I can;t figure out if this can be done or not, so thought I'd ask... I'd like to create an environment,hills, valleys, etc... but thought it would be more economical to keep it the size of the grid box and just shrink the camera down smaller and have it fly through, but it doesnt seem to matter nothign really changes when i change the physical size of the camera...and i dont really want to change the lens length... am i going to have to scale the environment up to 1000% or whatever? Thanks...

hmmmm... Check the manual, but I didn't think you could scale the camera size.A long time user I am sure will know.

Under the properties you can 'scale' the camera as far as the size..it just doesnt really make any difference as far as what youre seeing... so its not like you can build something, a tube or whatever, shrink the camera down and do a fly through... nothing about it in the manual...im still using version 3 too, so mayeb its something available in the newer versions...

The camera is really just a point in space. Since a single point would be hard to see as you're working, Carrara gives you that blue camera object. It's just a proxy, representing that single point in space. When you adjust the size of the camera, you're just adjusting that proxy to make it easier to find in your scene (but not so big it engulfs the other objects). It has no effect on the actual image the camera captures.

Ohhh..that makes sense...so ultimately I would have to build the scene much larger than Id need it if I wanted to do a fly through... but im guessing the larger the size of the scene the larger the file size and more processing power its going to take... hmmm...ok..thanks...

Well it depends. Actually you shouldn't need to build it bigger than you need. Just as large as you need. Will also depend on your camera travel. Carrara handles large amts of objects fairly well w/ the replicator I think you will be fine. How much of a fly thru, time wise are you talking about? Any movie is going to take some time.

Message edited on: 03/26/2006 20:55

It was something that Id thought about at one point then got sparked again recently when I went into the premade scenes and came across one of the mountain terrains that is already there. I was trying to play around with getting the camera much smaller so it could glide across the river and up the mountain, but I found I couldnt get the camera to actually shrink at all. Id meant to ask atthat point but had other work to do, so it got forgotten about again until today..

Why are you against changing the focal length? It's what you do in real life with a camera. The great thing about CG is you don't have depth of field problems.

Message edited on: 03/27/2006 00:03

changing the focal length distorts the picture, or can, at least in the instance im thinking of... but if i build the model small..like a tube or something....theres no real way to fly through it without scaling it up first... or at least thats what i trying to find out...i thought if you could just shrink the camera down to fit the scene or if there was that option... ive just been checking if theres a 'best' way to do it or not...

You should not see any distortions until you go VERY wide agle. Try it and see.

I think I got it or closer to it. Just have to switch to either the spherical or the isometric camera to get it. Using the conical at 6mm creates distortion, which isnt what I was thinking. Comparatively, what Im thinking would be, walking through a large pipe with a 35mm / 50mm camera...now if that pipe is reduced down 1000% how do you shrink the camera down 1000% to get the same view you had... Taking the conical camera down to 6mm distorts...although the other 2 options arent as intuitive its more like shrinking yourself down...

Why reduce the size of your tube? Just leave it at size and fly your cam thru.

You use the Zoom on the camera. Thats the standard idea I was talking of before. You just zoom from about 6mm through to I think 400mm.Just stick with the standard camera and use the zoom to suit. You can chamge the zooming setting as you are flying around also.

The only problem I ran into was that I couldn't illuminate the inside of the tube very well, but I think that's a matter of finding the "sweet spot" for the general light. I tried putting a spot and then a bulb "inside" the tube, and neither had any effect, but I'm guessing I didn't get their centers exactly inside.

Two other problems, which I found when I started fiddling with animation: 2. Image-based textures can't deal with this kind of close scrutiny. I put a "wall" at the end of the tube and when the camera arrive, the wall's texture was grossly pixelated. But that's to be expected if the camera moves in too close.

1. You do have to Zoom the camera to get the viewfinder entirely inside the tube. Under my original light conditions, I didn't know I could "see" outside the tube. When I added the wall, I noticed a "viewfinder" box at the center of my image, and moving to 200mm made it go away.

Theoretically, I should be able to reduce the tube diameter to .06" and still "see" down it, because the camera is a polar point. As long as I get that exact point inside the tube, the camera can see. I just tried .06" and I can still see inside the tube. I also tried animating the camera to run the length of the tube, and that worked.

Am I missing something here?
M

Ok, I think I see, so I'll try to describe the problem, mainly to be sure I get it.

The fact that the camera is a point is not the problem. The problem is that the camera's "viewfinder" is a rectangular port that creates a "pyramid" of visibility (like a cone of light) defined by the rectangle's distance from the point. You can increase that distance (reducing angle of view) by Zooming the lens. If you don't reduce the angle of view, then your viewfinder will be both inside and outside anything smaller than the rectangle.

I'm curious to know if the Production Viewer (which IS the rectangle, I think) can solve this problem, and I suspect that it can. Unfortunately, I gotta walk the dog and get to work. Maybe someone else will investigate. If the Production Viewer is not the answer, then I suspect Todd is right; he's stuck with increasing the size of the "set."
But again, if the set is small, isn't the limitation you are running into the detail of the set rather than the "size" of the camera? If I build the entire US as a 20" x20" plane, when I bring the camera down to the street level I have to have street level detail for it to see? That seems like a real problem to me.
M

Message edited on: 03/29/2006 10:02

The production frame or the type of camera being changed might help with this. Haven't really played with this so don't know.