Drilling further down the problem
First of all late me state that I am a coworker of the person who started this thread.
Reading the responses on this thread, and the confusion generated, I just can’t seem to avoid jumping in…
Depth of field is the range (not a point) along the lens distance axis where objects are in focus. It is a factor of the lens’ focal distance (mm), and aperture.
The lower the focal distance (wide angle) the larger the depth of field. Also, the smaller the aperture (i.e. FStop 32), also the larger the depth of field.
Hence, we usually don’t see out-of-focus objects in wide-angle shots. Also, if we use a larger aperture (i.e. FStop 2.8) we get more out of focus. Every photorapher nows that if you want to blurr the background of a portrait, use at least a 100mm lens.
Let me make clear that this is not only about “how much blurred” an object is, but also how wide, or long, the range is where everything is in focus.
I know that vRay lets you set the “how much blurr” (aperture), and the distance from the camera (focal dist.) where an object appears in focus.
The question still remains, then:
How do I control, based on the focal dist. as the center, how deep the range is (towards and moving away from the camera), where all objects are in focus?
It is my humble opinion that vRay does not seem to simulate a real camera correctly.
First, the aperture value is inverted, as i should get more depth of field with a lower apperture (an open diafragm) than with a bigger one (a closed diafragm).
Second, if vRay would simulate a real camera correctly, I should not be able to tell vRay “how much blurr i want”. That would be computed based on the lens’ focal range (mm, or field of view as it usually mentioned in CG), and the diafragm’s aperture.
For those who like to tweak real-life things, let me say that at least, i should be given the option to let vRay compute this correctly, which I don’t seem to have.
I have rendered images with vRay’s DOF with wide-angle and telephoto cameras, and the range of objects in focus appears to be the same. this is simply not possible in real life.
regarding to vlado’s comment, I am not a programmer, but I’ve certainly seen raytracers who could simmulate real cameras correctly.
The botom line is, I guess the problem was not well understood from the beginning… and I just wanted to clarify the idea. I hope it helped.
Cristian.-