Mousetester graphs

[widow13]

So I've been having issues with mouse feeling weird and unstable and I've been trying to get good mousetester graph just in case it helps.
Today I set affinity on 0 core on my mouse and also changed mouse/keyboard DataQueueSize to 50.
Should I change affinity on the headset/mic(chipset usbs) so they dont conflict(if they do),or do I leave them on default without affinity set ?
I also have both mouse/keyboard on CPU bus and each one has a hub of its own.
https://ibb.co/Cn7m35s Mouse on 4,keyboard on 7 and headset/mic on 6. B650 aorus elite ax, mouse is viper v2 pro.

This is what I'm at right now https://imgbb.com/B215ncJ with the new changes.
And this is the before https://imgbb.com/0rY8vNK

Don't know if this is any improvement even tho it felt a bit better ingame or if the mousetester graphs are any good.
MouseTester version is v1.5.3 and using windows 10.

EDIT to add image inserts

Before:


After:

[Chief Blur Buster]

I see some weird cyclic waves in both graphs -- can you do a 2nd test pass? Back-and-fourth?

Make sure nothing is running (no browsers, no apps).

Also maybe run again with the game running in fullscreen borderless in the background while mousetester is running in foreground. The graphs often change a lot when this is done (for games that do not autopause in this state).

[axaro1]

...or if the mousetester graphs are any good.
MouseTester version is v1.5.3 and using windows 10.

Dobagrab Mousetester v1.5.3 is not reliable especially when testing 1000+hz, do you get the same results with microe1 official branch 1.4 (https://github.com/microe1/MouseTester) or amitxv fork (https://github.com/amitxv/MouseTester) ?

The latest builds from amitxv provide a lot of useful statistics such as standard deviation, maximum/minimum and average polling rate values which makes comparison a little bit easier to quantify how much tweaks can improve/degrade performance.

[widow13]

I see some weird cyclic waves in both graphs -- can you do a 2nd test pass? Back-and-fourth?

Make sure nothing is running (no browsers, no apps).

Also maybe run again with the game running in fullscreen borderless in the background while mousetester is running in foreground. The graphs often change a lot when this is done (for games that do not autopause in this state).

Do you mean from left to right instead of circle movements?
It does not give me anywhere near good results because my hz do not right their peak with left to right movements from some reason(see picture below).
I even tried with a game open in the background but no difference.

1.png (145.13 KiB) Viewed 9757 times

...or if the mousetester graphs are any good.
MouseTester version is v1.5.3 and using windows 10.

Dobagrab Mousetester v1.5.3 is not reliable especially when testing 1000+hz, do you get the same results with microe1 official branch 1.4 (https://github.com/microe1/MouseTester) or amitxv fork (https://github.com/amitxv/MouseTester) ?

The latest builds from amitxv provide a lot of useful statistics such as standard deviation, maximum/minimum and average polling rate values which makes comparison a little bit easier to quantify how much tweaks can improve/degrade performance.

How do I run these ?

[axaro1]

How do I run these ?

Click on "Releases" -> Click on the .zip file -> extract the folder inside the .zip and double click on the .exe.

As for which one you should download, stick to amitxv's fork.

[widow13]

How do I run these ?

Click on "Releases" -> Click on the .zip file -> extract the folder inside the .zip and double click on the .exe.

As for which one you should download, stick to amitxv's fork.

Thanks!
This is what I get using the same exact test.
Don't know if these are any better. Also are the DataQueueSize settings and affinity good ?

[F1zus]

See how my mouse behaves at 8000Hz. Absolutely unstable frequency with jumps up to 200000 Hz.
I tried to bind different cores, 2, 4, 6, the result is the same.
Disconnected all possible usb ports. Disable interrupt moderation.
My amd 5800x processor is overclocked to a stable 4.6GHz, RAM 3800cl14 according to the formulas.
Here are the results at 8000hz




Here it is at 4000hz.





I always use 1600dpi

Perhaps this is due to the mouse buffer MouseDataQueueSize

In theory, the higher the polling rate, the more data you need to store in the buffer and the larger the buffer should be.
I have tried 25/50/100 options. Didn't notice much of a difference.

[widow13]

I'm actually confused and don't know how my graphs are looking compared to a good one because I don't really know how to read em.
Are they bad/really bad/average/good ?

[axaro1]

Your 1khz results look just slightly above average, most people with optimized setups get similar results in W10/W11.
It's also heavily dependent on the type of mouse that you are using or if you are wireless/wired. Some mice with powerful MCU and good firmwares can achieve better polling timings.

Core0 is generally the core that other applications tend to use as default, so it's wise to use a different one.
As for DataQueueSize, I've honestly never tried to mess with it.

Are you on the latest W11 build with the mouse new input service stack? It will have a positive impact even at 1000hz if you happen to play with discord/broswers/MSIafterburner/... open, in the background or minimized.

[widow13]

Your 1khz results look just slightly above average, most people with optimized setups get similar results in W10/W11.
It's also heavily dependent on the type of mouse that you are using or if you are wireless/wired. Some mice with powerful MCU and good firmwares can achieve better polling timings.

Core0 is generally the core that other applications tend to use as default, so it's wise to use a different one.
As for DataQueueSize, I've honestly never tried to mess with it.

Are you on the latest W11 build with the mouse new input service stack? It will have a positive impact even at 1000hz if you happen to play with discord/broswers/MSIafterburner/... open, in the background or minimized.

Thanks a lot for your answer!
I'm on the latest windows 10. Tried using proccess lasso to minimize the impact discord and other programms have when gaming by giving them lower priority.
Tried using core 2/4 didn't see any difference tho.

Are there any bios settings that are really important that I might be missing ? For example HPET is hidden for me in bios, do people still disable it ?
Been a long time since the last time I tried w11.
Honestly mouse been feeling better but I miss the snappy feeling I used to have, it feels unstable.