Can someone check my maths? (RE DoF + Brenizer)

Ksanti

Ksanti

Messages
548
Name
Matt
Edit My Images
No
WARNING: Lots of poorly organised maths thinking here

I'm a couple of months away from hitting my savings target for my next lens for my 5D; this one being probably my last for a while - a portraiture lens. I'm deciding between a Sigma 85 f/1.4 and Canon 135 f/2L. I realise the Sigma will walk all over the Canon in low light.

Been using DOFmaster to work out a couple of DOF things:
The Sigma would give shallower DOF for like-for-like images (i.e. more or less same composition).
Sigma:
85mm
f/1.4
8.5 feet subject distance
Total DoF: 0.25 feet

Canon:
135mm
f/2.0
13.5 feet subject distance
Total DoF: 0.35 feet

Perspective aside, is the assumption that focal length A/focal length B = subject distance A/subject distance B correct?

Now I do like bokeh, but this lens will most likely be used more for planned shooting than spontaneous shooting so it would be possible to set stuff up specifically if I want effect; namely the Brenizer "bokeh panorama" method. I've been toying with these for a while, and I'm wondering if the following maths is also right (for working out what DoF would be in a 135mm-shot image, using the Brenizer method to bring it back to an 85mm perspective):

Sigma:
85mm
f/1.4
8.5 feet subject distance
Total DoF: 0.25 feet

Canon:
135mm
f/2.0
8.5 feet subject
Total DoF: 0.14 feet

This would give the 135mm's 85mm equivalent images the same look as an 85mm f/0.x lens (DoFMaster only goes down to f/1.0 which give 0.18ft DoF).

To me the maths makes sense, as focusing at the same distance means the same subject distance which should in theory mean the same equivalent focal length once the panorama is stitched together. BrettMaxwell's calculator says that using a 135 f/2.0 lens at an equivalent of 85mm, gives an equivalent aperture of f/1.2xx, but I /think/ that might be different in that it doesn't adjust subject distance for each focal length i.e. you'd have a normal 135 frame with extra around the outside, and not move closer to get the panorama.
 
What's the Brenizer method?

Not checked your figures, but for the same framing (on the same format) DoF is essentially the same regardless of focal length, except at very close distances.

If it's bokeh you're after, focal length can have almost as much effect as f/number as the field of view is narrowed and the background appears larger, therefore shapes appear more simplified and less cluttered, which makes the subject stand out more.

TBH I'm unclear about what it is you want to know, but all the asnwers to DoF questions are on that DoFmaster site.
 
Ksanti said:
Using panoramas with long, fast lenses to recreate the results that would be achieved by hypothetical super fast wides e.g. 24mm f/0.5
Click to expand...

Oh right. Just googled it - very interesting. For anyone else, link here http://www.youtube.com/watch?v=UsMnRxmeJ74 and lots of examples on Flickr.

Still guessing at exactly what it is you want to know, but if it's which lens would give you shallowest DoF, shorthand maths is to divide focal length by f/number and take the highest number (diameter of aperture) ie 135 f/2 would be 67.5mm and give shallower DoF than 85 f/1.4 = 60.7mm.

As you say, the effect is like a wide angle lens with an impossibly low f/number, or a large format camera with say, an f/2 lens, which is equally impossible. And patching together multiple frames like that gives utterly fabulous image quality.

I also wonder what it would look like if you shifted focus during the patchwork process. You could maybe create some interesting paths of sharpness through the picture :)

Edit: From a practical point of view, for an equal diameter aperture, the shorter focal length would cover the same field of view in less frames. And the same rules of bokeh apply, in that bringing the subject forward towards the camera and away from the background as far as possible enhances the differential effect of DoF.
 
Last edited:
I've tried this with 50 and 85mm lenses but gave up because...

I wasn't very good at stitching in CS5 and I kept getting faults. I do feel better after following that link though as you can see what appears to be a fault at the guys left foot. Also when doing this you don't get the same perspective you'd get with a wide angle lens. Oh, and the file sizes end up being massive.

Having said all that I will give this another go sometime.
 
You must log in or register to reply here.
Back
Top