• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

Stutter with jitter effect/ frame drops?

Sorry fell asleep last night didn't realize I had to have the game running, this is with a game running for maybe 5 mins, and I did send a bug report to Mordhau also emailed AMD directly to see what they could tell me, appreciate it as always fellas.
1630502492093.png
 
Just installed keyboard inspector the keyboard comes up as 125hz and the mouse comes up as 500-450hz (roughly)
Nothing to worry about then , I was asking cz I’ve seen similar behavior when mice a keyboards were set to very high polling rates(above 1500) once polling rates were set at around 1000 hz stuttering was gone . If you have the ability to test this gpu in another know good system that would speed up the diagnostic, otherwise I can’t think of anything else , good luck :)
 
Have you tested any other games, btw?
 
Have you tested any other games, btw?
Yeah I've tested a few (still waiting on some more to install) but so far I tested Mordhau, DOOM Eternal, Team Fortress 2, RDR2 and CoD WaW

Before doing configuring anything all those games were dropping frames and stuttering like crazy, after all the fixes it seems Mordhau is just a pain to get running, DOOM felt fine, TF2 was fine for the most part but got little stutters here and there, RDR2 was almost the same as Mordhau really stuttery and WaW was decent as well.


I did notice one thing last night while running a few games and I know we went over this but my cpu got to 91c in the matter of maybe 10 minutes playing a game, I know max safe temp is 95c but I'm cutting close? Or am I still completely fine in the temp scene? Sorry for my cluelessness once more lol.

Nothing to worry about then , I was asking cz I’ve seen similar behavior when mice a keyboards were set to very high polling rates(above 1500) once polling rates were set at around 1000 hz stuttering was gone . If you have the ability to test this gpu in another know good system that would speed up the diagnostic, otherwise I can’t think of anything else , good luck :)
Thanks man appreciate the help ^^, if I ever do get this solved I'll be sure to post here in case someone else has the same kind of problem as me.

Sorry for posting again but I just wanted to confirm one last time it's not throttling, I was playing a few games today and within 20 minutes the cpu reached 90-91c and stayed there for the remainder of the game (till I closed out) I know max safe temp is 95c but I feel I'm cutting it too close, maybe I should try with an aftermarket cooler first before taking it back?
 
was playing a few games today and within 20 minutes the cpu reached 90-91c and stayed there for the remainder of the game (till I closed out) I know max safe temp is 95c but I feel I'm cutting it too close, maybe I should try with an aftermarket cooler first before taking it back?
I don’t think it’s very serious, it’s definitely not ideal either , as long as the cpu is turbo boosting without throttling , though if you keep this system you should improve the cooling, could be an airflow issue and has nothing to do with heatsink performance ...
you can run hwinfo “sensors only “ and share a screenshot of all the sensors ( preferably while the cpu hits 90C ) the software will capture the max temps
 
The problem actually is when your CPU is reaching that high temps for a long time what are the temps of the VRM then?More often is that motherboard VRM have higher temps then the CPU and that Asus Prime B550m-a doesn't look that promising.....
Also Point of the LatencyMon is to stay turned ON for a while and when your stuttering/Laggs happens maybe LatencyMon could possibly pick up that and register what was the actual cause.....
 
Last edited:
I believe the psu is a evga 650w bronze 80+ but I have to crack open the case to find out for sure.
Being a bronze-rated PSU, you better open the PSU case without touching anything while energized and check for leaking or bulging caps!
 
Being a bronze-rated PSU, you better open the PSU case without touching anything while energized and check for leaking or bulging caps!
Wait, what? Unless the PSU is old or has been severely abused, there is zero reason to suspect bulging or leaking caps.

Also, a Bronze efficiency rating tells us next to nothing about the quality of the capacitors in use.
 
The problem actually is when your CPU is reaching that high temps for a long time what are the temps of the VRM then?More often is that motherboard VRM have higher temps then the CPU and that Asus Prime B550m-a doesn't look that promising.....
Also Point of the LatencyMon is to stay turned ON for a while and when your stuttering/Laggs happens maybe LatencyMon could possibly pick up that and register what was the actual cause.....
(Had game in windowed and latencymon on the other side to compare) The current measured interrupt jumps between 11-90ms but couldn't really equate it to stuttering in game, never seen that high dpc latency before so that's new.

I'm not sure if I can monitor my vrms, in hwinfo at least (correct me if I'm wrong I just can't find it in sensors)
 

Attachments

  • LatMon_SzWxu2ACyx.png
    LatMon_SzWxu2ACyx.png
    57.7 KB · Views: 104
  • LatMon_jjBxWsyQXS.png
    LatMon_jjBxWsyQXS.png
    154 KB · Views: 88
(Had game in windowed and latencymon on the other side to compare) The current measured interrupt jumps between 11-90ms but couldn't really equate it to stuttering in game, never seen that high dpc latency before so that's new.

I'm not sure if I can monitor my vrms, in hwinfo at least (correct me if I'm wrong I just can't find it in sensors)
Depends of the motherboard some of them have the sensors on the VRM or possibly in HWinfo VRM temps could be registered as system temps...also you maybe have temp sensor(on wire) that you could place near to the VRM and see the actual temps...your motherboard VRM can "handle" up to the 150c BUT thing is that everything above 100c on VRM ain't good especially if it stays on those temps for longer period of time....same thing goes for the CPU's it's not a big problem if your CPU reach 90c and works like that for a min or two BUT it could be a issue if it works on that temps for longer period of time.......Also IF your CPU reach that temps in gaming I assume it will easily reach 100+c if you do some CPU benchmark or some more intensive app that can push much more load on CPU....do me a favor and run Cinebench R23 then see if your CPU can finish that test at all what are the temps and compare your results to the others(same CPU) also you can leave latencMon run to see when pressure is only on your CPU and not on the GPU are you going to have any latency spikes....GL
 
Last edited:
Depends of the motherboard some of them have the sensors on the VRM or possibly in HWinfo VRM temps could be registered as system temps...also you maybe have temp sensor(on wire) that you could place near to the VRM and see the actual temps...your motherboard VRM can "handle" up to the 150c BUT thing is that everything above 100c on VRM ain't good especially if it stays on those temps for longer period of time....same thing goes for the CPU's it's not a big problem if your CPU reach 90c and works like that for a min or two BUT it could be a issue if it works on that temps for longer period of time.......Also IF your CPU reach that temps in gaming I assume it will easily reach 100+c if you do some CPU benchmark or some more intensive app that can push much more load on CPU....do me a favor and run Cinebench R23 then see if your CPU can finish that test at all what are the temps and compare your results to the others(same CPU) also you can leave latencMon run to see when pressure is only on your CPU and not on the GPU are you going to have any latency spikes....GL
I was able to run Cinebench 23 twice without any problems, CPU temps reached 81-82c and stayed there till the test was over (both times), here are screenshots of them running at once. as before the current measured interrupt was around 11-106 it just bounces back and forth from low to high numbers, managed to get directx kernel down a tad, I appreciate you trying to help.
 

Attachments

  • HWiNFO64_TMYOEC8U08.png
    HWiNFO64_TMYOEC8U08.png
    148.1 KB · Views: 87
  • Cinebench_ZCGyzfYoDN.jpg
    Cinebench_ZCGyzfYoDN.jpg
    227.5 KB · Views: 85
  • LatMon_A1TFVZF65M.png
    LatMon_A1TFVZF65M.png
    40.9 KB · Views: 78
I think in your screenshot you left some more sensors on the bottom (hwinfo) right ?
 
I think in your screenshot you left some more sensors on the bottom (hwinfo) right ?
Sorry here's the rest of the hwinfo tab (cinebench still running)
 

Attachments

  • HWiNFO64_7s5rPqQrFC.png
    HWiNFO64_7s5rPqQrFC.png
    97.5 KB · Views: 88
  • HWiNFO64_9QWR3O20St.png
    HWiNFO64_9QWR3O20St.png
    110 KB · Views: 74
  • HWiNFO64_TMYOEC8U08.png
    HWiNFO64_TMYOEC8U08.png
    148.1 KB · Views: 75
I was able to run Cinebench 23 twice without any problems, CPU temps reached 81-82c and stayed there till the test was over (both times), here are screenshots of them running at once. as before the current measured interrupt was around 11-106 it just bounces back and forth from low to high numbers, managed to get directx kernel down a tad, I appreciate you trying to help.
Hmm so then it seems like that CPU temps rise really high only when your GPU is also involved....well possibly it could be to much for that motherboard and it starts to heat when GPU&CPU working under heavier load which makes even more pressure after longer period of time.....Not sure maybe if you have some bigger FAN that you can mount on the VRM and see how that goes and if its going to help.....
 
Sorry, only skim-read three pages but have we nuked the OEM windows install yet and done a fresh install from scratch?

If no, then that is the first thing to try. As a low-volume corporate SI, I can only begin to imagine how many horrible things are embedded in the system image of larger scale SIs.

By all means backup your existing install with Macrium before wiping/reinstalling, but even if it solves nothing, I'd be much happier running a clean install of Windows than some SI image.
 
Sorry, only skim-read three pages but have we nuked the OEM windows install yet and done a fresh install from scratch?

If no, then that is the first thing to try. As a low-volume corporate SI, I can only begin to imagine how many horrible things are embedded in the system image of larger scale SIs.

By all means backup your existing install with Macrium before wiping/reinstalling, but even if it solves nothing, I'd be much happier running a clean install of Windows than some SI image.
Forgive me but I'm new to the tech related side of pc, is nuking the same as clean installing via a usb? If not I'll gladly back everything up and continue with that.
 
Sorry yes. When I say nuke I just mean wiping the drive which the Windows Installer can do for you as part of a clean install.
 
Sorry here's the rest of the hwinfo tab (cinebench still running)
All looks within the norma, I quickly checked some reviews on the mobo seems like it’s vrm can handle that cpu with no issues, my guess is that the hardware is fine , this is probably down to firmware issues bad bios bad drivers , and seeing gpu not being fully utilized could be game optimization issues, I would just return it and save some time , you can’t fix them all :(
 
Sorry yes. When I say nuke I just mean wiping the drive which the Windows Installer can do for you as part of a clean install.
Sorry for the late response I managed to nuke the drives during the clean install didn't seem to help, I did notice the blank screen glitch is due to amd gpu drivers (using ddu barebones I don't get that blank screen but ofc bad performance) starting to think its just amd?
All looks within the norma, I quickly checked some reviews on the mobo seems like it’s vrm can handle that cpu with no issues, my guess is that the hardware is fine , this is probably down to firmware issues bad bios bad drivers , and seeing gpu not being fully utilized could be game optimization issues, I would just return it and save some time , you can’t fix them all :(
Yeah I think that's the best case scenario here, would you even bother having MC look at it or just return/get a new model? I would love to build my own but getting a gpu is like getting the holy grail lol
 
What's the Seagate that doesn't work that you mentioned, is it plugged in, if so unplug it, if it's partially INOP it could be interfering with the Io from other drive's.
 
What's the Seagate that doesn't work that you mentioned, is it plugged in, if so unplug it, if it's partially INOP it could be interfering with the Io from other drive's.
I actually got that to work, dumb me forgot it wasn't formatted so I had to go into disk management and reformat. but it is unplugged the only things I have plugged into the rig is keyboard and mouse and my pair of headphones.
 
I actually got that to work, dumb me forgot it wasn't formatted so I had to go into disk management and reformat. but it is unplugged the only things I have plugged into the rig is keyboard and mouse and my pair of headphones.
Fair enough, I have had failing data drives do weird things.
Other AMD users don't have this though, have you tried messing with memory timings, loosen them perhaps.
 
Sorry for the late response I managed to nuke the drives during the clean install didn't seem to help, I did notice the blank screen glitch is due to amd gpu drivers (using ddu barebones I don't get that blank screen but ofc bad performance) starting to think its just amd?
So you've ruled out software and third party utilities with a clean install so now we're talking either a hardware or BIOS issue.
  • I'd be tempted to disable XMP and test the system for stutters at 2133MHz RAM speeds
  • I'd also be tempted to update to the latest Prime B550 firmware as the GPU is newer than your board. Black screen glitch shouldn't happen as the Powercolor card should work fine with the windows default display adpater drivers and anything downloaded from Windows Update after that.
  • Unplug anything that isn't 100% necessary to test to make sure it's not some weird perhipheral problem giving you grief. USB headphones, USB hubs, drives, gamepads etc.
If you still have issues with it after that then it's likely a hardware fault. The stock cooler may throttle but it shouldn't cause the issues you're seeing even if it throttles. The PSU would result in crashes, lockups, restarts - not intermittent pauses and stuttering. Either way, you've gone beyond due diligence at that point and should just return it for a refund.

So you've ruled out software and third party utilities with a clean install so now we're talking either a hardware or BIOS issue.
  • I'd be tempted to disable XMP and test the system for stutters at 2133MHz RAM speeds
  • I'd also be tempted to update to the latest Prime B550 firmware as the GPU is newer than your board. Black screen glitch shouldn't happen as the Powercolor card should work fine with the windows default display adpater drivers and anything downloaded from Windows Update after that.
  • Unplug anything that isn't 100% necessary to test to make sure it's not some weird perhipheral problem giving you grief. USB headphones, USB hubs, drives, gamepads etc.
If you still have issues with it after that then it's likely a hardware fault. The stock cooler may throttle but it shouldn't cause the issues you're seeing even if it throttles. The PSU would result in crashes, lockups, restarts - not intermittent pauses and stuttering. Either way, you've gone beyond due diligence at that point and should just return it for a refund.
Just read back through this thread, looks like @Czaxi has already covered the first two, so if it's not a connected device causing the stuttering you probably should take it back to Microcenter and try another one.

I've had several Asus Prime B550 boards running Zen3 without issues. Can't speak for the 6700XT as I'm bulk-buying 3060s at the moment but if this was a widespread problem with 6700XTs it would be far more prevalent on Reddit and other forums for sure.
 
So you've ruled out software and third party utilities with a clean install so now we're talking either a hardware or BIOS issue.
  • I'd be tempted to disable XMP and test the system for stutters at 2133MHz RAM speeds
  • I'd also be tempted to update to the latest Prime B550 firmware as the GPU is newer than your board. Black screen glitch shouldn't happen as the Powercolor card should work fine with the windows default display adpater drivers and anything downloaded from Windows Update after that.
  • Unplug anything that isn't 100% necessary to test to make sure it's not some weird perhipheral problem giving you grief. USB headphones, USB hubs, drives, gamepads etc.
If you still have issues with it after that then it's likely a hardware fault. The stock cooler may throttle but it shouldn't cause the issues you're seeing even if it throttles. The PSU would result in crashes, lockups, restarts - not intermittent pauses and stuttering. Either way, you've gone beyond due diligence at that point and should just return it for a refund.


Just read back through this thread, looks like @Czaxi has already covered the first two, so if it's not a connected device causing the stuttering you probably should take it back to Microcenter and try another one.

I've had several Asus Prime B550 boards running Zen3 without issues. Can't speak for the 6700XT as I'm bulk-buying 3060s at the moment but if this was a widespread problem with 6700XTs it would be far more prevalent on Reddit and other forums for sure.
Yeah I just clean installed one more time, made absolutely sure I had all mobo drivers installed tried the most recent gpu driver (driver only no radeon software this time) and it still stutters not even two minutes into gameplay, I'm now noticing a blank screen glitch on restarts and sometimes in desktop, safe to say this one is just a dud, I'm unsure whether I should try another of the same model or take a performance hit on the 3060 rig, regardless I appreciate your help and everyone else here who pitched into this post I would've never been able to troubleshoot all these steps alone.
 
Back
Top