Inconsistent mouse movement for years over multiple pcs.

[BumFlannel]

Hey.

I've been having this issue now for a few years and its driving me crazy. Let me describe the symptoms first. From day to day, boot to boot, the mouse input feels different and not as responsive as it should be. It's not mouse lag though, it appears to be more like the mouse is deviating in how far it moves based on how far/fast you moved it and changing directions makes it feel like the mouse has the mass of a house brick so my cursor is more waving left and right rather than responding snappily.

It ruins aiming in fps games. It ends up with me not being able to get my crosshair on target, undershooting, overshooting every time, swinging the mouse left and right when trying to track targets because I'm constantly overshooting or not being able to get the mouse to catch up. When I have my crosshair just off an enemy and need to make a micro correction the mouse doesn't seem to move enough, which make me try to move it a little quicker and then it overshoots.....it's so damn annoying.

Nothing seems to fix it. I've had the issue for years over multiple high end builds. Current specs are:

X670 GAMING X AX 7950x
7900xtx
Patriot Viper 32gb 6000mt RAM
Alienware AW2725DF QD-OLED 360hz 1440p

It's not a display issue as it's the same on all displays I've tried. Currently running a 3 display setup with the alienware, an asus tuf 240hz and a huion kamvas 24 screen tablet doubling up as an extra monitor.

Visually, when I move the mouse quickly on the desktop so you can see the ghosts of the cursor it looks like the cursor is broken up with gaps. I know gaps are normal but it's as if the cursor is displaying in bursts or groups. At 500hz which I'm using right now I'd expect to see the mouse trails across the screen like so...

Code: Select all

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

but how i actually see them is like so...

Code: Select all

. .  . . .  . . .  . .  . . .  . .  . . . .  . . .  . . .

In groups of 2 or 3 with a consistently spaced gap between each group as if a cursor/update is missing. At 1000hz I see the same thing just with smaller space between cursors, smaller gaps between the cursor "groups" and groups of 3 or 4 like so...

Code: Select all

.... ... ... .... ... ... .... .... .... ... ... .... ... .... ...

I recorded a 360fps video of the cursor on a black screen and then overlayed all the frames to show what I mean...



Here are some graphs from mouse tester 1.53.

Interval Vs Time



Frequency Vs Time



X And Y Velocity Vs Time



X And Y Count Vs Time



These tests are done at 500hz. Can anyone who understands these graphs better than me explain what's happening here? To me it looks like I have a base update interval of 8ms with 4ms dips followed directly by 2ms spikes? At 500hz shouldn't it be around 2ms baseline with sub-ms deviation either way and at 1000hz 1ms baseline with sub-ms deviation either way?

Also looks to me like I have a baseline of 500hz polling with weird spikes between 8 and 16khz and spikes in velocity between 20 and 40%? Am I reading the graphs right/am I assuming right with what is expected?

I've tried everything. Windows reinstallation(which was done again recently), every combination of bios settings you can think of including changing hidden bios settings. System timer settings, although I can only change this is windows, I don't have any HPET setting in my bios, even hidden. All power saving options disabled across the board both in windows and bios. Bit sum performance power plan, all drivers system wide are updated to the latest, windows is fully up to date, no malware. I've done literally everything and this issue won't go away. I only have this mouse for the time being but I've had these issues across many mice and systems for years. I've done mouse tester tests on other mice and systems and had similar results and have always noticed the gaps in the mouse cursor and the inconsistent mouse movement. Different DPIs and polling rates just tighten the inconsistencies, not remove them. I've tried every USB port and with all other USB devices unhooked. Latency Mon shows no issues. I have no input lag, the mouse is snappy and responsive for initial movements, it just doesn't feel like it's doing what it's meant to be doing over time. I have all acceleration off. EPP, driver, game, used mark c mouse fix, zeroed out SmoothMouseX/Y in reg settings. Below is a shot of my reg settings.



As I said from day to day the mouse feels different and there is no way to be consistent or even build any muscle memory with it. Some days when gaming it's as if my sensitivity has gone up because the mouse feels floatier/swingier but the sensitivity hasn't actually changed and general distances covered by the the mouse hasn't changed. It feels more like my mouse inputs have inertia or something, but no input delay. On rare occasions, without changing anything, the mouse feels crisp and perfect, but that is very rare and it always goes back to feeling like crap either within minutes, days, sometimes a week or two, but it always goes back.

This is driving me crazy, does anyone have any ideas or suggestions?

[RealNC]

I recorded a 360fps video of the cursor on a black screen and then overlayed all the frames to show what I mean...

This is normal. I get this too with my 1000Hz mouse at 240Hz. A 2000Hz mouse can fix this up quite a bit.

The higher the refresh rate gets, the more obvious the mismatch between mouse rate and refresh rate will become. 1000Hz polling rate is perfectly fine for a 165Hz display. At 240Hz though, I start seeing the gaps. And 360Hz is obviously very noticeable. The only ways to fix are: a) enable mouse smoothing in games that have that option, but it adds a bit of lag, or b) use a 2000Hz mouse.

A 500Hz mouse is not even good enough for 165Hz, btw, let alone 360Hz. If you wonder why that is, it's because mouse polling and display refresh rate are not synced. That means some mouse updates simply miss the current display refresh, resulting in missing mouse input for that frame. Bumping the mouse polling rate up will make those gaps occur less often.

[Slender]

I recorded a 360fps video of the cursor on a black screen and then overlayed all the frames to show what I mean...

This is normal. I get this too with my 1000Hz mouse at 240Hz. A 2000Hz mouse can fix this up quite a bit.

The higher the refresh rate gets, the more obvious the mismatch between mouse rate and refresh rate will become. 1000Hz polling rate is perfectly fine for a 165Hz display. At 240Hz though, I start seeing the gaps. And 360Hz is obviously very noticeable. The only ways to fix are: a) enable mouse smoothing in games that have that option, but it adds a bit of lag, or b) use a 2000Hz mouse.

A 500Hz mouse is not even good enough for 165Hz, btw, let alone 360Hz. If you wonder why that is, it's because mouse polling and display refresh rate are not synced. That means some mouse updates simply miss the current display refresh, resulting in missing mouse input for that frame. Bumping the mouse polling rate up will make those gaps occur less often.

he told about instability. If he start use 2000hz + mouse it start feel like 250hz mice.

[Hyote]

I won't say these are the problems you have but from personal experience any change in Windows affects the mouse greatly. First, you get different process IDs every time you restart Windows. This already changes everything as I found that some conbinations provide good input and good performance and others will feel like there is something wrong with the computer. Then a bunch of things load in, for example sppsvc after a few minutes, things like WMIADAP can pop up after certain actions or even just by itself. My "solution" for this is to run 16 services when playing games. For a long time now, which can be counted in months I haven't had problems with inconsistency as I was doing the same things the same way.
The mouse heaviness issue that is mostly discussed here is still present in some way. What I noticed is that every change I make which has not been done to the pc before suddenly makes the monitor clearer and inputs snappier. Yesterday I entered the factory menu of my monitor to change my OD settings and it caused an unbelievable amount of change but only for about an hour. Now that happened a few months back with RAM timings and SCEWIN settings. The first time changes are done is the only time they are going to give that effect. It's like everything starts working properly then over the course of minutes or hours it goes back to being lazy or something.

[Slender]

I won't say these are the problems you have but from personal experience any change in Windows affects the mouse greatly. First, you get different process IDs every time you restart Windows. This already changes everything as I found that some conbinations provide good input and good performance and others will feel like there is something wrong with the computer. Then a bunch of things load in, for example sppsvc after a few minutes, things like WMIADAP can pop up after certain actions or even just by itself. My "solution" for this is to run 16 services when playing games. For a long time now, which can be counted in months I haven't had problems with inconsistency as I was doing the same things the same way.
The mouse heaviness issue that is mostly discussed here is still present in some way. What I noticed is that every change I make which has not been done to the pc before suddenly makes the monitor clearer and inputs snappier. Yesterday I entered the factory menu of my monitor to change my OD settings and it caused an unbelievable amount of change but only for about an hour. Now that happened a few months back with RAM timings and SCEWIN settings. The first time changes are done is the only time they are going to give that effect. It's like everything starts working properly then over the course of minutes or hours it goes back to being lazy or something.

I also thought that changing the settings affected mouse input, but then I realized that all the changes took place at night.

[RealNC]

First, you get different process IDs every time you restart Windows. This already changes everything as I found that some conbinations provide good input and good performance and others will feel like there is something wrong with the computer.

wat

[BumFlannel]

I recorded a 360fps video of the cursor on a black screen and then overlayed all the frames to show what I mean...

This is normal. I get this too with my 1000Hz mouse at 240Hz. A 2000Hz mouse can fix this up quite a bit.

The higher the refresh rate gets, the more obvious the mismatch between mouse rate and refresh rate will become. 1000Hz polling rate is perfectly fine for a 165Hz display. At 240Hz though, I start seeing the gaps. And 360Hz is obviously very noticeable. The only ways to fix are: a) enable mouse smoothing in games that have that option, but it adds a bit of lag, or b) use a 2000Hz mouse.

A 500Hz mouse is not even good enough for 165Hz, btw, let alone 360Hz. If you wonder why that is, it's because mouse polling and display refresh rate are not synced. That means some mouse updates simply miss the current display refresh, resulting in missing mouse input for that frame. Bumping the mouse polling rate up will make those gaps occur less often.

But this is from a video recording at 360fps (on system, with OBS, not external with a camera) and it shows exactly what I see with the naked eye at 360hz. On system video recording doesn't conform to monitor refresh rate, there are literal gaps in sampling it seems.

I have my mouse now at 8k polling rate and 30000dpi.....excessive I know....but I still see gaps in the cursor movement with the naked eye, just smaller ones and more "cursor groups" in-between.

It looks like the input is cutting out a sample at a certain rate regardless of mouse settings....pushing high hz just sort of drowns it out, akin to trying to push higher fps to fix inconsistent frame times.

[BumFlannel]

First, you get different process IDs every time you restart Windows. This already changes everything as I found that some conbinations provide good input and good performance and others will feel like there is something wrong with the computer.

wat

I know what he's talking about. Threads get assigned to do things...some thread combos are better than others for certain things. A thread being shared by 2 things that don't "work well" together or aren't "sharing fairly" can cause issues. That's done dynamically. It was talked about in the CS2 community quite recently, and having to use process lasso to force affinity for CS2 so you get good performance/expected fps/fix stutters.

[RealNC]

But this is from a video recording at 360fps (on system, with OBS, not external with a camera) and it shows exactly what I see with the naked eye at 360hz. On system video recording doesn't conform to monitor refresh rate, there are literal gaps in sampling it seems.

I have my mouse now at 8k polling rate and 30000dpi.....excessive I know....but I still see gaps in the cursor movement with the naked eye, just smaller ones and more "cursor groups" in-between.

It looks like the input is cutting out a sample at a certain rate regardless of mouse settings....pushing high hz just sort of drowns it out, akin to trying to push higher fps to fix inconsistent frame times.

Which is normal. The higher the refresh rate, the more obvious it becomes. At 144Hz or 120Hz you shouldn't be seeing odd gaps even at 1000Hz mouse polling.

(Btw, I don't think DPI has anything to do with it. Only polling rate does.)

[Brainlet]

At 144Hz or 120Hz you shouldn't be seeing odd gaps even at 1000Hz mouse polling.

144 Hz barely, but 120 Hz will have plenty of odd gaps