Depth of field on FX compared to DX

[HoppyUK]

Or is there a way to edit/evaluate images based upon final print/viewing size? Without printing that is...and if it's not for print, then how could you ever account for display resolutions?

The DoF standard only works if all the assumptions and parameters that go into it are adhered to - from sensor size through to final viewing size and distance. It works very well then, and importantly, it is universally accepted and applied so we all know where we are with it.

It's easy enough to modify, to account for say cropping or different viewing size/distance etc, if you know what you're doing and do the maths.

You make an interesting point about sharpening, and that reveals the Achilles heel surrounding most things to do with sharpness, and DoF calcs, that only concern themselves with resolution and convenient linear measures like CoC or lines-per-mm and pixels.

For an image to be considered as sharp, the resolution standard required is very modest, and the thing that contributes most to perceived sharpness is contrast - and that is never quantified outside of MTF tests. Take the example of sharpening in software, that is very effective, but it doesn't increase resolution at all, only increases edge contrast.

 

[sk66]

Take the example of sharpening in software, that is very effective, but it doesn't increase resolution at all, only increases edge contrast.

Yeah, just like the increased DR/Contrast of a lower MP sensor can outweigh higher MP's...

But it makes a point...if you really want "full control" you really can't edit an image fully/correctly until it's final size/use has been determined. And then it needs to be optimized for that presentation only (size/medium).

Who does that? I (almost) *never* do. On the other hand, I don't notice DOF shifting much between minor presentation size changes, even viewed from the same distance. (800 vs 1024, 8x10 vs 11x16).

 

[woof woof]

Yeah, just like the increased DR/Contrast of a lower MP sensor can outweigh higher MP's...

But it makes a point...if you really want "full control" you really can't edit an image fully/correctly until it's final size/use has been determined. And then it needs to be optimized for that presentation only (size/medium).

Who does that? I (almost) *never* do. On the other hand, I don't notice DOF shifting much between minor presentation size changes, even viewed from the same distance. (800 vs 1024, 8x10 vs 11x16).

In my thinking the final use / size / viewing should ideally be known first before you select the kit and the settings you shoot at as once you've taken the shot there are things that you can do little about.

I've long since accepted that my images are very unlikely to be printed twenty feet wide and displayed on a gallery wall and with that comes the freedom to use kit that's cheap and portable yet will still produce a good image that can be printed A3, which is the largest I've ever printed, or viewed on screen at even 100% and still look ok.

 

[sk66]

In my thinking the final use / size / viewing should ideally be known first before you select the kit and the settings you shoot at as once you've taken the shot there are things that you can do little about.

I'd have to agree. So "ideally" an assignment/contract wouldn't be for "head shots" or "web display" images it would be for "8x10 print" or "1024x display."

I've never done that, and I don't know anyone who does...But maybe we (I) should, and maybe those at the very top already do.

 

[woof woof]

No one pays me to take pictures, plenty of people have my pictures from the film days to today but they were given freely

If someone was to ask me to shoot for money I'd want to know the output size they wanted. If they wanted anything from a passport photo sized print to A3+ I know my current kit (cameras, lenses, software and my ability) are ok but anything bigger and I'd have to say No Can Do as it would be beyond anything I've done.

I think that if I knew an image would be printed very large I wouldn't use my MFT kit and I'd probably think my 5D kit wasn't suitable either. I think that once you get into very large prints and very critical viewing the advantages of larger formats probably come more into it.

At smaller sizes like A3+ and smaller and "normal" viewing I honestly think that almost any kit with a sensor from MFT size and upwards (and possibly even smaller too) is probably good enough.