Lens Physics 101 (Warning: long but hopefully informative)

Greg83 said:

Do you consider Canon should write on the box of my 28/1.8 lens
info that this data refers to 35mm, while for cropped cameras the
lens has the characteristics of 45/2.9 (actually being this)?

Click to expand...

ljfinger said:

Yes. And the same of the compacts. The S3IS has a 6-72/2.7-3.5
lens. It says 36mm-432mm (35mm-equivalent) on the box. It should
also say f16-f21 (35mm-equivalent).

Click to expand...

Lee Jay said:

They're similar in DOF too. You chose subject distances close to
the hyperfocal distance. That causes a huge increase in
sensitivity to actual numbers and roundoff error starts to matter.

For the first case, 16m was chosen and the hyperfocal distance is
22.8m.
For the second case, 10m was chose and the hyperfocal distance is
22.8m.
For the third case, 10m was chose and the hyperfocal distance is
9.13m.

Plus, you didn't hold perspective constant (subject distance).

Let's try another example.

30D, 35mm, f2.8, 5 meters, DOF = 2.3m
5D, 56mm, f4.5, 5 meters, DOF = 2.22m

There's a roundoff error we've been neglecting here. We call 1.6x
1 1/3 stops. It's actually 1.357 stops, not 1.333. These little
differences can have an effect on the calculations when you are in
the hyper-sensitive region right around the hyperfocal distance.
They have little to no effect on actual images because the blur is
changing so slowly with distance.

--
Lee Jay
(see profile for equipment)

Click to expand...

Mokes said:

If it said that on the box for the S3IS, then I would expect for
ISO 100 (the only practical ISO on that camera) the following (for
example):
DSLR: f/4 1/125s
S3IS: "f/16" (i.e. f/2.7) 1/8s
Boy that IS had better work well....
Are you sure you don't want to qualify all this, LJ?!?

Click to expand...