The GetDPI Photography Forum

Great to see you here. Join our insightful photographic forum today and start tapping into a huge wealth of photographic knowledge. Completing our simple registration process will allow you to gain access to exclusive content, add your own topics and posts, share your work and connect with other members through your own private inbox! And don’t forget to say hi!

Advise on Tilt

jagsiva

Active member
This would be from the same exact place. The larger format with the same focal length lens at the same subject distance would have greater DoF.
So are you saying that if I take my Sony A7R, FF camera, stick it on a tripod with a 55mm lens and shoot a subject. Then put the camera in crop mode (APS-C) - same camera, same lens, same camera position, same subject position. Only difference now is that the sensor is a different format, as in cropped to 1.6x), I am going to have a different DoF in the resulting image?
 

danlindberg

Well-known member
Tricky image. As others have mentioned I too believe in this case a focus stack is the better solution at f-stop just before diffraction. The one 'tip' I can share is that you say the cam is close to the ground with rise. I would go the other way around and put cam higher and introduce fall instead until I get a similar perspective - this would mean a greater distance to the first objects and a lot less exposures for the stack! With my SK28 I would probably only need three exposures in this scene. (not counting exposure bracketing)
 

torger

Active member
I think Shashin is making it more confusing than it needs to be, although what he says is correct. Anyway, there's no free lunch. You can't "win" depth of field by shooting a smaller or larger format.

The only thing you gain by shooting a small format is that you can get the same FoV and DoF with a shorter focal length and larger aperture combination which means shorter shutter speed. In some cases short shutter speed is important of course.

The traditional view on DoF is based on a specific viewing distance and a model of vision, and to make the story short everything above 4 megapixels or so is a waste with the old-school model. However, when people print large and watch close you resolve a lot more than 4 megapixels. For 645 film the CoC was traditionally 47um, which is a blur corresponding to the diffraction you get when you shoot at about f/32.

The depth of field calculators becomes pretty useless as a tool for optimizing your deep depth of field focusing if you put in that large CoC, you end up shooting with less than optimal settings.
 

Shashin

Well-known member
So are you saying that if I take my Sony A7R, FF camera, stick it on a tripod with a 55mm lens and shoot a subject. Then put the camera in crop mode (APS-C) - same camera, same lens, same camera position, same subject position. Only difference now is that the sensor is a different format, as in cropped to 1.6x), I am going to have a different DoF in the resulting image?
Yes.
 

torger

Active member
Tricky image. As others have mentioned I too believe in this case a focus stack is the better solution at f-stop just before diffraction. The one 'tip' I can share is that you say the cam is close to the ground with rise. I would go the other way around and put cam higher and introduce fall instead until I get a similar perspective - this would mean a greater distance to the first objects and a lot less exposures for the stack! With my SK28 I would probably only need three exposures in this scene. (not counting exposure bracketing)
Would that actually change anything? If you see the same objects in the foreground, it's the same distance regardless if you have rise or fall. You would need to have field curvature working with you to make a difference, and I don't think the HR32 has much of that :)

Sure if you change field of view to cut away foreground it becomes less problematic, but then it's not the same framing.
 

jagsiva

Active member
I think Dan meant to drop the lens (or raise the back in the of case RM3Di). This would frame the lower part albeit with a different perspective at a longer distance.
 

torger

Active member
I think Dan meant to drop the lens (or raise the back in the of case RM3Di). This would frame the lower part albeit with a different perspective at a longer distance.
Ahhh... you're right I was not thinking straight :). As it's an ultra-wide it can have a significant effect.
 

Shashin

Well-known member
I think Shashin is making it more confusing than it needs to be, although what he says is correct.
What am I making confusing? Someone said that sticking a particular focal on any format camera would result in the same DoF. That is not true. What is actually confusing? Especially, since you just admitted it was true.

Anyway, there's no free lunch. You can't "win" depth of field by shooting a smaller or larger format.
Well, you need to make up your mind. You can't in one sentence state DoF is a product of format and then say it is not.
 

Shashin

Well-known member
The traditional view on DoF is based on a specific viewing distance and a model of vision, and to make the story short everything above 4 megapixels or so is a waste with the old-school model. However, when people print large and watch close you resolve a lot more than 4 megapixels. For 645 film the CoC was traditionally 47um, which is a blur corresponding to the diffraction you get when you shoot at about f/32.

The depth of field calculators becomes pretty useless as a tool for optimizing your deep depth of field focusing if you put in that large CoC, you end up shooting with less than optimal settings.
Sorry, that is not what the model actually says. Many film-based systems out resolved 4MP and the modeling of DoF was still valid (actually, resolving power is not part of the model). All of this works at any defined viewing distance. And there are several standards. And you can make your own, like new-fangled digital viewing distance of pixel peeping, which is not a real-world viewing condition.

Photography is more than an exercise in resolving power.
 

torger

Active member
Well, you need to make up your mind. You can't in one sentence state DoF is a product of format and then say it is not.
Because the example of just changing the format and keeping all other parameters the same is useless, you change the field of view and get a different picture. If you change format you need to change focal length to get same FoV and voila you end up with the same DoF challenge.

Traditional CoC was set so you fit the same amount of CoC's on the format area, regardless of the resolving power capability of the format, ie the traditional view on DoF says you would have no more difficulties with DoF with 8x10 (220um CoC) than 135 (29um CoC). The traditional DoF model I'd say that is just as relevant as saying that 8 bit sound is all you need.

If you want to maximize resolution your system is capable of the old way of defining CoC won't do you much good. Anyone that uses a DoF calculator in the field knows that. Jagsiva's challenge is pretty clear, maximize resolution in the image in the best possible way, not lower the standards by bringing in a 47 um CoC from back in the days.
 

torger

Active member
Photography is more than an exercise in resolving power.
I agree. In general I think people should dare to shoot at smaller apertures than they do.

However, if you have very expensive gear which one of it's key advantages actually is resolving power I think it's a good idea to perfect your shooting technique to make the best out of that.

DoF calculators can be good at optimizing your focus distance and figuring out if tilt is going to help or not, but not without adapting the CoC to something that's better suited for your system's capabilities. If you feed your DoF calculator with a traditional 47um CoC for your 80 megapixel 645 back, the results out of your calculator won't give you much guidance at all, it will practically say it won't matter much where you focus. That's not contributing to good shooting technique. MF tech photographers have of course noticed this and have therefore figured out alternate models that actually help them to make sharpest possible images, and this includes having a smaller CoC.
 

timparkin

Member
...

With downward tilt, I was losing the focus at the top of the image (distant part of the image). I ended up bracketing for focus and will be stacking. However, is there a solution with tilt in this case?

On exposure, some of the scenes were 6-8 stops off, so I don't think there is any way around it than bracketing and blending.

Appreciate your thoughts.
It's a tough one but if you had to use tilt instead of focus blending you can think about the problem in terms of 'psychological sharpness'.

In other words you take a combination of where people's attention will be in the picture and what 'edges' you have to play with to signal sharpness to a viewer.

In this picture the sharpness is indicated by the pebbles in the foreground and where the river crosses rocks and possibly by the right hand side wall. I would use a combination of tilt and swing to bring these areas into focus with the depth of field projecting at about 45 degrees from horizontal...



I would probably do this AND stop down to f/11-f/16

You could alternatively use tilt and then take one shot at f/8 and one at f/16 or more. Blend these two together and you end up with your important stuff on the right plane and sharp at f/8 and everything else at f/16.

Tim
 

Shashin

Well-known member
Because the example of just changing the format and keeping all other parameters the same is useless, you change the field of view and get a different picture. If you change format you need to change focal length to get same FoV and voila you end up with the same DoF challenge.
I do not NEED to do anything. I could have a larger field of view. There is not "correct" way of taking any particular image. The is no pre-determine variable. Photography is not an exercise in comparison images.

Traditional CoC was set so you fit the same amount of CoC's on the format area, regardless of the resolving power capability of the format, ie the traditional view on DoF says you would have no more difficulties with DoF with 8x10 (220um CoC) than 135 (29um CoC). The traditional DoF model I'd say that is just as relevant as saying that 8 bit sound is all you need.
DoF models the human perception of sharpness of an image. It does not say that higher frequency information is not significant nor imperceptible--you can resolve details smaller the the resolving power of a system, you do not conclude that that information is irrelevant. It is false to say DoF and CoC is saying there is a base minimum like 8-bit sound or 4MP images--that is your false assumption.

If you want to maximize resolution your system is capable of the old way of defining CoC won't do you much good. Anyone that uses a DoF calculator in the field knows that. Jagsiva's challenge is pretty clear, maximize resolution in the image in the best possible way, not lower the standards by bringing in a 47 um CoC from back in the days.
You are just maximizing the resolving power of the focal plane irrespective of what that context is to the entire image. That is not the same as maximizing the resolving power of the image--DoF actually adds resolution to parts of the image.

CoC does not indicate that you need to use any particular aperture. All it is doing is define what is being perceived as sharp--the part of the image smaller than the CoC is still significant and still perceptible. Your method of saying the only definition is at the focal plane when viewing at 100% is just as false as how you are trying to portrait what you think DoF and CoCs are about. And if you knew anything about DoF calculators, you would know you can actually put in your own CoC definition that you think models the DoF for you. But it all comes back to how an image is perceived and those relationships don't change.
 

torger

Active member
Tim, now you're thinking like a large format photograper, I like that :). But unfortunately the RM3Di can't do tilt and swing in combination.
 

torger

Active member
And if you knew anything about DoF calculators, you would know you can actually put in your own CoC definition that you think models the DoF for you
Well, I have actually coded a few of them, always with custom CoC of course... far from all DoF calculators have that though

It comes down to how you use the tools. Some just put the focus where they want it sharpest and stop down with some sutiable amount. Traditional DoF/CoC would work for that. It seems to me that is your way to go. With this way of work you typically can do without a DoF calculator at all and just go by experience.

If you however use things like hyperfocal and want "everything about equally sharp", and want to know which distance to focus at then there's a great need to adjust the CoC. This is the context I'm referring to, and in this context the traditional CoCs don't work very well.
 

Shashin

Well-known member
I agree. In general I think people should dare to shoot at smaller apertures than they do.

However, if you have very expensive gear which one of it's key advantages actually is resolving power I think it's a good idea to perfect your shooting technique to make the best out of that.

DoF calculators can be good at optimizing your focus distance and figuring out if tilt is going to help or not, but not without adapting the CoC to something that's better suited for your system's capabilities. If you feed your DoF calculator with a traditional 47um CoC for your 80 megapixel 645 back, the results out of your calculator won't give you much guidance at all, it will practically say it won't matter much where you focus. That's not contributing to good shooting technique. MF tech photographers have of course noticed this and have therefore figured out alternate models that actually help them to make sharpest possible images, and this includes having a smaller CoC.
There is not much I disagree with here. But just as I would not say only shoot at f/8 because you have 5um pixels, I would not say f/32 is irrelevant either. Aperture is a tool. It has effects. Understand the process and make the compromises with that.
 

Shashin

Well-known member
If you however use things like hyperfocal and want "everything about equally sharp", and want to know which distance to focus at then there's a great need to adjust the CoC. This is the context I'm referring to, and in this context the traditional CoCs don't work very well.
Macro at f/16, effective aperture much more.



100% crop.



I guess I don't agree.
 

jagsiva

Active member
I guess I don't agree.
try to compare an 80MP back at f/9 or f11 against f16 or f/22. You may think differently.

Take a look at the post at f22 where the poster was losing resolution. Also, this was with a 60MP back, it only gets works with an 80. The original images have been taken down, but you can get the conclusion from he comments.

http://www.getdpi.com/forum/medium-...s/47332-my-thoughts-after-48-hours-iq260.html

Torger mentioned expensive gear and getting the most out of it with optimal f-stops, I would add to this, the effort of getting to a location. Specifically, I wanted to get the best I could given the trouble I had taken to get there. Had this not been the case, I would have taken my A7R or D800E. I didn't have to lug an MFDB kit and heavy tripod with CUBE. So while all the advise on resolution not being the be-all and end-all is great, it is a little motherhood and apple pie without context. Surely, we are all beyond a first year college course in photography. We all realize that we would still rather have the shot.

My question was how to maximize DoF in as few shots as possible given the subject.

In any case, we are way off topic. To wrap up the recommendations here:

1. Focus stack is likely still the best option
2. Could play with tripod hight and use back rise to get similar framing with longer subject distance
3. Play with multiple swing planes (vertical focus planes on the sides) and stack in post.
 
Top