cemu-project
Wiimote v1 not working properly under linux
I use Cemu to play Mario Kart 8. My Cemu version is 2.0-20230129gitf3ff919 (Fedora 37)
I have a wiimote (v1 without Motion Plus) and to connect it to Cemu, I use linuxmotehook2.
Unfortunately, there are 2 problems:
- the first comes from the fact that I have to turn the wiimote around (have the back of the wiimote in front of me) so that the kart turns correctly left and right.
- the second comes from the fact that when the wiimote is flat, the kart turns slightly.
To go to the end of a race, I therefore have to turn the wiimote over and turn it slightly to compensate for the neutral position problem.
I talked about the problem on discord.
After discussion with the developer of linuxmotehook2, it seems that the bug comes from Cemu. See: https://github.com/v1993/linuxmotehook2/issues/9 for more details.
Note that my wiimote works fine with dolphin.
Wii U only officially supports Wiimotes with Motion Plus, Mario Kart 8 is noted and known to be less precise and more twitchy with a non-motion plus controller.
the only reason for compatibility with V1's on a wii u is with games running in Wii mode.
The main issue here is axis mapping apparently being wrong. While low precision is expected, axles being mapped wrong/inverted are not. I don't have MK8 on WiiU or gen1 wiimote to test this myself, but this sounds like a rather unlikely bug to slip into released game, so Cemu would be the main suspect here.
Ok it explains my problem.
Anyway, is there a way to change neutral point in Cemu ?
Because Mario Kart is playable with some modifs of linuxmotehook2 but neutral point is the most annoying.
The main issue here is axis mapping apparently being wrong. While low precision is expected, axles being mapped wrong/inverted are not. I don't have MK8 on WiiU or gen1 wiimote to test this myself, but this sounds like a rather unlikely bug to slip into released game, so Cemu would be the main suspect here.
this will of course come down to not directly using the wii mote api to interact with the BTHID capabilities reported but instead having an intermediate driver between the control and cemu.
So at short term, the solution is to find a wiimote motion plus. Right ?
And at medium (or long) term, is there a solution that would go through a modification of the code to better take into account a wiimote v1 for mario kart 8 (for example) ?
You also can try ticking "MotionPlus" box in cemu and see if that makes any difference.
I borrowed a motion plus wiimote but the situation is worse.
When I check the "Motion plus" box in cemu, I have the impression that the movements of the kart are random. There is no longer any link between the movement of the wiimote and that of the kart.
I don't know if the problem come from cemu or linuxmotehook2.
I have tested with dolphin (with linuxmotehook2) and accelerometers and gyroscopes seems to be ok.
Accelerometers are not well "calibrate" but when I test my wiimote with only dolphin Mario Kart Wii works fine.
I don't know if I can test something else ?
Hello. How did you use linuxmotehook2? I've been trying to start it up on my fedora machine but it crashes on me instantly. any help appreciated
Hello, if you encounter a crash with linuxmotehook2, you might post issue in linuxmotehook2 github: https://github.com/v1993/linuxmotehook2/