Paul Timpa writes...
"Great note on the Depth of Field comparison between the crop bodies and full frame bodies.
Everything you wrote was 100% accurate. I would say that Joey's point is a valid one, and is really what the "debate" is about. In practical use (say for a portrait of someone's face from forehead to chin) a photographer using a full-frame body with 105mm prime lens would step much closer to the subject than someone with a crop body and the same lens. This change in distance affects the depth of field, as you stated (making it shallower). So for any given framing, an identical lens/aperture combo on a crop body vs. full-frame body will in fact result in less depth of field for the full-frame body. If we were to watch two wedding photographers trying to get the same exact shots ("framing"), the one with the full frame body would always be closer to the happy couple, and the resulting shots would have less depth of field."
Best regards,
Paul
Dear Paul,
"As a pun, it is all a matter of perspective isn't it?
When working, I seldom worry about the field of view of one body vs another, as I use the lens that will generate the depth of field I want for the perspective I have chosen. I think sometimes people try to complicate things too much."
Thanks for the mail Paul.
***
John Write writes (no pun intended)...
"I still have not found a consistent set of steps for sharpening the raw images from the 50D. This is in contrast to the 20D, which seemed very consistent in what I considered the best steps.
Do you have a starting point you use in photoshop for sharpening the images?"
thanks,
John
Dear John (no pun intended),
"My starting point is what I call a capture sharpen and I do it in Lightroom at the time of import. I set a radius of .7, sharpen to 50 on the slider and a detail of 20. From there on out, I do all sharpening on an as needed and image by image basis. I would not recommend taking a blanket approach to sharpening your images. Some lenses are sharper than others, some images don't require as much sharpening. My standard practice is to not sharpen images beyond minor touch up from import sharpening until I make a print and then I adjust my sharpening to the print size I'm making based on the file I'm working with.
Contrary to popular digital post processing culture, I don't like my images to look over sharpened when printed. I sharpen only to the point that fine detail is visible where I want it to be visible and not beyond. I rely on my lenses to be the primary source of sharpness in the image. With post processing sharpening, I believe less is more. Your mileage may vary."
Thanks again John
***
Steve Lin writes...
"Having been a photographer since the mid-80s and an engineer by education, I'd like to think I'm both a photographer and a techno-geek. I still remember doing experiments with lenses in high-school physics. I'll admit that it does bother me a little when I see statements that, in my scientific opinion, are incorrect. Back in the film days, didn't you ever wonder where the depth of field marks on the lenses came from? From those equations with "circles of confusion...
Anyway, I suspected you had this discussion with many before me. I figured it was worth a shot to see if I could bring something new to the table. .. I know all about the futility of trying to win an argument over the Internet, so I don't take things personally...
...By my aging but not so old, yet, eyes, the DOF in the 30D appears to extend from about the 19.5 in. mark to the 20.5 in. mark. On the 5D, the DOF appears to extend from the 19 in. mark to the 20.5 in. mark. That's a difference of about 1.5x which roughly matches the difference in sensor sizes. FWIW, you might want to check out this article on the effect of sensor size on DOF. I found it highly accessible and logically coherent..."
http://photo.net/learn/optics/dofdigital/
Thanks for your time. I appreciated our short discourse.
Steven
Dear Steve,
"...A crystal can be viewed from many angles. My emphasis is on teaching people how to make photographs, not on how to understand optical mathematics. If you start a conversation like this around most successful and high-end photographers, they'll throw a brick at you and tell you to shut up, or at the very least look at you like you're a complete moron. There's a dividing line out here. People who take pictures for a living and people who don't. Most of my colleagues who take pictures for a living ignore this crap because they don't want to be associated with the gear-heads and they have little patience for the gear head discussions in their presence. Some of us are gear heads too though, like myself, but we've learned to keep it in the closet. Being a gear head first doesn't win you any friends in the photography business.
Extinction point, the point where there is no longer a change in discernible detail. It's relative of course, magnification will change the extinction point, as will reduction.
But, to illustrate that there is no functional difference between the bodies when all else is constant, here are 4 crops, with the 30D image as the base shot and the 5d image as an overlay. This is a sequence with the 5d brought out on a second layer in increments of 0% opacity, 33% opacity, 66% opacity and finally 100% opacity. To my eye, the depth of field does not change as you sequence through the overlays, only the resolution of the cropped 5d image which produces a slightly less detailed image due to the lower pixel count (which I compensated for by transforming it to the same effective size as the 30D shot, but that's another issue all together).
Funny you mention the dof indicator on lenses, as if you test this yourself using such a lens, you'll find the dof indicator stays at the same position between the bodies. The reason for this, is nothing with the lens has changed. Circle of confusion is controlled by the design and build quality of the lens. The pixel density of the sensor will have an effect on detail rendered (due to a few different factors), none of which have a control over DOF to any meaningful degree.
Canon EF lenses are all designed with the same circle of confusion factor for a 35mm format (though they may vary from copy to copy due to construction variations), I can't nor do I even want to remember the value they use, I'm sure it's somewhere in their technical documentation on their web site. I don't know if the EF-S lenses are designed with a different value or not, but I don't use those lenses, so I've never really compared their characteristics. My guess is there may very well be a slight difference due to the lens designs and distance to the sensor...
...The functional depth of field will stay the same if the lens doesn't change and the distance to subject stays the same and the aperture doesn't change. Where you factor in the different circle of confusion (and this is the real circle of confusion IMO) is when you adjust the field of view to accommodate the difference between the crop and full frame sensor by repositioning. The dof calculation applies to a situation where you are changing the distance to subject to achieve the same FOV. In that case, dof is different, no doubt about it.
In the field, I don't change position to accommodate a different body because I've already selected my POV. I change lenses to obtain a different FOV and use the same dof calculation process for that lens regardless of the body because no matter what body I'm using a 105mm lens is going to give me a fairly specific DOF at a certain distance to the subject with a specific aperture. Besides, changing position may scare the moose off or make it charge, or change the composition or light.
I preach photography over mathematics to my students. It teaches them to think about the photo and not about the engineering. Not always the easiest thing to do I might add. When a student goes off into gear-head mode, I politely shut them down. We're learning the art, not the science and most gear-head amateurs understand the science long before they grasp the art. If they want to get involved with designing lens setups, I'll point them to the local university."
Thanks for the feedback Steve. It's always a pleasure to hear different viewpoints.