Mousekeys inertia zero friction improvements #23677
Conversation
Velocity multiplication works the same regardless of sign. Removing this unnecessary +1 makes it so that left/up & right/down decelerate consistently with each other.
This new option controls the threshold at which velocity snaps to 0 whilst decelerating. Only applicable to the inertia mode
quantum/mousekey.c
Outdated
| else if (velocity <= MOUSEKEY_STOP_SPEED && velocity >= -MOUSEKEY_STOP_SPEED) | ||
| velocity = 0; |
There was a problem hiding this comment.
I wasn't quite sure what to do here regarding the styleguide. The guide is unambiguous in saying that braces must always be used... but since I am extending an existing unbraced if/else chain it felt wrong to differently format only my lines in the function.
I suppose I could have reformatted the entire function, but I worried that doing so would be considered out-of-scope for the purposes of this PR.
| @@ -191,6 +191,7 @@ Recommended settings in your keymap’s `config.h` file: | |||
| |`MOUSEKEY_MAX_SPEED` |32 |Maximum cursor speed at which acceleration stops | | |||
| |`MOUSEKEY_TIME_TO_MAX` |32 |Number of frames until maximum cursor speed is reached | | |||
| |`MOUSEKEY_FRICTION` |24 |How quickly the cursor stops after releasing a key | | |||
| |`MOUSEKEY_STOP_SPEED` |1 |Lowest possible decelerating velocity before it snaps to 0 | | |||
There was a problem hiding this comment.
My Japanese isn't particularly good. Is it alright that I only updated the English documentation?
|
Regarding tests: I couldn't seem to find any existing tests for mousekeys... so my takeaway was that this area of the code isn't considered critical enough to require tests. If I'm wrong about that, I can give writing some tests a shot... but I'll need someone to point me towards a test file which would be workable as a template for mouse stuff. |
|
Well... I suppose that I should have anticipated I'd get dinged by the linter regardless of existing code. I've revised the surrounding calc_inertia function so that it's up to spec with the styleguide. If I wasn't supposed to do that, then please by all means feel free to drop f3d9898. Thanks |
The reason for a lack of tests, is because the framework isn't written for them. Currently, the test framework only supports HID keyboard reporting, it doesn't support HID mouse reporting. It's something that does need to be built out, but nobody has stepped up to do it, yet. Ideally, it is something that needs to be done and there should be tests for it. |
|
Ah, lucky me that I get to skip eating my vegetables today! |
I like using Mousekeys in inertia mode with the friction turned off. It's a great way to get analog-like movements when piloted well, though there are some tweaks which I had to make in order to get things to properly behave when
MOUSEKEY_FRICTIONis set to zero.These changes do very slightly impact the current behavior of the inertia mode, but only imperceptibly so.
Description
There are two changes included in this PR:
MOUSE_STOP_SPEEDwhich allows configuring a threshold below which velocity will snap to zero (if currently decelerating). A value of 0 is identical to the current behavior, though I've chosen a default of 1 instead because it makes stopping through reverse acceleration substantially easier regardless of the friction setting.Types of Changes
Issues Fixed or Closed by This PR
Checklist