VR does not work

no. not even a little.

that document is innocent. devoid of any of the brutal and insane detail required to get a headset running.

I had to figure it all out on my own.

like the fact that steamVR will not work at all unless you install the secret mac version of SteamVR.

the fact that there is no way to update the firmware in all of your various devices. because the mac client doesn't actually support support firmware updating.

but the problems I am seeing after getting through all of that is the most unforgivable kind of mistake. It's just pure negligence. poor quality control.

fox_m,

yes. project is set up properly. I can navigate it just fine in the 360 viewer, I can export it and upload it to youtube and it works the way you'd expect. I am producing appropriate content.

in fact, fcpx and steam VR's mirroring function, both have the same, and correct output. the output you'd expect to see in the glasses.

the glasses themselves are not showing any of that. the glasses - and I can't figure out how to get a picture of this - are showing a nonsensical subsection of the raster image.

I can tell, because there's a wireframe in the background, that the hardware is tracking correctly. tracking the controls, the head unit... and producing correct results. because both FCPX and the SteamVR mirroring is showing the same and correct results, we know that the software is both capable of producing the correct output from the footage, and that at some level it in fact does.

what I'm getting at here, is everything in the pipeline, before the glasses displays the video, is clearly setup properly, and doing the right thing. The glasses themselves, show the background wireframe correctly tracking and correctly stereoscopic. It's just the content that is being displayed that is jacked up.

similarly, the controls are tracking perfectly. the rays coming out of the controls point exactly where you aim them. It's as intuitive as it sounded. Except... and this can only be tested in one place, the system menu... the hit testing for the controls is wrong. when you click, the hit is transformed on the Y axis... it's as if you're pointing in the opposite direction on the Y axis.

so wherever you are pointing it's magically transformed to x = x, y = -1 * y, this affects polling as well...

so let's stop looking for problems with the user here. the problem is not operator error. it's one of the three parties involved in this product: Steam, HTC, or Apple. Some terrible programmer has inserted observable, predictable, bad math into the system somewhere. One of those 3 parties.