1600 DPI can't loot in PUBG, Apex etc. menu is to fast

[daviddave1]

Hello everyone. I have the Razer 8K. ( with a Alienware 360 hz monitor, a 19-12900K and a 3080ti)
I Mainly play PUBG. When i put the mouse on 1600 DPI I cant loot cause the menu is to fast. Its not worth it. I rather leave it on 800 DPI then. I can use 1600 DPI when i put the Windows 10 mouse sensitivity on 4/11. But does PUBG uses raw input when the loot screen is dependable on your windows sensitivity settings?
Shall i play A :PUBG with 800 DPI, 6/11 Windows sensitivity or B: 1600 DPI, with 4/11 Windows sensitivity? ( both on 8k pollingrate) or C: 1600 DPI, with 6/11 Windows sensitivity and rawaccel?

[TheNoobPolice]

Firstly, congrats on the beast rig!

Most games expose the windows cursor for menu activity even when using raw input for the 3d world, but rarely they will use also use a rawinput cursor (like the Doom games). I don't own PUBG but it's very easy to check, just change the windows pointer speed and see if there is any change in the menu cursor sensitivity.

There is no issue whatsoever in using any windows pointer speed less than 6/11, if you go above 6/11 though it does some wonky interpolation where it doesn't carry remainders properly and you get slightly wonky input. So basically just stay at or below 6/11.

Using rawaccel to downscale 1600 DPI would be functionally identical to using 800 DPI so no point in that.

Which begs the question; if 800 DPI works and feels comfortable, why even bother with 1600 DPI?

[daviddave1]

8 minutes ago, TheNoobPolice said:

Tnx for the answer! Mainly cause of 

but that's my question. Shall I bother? 

Which begs the question; if 800 DPI works and feels comfortable, why even bother with 1600 DPI? 

 

[TheNoobPolice]

No, you don't need to worry about that. Whilst it may appear that is a sound testing methodology (reproducible action / results over multiple runs etc) it is fundamentally testing the wrong thing. I wrote a reply about why it is not correct / useful here:

https://www.reddit.com/r/MouseReview/comments/p49dld/gaming_mice_and_sensor_lag_new_testing/h96ctoj/

 

[MacSquirrel_Jedi]

It was really painful 7 minutes for me  I've spent thousands of hours measuring mouse distances. For a short time I also played with acceleration in Windows, which works with polling rate. So i know that at low speeds you do not reach 1000 Hz. But something about 0-250Hz. In video he have in every graph 1000 Hz for all speeds. So it is clear that this does not reflect reality. Also i didn't hear in video anything about sensor reaction time. Because when the mouse is idle (not moving). Sensor is not working at his full potentional. That's why you should make a small movements by mouse even when only watching a spot. To make the sensor active (ready for fast move).

You should more care about overall input lag. This is so far the best article about input lag. Lots of articles about input lag are based on this. Unfortunately, now that I've reviewed it, the videos related to this article are gone. Such a loss... Or something more recent here.

[heckminth]

CPI only affects input lag in arbitrary circumstances by a few milliseconds at worst. I really wouldn't worry about it. That being said, a higher sensor resolution is objectively going to be more precise. Here is some information on how RawAccel produces linear acceleration and input velocity based on input vectors, CPI, polling rate, an acceleration coefficient, and your sensitivity. You can get a great sense of how it all works by looking at the mathematics and playing around with the software yourself. You don't have to use acceleration at all to understand this, it's just a more fun way to show how it all adds up. If you can understand how CPI and polling rate impacts a positive linear acceleration curve, then a flat one is no problem either. As a bonus, this makes getting into acceleration aiming easier.

Edited December 4, 2021 by heckminth
new information