Hi, thanks for the comments!
The “wrong scale” is a feeling that I suspect is not real. You should measure it exactly (e.g. add to your model a line that is exactly 2 meter long, somewhere on the floor; then in VR, look at the line and leave each controller at the end; then outside VR measure the real distance). Maybe the problem comes from the ground not being at the right height in VR?
Of course it may be the case that there is really a systematic and measurable difference. If so—and while I don’t understand why there should be one—I could think about adding a configuration parameter.
EDIT: ah, maybe adjusting the distance of the lens between the eyes would have an effect on the perceived distances, too.