@patrick, and anybody else who is/was working on the graphic engine:
Nowic mention something, that transparent textures applied to 0 thickness "wall" are not possible, as the engine does not know which part to draw, and which not to...
As for the blaster shots, one possibility was to use a multiply plane, an smear a texture onto it. Same goes for the missile trails. while working only in 2D, only one plane is really needed (and only one topside texture); But in 3D at least 2, if not 3 intersecting planes are needed...
Possible solution:
Is is somehow possible to draw (resp. add textures) only those parts, in which the normal vector points toward the camera, and ignore the rest. or does this produce too much checking and too little drawing?
cheers
nico
Transparancy Impossible?!
Moderator: PPS-Leaders
-
- Baron Vladimir Harkonnen
- Posts: 258
- Joined: Wed Nov 01, 2006 7:58 pm
- Location: your mind
- Contact:
Transparancy Impossible?!
BOFH Excuse #212: Of course is doesn't work. We've performed a software upgrade.
It might be possible to make a 3D imposter model, that has a longitudinal, transversal and sagital plane (have this from a medical student ) e.g. xy, yz, xz plane..
This is probably something, that many shots are supposed to do, and it might be (might be) wise to make an own model-type for this kind of stuff (might be...). This model would only draw the plane, with the smallest normal angle to the camera -> problem: what's with the switches between planes -> fading -> occlusion-problem.
But anyways, what might the model look like this:
DIA-file
The difference to the other model would be, that this one knows the position of each model for the sake of a higher performance.
Additionally (for performance reasons) this model should be able to hold groups, so that material changes are negligible.
This is probably something, that many shots are supposed to do, and it might be (might be) wise to make an own model-type for this kind of stuff (might be...). This model would only draw the plane, with the smallest normal angle to the camera -> problem: what's with the switches between planes -> fading -> occlusion-problem.
But anyways, what might the model look like this:
DIA-file
The difference to the other model would be, that this one knows the position of each model for the sake of a higher performance.
Additionally (for performance reasons) this model should be able to hold groups, so that material changes are negligible.
Who is online
Users browsing this forum: No registered users and 1 guest