Originally posted by pschlute Except that the more you crop an image, the more you increase the "circles of confusion", and the less likely the image will have acceptable focus. Take your image in post #8. Crop the back three subjects so they are the same size as the front three and see if they are in acceptable focus
There are really no short cuts to stopping down the lens if you want a wide DOF
Smaller sensor with denser pixels and shorter lens does it every time.
Too funny, I shot that image the way I did because I was tired of the typical "Straight line parallel to the lens method fo shooting group portraits. I wanted something different. So, you're asking me to do something I specifically chose not to do, to prove your point.
Because I've worked on the images at full size furthest away images I know they are slightly soft and slightly out of focus. A calculation made consciously to achieve a compositional goal. But at that size in acceptable focus. So I achieved my goal.
I'm not sure if you read my post because I pointed out in that post that stopping down would have reduced my shutter speed so much that I would have had to ask the people to consciously try and stand still. That's not a good strategy for successful photography.
If your theory about expanding circles of confusion were correct, then the whole image would be soft.
Essentially what I've done is shot a 9mm and enlarged the circles of confusion recorded on the small sensor and yet, there Is no image degradation caused by enlarged circles of confusion.
I will believe this might be possible when I see an example. In my experience circles of confusion are a non factor in photographic enlargement. I have seen instances where the subject was slightly out of focus, like the ones in that image, and having the lack of sharpness enlarged to visible levels. But DoF is normalized for 8x10 images, so it should the same enlargement at any size of sensor for a given image. I have not seen is a circle of confusion enlarged, where it wasn't the softness used is invisible in one image, and visible when enlarged.
If enlarged circles of confusion is a thing, how is my Lumix ZS100 able to beat my K-5 and almost equal my K-3 in resolution? Why do those enlarged circles of confusion not degrade the small sensor image? After all, the APS-c sensors are 4 times the size. The circles of confusion on the small sensor image should be 4 times the size. And yet there is no visible difference. The issue being, if you are in sharp focus, the circles of confusion are so small, 4 times the size still isn't an issue. You need to get well beyond acceptable focus before enlarging your circles of confusion becomes an issue. That's part of why it's called acceptable focus. It's all based on the size of the circles of confusion. The whole idea of depth of field is defined by acceptable focus. SO DoF already incorporates the circles of confusion in the way it's calculated. If you get more depth of field from a wider lens, the enlargement of circles of confusion is already compensated for. You don't get to factor it in a second time to make a point.