Question Bad stuttering in games with high consistent fps, no temps above 60c under load, no power issues (from what i can tell) and average cpu,gpu, ram usage

[mojaRx]

Like the title says, spent days now trying everything software related to help, computer is completely clean and been put back together twice. Have tried a clean reinstall of graphics drivers and windows itself, both from recovery settings and from a usb drive. Ive changed just about any setting that can be found and dozens of combinations of them. FPS is locked at 120 to match refresh rate and settings are at the bare minimum. The only part im able to swap out is the ram but doesnt seem like much of a hardware problem to me. The problem happens in pretty much every game ive tried except for lost ark and apex legends which only has a small one every 10-15 sec. Im at a loss, ive searched every issue i can come up with or find from digging in other forums but nothing really seems to fix it. However in some games like rust and the division 2, its not so bad if you just dont look around with your mouse, which i have also explored by trying a different mouse, different usb port, reinstalling drivers and letting windows install its own.

Would just like to get some direct input rather than trying to figure it out on my own before i start dumping money i dont have into new parts.

Games ive tried: The division 2, Rust, elden ring, lost ark, world war z, sea of thieves, apex legends (I can install others to try if need be.)

Sorry for confusion, heres a list of everything ive tried, will keep adding to it as i go along.
https://docs.google.com/document/d/1DGg0u7JpFkQDSFX4Iih6hTpbU68iRfCeycZJOnPST_o/edit?usp=sharing

Specs:
rog b550-f, not the wifi version
EVGA 3080ti xc3
r9 5900x
Corsair vengeance lpx 32gb 3600mhz ram
wd black 500gb nvme ssd for windows, the games are on a seperate wd green and crucial mx500 sata ssds
thermaltake 850w power supply
coolermaster ML240L AIO
case is a corsair 4000d if it matters

 

[Aeacus]

The only part im able to swap out is the ram but doesnt seem like much of a hardware problem to me.

When it comes to stutters, RAM is the most likely suspect, especially when using Ryzen CPU. But there can be other suspects as well. Still, lets explore the RAM issue 1st since there are two simple test you can do.

You say you have 3600 Mhz RAM. Does it also run at 3600 Mhz? Also what timings the RAM has? And does it have any other XMP profiles in BIOS, other than 2133 Mhz and 3600 Mhz?

When it comes to testing, idea is to run your RAM at slower frequency (with preferably tighter timings) to see if your stutters either lessen or get worse. Or if there is any change at all.

Frequencies to test: 2133 Mhz and 3000/3200 Mhz.

Your RAM for sure has 2133 Mhz profile when you look at it in BIOS. With 3000/3200 Mhz profiles, i'm unsure about. You need to check it from BIOS. Also, when timings are listed with each profile, do list them down as well, so that i could calculate either tighter or looser timings that you could also try.

 

[mojaRx]

When it comes to stutters, RAM is the most likely suspect, especially when using Ryzen CPU. But there can be other suspects as well. Still, lets explore the RAM issue 1st since there are two simple test you can do.

You say you have 3600 Mhz RAM. Does it also run at 3600 Mhz? Also what timings the RAM has? And does it have any other XMP profiles in BIOS, other than 2133 Mhz and 3600 Mhz?

When it comes to testing, idea is to run your RAM at slower frequency (with preferably tighter timings) to see if your stutters either lessen or get worse. Or if there is any change at all.

Frequencies to test: 2133 Mhz and 3000/3200 Mhz.

Your RAM for sure has 2133 Mhz profile when you look at it in BIOS. With 3000/3200 Mhz profiles, i'm unsure about. You need to check it from BIOS. Also, when timings are listed with each profile, do list them down as well, so that i could calculate either tighter or looser timings that you could also try.

https://imgur.com/hXR5EaL

This is the only option ive got. Have tried setting the frequency to 3000/3200 before but not lower. Rust and apex got better before i had even done anything but its still so much worse than it was before.

(also it was set to 3600 in memory frequency, changed it first then decided to just take a picture)

 

[Aeacus]

Under "AI Overclock Tuner", what other options you have, other than DOCP?

Try using JEDEC default, which for your kit is: 2133 Mhz, 15-15-15-36 timings and at 1.2 V.
Look if you get any difference (better or worse), compared to the DOCP profile of: 3600 Mhz, 18-22-22-42 at 1.35 V.

Another option would be using manual OC, whereby you input the specs manually. E.g 3000 Mhz, 18-22-22-42 (or for example 17-21-21-42 to tighten the timings) at 1.35V.
Tightening timings takes a lot of time, since you'd need to validate if the tighter timings hold with Memtest86.

Speaking of Memtest86, i suggest that you also run it to validate if your RAM is actually sound.
link: https://www.memtest86.com/
how to use: https://www.memtest86.com/tech_creating-window.html

With Memtest86, one full pass (all 14 tests) is bare minimum, 2 full passes are better, while 4 full passes is considered acceptable.
Time wise, it takes a LONG time. For 8GB of RAM, it takes ~1 hour. 2x 8GB would take ~2.5 hours and not ~2 hours (due to RAM configuration. Namely being on two separate DIMMs.). Since you have 2x 16GB, expect one full pass to take ~5 hours. So, 4 full passes for acceptable results, would take ~20 hours.

To test every possible data pattern inside RAM, it will take 32 full passes. (Test #7 is the one that needs 32 passes for all possible combinations.) But no-one in their right mind is going to to 32 full passes since it takes forever to complete, especially when you have high RAM amount to begin with. So, 4 full passes is enough.
Test descriptions: https://www.memtest86.com/tech_individual-test-descr.html

Recap:
* memtest86 4 full passes with DOCP profile of: 3600 Mhz, 18-22-22-42 at 1.35 V.
If errors are found, not all is lost, since you can also use JEDEC default. It could be possible that DOCP profile itself has issues and you need to reduce frequency and/or change the timings. After which;
* memtest86 4 full passes with JEDEC default of: 2133 Mhz, 15-15-15-36 at 1.2 V.
If errors are still found even with JEDEC default, buy new RAM. Preferably the one with lower CAS Latency that what you currently have (e.g 3600 Mhz, CL16).
If no errors are found, you could increase frequency to e.g 3000 Mhz with timings of 17-21-21-42 at 1.35V. But to validate if the new frequency/timings are sound, you have to run memtest86 again.

So, testing RAM doesn't cost any money, but it takes a TON of time to validate if frequency/timings are sound.

 

[mojaRx]

Under "AI Overclock Tuner", what other options you have, other than DOCP?

Try using JEDEC default, which for your kit is: 2133 Mhz, 15-15-15-36 timings and at 1.2 V.
Look if you get any difference (better or worse), compared to the DOCP profile of: 3600 Mhz, 18-22-22-42 at 1.35 V.

Another option would be using manual OC, whereby you input the specs manually. E.g 3000 Mhz, 18-22-22-42 (or for example 17-21-21-42 to tighten the timings) at 1.35V.
Tightening timings takes a lot of time, since you'd need to validate if the tighter timings hold with Memtest86.

Speaking of Memtest86, i suggest that you also run it to validate if your RAM is actually sound.
link: https://www.memtest86.com/
how to use: https://www.memtest86.com/tech_creating-window.html

With Memtest86, one full pass (all 14 tests) is bare minimum, 2 full passes are better, while 4 full passes is considered acceptable.
Time wise, it takes a LONG time. For 8GB of RAM, it takes ~1 hour. 2x 8GB would take ~2.5 hours and not ~2 hours (due to RAM configuration. Namely being on two separate DIMMs.). Since you have 2x 16GB, expect one full pass to take ~5 hours. So, 4 full passes for acceptable results, would take ~20 hours.

To test every possible data pattern inside RAM, it will take 32 full passes. (Test #7 is the one that needs 32 passes for all possible combinations.) But no-one in their right mind is going to to 32 full passes since it takes forever to complete, especially when you have high RAM amount to begin with. So, 4 full passes is enough.
Test descriptions: https://www.memtest86.com/tech_individual-test-descr.html

Recap:
* memtest86 4 full passes with DOCP profile of: 3600 Mhz, 18-22-22-42 at 1.35 V.
If errors are found, not all is lost, since you can also use JEDEC default. It could be possible that DOCP profile itself has issues and you need to reduce frequency and/or change the timings. After which;
* memtest86 4 full passes with JEDEC default of: 2133 Mhz, 15-15-15-36 at 1.2 V.
If errors are still found even with JEDEC default, buy new RAM. Preferably the one with lower CAS Latency that what you currently have (e.g 3600 Mhz, CL16).
If no errors are found, you could increase frequency to e.g 3000 Mhz with timings of 17-21-21-42 at 1.35V. But to validate if the new frequency/timings are sound, you have to run memtest86 again.

So, testing RAM doesn't cost any money, but it takes a TON of time to validate if frequency/timings are sound.

Under ai overlock tuner theres also auto and manual.

One of the first things i did after stress testing for temp issues was a memtest since the biggest issue game was rust (purely because its the most played currently), however i only did one pass. Can do the full one when i go to bed, have to use the computer to work still.

 

[mojaRx]

Under "AI Overclock Tuner", what other options you have, other than DOCP?

Try using JEDEC default, which for your kit is: 2133 Mhz, 15-15-15-36 timings and at 1.2 V.
Look if you get any difference (better or worse), compared to the DOCP profile of: 3600 Mhz, 18-22-22-42 at 1.35 V.

Another option would be using manual OC, whereby you input the specs manually. E.g 3000 Mhz, 18-22-22-42 (or for example 17-21-21-42 to tighten the timings) at 1.35V.
Tightening timings takes a lot of time, since you'd need to validate if the tighter timings hold with Memtest86.

Speaking of Memtest86, i suggest that you also run it to validate if your RAM is actually sound.
link: https://www.memtest86.com/
how to use: https://www.memtest86.com/tech_creating-window.html

With Memtest86, one full pass (all 14 tests) is bare minimum, 2 full passes are better, while 4 full passes is considered acceptable.
Time wise, it takes a LONG time. For 8GB of RAM, it takes ~1 hour. 2x 8GB would take ~2.5 hours and not ~2 hours (due to RAM configuration. Namely being on two separate DIMMs.). Since you have 2x 16GB, expect one full pass to take ~5 hours. So, 4 full passes for acceptable results, would take ~20 hours.

To test every possible data pattern inside RAM, it will take 32 full passes. (Test #7 is the one that needs 32 passes for all possible combinations.) But no-one in their right mind is going to to 32 full passes since it takes forever to complete, especially when you have high RAM amount to begin with. So, 4 full passes is enough.
Test descriptions: https://www.memtest86.com/tech_individual-test-descr.html

Recap:
* memtest86 4 full passes with DOCP profile of: 3600 Mhz, 18-22-22-42 at 1.35 V.
If errors are found, not all is lost, since you can also use JEDEC default. It could be possible that DOCP profile itself has issues and you need to reduce frequency and/or change the timings. After which;
* memtest86 4 full passes with JEDEC default of: 2133 Mhz, 15-15-15-36 at 1.2 V.
If errors are still found even with JEDEC default, buy new RAM. Preferably the one with lower CAS Latency that what you currently have (e.g 3600 Mhz, CL16).
If no errors are found, you could increase frequency to e.g 3000 Mhz with timings of 17-21-21-42 at 1.35V. But to validate if the new frequency/timings are sound, you have to run memtest86 again.

So, testing RAM doesn't cost any money, but it takes a TON of time to validate if frequency/timings are sound.

I can also try swapping out the ram for my old set, I would just have to take a good while to go find it, its buried in boxes since i moved earlier this year. Same set pretty much, vengeance lpx cl18 3600 except its a 2x8 instead

 

[Aeacus]

Under ai overlock tuner theres also auto and manual.

With "Auto" option, i'm unsure to which frequency/timings the MoBo would put the RAM. Thus, for RAM testing, select "Manual" option and manually define frequency/timings/voltage.

however i only did one pass. Can do the full one when i go to bed, have to use the computer to work still.

Yes, it would be best to let the memtest86 run overnight, when you're sleeping.

I can also try swapping out the ram for my old set, I would just have to take a good while to go find it, its buried in boxes since i moved earlier this year. Same set pretty much, vengeance lpx cl18 3600 except its a 2x8 instead

Well, you could. It would be faster time wise, than running 4 full passes on memtest86 for your 2x 16GB kit. Maybe you can even get rid of stutters (or at least, lessen them). Won't hurt to test.

 

[mojaRx]

With "Auto" option, i'm unsure to which frequency/timings the MoBo would put the RAM. Thus, for RAM testing, select "Manual" option and manually define frequency/timings/voltage.


Yes, it would be best to let the memtest86 run overnight, when you're sleeping.


Well, you could. It would be faster time wise, than running 4 full passes on memtest86 for your 2x 16GB kit. Maybe you can even get rid of stutters (or at least, lessen them). Won't hurt to test.

Well, good news is it seems to have resolved itself, turned the cpu undervolt off and kept the gpu undervolt (have done this exact thing multiple times keep in mind, by fully restoring to default in bios not changing them manually). This time it kept crashing in games, switched to a more stable undervolt at 925mv @ 1850mhz and it seems fine. Genuinely couldnt tell you whats changed, thats all I did. Going to keep testing for a while to see if it starts up again, but 30mins into the division 2, which was the biggest problem and its alright.

However whenever i tab out and tab back in from my second monitor it freezes, have to hit windows key twice and itll be fine, any ideas on that? Some micro stutters here and there as well, but could just be cause its a ubisoft game, you never know.

edit: another 30mins later and its started to stutter again, nowhere near as bad but it most definitely shouldnt be like this. temps are fine even with taking the undervolt off so im not sure.

this is what everything looks like right when it starts, can give you anything else you need as well. Tabbing out and back in fixes it for about 10 seconds, and its not temps as far as im aware cause its not even enough time for them to drop.

https://imgur.com/6OpjSBd

View: https://imgur.com/6OpjSBd

 

[Aeacus]

temps are fine even with taking the undervolt off so im not sure.

If temps are within reason, why undervolt CPU/GPU in the first place?

the games are on a seperate wd green sata ssd

I'm starting to suspect this one, since WD Green SATA SSD, while cheap, isn't nothing to brag about and can have reliability issues.

To test it, install the most severely impacted game(s) to your M.2 NVMe SSD and look if it makes a diff.
It could be the issue, since compared to PCI-E SSD, SATA SSD is far slower in terms of read/writes. Could be enough to cause the stutters you're experiencing.

 

[mojaRx]

If temps are within reason, why undervolt CPU/GPU in the first place?


I'm starting to suspect this one, since WD Green SATA SSD, while cheap, isn't nothing to brag about and can have reliability issues.

To test it, install the most severely impacted game(s) to your M.2 NVMe SSD and look if it makes a diff.
It could be the issue, since compared to PCI-E SSD, SATA SSD is far slower in terms of read/writes. Could be enough to cause the stutters you're experiencing.

The undervolt is because it gets very hot very quickly in my room haha, purely a comfort thing. I dont mind the temps being 70c+ its just not a good time for me personally. Those temps in the screenshot i sent are with the fan speed at 65% in msi afterburner and the window open while its 40F outside, under about half load. Still lagging in rust even at 60c and 30% usage on cpu and gpu, about 20 of the 32gb of ram used, so theres still an issue somewhere. (This is with gpu undervolt also off, rust doesnt really stress my pc too much, but the division 2 is a very different story.)

Someone else already had me try to install it on my windows ssd and completely unplug all of the sata ssds, power and all. I never moved the 2 games off of there but i can try it unplugging them again now that its running alright enough.

Maybe try moving windows over to the crucial mx500 instead?

Have also noticed that ram usage just caps out at 70% now, 10gb stuck in standby and about 2mb free right at this moment

 

[Aeacus]

Someone else already had me try to install it on my windows ssd and completely unplug all of the sata ssds, power and all. I never moved the 2 games off of there but i can try it unplugging them again now that its running alright enough.

At this point, it would be nice if you could list each and every troubleshooting step you've tried thus far, so that we don't have to guide you over those that you've already tried.

Maybe try moving windows over to the crucial mx500 instead?

Unlikely that it would solve the issue, unless your OS drive is the culprit. Still, doesn't hurt to test it as well.

and about 2mb free right at this moment

This can actually be the main culprit since when you're out of RAM amount all kinds of narly things can happen. Stutters are just one symptom.

Here, i'd look into ways to free up the RAM.

Standby RAM usually means that you have plenty of background programs in use. Those also are included that did run at some point but are put to "sleep". They still keep some RAM for themselves, just in case you'd activate the program again.

To see which programs hog your system resources, download and run Process Explorer,
link: https://docs.microsoft.com/en-us/sysinternals/downloads/process-explorer

Guide on how to understand it,
link: https://www.pcworld.com/article/406...ee-supercharged-task-manager-alternative.html

 

[mojaRx]

At this point, it would be nice if you could list each and every troubleshooting step you've tried thus far, so that we don't have to guide you over those that you've already tried.


Unlikely that it would solve the issue, unless your OS drive is the culprit. Still, doesn't hurt to test it as well.


This can actually be the main culprit since when you're out of RAM amount all kinds of narly things can happen. Stutters are just one symptom.

Here, i'd look into ways to free up the RAM.

Standby RAM usually means that you have plenty of background programs in use. Those also are included that did run at some point but are put to "sleep". They still keep some RAM for themselves, just in case you'd activate the program again.

To see which programs hog your system resources, download and run Process Explorer,
link: https://docs.microsoft.com/en-us/sysinternals/downloads/process-explorer

Guide on how to understand it,
link: https://www.pcworld.com/article/406...ee-supercharged-task-manager-alternative.html

I'll put together a list and put it in a google doc, meant to do that already but I've been quite busy. I did notice that my cpu has been hovering around 3.7-4.35ghz instead of the normal 4.6-4.8. Changed PBO settings and its running at the normal with good temps still. Im thinking cpu cooler isnt as strong still as i thought or something in the settings needs to be changed to accomidate. Its got fresh kryonaut on it so thats not it, remounted multiple times incase its mounting pressure. Everything is running so much better since changing those settings but its still stuttering a bit here and there. Sorry to drag you around in circles, I'm super anal about it being so smooth since I spent so much lol.

HWinfo had been giving bad readings on the clock speed this whole time, switched to putting it on rivatuner instead and found that out. I'll put the video I followed to change the PBO settings to confirm its not worse than it should be.


No clue whats with the ram though. This is a completely fresh barebones install of windows but I can keep trying to isolate it.

Heres the video.

View: https://www.youtube.com/watch?v=dfkrp25dpQ0&t=2s

 

[Aeacus]

I did notice that my cpu has been hovering around 3.7-4.35ghz instead of the normal 4.6-4.8. Changed PBO settings and its running at the normal with good temps still.
Heres the video.

Sorry to drag you around in circles, I'm super anal about it being so smooth since I spent so much lol.

I don't do well with OC or UV troubleshooting.

My take on this OC/UV is simple: if one needs better CPU, one should buy better CPU. OC as such, isn't worthwhile on latest chips since those run so efficiently off the bat, that the little gains in frequency one is able to achieve, isn't worthwhile.

For example;
Usually when CPU OC is done, it is all core OC, negating the effects of different Turbo Boost ratios on different core amounts.
But on the flip side, the latest CPUs are so fine tuned that they have little, if any OC headroom (frequencies over max turbo ratio).

Back in the day, with older CPUs, CPU OC was worthwhile.
E.g i have i5-6600K with 3.5 GHz base and 3.9 Ghz boost. With CPU OC, i could get it 4.5 Ghz all core (increase of 600 Mhz over boost), or with delid, ~4.7 Ghz all core (800 Mhz over boost). And there have been some delidded i5-6600K CPUs, that can hold 5 Ghz all core.

Essentially from Intel 12th gen and onwards, most chips out there can only hold all core stable 100-300 Mhz over max boost. That gain is so little, that CPU OC with current, highly efficient chips, isn't worthwhile. There won't be any meaningful performance increase.
If the headroom would be bigger, like it is with my 6th gen CPU, where on minimum, i look towards 600 Mhz increase over boost clocks (or up to 1.1 Ghz over boost, if very lucky with delidded chip), then CPU OC makes sense.

All-in-all, CPU OC is dying niche and outside of record breaking, isn't worth the effort anymore. Better to run stock clocks and let CPU to decide when to turbo up. Less energy waste and less heat production this way also.


And while the idea of undervolting is to get lower temps, one can easily just buy better CPU cooler and/or increase airflow inside their PC. Choking CPU of power (too low voltage) can cause all sorts of issues. From stutters all the way to complete system instability.

Intel chips are robust and can handle UV without much issues. But Ryzen chips are capricious and with those, i wouldn't poke around in BIOS. Better run default/auto settings and cross your fingers that RAM DOCP profile (aka XMP) is stable and doesn't cause any issues.

I kinda wonder, when you 1st built your system, did you run auto/default settings for CPU, and if you did, was it smooth then? Or did you jumped straight into BIOS to poke around in there (e.g UV)?

 

[mojaRx]

I kinda wonder, when you 1st built your system, did you run auto/default settings for CPU, and if you did, was it smooth then? Or did you jumped straight into BIOS to poke around in there (e.g UV)?

Used as is with near perfect performance with nothing in bios touched, not even XMP/DOCP enabled. Few weeks ago it slowly started to fall off until about 5 or 6 days ago when it just began stuttering unbearably in a majority of games, after a few days of troubleshooting everything seemed normal except for how games ran which led me to just ask someone. My one and only guess left is this cooler master aio just couldnt hold up to how long I use my computer while its stressed, though I've only had it for a little over a year and a half so im not sure about that either.

Makes no sense why changing that and forcing it up to 4.7 has helped but left a bit of stuttering behind, it always ran at a steady 4.6-4.8 with no issues so my only guess is cooling. Very roundabout way of saying its out of my scope pretty much lmfao.

Heres the list, its most definitely missing a bunch of things, its been almost a week of this and ive got the memory of a dead rock lol
https://docs.google.com/document/d/1DGg0u7JpFkQDSFX4Iih6hTpbU68iRfCeycZJOnPST_o/edit?usp=sharing

 

[Sassy51xx]

Might sound stupid.. But how full is your drive? and is it defragged, etc..? I had this problem few times.. 1st it was ofc drive was full specially when in a fight.. and 2nd time, drive was failing, constantly using 100% on anything... for something it should even be using 50% on..

 

[mojaRx]

Might sound stupid.. But how full is your drive? and is it defragged, etc..? I had this problem few times.. 1st it was ofc drive was full specially when in a fight.. and 2nd time, drive was failing, constantly using 100% on anything... for something it should even be using 50% on..

Not stupid at all, every drive has 100gb+ free minimum and the windows drive has 400gb free, OS and nothing else. Was just about to add a list with everything ive tried to the previous comment i made

 

[Sassy51xx]

Not stupid at all, every drive has 100gb+ free minimum and the windows drive has 400gb free, OS and nothing else. Was just about to add a list with everything ive tried to the previous comment i made

Possibly reinstalling direct x or microsoft visuals? from memory, was experiencing a similar gaming difficulty after doing something and this time was getting error codes.. But maybe direct x or microsoft visuals is bugging up on you.. Also Install latest version of java.. cant remember if it fixed a pass problem or not..

Worst case, reinstalling Windows 10 again could fix your issue. I know windows itself can get buggy and <Mod Edit> with the pc.

 

[mojaRx]

Possibly reinstalling direct x or microsoft visuals? from memory, was experiencing a similar gaming difficulty after doing something and this time was getting error codes.. But maybe direct x or microsoft visuals is bugging up on you.. Also Install latest version of java.. cant remember if it fixed a pass problem or not..

Worst case, reinstalling Windows 10 again could fix your issue. I know windows itself can get buggy and <Mod Edit> with the pc.

Im on my 4th reinstall of windows 11 in the last 3 weeks at this point, both from the recovery settings within windows and just wiping the drive and reinstalling it from a usb. Havent tried reinstalling DX or MVs yet, figured it wouldve installed with windows but its worth a shot, willing to try anything at this point

 

[Aeacus]

Heres the list, its most definitely missing a bunch of things, its been almost a week of this and ive got the memory of a dead rock lol
https://docs.google.com/document/d/1DGg0u7JpFkQDSFX4Iih6hTpbU68iRfCeycZJOnPST_o/edit?usp=sharing

At least it's something.

I'll ask around if someone has ideas as of what to troubleshoot next, since i'm running out of ideas as well. Since at this point, i'm unsure if it's software or hardware.

Clean Win install usually gets rid of all software issues, thus, it would indicate hardware issue. But the fact that your system ran smoothly for some time and it gradually got worse (especially gaming), it would point towards software issue, since hardware didn't change during that (but software most likely did, updates and such). Unless there was severe hardware degradation (very unlikely).

 

[Darkbreeze]

What is your current motherboard BIOS version? If it's been mentioned here already, I didn't see that at all. This should have been the first question anybody asked regarding anything that might even remotely be memory related.

Also, if you have not already, because I'm not SPECIFICALLY seeing this asked either, I would recommend going to the AMD website and downloading/installing the latest chipset driver package for your motherboard's B550 chipset. On AMD systems I recommend using the AMD supplied drivers because they update far more frequently than the motherboard manufacturer does. I would also go to the product page for your motherboard and make sure you have the latest network adapter(s) and audio controller drivers. If you've played around with various drivers along the way, I'd just go ahead and download and reinstall those anyhow because they are the only ones you should be using anyhow.

Also, if you have any specialized peripherals be sure to go to THAT manufacturer's website and get the latest drivers for those USB devices and install them as well. You'd be surprised how often the driver for a specialized keyboard or mouse can cause problems that make people tear their hair out thinking it's a hardware issue with the board or CPU or something else.

 

[Eximo]

Ryzen chips have been known to have memory channels go bad. Have you tried running both sticks on a single channel, then swapping to the other channel to see if the problem persists.

Then it would just be a matter of RMA through AMD.

 

[Darkbreeze]

Ryzen chips have been known to have memory channels go bad. Have you tried running both sticks on a single channel, then swapping to the other channel to see if the problem persists.

Then it would just be a matter of RMA through AMD.

This is also a thing. But, eliminate the basics first. Always.

 

[Sassy51xx]

Im on my 4th reinstall of windows 11 in the last 3 weeks at this point, both from the recovery settings within windows and just wiping the drive and reinstalling it from a usb. Havent tried reinstalling DX or MVs yet, figured it wouldve installed with windows but its worth a shot, willing to try anything at this point

Windows 11... issue right there mate. had my pc on windows 11 while back and nothing but problems on games.. cant remember the problem exactly only to know I was on my pc 6 hours to fix it and ended in redoing the whole system back to windows 10.. which wouldn't let me from settings cause past 7 days.. But something with windows 11, idkw I believe its certain combinations of specs that glitch/bug out idk...

 

[mojaRx]

What is your current motherboard BIOS version? If it's been mentioned here already, I didn't see that at all. This should have been the first question anybody asked regarding anything that might even remotely be memory related.

Also, if you have not already, because I'm not SPECIFICALLY seeing this asked either, I would recommend going to the AMD website and downloading/installing the latest chipset driver package for your motherboard's B550 chipset. On AMD systems I recommend using the AMD supplied drivers because they update far more frequently than the motherboard manufacturer does. I would also go to the product page for your motherboard and make sure you have the latest network adapter(s) and audio controller drivers. If you've played around with various drivers along the way, I'd just go ahead and download and reinstall those anyhow because they are the only ones you should be using anyhow.

Also, if you have any specialized peripherals be sure to go to THAT manufacturer's website and get the latest drivers for those USB devices and install them as well. You'd be surprised how often the driver for a specialized keyboard or mouse can cause problems that make people tear their hair out thinking it's a hardware issue with the board or CPU or something else.

Sorry forgot to add the list of stuff ive already done to the main post, bios version is the latest, 3404 (unless theyve updated again in the last few weeks). Same goes for chipset drivers and the ones for audio/network.
Have done the drivers for my keyboard but razer didnt want to make it easy for my mouse, Ive got the non-wifi version of my motherboard so no bluetooth, the driver installer wont let you continue without bluetooth 4.0 enabled.

Also meant to add that replacing the thermal paste on the cpu again, which was done only about 3 or 4 months ago, and remounting the cooler a couple times to get the pressure right made a huge difference, runs pretty good for about 20-30mins and it goes back to feeling sluggish with micro-stutters every 10-15 seconds or so. Thats why I was going on about this probably mediocre AIO starting to give out. Jumps up to 70c+ wayyyy faster than it should in a loading screen (Division 2 and rust in most of this context) as well.

https://docs.google.com/document/d/1DGg0u7JpFkQDSFX4Iih6hTpbU68iRfCeycZJOnPST_o/edit?usp=sharing

 

[Darkbreeze]

Windows 11... issue right there mate.

No.

cant remember the problem

Can't remember the problem, but you're so sure that it was Windows 11 itself, rather than something else going on IN Windows 11, that you're willing to make recommendations to other people based on that belief? That's just not reasonable and really not how we do things around here. We base recommendations on things that are realistic and can actually be pointed directly at, or at least semi-directly at, with some kind of evidence or probability based on an accumulation of prior proof from similar situations. Plural. A sample of one, even when it IS the thing that went wrong, is usually not very effective as a scapegoat.