I was looking at images of the output between different OVR apps, and
was surprised to find that many of them differ in the size and shape of
the output for each eye. It makes sense that different implementations
might take advantage of different opt...
I've noticed that with the
new SDK [note: I haven't tested extensively with older SDKs, so it might
not be a new bug] I am consistently seeing a breakpoint exception being
tripped in the following code upon
exit:HIDDeviceManager::~HIDDeviceManager(){...
OK, I've done some experimentation, and the issues I am seeing go away
if I eliminate the view transform shift (i.e. set it to 0) meaning that
it seems like the distortion and projection center shift are
correct.Offsetting by +/- 0.5*IPD in the left/...
"jherico" wrote:"nhoobler" wrote:I'm not sure I agree. As far as I
understand the system, using the full display is going to be wasteful
(forcing your undistorted render target to be larger and forcing you to
read and write more texels during the dis...
"jherico" wrote:No, I haven't seen anything that uses the functionality.
Probably the reasoning is that any new system you're building would want
to use the maximum display area. The only real reason I could see for
changing the value would be if you...
Yeah, my guess is that the IPD calculation isn't accounting for
something properly. "jherico" wrote:"nhoobler" wrote:Note that instead
of scaling so that the projection lines up with (-1, 0) I use the OVR
Profile to find the maximum visible extent fo...
"cybereality" wrote:I briefly looked through the code you posted, looks
normal.I also looked at the image you posted (free-view, not in the
Rift) and it doesn't look abnormal.However, even small errors in the
math can produce incorrect results so it'...