It's quite amazing how one can forget all the basics by working with abstractions. More specificly… I had been creating a series of auxiliary classes for using them in my demos, so I didn't have to think of gluLookAt's and similar, and concentrate instead in specifying cameras in human terms, like, look to that point, etc.
The problem arises when I look at what did exactly each class and how it interacts with OpenGL - I just can realise how much I have forgotten about everything. Did OpenGL setup a default projection matrix? Which values did it have? Did I need to push that projection matrix and load an identity one? And what about the modelview matrix? Why can't I do a decent particle billboarding even looking at three tutorials and finally even copying the same code that works?
I am trying to avoid the kind of particles which appeared on vslpx [mov video], i.e., not properly oriented. But no matter what I do in regards to the projection or modelview matrix, the supposedly billboarded particles always appear like if they had a mind of their own, putting their front face wherever they decide to, and obviously not being perpendicular to the camera's Look vector, and I feel a bit lost >_<
Luckily it is raining so it might be time for taking advantage of the bad weather and stay home while guessing how my not-engine worked ;)
