I would like to know if there is any problem with the method Im thinking for rendering multiple quads
for each quad create their 4 vertices based on their w and h values
translate quad to own location (x,y)
apply rotation matrix (if needed)
once this is done send the vertex data to the VBO
apply projection and camera transforms
call glDrawArray 4*number of quads
usually you don’t translate/rotate your model data beforehand. You build the object in modelspace (usually centered around the origin), use a world matrix to get the object positioned somewhere, and then have the view matrix transform that into screenspace, where openGL does its magic.
also, I think it would be a little faster if you added almost all your mesh data to a VBO, rather than have a ton of VBOs storing 1 quad each xD
Okay but if that is the case how do I change vertex data for one particular object in case I want to do some vertex modifications on it?