[baw]

I feel the DOF concept isn't overly complex, especially if you separate the optical and the imaging/viewing part.
Optical results in CoC's on the sensor/film, imaging/viewing in a CoC limt, that decides which CoC's fall within DOF.

That's the gist of the "Back to the basics" post above. I'm interested in comments pro or contra this post to see if I have to change something in my reasoning

In practice you may be interested in this concept. It's a bit different from the hyperfocal focussing thing: focus on the background and stop down as needed to resolve foreground objects. useable for landscapes etc.

Problem with the DOF preview button to me is the small size of the LCD screen. Allmost everything seems to be sharp on that screen (or in the viewfinder, but then everyhing turns allmost dark when stopping down.

[sewebster]

I feel the DOF concept isn't overly complex, especially if you separate the optical and the imaging/viewing part.
Optical results in CoC's on the sensor/film, imaging/viewing in a CoC limt, that decides which CoC's fall within DOF.

That's the gist of the "Back to the basics" post above. I'm interested in comments pro or contra this post to see if I have to change something in my reasoning

Like the other guys I have to say that I am tiring of the subject.

Your Back to basics post seems generally good, but I don't think your last example (25mm f/2 vs 100mm f/8) works. The DOF for those two cases is very different (around 20m for the 25mm and 3m for the 100mm). I don't think the final crop and enlarge on the 25mm lens photograph will actually decrease the perceived depth of field down to 3m.

[stevebrot]

Like the other guys I have to say that I am tiring of the subject.

I...am...soooo...tired...zzzzzzzzzzzzzz :ugh:

Steve

[brownargus]

Fascinating discussion. I printed off all the posts to read and digest. Didn't realise that DOF could be so complicated. Accept that CoC is dependent on the final image (print) and agree that DoF can be difficult to assess in-camera using the DoF preview.

I shall continue to use the technique I have used for many years - ensure that the key point in the image is at the point of sharpest focus, e.g. the eye of the bird or head of a dragonfly and make sure that I maximise the DoF by using the smallest aperture commensurate with acceptable shutter speed. And use viewfinder magnifier and a tripod wherever possible.

Thanks all
John

[baw]

To check my example, use DOFMaster. It has the option to select a CoC limit iso camera.
Just use a 4 x higher CoC limit for the 100mm because you magnify the 25mm crop 4 times. so .001 and .004 or .002 and .008 etc.

I still had this Excel sheet on my harddisk.
All distances are calculated using DOFMaster.
For those of you still interested:

Code:

Distance from lens where diff. CoC's originate from.		
Both lenses focussed at 10m.		
		
CoC (mm)	25mm f/2	100mm f/8
		
0.12	2.1	5.1
0.11	2.2	5.3
0.10	2.4	5.6
0.09	2.6	5.8
0.08	2.8	6.1
0.07	3.1	6.4
0.06	3.4	6.8
0.05	3.9	7.2
0.04	4.4	7.6
0.03	5.1	8.1
0.02	6.1	8.6
0.01	7.6	9.3
		
0.00	10.0	10.0
		
0.01	14.7	10.9
0.02	27.7	11.9
0.03	235.8	13.1
0.04	infinity	14.6
0.05	infinity	16.6
0.06	infinity	19.1
0.07	infinity	22.4
0.08	infinity	27.3
0.09	infinity	34.8
0.10	infinity	48.1
0.11	infinity	77.6
0.12	infinity	210.6

Compare CoC 0.01 for 25mm vs. 0.04 for 100mm
Or 0.02 vs 0.08 or 0.03 vs 0.12

[sewebster]

To check my example, use DOFMaster. It has the option to select a CoC limit iso camera.
Just use a 4 x higher CoC limit for the 100mm because you magnify the 25mm crop 4 times. so .001 and .004 or .002 and .008 etc.

I don't think it is that simple. Once the image is captured by the sensor, certain parts of it will be "permanently" sharp because the CoC was smaller than the pixel size. So you can enlarge the image without starting to detect out-of-focus.

I'm starting to get a little out of my comfort zone with this reasoning, but I would really doubt that your 25mm vs 100mm test works in practice. You could try it and post photos, but I'm not really super-interested at this point.