[normhead]

One of my most used features on the Super Program was the depth of field preview lever next to the lens.

After grappling with this for a while, I simply started bracketing for depth of field. if the shot looks like a keeper, shoot at intervals for the best shot. In my experience without the DoF preview, often F-16 is the best setting, sometimes 5.6 is, but, the thing is, it's not costing me to take lots of images. I usually shhot in 2 stop intervals, Wide open 5.6 11, 22 and so on. I think I can probably spin that Aperture dial faster than you can evaluate depth of field visually. I know I can just because my eyes have deteriorated with age. That's my solution. Digital is different, in that shooting 100 images and then tossing 95 doesn't cost me much. With your Super Program, that probably wouldn't be such a good strategy.

[dlacouture]

The DoF optical preview is just a joke anyway, as modern focusing screens are heavily optimized to handle kit lenses and, as such, cannot handle fast lenses or small apertures that well...
Using a f/1.4 lens, you won't see any differences in the viewfinder until f/2.4 (just like the camera itself, that's why stop-down metering is such a mess!).
And a f/1.4 shot is wildly different from what you see in the viewfinder...

[TomTextura]

I care based on what I see, not what the math tells me is happening. So while most of this esoteric discussion is true, how many of us think about this as we are pushing the shutter release?

Yes, when it comes down to it, it really is about what you see in the viewfinder. This stuff is not on my mind while I'm taking pictures but the math and such are things I like to understand and read about because I generally find it interesting and often enough practical. What I feel is the problem is that some, particularly beginners, get the wrong idea when they read about EFL of lenses on their APS-C sensor cameras: they end up thinking that they would see farther, more zoomed in, in their viewfinder. I've seen many times when EFL is talked about that way and that was my main motivation for posting.

Exactly... I don't know a single guy either from when I was in Photo Arts at Ryerson, to any of the working photographers I've known, who gave an owls hoot about the math or physics of lens performance. It's about getting your eye to the viewfinder and saying " I can work with this." That's a photographer. Analyzing charts for comparative depth of field etc. FOV etc. doesn't mean a thing, until you put your eye to the viewfinder and see what it looks like.

Maybe that's your experience but I personally wouldn't go that far. I definitely agree that taking great pictures doesn't necessarily rely on knowing the technical side of things, in this case the focal length of lenses and their respective AOV on different sensor formats. People who leave their cameras set at Auto but have a good sense of composition and timing can take great pictures. Pixel peeping is uninteresting to me but having a theoretical and technical understanding of all that makes up photography does not distract me from the main objective: taking great pictures. I believe that the more I know the better off I am. It's helpful when considering making purchases, when packing the bag for a photo outing, when deciding on a set up for a concept shoot, or when pre-visualizing a picture. And all the technical mumbo-jumbo and discussions about camera/lens performance is a huge part of this forum.

[TomTextura]

One of my most used features on the Super Program was the depth of field preview lever next to the lens. This is a feature missing from from the K-r. For cripes sake, the digital lenses don't even have the focus distance and DOF scales engraved! Sure, I found that the green button was reprogramable to preview DOF, which I have done - at the expense of other useful functions. I'm pretty sure this isn't true, but it almost seems to be a plot by the major camera manufacturers to not let us SEE what the lens will see during the exposure. Silly me. I thought the biggest reason for aperture priority automation was DOF control.

The DOF preview is combined with the off/on switch of the K-5. It's easy to use, which I do on occasion. I wouldn't be willing to give up the normal use of the green button for DOF preview though.

modern focusing screens are heavily optimized to handle kit lenses and, as such, cannot handle fast lenses or small apertures that well

That's why swapping out the stock screen for the matte S-type (Canon sourced) one from focusingscreen.com is a great option. I recently did and wish I had earlier. It makes a big enough difference when using fast lenses and/or when manual focusing to be worth it.

[TomTextura]

To add to the confusion, Canon saw fit to appropriate "APS-H" for its 1.3x crop 3:2 aspect sensor.

Yeah, what's the story behind that? Is that format even being produced anymore by Canon?

[TomTextura]

It is so rare that the wide open shallow depth of field is the one I want to keep that I shudder whenever I hear about people wanting to pay big bucks to get it. I often find the images of those who seem to think shallow depth of field is really important, irritating. Quite simply stated. Shallow depth of field is more often a problem, not a solution.

A bit off topic but to respond... I agree in as far as that large apertures are not always called for and that nailing focus when shooting wide open is difficult. However, I happen to really like pictures with shallow DOF. I also sort of enjoy the challenge of getting the focus to be right where I want it when using large apertures, more so when I am successful than when I am not of course. Shallow DOF helps with subject isolation and can be a real advantage when the background is cluttered with things you'd rather not have in the pic but can't get rid of other than making them go out of focus. I think RioRico said something to the effect that deep DOF is commercial and shallow DOF is artistic. That is a generalization but I can see how it applies. In the end, horses for courses.

[pingflood]

Yeah, what's the story behind that? Is that format even being produced anymore by Canon?

Yeah, it is being used in the Mark IV.

The story behind it is pretty simple (of the sensor size, not the name, which I have no idea about). The 1.3x crop sensor was the biggest that could be produced with a single exposure in the imager used to make the sensor. If you want a larger sensor they have to be "stitched" together with multiple exposures, significantly adding cost and difficulty.

So, basically, the 1.3x sensor was the biggest one that was economically viable at the time of the first 1D.

[dlacouture]

That's why swapping out the stock screen for the matte S-type (Canon sourced) one from focusingscreen.com is a great option. I recently did and wish I had earlier. It makes a big enough difference when using fast lenses and/or when manual focusing to be worth it.

Man, you slight me here! I'm the OP of the EE-S screen thread!
I'm glad to see so many adopters of this great screen!

[TomTextura]

@dlacouture My bad. Do you agree that having the EE-S screen makes the DOF optical preview less of a "joke"? Did you originally modify your own S-type screen or buy it pre-cut? I had enough trouble installing the pre-cut one (got it dirty, unfortunately; didn't need any shims though, luckily). I would never want to try modifying it myself! Also, I wonder what the new Natural-Bright-Matte III screen is like in the K-30, perhaps a marginal improvement.

[TomTextura]

@pingflood Thanks for the answer! I wonder if Pentax will consider some larger format sensor that isn't a full 35mm. It might make more of the DA lenses usable without cropping and bigger is bigger, right?

[starbase218]

Say you take two pictures of a beetle from the same distance with a 100mm macro lens (a true macro lens capable of 1:1 reproduction, to be specific). Imagine that the picture you take with the APS-C camera has the beetle filling the entire frame. Now imagine that you us a full frame camera using the same lens and from the same distance to get the same 1:1 magnification. The beetle on the full frame image will not fill the entire frame. You will still get the very same 100% magnification from both cameras. What you will get from the FF image though, that you will not get with the APS-C image, is more of the beetle's surroundings. View both those images at 100% and the beetle is going to be the same size on your screen, as long as the pixel count is roughly the same. You will not have any reduction in terms of magnification with the FF image. To have the beetle be the same size on your screen, when your viewing mode is set to fit the screen, then you would have to crop the FF image. But at a 100% viewing size, given that the images have the same number of pixels, the magnification will be identical.

By the same pixel count I assume you mean that both the fullframe and the APS-C sensor both have the same number of (mega)pixels, say 18. If that is the case, the pixels on the APS-C sensor will be smaller than those on the fullframe sensor. Thus, the images, when viewed at 100%, will be the same size. But as you said, the APS-C image captures less of the surroundings than the fullframe sensor does. So on the final APS-C image, the beetle is bigger than on the final fullframe image. And that final APS-C image corresponds to what I saw when looking through the viewfinder, so I would say the magnification does occur there.

But why do you care about the magnification factor? I mean, I don't care if it's 1:1, 1:1.5 or 1:2. If I can get the same framing on the final image with a crop sensor and 1:2 magnification as I do with a fullframe sensor and 1:1 magnification, why should I care about the sensor size (or magnification ratios for that matter)?

This is without going into the better detail that you would probably get from a fullframe sensor though, but that's a different story I think.

[ChrisGVE]

These questions whether the crop factor transform the FoV or the FL is not that much relevant. At the end of the day it's roughly the same. However DoF is a more important matter, I did some rough algebra a few years ago (don't have the details just remember the result) on the difference between DoF for a FF and for an APS. There was some approximation there, but I ended up with something like: DoF (APS) ~ DoF (FF) * crop. So say APS was exactly sqrt(2), using a f/2 lens on an APS would result in a similar DoF than a f/2.8... now these things are approximate but that gives some sense...

Now that explains why I am not very warm about micro-4/3... You get a f/2 lens (and pay for it) but your DoF will be closer to a f/4 on FF... sure it's fast but not that shallow...

[normhead]

RioRico said something to the effect that deep DOF is commercial and shallow DOF is artistic. That is a generalization but I can see how it applies. In the end, horses for courses.

It's a mistake to assume you can't achieve shallow depth of field on an APS-c camera. I didn't go looking through my files t prove a point. This is one of my files from yesterday's walk. I shot at 5.6, but, I would have liked the wings in focus. It is a common refrain of FF shooters, that we need shallower depth of field than this. Well, my point is, you don't always need shallower depth of field, even on your narrow DoF shots. APS-c does shallow depth of field quite well. The issue is, do you need more. A lot of people who favour FF say they do. Fine. But for the 99.9 % of us for whom APS-c is more than adequate, and that statement doesn't make any sense.

What you will get from the FF image though, that you will not get with the APS-C image, is more of the beetle's surroundings. View both those images at 100% and the beetle is going to be the same size on your screen, as long as the pixel count is roughly the same. You will not have any reduction in terms of magnification with the FF image. To have the beetle be the same size on your screen, when your viewing mode is set to fit the screen, then you would have to crop the FF image.

What you will get from the FF image though, that you will not get with the APS-C image, is more of the beetle's surroundings

.

Assumption 1. That the shooter actually wants more of the beetle's surroundings.
Assumption 2. That if the shooter did want more of the surroundings he wouldn't just step back a bit or change lenses, or more commonly, zoom out a bit.


A brief examination of the rampant assumptions in this post.

beetle is going to be the same size on your screen, as long as the pixel count is roughly the same.

But we already established that the pixel count on an 18 Mp APS-c is higher than on an 18Mp full frame, so that's not even worth mentioning.

Since getting into this area of discussion I am amazed at how creative FF users are at trying to frame situations where it would be an advantage to be shooting Full Frame, and how despite the math, the intellectual contortions etc. is very little imagery to back up their claims.

So just for the record...
This is the kind of thing I'd like to know. I don't care about mathematical formula.

I'd like a situation where a guy took some pictures with a full frame, with his full arsenal of lenses available, using an FF and an APS-c, what image would be the better image. Nothing anyone has said, no math, the math says nothing about which image is more pleasing, no images posted, has convinced me, that the APS-c image wouldn't be the best image 50% of the time. Different, as in a a 1.5 stop difference in f-dtp DoF values, does not mean better. IN fact it could also mean worse. On most shots, it does mean worse. To imply that you can't get narrow enough DoF on an APS-c camera is crazy, yet it's proposed on a regular basis.

I'm sorry I can't help out more here. I'd love to have an FF camera to fool around with. I'd be happy to throw up some shots to illustrate my point. However, laying out 3 or 4 grand for a system I don't need, just to make a point is a bit beyond my idea of what's worthwhile.

Imagining some scene where we have to use a 100 mm lens and assuming you'd get more of what you want with the FF is just crazy. You might actually prefer your beetle as rendered by the higher pixel per inch count in the APS-c.

[vonBaloney]

I think this was explained to me in another thread, but I can't get it through my thick skull, so maybe someone can try again. When people say that the DOF is thicker on APS-C (or shallower on FF), what are they talking about? I mean, we've established that a 50mm is a 50mm is a 50mm no matter the sensor size, but DOF does magically change with sensor size? This makes no sense to me. If I am using my FF camera, and am standing 10 feet away from a subject that is in-focus on my 50mm lens at f1.4 and there is some blurred background 3 feet behind the subject, and then I swap that same lens onto my APS-C cam without moving my position, still at f1.4, focus set the same, then the subject is still in focus and the background is still blurred exactly the same amount as before. Now I'll be able to SEE LESS OF IT because the image covers a smaller area, but that blurred background will not suddenly become sharper because I swapped camera bodies to one with a smaller sensor, ALL ELSE BEING EQUAL.

So what the heck is this about? Is it another equivalency formula or what where you are not comparing apples to apples? Because stuff that is out-of-focus can't just snap into focus by cropping out the edges of the image (even with greater pixel density in that leftover area), which is what it sounds like people are saying when they talk about DOF and sensor sizes.

[starbase218]

Well with fullframe you can achieve shallower DOF than with APS-C, but that wasn't his point.

Btw: of course a smaller sensor does not increase the focal length. Focal length has nothing to do with the sensor or its size. It is the distance at which light beams that enter the lens parallel to each other, either converge into a single point on the other side of the lens (for a positive diopter lens), or the point from which the lightbeams appear to originate (for a negative diopter lens). That's basic physics.