3D Cameras

[GALACTUS]

Hey guys

Is there a difference between real world and max cameras, ie is a 35mm digital camera the same as max's 35mm.

Regards

Galactus

[percydaman]

yes, AFAIK, the max camera does not reproduce the curved lens distortion of real life cameras. anyone else correct me if Im wrong.

[archprog]

i'm not sure if the max camera matches a reflex camera. But there is a difference when compared to digital cameras, because digital cameras are different from regular reflex cameras, a 35mm regular reflex camera is not the same as a 35mm digital camera. i don't know the technical explanation but digital cameras have less fov. There is a lens multiplier factor (LMF) for each type of digital camera, you have to find the LMF of the camera model the picture was taken with, and multiply that with the focal length and the result is the focal length that matches a reflex camera and max too in case max cameras match real cameras.

[dapeter]

Okay, let me see if I can explain this one. The MAX camera is basically a pinhole camera - you can't specify the "frame size" of the "film," and there's no lens in front of it. So, as light enters a real camera, the light is bent by the lens to a path that sends it straight down the lens. Then these rays converge on the film (which is 24x36mm). A digital camera typically has a smaller "film" area (for example, a Nikon D70's CCD is 23.7 x 15.6mm).

You can search for "pinhole camera how works" in Google and find some good explainations. There's a good analogy on the How It Works website. Basically, if you think of your rendered image as the "film frame" of the camera, instead of all the light rays being coverged on a path through a lens, they are linear the entire way from their point of origin thru an imaginary pinhole to the rendered frame. Technically, the image a pinhole camera creates on the frame is upside down, but the software flips the frame after being rendering.

Because they are linear from the point of origin to the frame, there is no depth of field on a pinhole camera, and all objects at every distance always appear to be in focus. The depth of field is created by the convergence of light rays by the lens and aperature - without the lens altering the path of the light as it approaches the film, there's no depth of field and everything is in focus.

This is how depth of field is created in a standard camera. To blur objects in the foreground, the aperature is opened wider, creating more refraction of the light rays in the lens. Because there is more refraction, the different wavelengths will not converge precisely on the film surface, basically creating the "blurring" depth of field effect. To decrease depth of field and make objects in the foreground and background sharp, the aperature of the lens is decreased in diameter - making the camera approach a "pinhole" type camera (which, as I said before, has no depth of field because there's no refraction).

So, the larger the film frame, the larger the aperature of the lens before it can be made, and the greater the depth of field effect can be because the light rays can be refracted more. (it's a little more complicated than that, because it really involves the cone of light that can approach the film). Therefore, a digital camera with a smaller "film" area (the actual CCD), can get less depth of field than a comperable film camera with the exact same lens. Some pro digital cameras match the frame size of a 35mm negative to match the depth of field of a film camera. Since most renderers compute in RGB colorspace and not physical wavelengths of light, depth of field effects are created using compound blurring and z-buffer depth. I'm not sure if this is how Vray calculates it, though.