[SOLVED] Frametime spikes/Stutters/FPS drops in all games.

[pikko]

I'm joining the stutter-crew

Computer Type: Desktop
GPU: AMD Radeon RX 7800 XT (Sapphire nitro+)
CPU: AMD Ryzen 7 5800X 8-Core Processor
Motherboard: MPG B550 GAMING PLUS
BIOS Version: 1.F0
BIOS Release Date: 10/11/2023
RAM: 32 GB @

• - 16 GB DDR4-2666, G Skill Intl F4-3600C14-16GVK
• - 16 GB DDR4-2666, G Skill Intl F4-3600C14-16GVK

PSU: Seasonic Focus GX-750W
Case: be quiet! PURE BASE 500DX Window
Operating System & Version: Windows 11 Pro
GPU Drivers: Adrenalin Edition-Version 23.12.1
Chipset Drivers: AMD CHIPSET DRIVER VERSION 5.11.02.217 (earlier 5.01.29.2026)
Background Applications: DISCORD, Steam, MSI/AMD Software

Description of Original Problem:
I'm looking now for a solutions since a long time and did a lot. I read a lot on websites and trouble shooting, but never found a solutions, so I wanna try it now here.
My issue is, that I have micro freezes in every game, that a game stops really for a very very short time and then it keeps going fluently.

• Rogue Trader: When selecting Menues, or AoE-Debuffs
• Age of Wodners 4: Scrolling over the map
• Pathfinder 2: I noticed it when I killed an enemy and when he exploded
  • Also scrolling at an area with fire on houses, though when pressing pause and scrolling then, there is no stutter
• Darktide, when I hit enemies in Melee, or sometimes with range.
• That are just some examples.

Troubleshooting
What I have tryed so far:

• New Monitor
• New Ram
• Windows upgrade from 10 to 11
• Complete wipe, installation of w11
• Intallation of w11 on another SSD
• A-XMP p1 and p2 and none
• Removed sound card, speakers in on board
• USB front panel disabled (and enabled again)
• When I had w10, I did DDU with the new graphic cardI did a lot things, that I dont even remember.
• Mouse polling on 500/1000
• Benchmark 3DMark Speedway -> smooth, no stutter on 40 fps
• MSI util
• Both SSD

Idk, if this also might help, I tryed shortly a round of Pathfinder, got a few micro freezes/stutters for the frame times

 

[P4-630]

What I have tryed so far:

• New Monitor
• New Ram
• Windows upgrade from 10 to 11
• Complete wipe, installation of w11
• Intallation of w11 on another SSD
• A-XMP p1 and p2 and none
• Removed sound card, speakers in on board
• USB front panel disabled (and enabled again)
• When I had w10, I did DDU with the new graphic cardI did a lot things, that I dont even remember.
• Mouse polling on 500/1000
• Benchmark 3DMark Speedway -> smooth, no stutter on 40 fps
• MSI util
• Both SSD

Try Intel and Nvidia.





j/k

 

[Macro Device]

I'm joining the stutter-crew

Computer Type: Desktop
GPU: AMD Radeon RX 7800 XT (Sapphire nitro+)
CPU: AMD Ryzen 7 5800X 8-Core Processor
Motherboard: MPG B550 GAMING PLUS
BIOS Version: 1.F0
BIOS Release Date: 10/11/2023
RAM: 32 GB @

• - 16 GB DDR4-2666, G Skill Intl F4-3600C14-16GVK
• - 16 GB DDR4-2666, G Skill Intl F4-3600C14-16GVK

PSU: Seasonic Focus GX-750W
Case: be quiet! PURE BASE 500DX Window
Operating System & Version: Windows 11 Pro
GPU Drivers: Adrenalin Edition-Version 23.12.1
Chipset Drivers: AMD CHIPSET DRIVER VERSION 5.11.02.217 (earlier 5.01.29.2026)
Background Applications: DISCORD, Steam, MSI/AMD Software

Description of Original Problem:
I'm looking now for a solutions since a long time and did a lot. I read a lot on websites and trouble shooting, but never found a solutions, so I wanna try it now here.
My issue is, that I have micro freezes in every game, that a game stops really for a very very short time and then it keeps going fluently.

• Rogue Trader: When selecting Menues, or AoE-Debuffs
• Age of Wodners 4: Scrolling over the map
• Pathfinder 2: I noticed it when I killed an enemy and when he exploded
  • Also scrolling at an area with fire on houses, though when pressing pause and scrolling then, there is no stutter
• Darktide, when I hit enemies in Melee, or sometimes with range.
• That are just some examples.

Troubleshooting
What I have tryed so far:

• New Monitor
• New Ram
• Windows upgrade from 10 to 11
• Complete wipe, installation of w11
• Intallation of w11 on another SSD
• A-XMP p1 and p2 and none
• Removed sound card, speakers in on board
• USB front panel disabled (and enabled again)
• When I had w10, I did DDU with the new graphic cardI did a lot things, that I dont even remember.
• Mouse polling on 500/1000
• Benchmark 3DMark Speedway -> smooth, no stutter on 40 fps
• MSI util
• Both SSD

Idk, if this also might help, I tryed shortly a round of Pathfinder, got a few micro freezes/stutters for the frame times


View attachment 327424View attachment 327425

It might be connected to your high CPU temperature and clock instability. Reconfigure your CPU and/or CPU cooling the way it's 100% safe from throttling.

 

[pikko]

It might be connected to your high CPU temperature and clock instability. Reconfigure your CPU and/or CPU cooling the way it's 100% safe from throttling.

Hmm, that could be a thing. I just read, that it's normal for that specific cpu to get very hot. I changed the PBO, now it's cooler, though I think I have to change the "clock instability", but idk how

I don't understand it, I have stutter in all the games, but the Benchmark runs clean, even though the fps are kinda low. But it's smooth and 0 stutter:

A friend told me to disable PBO. I did that, and it seems, that I have no less micro freezes. Though they are not fully gone

 

[Macro Device]

less micro freezes. Though they are not fully gone

So it's definitely clock instability. Try lowering your CPU clock until it doesn't change whilst gaming at all.

 

[pikko]

So it's definitely clock instability. Try lowering your CPU clock until it doesn't change whilst gaming at all.

I got this while playing Darktide. And I will try to lower the clock, just have to figure out how ^^'
and btw thanks for helping
I also observered, I had the 200%+ values, after leaving the game. Could imagine, that's normal, since it says in the description, that it counts only during heavy load
In the game, it was in a test room in Darktide, it was most of the time at 110-130%, but I also had the 76% peak somewhere, but didnt realized when. Also the loading screen where high
I think , that one doesn't matter while gaming. The CPU Benchmark test was around 98%

 

[freeagent]

Looks like everything in your system is hot, your ram could use some attention.

 

[pikko]

Looks like everything in your system is hot, your ram could use some attention.

Themal Throttling is on "no" in HWInfo64, so it seems, it throttle didn't occur.
I had once some spikes with CCD1 at 94°C but in general the cpu is below 90°C. Also I have these micro freezes when the Temps are fine.
The Ram is switched out. I had 3200 before and now 3600 Mhz. The Ram itself can't be faulty (the settings in Bios still could be)
As far as I know, Ryzen CPUs boost hot all the time, so I'm not sure, if these hot temps are really an issue

Right now I try to figure out, how to lower the clocks like @Beginner Micro Device told me. It's set in Bios on 3800 Mhz and Auto, but can't change it :-|

Thanks for the answer

I did a CPU Benchmark test. And it seems fine. The Temps should be ok

 

[HD64G]

Just put a lower power limit in UEFI or Ryzen Master (100W?) until the hottest temp is below 85C and check again for stuttering.

 

[pikko]

Just put a lower power limit in UEFI or Ryzen Master (100W?) until the hottest temp is below 85C and check again for stuttering.

I will try it, though is this implying a thermal throttling? Because hwinfo tells me, that no temp throttling appears. Also I activated PBO again on Auto, because it was on Advanced on default (before I disabled it) and set power setting on "Balanced" in Windows. I got good temps, but still stutters appear. And thanks for the help

 

[Patterned1]

I just thought I’d chime in on this thread because it’s been so long. I’m convinced the core problem for most experiencing this issue is a “failing” graphics card. So yes, it was caused by excessive heat. I am going to try the “oven trick” to fix mine eventually and I’ll post back with my result. If that doesn’t work I’ll desolder and maybe use chipquik on it. So yeah, just my 2 cents on this super elusive issue.

 

[pikko]

So it's definitely clock instability. Try lowering your CPU clock until it doesn't change whilst gaming at all.

Could you help me with that? I was in Bios and saw the clock frequency at 3800 Mhz, it was set on Auto and greyed out. I couldn't change anything there. So I was wondering, do you mean to change it somewhere else in Bios? Or could you tell me, what exactly you mean or a guide or anything?

I just thought I’d chime in on this thread because it’s been so long. I’m convinced the core problem for most experiencing this issue is a “failing” graphics card. So yes, it was caused by excessive heat. I am going to try the “oven trick” to fix mine eventually and I’ll post back with my result. If that doesn’t work I’ll desolder and maybe use chipquik on it. So yeah, just my 2 cents on this super elusive issue.

I think in my case, it's not the graphic card, I think I had the same stutter all the time before I changed the gc. Before I just thought, it was my old 1070 ^^

 

[Macro Device]

do you mean to change it somewhere else in Bios?

Yes but I'm not fully aware of how to do it the most proper way with MSi AM4 boards. Will google the matter and come back in some time.

 

[Zach_01]

“Safest” way to keep a Zen2/3/4 temps lower than max operating value:

1. Set a platform max temp to a desired value in BIOS
2. (Optional), set a max PPT to a lower than stock value if all core workload is not a thing for you.

In particular a 5800X (all Zen3) has a 90C max operating temp. You can set 85 or 80C or whatever you like in BIOS.
142W is the default max PPT value of a 105W TDP Zen2/3 CPU.
If gaming is the main heavy usage of PC you can set it to something like 110-120W. It’s plenty for gaming.

Also you can use CurveOptimizer to alter the VF (Voltage/Frequency) curve of the core voltage by setting a negative value(s). Practically you are telling the CPU to clock the same with lower voltage, or clock higher with same voltage.

There is also an option for the max boosting clock of cores with steps of 25MHz (From -200MHz to +200MHz).
(Not wise to use high negative CurveOptimizer values with high positive boost clock limit =high system instability)

All the above settings are in the PBO Advanced section of BIOS.

———————————

EDIT:

PowerReportingDeviation has a meaning and a useful value ONLY under a constant 100% CPU load.
It’s completely useless for gaming loads.

The meaning of the value is as this:
We are loading a 142W PPT CPU with an all core load to 100% usage…
(PPT = PowerPackageTracking = The entire CPU socket power consumption)

1) If PRD is for example 110% while/during PPT is reporting 142W then the true power consumption of the entire CPU is:
142 / 1.1 = 129W
2) If PRD is for example 90% while/during PPT is reporting 142W then the true power consumption of the entire CPU is:
142 / 0.9 = 158W

One should read PRD/PPT and calculate true CPU package power always during the test of 100% load. Not later, not after load drops and test ends.
Only during 100% CPU load.

EDIT2:

In HWiNFO settings (of sensor window) “Snapshot CPU Polling” must be enabled to avoid/minimize the observer effect of software polling the CPU and disturb its parameters (for all kinds of loads).

 

[pikko]

“Safest” way to keep a Zen2/3/4 temps lower than max operating value:

1. Set a platform max temp to a desired value in BIOS
2. (Optional), set a max PPT to a lower than stock value if all core workload is not a thing for you.

In particular a 5800X (all Zen3) has a 90C max operating temp. You can set 85 or 80C or whatever you like in BIOS.
142W is the default max PPT value of a 105W TDP Zen2/3 CPU.
If gaming is the main heavy usage of PC you can set it to something like 110-120W. It’s plenty for gaming.

Also you can use CurveOptimizer to alter the VF (Voltage/Frequency) curve of the core voltage by setting a negative value(s). Practically you are telling the CPU to clock the same with lower voltage, or clock higher with same voltage.

There is also an option for the max boosting clock of cores with steps of 25MHz (From -200MHz to +200MHz).
(Not wise to use high negative CurveOptimizer values with high positive boost clock limit =high system instability)

All the above settings are in the PBO Advanced section of BIOS.

———————————

EDIT:

PowerReportingDeviation has a meaning and a useful value ONLY under a constant 100% CPU load.
It’s completely useless for gaming loads.

The meaning of the value is as this:
We are loading a 142W PPT CPU with an all core load to 100% usage…
(PPT = PowerPackageTracking = The entire CPU socket power consumption)

1) If PRD is for example 110% while/during PPT is reporting 142W then the true power consumption of the entire CPU is:
142 / 1.1 = 129W
2) If PRD is for example 90% while/during PPT is reporting 142W then the true power consumption of the entire CPU is:
142 / 0.9 = 158W

One should read PRD/PPT and calculate true CPU package power always during the test of 100% load. Not later, not after load drops and test ends.
Only during 100% CPU load.

EDIT2:

In HWiNFO settings (of sensor window) “Snapshot CPU Polling” must be enabled to avoid/minimize the observer effect of software polling the CPU and disturb its parameters (for all kinds of loads).

and @Beginner Micro Device , I might have solved it. I switched the outlet and a different outlet strip xD
I played Darktide, I still had stutter, but actually only in loading scenes or zone changes. While gaming, everything was fine, just got a bit weak performance, but no stutter, or maybe one

 

[Macro Device]

and @Beginner Micro Device , I might have solved it. I switched the outlet and a different outlet strip xD
I played Darktide, I still had stutter, but actually only in loading scenes or zone changes. While gaming, everything was fine, just got a bit weak performance, but no stutter, or maybe one

Nice! Sorry for absence, I had a very crazy week and totally forgot about your issue.

 

[pikko]

Nice! Sorry for absence, I had a very crazy week and totally forgot about your issue.

No problem, how you are fine now
I thought the whole time, that it could be something "hardware" related, even ordered a new keyboard, since this one is very old. But I talked then a lot about voltage, it started with your recommendation of lowering the clock, so I looked more up about voltage, then I talked with a friend until I thought about "wait a minute, everything of my pc is plugged into one outlet strip, maybe thats the issue" xD

 

[Macro Device]

how you are fine now

I wasn't in trouble, just a lot of work and parties. Everything is cool.

it started with your recommendation of lowering the clock, so I looked more up about voltage, then I talked with a friend until I thought about "wait a minute, everything of my pc is plugged into one outlet strip, maybe thats the issue" xD

See? It's great to have your own brain! Marvellous work.

 

[Zach_01]

and @Beginner Micro Device , I might have solved it. I switched the outlet and a different outlet strip xD
I played Darktide, I still had stutter, but actually only in loading scenes or zone changes. While gaming, everything was fine, just got a bit weak performance, but no stutter, or maybe one

Very nice that you solved the issue!

What I’m recommending above can apply to any Zen2/3/4 based PC, troubled or not.
Even when the system is issue free it’s nice to have options and tunes that the user can regulate to its needs.
For instance I don’t like my CPU running to the upper operating temp limit of 90C, so I’ve set temp limit to 75C and PPT to 115W (142W default) as I’m not using it for any kind of 100% load. And I still get the 4.95-4.97GHz occasional boost with an average of 4.5-4.6GHz during gaming.

 

[pikko]

Very nice that you solved the issue!

What I’m recommending above can apply to any Zen2/3/4 based PC, troubled or not.
Even when the system is issue free it’s nice to have options and tunes that the user can regulate to its needs.
For instance I don’t like my CPU running to the upper operating temp limit of 90C, so I’ve set temp limit to 75C and PPT to 115W (142W default) as I’m not using it for any kind of 100% load. And I still get the 4.95-4.97GHz occasional boost with an average of 4.5-4.6GHz during gaming.

It's sadly not 100% gone the stutter, but some games are way better than before. I might try some stuff you told before, though need to figure out, how it works, also not sure about setting voltage or anything, don't wanna kill my cpu or anything ^^

though as far as I know, it's normal for a Ryzen to get very hot, because it does these boosts. Wouldn't it be very limiting the cpu perfomance to set it very low?

 

[f-14]

tech yes city already did a video on this This is Why Ryzen 5 3600s Are FAILING!... as well as other youtubers years ago... not sure why nobody doesn't check those out: Ryzen 5000 Undervolting with PBO2 – Absolutely Worth Doing even tech tubers jayz2cents, linus, hardware unboxed made videos on this issue
the answer is the cpu cooling isn't keeping up, your mother board polling rates are set too low for hardware temp monitoring to detect the micro temp spikes before on die thermal throttling kicks in.
if you don't get a better cpu cooler or fix it's contact issue or socket bend and some other factors that happen either out of the box or over time, another solution is to thermal limit...most of this problem starts at 76ºC with the micro spikes going to around 94ºC (tomshardware when pabst blue ribbon was editor noted this thermal limit when intel core2 duo was using 95W. part of an old problem resurfacing that had forced intel and amd to abandon ceramic die cover after pentium and K6.3) which causes the stutter. and it's not just your cpu that has this thermal issue NVMEm.2 suffers from this also.
best solution i have seen was to under volt any where from -.1mv to -.2 and .2-.3 is silicon lottery if you are still able to push a PBO of +400mhz to 600mhz over stock speeds on lower tier cpus.
(now if you have a samsung nvme..there is a firmware issue you need to look into alsoif you havent heard about this by now)
you get more bang for your buck because factory binned silicon gets top ratings for clearing manufacturing as planned with little to no defects in waffer production. better production = better binning = higher% chance of over clocking ...but not always as the chip was graded at being able to meet the threshold in testing for that bin label.
every single waffer is designed to be the top of the lines...the ones that can't make it, instead of getting trashed get disabling 'fixes' to recoup costs rather than pay to send to the garbage dump just because certain parts of the silicon don't work, like cores failures or cache failures and worst smd interconnect failures which is how you get the low power grade chips still good enough to run the entertainment system in an automobile or run a mall kiosk

i thought by now every one knew this these days. i don't forum/social media much, more of a bob marley ghost from xmas's past this decade
it was fun when you could upgrade your cpu with a #2 pencil on amd athlon XP cpus despite it's binned rank
do the diligence in researching your parts and you can do things like run a 5800x3d with ddr4000 at 18Cas. the more performance and simpler/easier you want it to be means there is a $$$$$ value hurdle. spending the most doesn't get you more than confirming you have no value for others blood sweat and tears, time and energy that money represents but alot of times, if you want the best, your not paying for just a brand name when it comes to machines unless barbie and ken are involved. team hasbro btw...eff mattel<-- that joke is probably older than 98% of you

oh and one more thing...power supply matters...sine wave ripples...it's the huge problem with every shrink...it's 90% of the problem with 5nm and 3nm and 1nm becauswe of the cross voltagte 'talk'...radio waves external electrical heat production and static and solar flare radiation is a huge issue again.

 

[Zach_01]

It's sadly not 100% gone the stutter, but some games are way better than before. I might try some stuff you told before, though need to figure out, how it works, also not sure about setting voltage or anything, don't wanna kill my cpu or anything ^^

though as far as I know, it's normal for a Ryzen to get very hot, because it does these boosts. Wouldn't it be very limiting the cpu perfomance to set it very low?

R7 5800X is literally THE most difficult to cool Zen3 CPU from the entire desktop 5000series. It has the highest power consumption (142W) among other SKUs like 5900X, 5950X. But compared to those R9s there is a decisive difference.
5800X has 8cores and a single CCD (CoreComplexDie = core die, chiplet) meaning that whole CCD is active and almost(*) all that heat is coming from this 1 small CCD.
5900X is a 12core and has 2 CCDs each one with 6 active cores.
5950X has 16cores on 2 full active 8core CCDs. So the same almost(*) 142W heat is coming from a double the area compared to a 5800X
So 5800X has double heat density at best against the higher(core) 142W R9 SKUs.
This means that it’s very hard for any cooler (air or liquid) to be able to transfer all that heat fast enough, coming from a very small area under the CPU lid (IHS, Integrated Heat Spreader).

Ryzen CPUs boost until they hit max operating temp but the closer temp gets to upper limit the less they boost. You may see it hitting the max advertising clock but I assure you it’s less and less frequent as it’s closer and closer to that 90C limit.
So the avg boosting clock gets lower.

Of course, by installing a better cooling device you can improve avg clocks but this isn’t always an option for every user, and even some times what you spend for better cooling has very low returns in terms of boost.

So let me give you an example of my 5900X.
I use it primarily for gaming, other than browsing, watching movies and so. The reason I got a 12core is because I run a few VMs occasionally and I need the extra cores. I never use it for any kind of productivity tasks. So the single/few core(s) boost is important to me.
1. I’ve set the max platform temp from BIOS to 75C. A whole 15C below max operating temp.
2. I have set max CPU PPT from 142W to 115W, the EDC from 140A(ampere) to 105A.
3. Gave the VoltageFrequency curve a -7 (negative steps) to the primary CCD and -13 steps to secondary CCD.
4. Also I’ve set max frequency to +25MHz.

Results in gaming (1example)
Yesterday I was playing for about 4hours FarCry6.
After, I saw HWiNFO64 sensors to observe some stuff I’m interested in. (I had it reset values before I started gaming)
5900X has a max advertised boost clock of 4.8GHz.
The CPU had boosted cores to a max of 4.85-4.97GHz for primary CCD and a max of 4.8GHz for secondary. The avg boost clock for a couple of cores on primary CCD was above 4.8GHz and the rest around 4.5-4.6GHz.
Anyway games, even the demanding ones are run primarily on 1-2cores and have a bunch of secondary threads to a few other.
Avg. CPU PPT was around 100-105W.

As you can see I’m not loosing any performance by setting temp limit to 75C.
To be completely honest here I should declare that I have a 280mm CPU AIO cooler and use Liquid Metal for the heat transfer material from CPU to cooler.
I do not openly recommend LM as it’s a dangerous substance and can easily damage the system if misused.
On Ryzen CPUs with the small and off-center chiplets, thermal paste is crucial for cooling. There are few pastes slightly worse than LM that can improve heat transfer from the more “regular” ones.
Even some very good thermal pads exist nowadays.

I understand very well your concerns about CPU voltage.
But in curve optimizer when you set it to negative values you are subtracting voltage (undervoltage) from default. The worst thing that can happen if you apply too much negative steps is to create instability. No harm done. If it’s too much you go a couple of steps back.

—————
(*) almost because around 10-15% of that heat is coming from the I/O Die

 

[Mussels]

I'm joining the stutter-crew

Computer Type: Desktop
GPU: AMD Radeon RX 7800 XT (Sapphire nitro+)
CPU: AMD Ryzen 7 5800X 8-Core Processor
Motherboard: MPG B550 GAMING PLUS
BIOS Version: 1.F0
BIOS Release Date: 10/11/2023
RAM: 32 GB @

• - 16 GB DDR4-2666, G Skill Intl F4-3600C14-16GVK
• - 16 GB DDR4-2666, G Skill Intl F4-3600C14-16GVK

PSU: Seasonic Focus GX-750W
Case: be quiet! PURE BASE 500DX Window
Operating System & Version: Windows 11 Pro
GPU Drivers: Adrenalin Edition-Version 23.12.1
Chipset Drivers: AMD CHIPSET DRIVER VERSION 5.11.02.217 (earlier 5.01.29.2026)
Background Applications: DISCORD, Steam, MSI/AMD Software

Description of Original Problem:
I'm looking now for a solutions since a long time and did a lot. I read a lot on websites and trouble shooting, but never found a solutions, so I wanna try it now here.
My issue is, that I have micro freezes in every game, that a game stops really for a very very short time and then it keeps going fluently.

• Rogue Trader: When selecting Menues, or AoE-Debuffs
• Age of Wodners 4: Scrolling over the map
• Pathfinder 2: I noticed it when I killed an enemy and when he exploded
  • Also scrolling at an area with fire on houses, though when pressing pause and scrolling then, there is no stutter
• Darktide, when I hit enemies in Melee, or sometimes with range.
• That are just some examples.

Troubleshooting
What I have tryed so far:

• New Monitor
• New Ram
• Windows upgrade from 10 to 11
• Complete wipe, installation of w11
• Intallation of w11 on another SSD
• A-XMP p1 and p2 and none
• Removed sound card, speakers in on board
• USB front panel disabled (and enabled again)
• When I had w10, I did DDU with the new graphic cardI did a lot things, that I dont even remember.
• Mouse polling on 500/1000
• Benchmark 3DMark Speedway -> smooth, no stutter on 40 fps
• MSI util
• Both SSD

Idk, if this also might help, I tryed shortly a round of Pathfinder, got a few micro freezes/stutters for the frame times


View attachment 327424View attachment 327425

Use HWinfo, not HWmonitor. It's trash.

You're hitting 94c which is WAY past thermal throttle on these.
You overheated at 111.5W which isn't near the 142W maximum so your cooling needs to be replaced, or ECO mode needs to be used.

That's DDR4 3600 RAM which isn't officially supported but can be made to work - enable XMP then set SoC voltage to 1.10v (1.15v if instability lingers around) and it should be stable.

7800XT hotspot is at 90c, sounds like you need to tighten the screws on the GPU (washer mod?) or replace the thermal paste -the average is cold enough but high hotspots indicates uneven cooling.

 

[Axels_70]

I have tried that program before, but never got any red spikes. Highest spike was to about 700.

Hi, I know this was way back but did you ever fix this issue? I’ve been having these spike issues and it’s like the Bermuda Triangle, I’ve attempted everything I can think of and have found and nothing makes sense, I even replace every single part of my pc except power supply, I’m running a 3080ti with I7 13700k and 6000mhz ddr5 32gb ram. I’m thinking it has to do with power. You never replied back to this so want to know if you were able to solve this.

 

[Ali96A]

hi guys i have been having sttuters since i got my pc 6 months ago but it wasnt noticable until like a month from now.
i dont have alot of knowledge in pc stuff so i searched alot in the internet i did everything i could find but still no avail.



i tried
-sfc
-DISM
-uninstalling gpu drivers using DDU
-reinstalling gpu drivers
-clean win11 reinstall
-updating all drivers
-updating bios
-reseting ram
-reseting gpu
-using some optimaztion programs
-disabling devices i dont need
-disabling some services
-memory diagnos on extended mode
-and alot of stuff that i dont remember


my pc also having some latency spikes issues that i think could cause sttuters i tried a lot of things to fix this issue but still the same


im also considering to install and try win10 because i saw alot of people suggesting that it will fix some latency issues

i also have constant kernel latency

info about my pc

Operating System	Microsoft Windows 11 Pro
Processor	12th Gen Intel(R) Core(TM) i5-12400F
Graphics Card	NVIDIA GeForce RTX 3060 Ti (8.0 GB)
Memory	11.5 GB Free (15.8 GB Total)
Monitor	Generic PnP Monitor (1920 x 1080 @ 165 Hz)
Disk Storage	369.7 GB Free (2339.1 GB Total)
Audio	Realtek High Definition Audio
Motherboard	ASUSTeK COMPUTER INC. (PRIME B760M-A D4)
Mouse	USB Input Device
Keyboard	USB Input Device

Top

Operating System​
Computer System
Computer Name
Organization
Operating System
OS Name	Microsoft Windows 11 Pro
OS Version	10.0.22635
Product ID
System Up Time	1/31/2024 1:10:28 AM
Internet Explorer Version	11.1.22621.0
Microsoft DirectX Version	DirectX 12
OpenGL Version	10.0.22621.2706 (WinBuild.160101.0800)
Registry
Maximum Size	4095 MB
Current Size	114 MB
Status	OK

Top

Processor MainBoard​
Central Processor
CPU Name	12th Gen Intel(R) Core(TM) i5-12400F
Number of logical processors	12
Code Name	N/A
Manufacturer	GenuineIntel
Current Clock Speed	2500 Mhz
Max Clock Speed	2500 Mhz
Voltage	1.1V
External Clock	100 Mhz
Serial Number
CPU ID	Intel64 Family 6 Model 151 Stepping 5
Socket Designation	LGA1700
L1 Cache	288 KB
L1 Cache	192 KB
L2 Cache	7680 KB
L3 Cache	18432 KB
Motherboard
Model	PRIME B760M-A D4
Manufacturer	ASUSTeK COMPUTER INC.
Serial Number
BIOS Name	1604
BIOS Vendor	American Megatrends Inc.
SMBIOS Version	1604
BIOS Date	12/15/2023
BIOS Features
PCI is supported	Yes
BIOS is Upgradable (Flash)	Yes
BIOS shadowing is allowed	Yes
Boot from CD is supported	Yes
Selectable Boot is supported	Yes
BIOS ROM is socketed	Yes
EDD (Enhanced Disk Drive) Specification is supported	Yes
Int 13h - Japanese Floppy for NEC 9800 1.2mb (3.5, 1k Bytes/Sector, 360 RPM) is supported	Yes
Int 13h - Japanese Floppy for Toshiba 1.2mb (3.5, 360 RPM) is supported	Yes
Int 13h - 5.25 / 360 KB Floppy Services are supported	Yes
Int 13h - 5.25 /1.2MB Floppy Services are supported	Yes
Int 13h - 3.5 / 720 KB Floppy Services are supported	Yes
Int 13h - 3.5 / 2.88 MB Floppy Services are supported	Yes
Int 5h, Print Screen Service is supported	Yes
Int 14h, Serial Services are supported	Yes
Int 17h, printer services are supported	Yes
Int 10h, CGA/Mono Video Services are supported	Yes
ACPI supported	Yes
USB Legacy is supported	Yes

Top

Memory Device​
Memory Resource
Total Memory	15.8 GB
Used Memory	4.3 GB
Free Memory	11.5 GB
Memory Usage	27%
Physical Memory
Memory Bank	BANK 0
Description	Physical Memory 1
Device Locator	Controller0-ChannelA-DIMM1
Capacity	8.0 GB
Speed	3600 Mhz
Manufacturer	Corsair
Data Width	64 bit
Memory Type	Unknown
Form Factor	DIMM
Physical Memory
Memory Bank	BANK 0
Description	Physical Memory 3
Device Locator	Controller1-ChannelA-DIMM1
Capacity	8.0 GB
Speed	3600 Mhz
Manufacturer	Corsair
Data Width	64 bit
Memory Type	Unknown
Form Factor	DIMM

Top

Drives​
Disk Drives
Name	ADATA HD710M PRO USB Device
Media Type	External hard disk media
Capability	1863.0 GB
Interface Type	USB
Partitions	1
Total Cylinders	243201
Total Heads	255
Total Sectors	3907024065
Total Tracks	62016255
Tracks Per Cylinder	255
Bytes Per Sector	512
Sectors Per Track	63
Disk Drives
Name	Micron_2400_MTFDKBA512QFM
Media Type	Fixed hard disk media
Capability	476.9 GB
Interface Type	SCSI
Partitions	3
Total Cylinders	62260
Total Heads	255
Total Sectors	1000206900
Total Tracks	15876300
Tracks Per Cylinder	255
Bytes Per Sector	512
Sectors Per Track	63
IDE Controller
Name	Standard SATA AHCI Controller
Manufacturer	Standard SATA AHCI Controller
Status	OK

Top

Display​
Video Adapter
Name	NVIDIA GeForce RTX 3060 Ti
Video Processor	NVIDIA GeForce RTX 3060 Ti
Manufacturer	NVIDIA
Video Architecture	VGA
DAC Type	Integrated RAMDAC
Memory Size	8.0 GB
Memory Type	Unknown
Video Mode	1920 x 1080 x 4294967296 colors
Current Refresh Rate	165 Hz
Driver Version	31.0.15.4633
Driver Date	12/6/2023
Monitor
Name	Generic PnP Monitor
Screen Height	1080
Screen Width	1920
Status	OK

Top

Network​
Ethernet
Product Name	Realtek Gaming 2.5GbE Family Controller
Service Name
Manufacturer	Realtek
MAC Address

Top

Other Device​
Sound Device
Name	Realtek High Definition Audio
Manufacturer	Realtek
Status	OK
Sound Device
Name	NVIDIA Virtual Audio Device (Wave Extensible) (WDM)
Manufacturer	NVIDIA
Status	Error
Sound Device
Name	NVIDIA High Definition Audio
Manufacturer	NVIDIA
Status	Error
Sound Device
Name	FxSound Audio Enhancer
Manufacturer	FxSound
Status	OK
Mouse
Name	USB Input Device
Manufacturer	(Standard system devices)
Buttons	0
Status	OK
Keyboard
Name	USB Input Device
Description	Enhanced (101- or 102-key)
Function Keys	12
Status	OK
Keyboard
Name	USB Input Device
Description	Enhanced (101- or 102-key)
Function Keys	12
Status	OK
Keyboard
Name	HID Keyboard Device
Description	Enhanced (101- or 102-key)
Function Keys	12
Status	OK
USB Controller
Product Name	Intel(R) USB 3.20 eXtensible Host Controller - 1.20 (Microsoft)
Manufacturer	Generic USB xHCI Host Controller
Protocol Supported	Universal Serial Bus
Status