[SOLVED] Could my PSU Cause micro-stutters?

Apr 12, 2021
28
0
30
I went from a 3070 to a RX 6900 XT and since day one I have been getting micro-stutters every few minutes when I game. I have tried a fresh install and format of my drive, undervolting, disabling resizable bar and numerous other things. The only thing I have left to try is upgrading my PSU (Currently a Thermaltake 850W Gold). I am wondering if this could be the culprit? Temps when gaming are great... 70-80 degrees tops, I have a lot of components in my rig and when I did the calculator, it said that I could peek at 740w. I've seen the card spike to 271w on a few occasions. This is my setup:
Gigabyte x570 Master
5900x
RX 6900
4 Dimms or 3600mhz ram (32gb)
2 NVME SSDs
11 Corsair Fans
Kraken AIO
2 Controllers for the fans
2 monitors About 8 USB devices including Corsair KB and mouse
4 LED strips.
I am out of options after this and I am just trying to get some feedback on if getting a 1000w 80 Gold PSU might help.
 
I went from a 3070 to a RX 6900 XT and since day one I have been getting micro-stutters every few minutes when I game. I have tried a fresh install and format of my drive, undervolting, disabling resizable bar and numerous other things. The only thing I have left to try is upgrading my PSU (Currently a Thermaltake 850W Gold). I am wondering if this could be the culprit? Temps when gaming are great... 70-80 degrees tops, I have a lot of components in my rig and when I did the calculator, it said that I could peek at 740w. I've seen the card spike to 271w on a few occasions. This is my setup:
Gigabyte x570 Master
5900x
RX 6900
4 Dimms or 3600mhz ram (32gb)
2 NVME SSDs
11 Corsair Fans
Kraken AIO
2 Controllers for the fans
2 monitors About 8 USB devices including Corsair KB and mouse
4 LED strips.
I am out of options after this and I am just trying to get some feedback on if getting a 1000w 80 Gold PSU might help.
How old is the psu in question?
 

FoxVoxDK

Distinguished
Did you try unplugging absolutely everything not essential? Just one monitor, keyboard and mouse, no other USB gadgets.
Have you disabled all start-up programs except Windows?
Have you updated BIOS, AMD chipset and GFX drivers?
The hang looks like something got worked on in the background, that is definitely not a micro-stutter, but almost a complete i/o freeze, like a disk write or something. xD
 
Apr 12, 2021
28
0
30
Did you try unplugging absolutely everything not essential? Just one monitor, keyboard and mouse, no other USB gadgets.
Have you disabled all start-up programs except Windows?
Have you updated BIOS, AMD chipset and GFX drivers?
The hang looks like something got worked on in the background, that is definitely not a micro-stutter, but almost a complete i/o freeze, like a disk write or something. xD
I've tried everything but unplugging all my gizmos. And the other monitor. I did Disable all background apps. I literally started a game right after doing a fresh install and not putting anything else on there. I thought the same thing as you though. Looks like it's processing something
The only things I have as usb besides my fan controllers and usb pump are my kb, mouse and speakers. I'm going to do some more troubleshooting and focus on that though. Is that normal? Even on high end gear?
 

FoxVoxDK

Distinguished
With computers, someone farts in another galaxy and you get a BSOD. Sometimes the weirdest and most obscure small detail is the culprit.

In your case, there's also the overhead issue, are you letting the 6900XT fly free? Unrestrained? Are you using AMD Freesync? Try setting Radeon Chill to something like 144Hz.
Try setting max 6900XT clock to min. 500 MHz and max 2500 MHz, if you keep smacking your head into the PPT limit of the card, it will clockdown and you'll see a small pause.
 
Apr 12, 2021
28
0
30
With computers, someone farts in another galaxy and you get a BSOD. Sometimes the weirdest and most obscure small detail is the culprit.

In your case, there's also the overhead issue, are you letting the 6900XT fly free? Unrestrained? Are you using AMD Freesync? Try setting Radeon Chill to something like 144Hz.
Try setting max 6900XT clock to min. 500 MHz and max 2500 MHz, if you keep smacking your head into the PPT limit of the card, it will clockdown and you'll see a small pause.
I'm sure my farts caused a few bsods. Honestly, amd software acts funny for me.. A lot of times it won't even open
I just have drivers only.
I've tried overclocking it and upping the power to 7%. Still does it. But I had 3 monitors on a 3070 fe with no issues 🤷‍♂️ I set the max frames in my game to 130... +10 above my monitor.
 

FoxVoxDK

Distinguished
Not really, Win 10 should be all good with the 5000 series power scheduling, there's no harm in choosing a higher plan like Performance or whatever is available.

Did you ever try the NVME drives one by one?
Last option would be RMA'ing the card, but man, during these times that would really be a poor situation to be in.
 
Apr 12, 2021
28
0
30
No reason not to, and also if at all possible, try with one drive.
Is there any way to run a test to see how the drives read and write under load? I ran a CrystalDiskMark check and got 6900 read 5100 Write. Whats my best option to run a test to see if its deffinitly something else processing in the background? I had task manager open while I ran a game and "System" used a log of power for a second right when the game stuttered.
 

Phaaze88

Titan
Ambassador
There should be 2 drive temperature readings under hwinfo. One is for the memory, the other should be the ASIC controller.
The controller normally is the hotter of the 2, and they appear to get real toasty on the Gen 4 drives.

Since it's a game drive, you should check it in game.
 
Apr 12, 2021
28
0
30
There should be 2 drive temperature readings under hwinfo. One is for the memory, the other should be the ASIC controller.
The controller normally is the hotter of the 2, and they appear to get real toasty on the Gen 4 drives.

Since it's a game drive, you should check it in game.
Oh check the drive temp in game? Never thought of that. I'd be shocked if I get heating issues in this system. I have 6 fans for intake, and a push /pull aio with 140mm
But it's definitely worth checking. I have to do relationship stuff right now but as soon as I'm let off the chain, I'll check it.
 
Apr 12, 2021
28
0
30
There should be 2 drive temperature readings under hwinfo. One is for the memory, the other should be the ASIC controller.
The controller normally is the hotter of the 2, and they appear to get real toasty on the Gen 4 drives.

Since it's a game drive, you should check it in game.
Ugh, I totally forgot to check the drive temps today but I Uninstalled one if the games, put it on another drive and same issue. Even switched the pci port to gen 3, and disabled above 4g decoding and resizable bar. I'm beginning to just come to the rationalize that I'm going to have to accept this 😢