Anybody know of a calculator or simulator for size of bokeh balls?

This note from the link:

"Y[FONT=Sintony, Arial]ou can make the little balls of light bigger by increasing the distance between your in focus subject and the out of focus lights in the background."[/FONT]

So now somebody can experiment and do some measuring, but I still think the apparent size will depend on the actual size of the lights, as modified by the distance to the lights.
 
This note from the link:

"Y[FONT=Sintony, Arial]ou can make the little balls of light bigger by increasing the distance between your in focus subject and the out of focus lights in the background."[/FONT]

So now somebody can experiment and do some measuring, but I still think the apparent size will depend on the actual size of the lights, as modified by the distance to the lights.

Yeah, I just provided the link to show people do care about broken balls and how to alter them. I like your response much better than "Deerrp I dunno. Just a dumb question, nobody cares."
 
This note from the link:

"You can make the little balls of light bigger by increasing the distance between your in focus subject and the out of focus lights in the background."

So now somebody can experiment and do some measuring, but I still think the apparent size will depend on the actual size of the lights, as modified by the distance to the lights.
Yes I think the circle of confusion equations cover it pretty well.

A non-point-source light would simply create a ball equal to the diameter of what a point source would have produced + the diameter of the apparent light itself. E.g., if you have a light source that itself would be taking up 50 pixels in focus, then (assuming no focus breathing), out of focus it would just be whatever diameter the out of focus bokeh would be for a point source + 50 pixels more.
 
About seven years ago, I did a couple of test sessions, using mini-size Christmas tree lights...the NON-LED, miniature bulb strings, multiple strings of which I suspended from a boom arm in front of the living room's gold-colored curtains. I started with gallery 1, using a Nikon D2x APS-C sized sensor with the 70-200 2.8 VR-I lens. minilights gallery 1 Photo Gallery by Derrel at pbase.com

A few days later, I did gallery 2, using the 70-200 zoom, and also some shots with a 200mm prime lens. Using an aperture of f/2.8 gave the largest, and most perfectly-rounded shapes to the out of focus highlights. You can look at some of these uncropped images, and get a sense for size of bokeh balls versus camera to subject distance. I am going to estimate that the subject to BACKGROUND distance was around 10 feet. Camera-to-subject was from about eight feet to 20 feet, using the zoom to frame.

minilights gallery 2 Photo Gallery by Derrel at pbase.com
 
3rd shot down, tree with feet in socks. STRONG 5-sided bokeh balls. Guess? CANON 50mm f/1.8 EF, one of the very few modern lenses with a 5-blade diaphragm.

Which begs the discussion of shaped bokeh balls, doesn't it? If you don't like five-sided bokeh balls, just make a template to fit over the front of your lens and you can have bokeh balls shaped like stars or diamonds or even hearts, and who doesn't love heart-shaped bokeh balls?
 
This note from the link:

"You can make the little balls of light bigger by increasing the distance between your in focus subject and the out of focus lights in the background."

So now somebody can experiment and do some measuring, but I still think the apparent size will depend on the actual size of the lights, as modified by the distance to the lights.
Yes I think the circle of confusion equations cover it pretty well.

A non-point-source light would simply create a ball equal to the diameter of what a point source would have produced + the diameter of the apparent light itself. E.g., if you have a light source that itself would be taking up 50 pixels in focus, then (assuming no focus breathing), out of focus it would just be whatever diameter the out of focus bokeh would be for a point source + 50 pixels more.

Not quite - the magnification changes with the distance from the rear nodal point to the image plane (just like normal focus breathing). The significance of the change in magnification compared to the circle of confusion will vary with f-number and it will also depend on the position of the exit pupil (the image appears to be projected from the exit pupil, not from the rear nodal point, but the magnification is determined partly by the distance from the rear nodal point to the image plane). Remember that you may need one lens-specific value, the position of the exit pupil, for a rigorous approach, but not if you simply want to assume thin lens properties.
 
About seven years ago, I did a couple of test sessions, using mini-size Christmas tree lights...the NON-LED, miniature bulb strings, multiple strings of which I suspended from a boom arm in front of the living room's gold-colored curtains. I started with gallery 1, using a Nikon D2x APS-C sized sensor with the 70-200 2.8 VR-I lens. minilights gallery 1 Photo Gallery by Derrel at pbase.com

A few days later, I did gallery 2, using the 70-200 zoom, and also some shots with a 200mm prime lens. Using an aperture of f/2.8 gave the largest, and most perfectly-rounded shapes to the out of focus highlights. You can look at some of these uncropped images, and get a sense for size of bokeh balls versus camera to subject distance. I am going to estimate that the subject to BACKGROUND distance was around 10 feet. Camera-to-subject was from about eight feet to 20 feet, using the zoom to frame.

minilights gallery 2 Photo Gallery by Derrel at pbase.com

Kewel!
 
Not quite - the magnification changes with the distance from the rear nodal point to the image plane (just like normal focus breathing). The significance of the change in magnification compared to the circle of confusion will vary with f-number and it will also depend on the position of the exit pupil (the image appears to be projected from the exit pupil, not from the rear nodal point, but the magnification is determined partly by the distance from the rear nodal point to the image plane). Remember that you may need one lens-specific value, the position of the exit pupil, for a rigorous approach, but not if you simply want to assume thin lens properties.
Is that likely to change the size by more than 5-10% of what I would expect if I ignored this, for any normal lens on the market? If not, I am happy to ignore it still.
 

Most reactions

Back
Top