Hi!
I wonder what is the "up" vector in [view( for [gemwin]? It looks like a vector normal to the upper plane of the camera, but I'm not sure about that (this info can be derived from vectors view and target as well)... Can someone help?
r; Piotr Majdak
Piotr Majdak wrote:
Hi!
I wonder what is the "up" vector in [view( for [gemwin]? It looks like a vector normal to the upper plane of the camera, but I'm not sure about that (this info can be derived from vectors view and target as well)... Can someone help?
well, the "up" vector shows you what is "up" in the view.
imagine you have the camera at 0,0,0 and an object at 1,0,0 if you want to rotate the camera along the (1,0,0) axis, you have to define where the up-direction is (there is no such thing as gravity in Gem-space)
if you turn your head by 90° around your nose-axis, your up-vector will turn from (0,1,0) to (1,0,0)
i hope this is clear (it is rather simple, but i don't know how to explain it well)
mfg.asdr. IOhannes
Hi!
IOhannes m zmoelnig wrote:
well, the "up" vector shows you what is "up" in the view.
imagine you have the camera at 0,0,0 and an object at 1,0,0 if you want to rotate the camera along the (1,0,0) axis, you have to define where the up-direction is (there is no such thing as gravity in Gem-space)
if you turn your head by 90° around your nose-axis, your up-vector will turn from (0,1,0) to (1,0,0)
i hope this is clear (it is rather simple, but i don't know how to explain it well)
thanks, your explanation is very clear. I wondered what happens setting the "up" vector to (0,0,1) in your example above. Generally, this should interfere with "target" and "view" vectors. But it doesn't - I think that only the "up" components, which are in the plane normal to the line defined by "view" and "target", are taken into account. The component parallel to the [view,target]-line seems to be ignored (what, of course, makes sense). Do you agree?
Thanks,
Piotr
Piotr Majdak wrote:
Hi!
thanks, your explanation is very clear. I wondered what happens setting the "up" vector to (0,0,1) in your example above. Generally, this should interfere with "target" and "view" vectors. But it doesn't - I think that only the "up" components, which are in the plane normal to the line defined by "view" and "target", are taken into account. The component parallel to the [view,target]-line seems to be ignored (what, of course, makes sense). Do you agree?
yep. for a reference explanation see "man gluFrustum"
mf.g.a IOhannes