if I can send it off to potentially get fixed. Worst case scenario is that they send it back unchanged I guess.
 
No luck with adding another cable unfortunately, I've done it and there's no difference in performance at all.

Think I'm going to have to send my GPU off and see if ASUS can diagnose something that I can't.