Sorry for all the confusion. But it seems to be a healthy one as this theme is recurring all the time.
This is why I kindly asked to follow the link before replying, an advice not taken. I considered writing an own blog article about the topic but if nobody is ever reading ...
Originally posted by Pål Jensen This is faulty logic because formats aren't equivalent
Pål, you are the last I would have expected this comment from. I thank drougge for explaining it.
Maybe, a general comment about the words equivalence and equality. Two things can be equivalent w/o being equal. You then have to define by what properties you go to define equivalence.
And here we go:
Two camera systems (lens, sensor, resolution, exposure/aperture/ISO setting etc.) are
equivalent if they deliver
equal images; equal in the sense that you cannot tell them apart w/o inspecting exif.
In any format discussion, it is important to first understand what are equivalent cameras. As otherwise, you focus your discussion on trivial differences.
Originally posted by Pål Jensen They are not equivalent lenses. A 40/2.8 and a 27/2.8 are equivalent lenses.
This is why I kindly asked to follow the link before replying, an advice not taken.
Originally posted by drougge Falk has it right, for the definition of equal that he uses. You seem to disagree on what is equal. I'll side with Falk. He describes what is required to get an identical image. That seems like a reasonable definition of equal to me. You describe what is needed to get the same exposure parameters but different images. That seems less equal.
Your definition gives the same amount of light per sensor area, Falk gives the same amount per whole sensor. So you give a noise advantage to the bigger format, Falk doesn't. Again, he seems to get it more equivalent.
Note that he's not anywhere suggesting that you need the equivalents, he's just describing what they are.
Thanks for your explaination which I fully endorse.
The following have to scale for equivalent cameras but different crop factors:
focal/crop, f-stop/crop, ISO/crop^2
Only scaling focal w/o scaling the other two parameters leads to confusion which is what is constantly hapenning.
One more thing:
Two equivalent camera systems but with different crop factors have interesting properties:
- same FoV, DoF, #pixels and pixel noise
- same blur due to diffraction
- same physical aperture in mm; i.e. same lens size/cost
But: the higher crop factor has
- a higher lines/mm requirement on optical quality,
- to be delivered at a lower f-stop figure
which work against each other and making equivalent lenses physically impossible for too high a crop factor. Moreover, the registration distance better scale with registration/crop too; which it normally doesn't except for some mirrorless designs. And last but not least, the crop-factor sensor
- needs ISO/crop^2
which (if FF is ISO100) needs ISO50 on the crop camera typically not available putting the crop camera onto a dynamic range disadvantage.
So, the real concerns with a crop camera are:
- registration distance.
- optical flaws running out of control.
- AF accuracy if tolerances not scaled by /crop.
- unavailable lens options like no APSC 32/0.9 or µFT 25/0.7 which used to be the standard prime on FF.
- (one stop) less dynamic range.
- VF magnification.
These real concerns remain completely un-understood if one isn't getting the concept of lens equivalence, and getting it right.
A typical sign are discussions centered around low light performance in the context of FF, or claims that FF is the more expensive system. The opposite is true considering that sensor costs are going down and lens costs are going up over time. FT used to be the sweet spot wrt price, now it is APSC and soon it will be FF.