OpenVR Driver documentation is very limited

[critrun]

The current documentation of driver creation is extremely limited. I feel like there's way too little explaining. Personally, I'm having great trouble even getting close to having a working driver. I was hoping for it to be a simple pointer situation, but having to go around all over the web to even just to find a small trace of what I'm supposed to do isn't very optimal, and really hinders any enthusiast from creating their dream project.

[tjhorner]

Have you checked out the driver sample? I agree that the documentation is really sparse and could use a lot of work, but the sample driver works out of the box and is pretty informative.

The header file is pretty decently commented (and lots of things are obvious enough to be self-documenting).

[TheDeveloperGuy]

I sometimes wonder if the lack of documentation is intentional to raise the barrier to entry. It's not like a billion dollar company could not afford to hire someone. In any case, it doesn't exist, probably won't ever exist, and piecing together tidbits of information and doing your own trial and error is a prerequisite for OpenVR driver development. If it's any consolation there is a lot more documentation available now than there was 4 years ago (even if that isn't saying much)!

[TheWhiteAmbit]

It is very frustrating. On Hmd level there is more documentation, but half of it is either outdated, misleading or just wrong! Frustrating.

[oneday1992]

Replying to https://github.com/ValveSoftware/openvr/issues/1463#issuecomment-716266435

The case can run, but it is of no use. For example, there are no records in the code and documents for the implementation of trackingoverrides. It is impossible to succeed after operating according to the official documents