Large format camera movements. What happened here?

[Steven Lee]

A couple of months ago I have acquried the Chamonix 45F-2 and I am on my 2nd box of HP5+. Initially I have not been using the movements much, just practicing the mechanics of shooting large format. Today I finished developing my first batch of sheets exposed while using camera movements. The results are wildly inconsistent! On most images I was able to troubleshoot what happened, but the one below is puzzling.

Here's a reasonably high-resolution scan. Sorry, my scanning setup for 4x5 is shit at the moment, but this should be enough to illustrate the problems.

• The brick wall on the right is fairly sharp
• The railing on the roof on the left is fairly sharp
• The bridge directly behind the light tower is sharp
• But everything else is not. Basically I have sharp areas in front, sharp background, and blurry mess in the middle!

I remember using the rear tilt to correct for the keystone effect somewhat, and I focused on the bridge and also on the side (the areas I focused on are indeed sharp), but how the hell I ended up with all other parts of the image being blurry? Where is the plane of focus in this image?

What movements would you have used?

 

[Ian Grant]

What aperture did you use. The only movement I'd have used is front rise to prevent the convergence, you focused on the distant part, you need to focus 1/3 in and then stop down to around f32 to allow DOF to give you better overall sharpness.

Ian

 

[Don_ih]

Looks like the bottom of your sheet is out of focus. When you tilted the back of your camera, did you also tilt the front to be parallel with the back? Without doing that, you will throw some part of the image out of focus.

You would want the camera back to be parallel to the subject. The front should be parallel to the back. If your camera is angled up, with the back tilted forward to be parallel to the scene, you'll notice that making the front of the camera parallel to the back is exactly the same as using front rise with the camera not angled up.

 

[neutron450]

Any chance the film was warping or bowing in the holder?

 

[ic-racer]

Wide angle with a view camera has its own challenges. What I do is first make sure the front and rear standards detents are parallel with a simple laser test.

Then, to obtain the best depth of field and sharpness, I use the popular focusing formula which optimizes both circles of confusion and Airy disks. Link here "How to set the F-Stop"

 

[gone]

It doesn't appear to be a focus issue to me, it looks like an exposure problem. You'll have to make some exposure compromises if you don't want that bright sky and bridge blown out.

 

[koraks]

I remember using the rear tilt to correct for the keystone effect somewhat, and I focused on the bridge and also on the side (the areas I focused on are indeed sharp), but how the hell I ended up with all other parts of the image being blurry?

The plane of focus in your image seems to run from about halfway along the vertical axis in the foreground to about 1/4 from the top in the background.
The image looks like excessive tilt has been used to me, which might match your description quoted above.
To make verticals vertical, just level the camera with the back vertical, then use front rise (or fall) to get the framing right. If you run out of image circle, you run out of luck and will have to live with non-vertical verticals

What movements would you have used?

A little front swing and perhaps a tiny bit of front tilt, but not much. I'd have handled as much as possible by just stopping down.
The more movements you use, the less benefit you get from stopping down - technically this assessment is debatable if you're nitpicky, but in practice it tends to work that way, at least for me.

 

[Andrew O'Neill]

I think when people first start with LF cameras, they want to use and think that they have to use movements (and lots of them) all the time. For this particular image, I would have used almost no movements... maybe a bit of back forward tilt (I'm assuming you have the camera tilted up), the front standard parallel to back... and probably lens raised a bit. No swings.

 

[GLS]

It looks like you've used too much tilt, and possibly too much swing too, without stopping down sufficiently.

As others have said, as a general rule you should level the tripod head first to get the back of the camera level and eliminate vertical convergence, then use front/rear rise/fall to get the framing you desire.

With a subject like this having compositional elements very near the camera which run the entire vertical length of the frame (i.e. the structure on the right), tilts are typically counter-productive because they will quickly throw a large part of that vertical element out of focus. For this scene I would have avoided using any tilt, and instead just used a tiny bit of swing with focus points on the bricks at frame edge right, and at a point on that lattice of girders under the bridge about one third into the frame from the left, then stopped down to get everything else in focus.

In most scenes it really doesn't take much tilt/swing to have the desired effect, especially with a wide angle lens. Usually 5 degrees or less is sufficient.

 

[Steven Lee]

This awkward situation when you're grateful to every person responding, but posting 8 thank you notes will be tiring for everybody. While I need some time to proecess and research some of your answers and provided links, let me thank you all for replying and answer (and ask) some questions:

A little front swing and perhaps a tiny bit of front tilt, but not much.

That would be a swing to the right to help balance the focus on the right-side wall?

you should level the tripod head first to get the back of the camera level and eliminate vertical convergence

Do you mind explaining this a bit deeper? Leveling would not have allowed me to get as much of the foggy bridge in the frame as I wanted, so I was trying quite hard (with rear tilt) to do it and I was surprised how ineffective it was, and - according to the comments here - that's mostly explains my focusing struggles.

it looks like an exposure problem. You'll have to make some exposure compromises if you don't want that bright sky and bridge blown out.

Actually the sky is not blown out, what you are looking at is a low hanging thick white fog. I did not get the desired effect though because it was quickly moving and its texture disappeared due to motion blur. I want to retake this image.

TLDR is that I used too wide of an aperture, it was f/16 and my tilt was too aggressive - I should have tried using a front raise with a less aggressive camera angle?

 

[koraks]

That would be a swing to the right to help balance the focus on the right-side wall?

In the direction of the plane of the bridge, as it were. I think that's what you mean?
The rationale behind it is that the left wall is further away from the camera than the right wall, and the bridge also is at an angle compared to the viewing line, so combined it makes sense to move a little along with that. Hope it makes sense.

it was f/16

Those old guys called themselves 'f/64' for a reason f/16 is pretty far open in large format terms.

 

[Steven Lee]

@koraks yep, makes sense. Just clarifying.

 

[GLS]

Do you mind explaining this a bit deeper? Leveling would not have allowed me to get as much of the foggy bridge in the frame as I wanted, so I was trying quite hard (with rear tilt) to do it and I was surprised how ineffective it was, and - according to the comments here - that's mostly explains my focusing struggles.

In order to avoid convergence of vertical lines in the subject, the plane of the rear standard must be also be oriented vertically. Once it had been set thus, to achieve the framing you wanted in this case front rise and/or rear fall could have been used. The rise/fall limits of your camera may not be sufficient to achieve the required framing in all cases. If so, additional rise/fall can be achieved by tilting the camera bed/rail up or down, then using tilts of the standards to bring them back to the vertical; this latter technique is known as indirect rise or fall. Of course, all of this assumes the lens you are using provides a large enough image circle to accommodate the movements you are trying to perform.

Eyeballing this composition, I imagine you should have had enough direct front rise to still get the bridge in. What lens was used on this? 90mm?

 

[Steven Lee]

@GLS yes, this was taken with the Nikkor 90mm f/8. Thank you!

 

[GLS]

Ah, OK. That lens has a very generous IC of 235mm at f22, so would have easily handled the necessary movements.

Of course you may not always want to completely eliminate convergence, but that is down to artistic intent and personal taste.

 

[Ian Grant]

Of course you may not always want to completely eliminate convergence, but that is down to artistic intent and personal taste.

A valid point, sometimes total correction when looking up actually has the opposite effect and looks decidedly weird.

TLDR is that I used too wide of an aperture, it was f/16 and my tilt was too aggressive - I should have tried using a front raise with a less aggressive camera angle?

Also my point, in the first reply, about where you place focus, DOF will easily bring distant objects into focus, it struggles more as you get closer, so the old adage is focus 1/3 of the way in.

Don't be afraid of stopping down past f22, f32, f45, or even f64. Someone analysed John Sexton, one of our member's, images in a Publication and his image quality at f64 was superb. The book gave exposure data, and it was surprising how often Sexton went to f32 and f45.

One of my Schneider 150mm lenses has an aperture scale to f90, it's the late version of the 150mm Xenar, the sharpest of them all the f5.6.

Ian

 

[cliveh]

I think when people first start with LF cameras, they want to use and think that they have to use movements (and lots of them) all the time. For this particular image, I would have used almost no movements... maybe a bit of back forward tilt (I'm assuming you have the camera tilted up), the front standard parallel to back... and probably lens raised a bit. No swings.

Amen. You don't need any movements for that image.

 

[MTGseattle]

Steven. Where the heck is that? Is it a place you can get back to easily? I'm not going to chime in regarding movements. It's been years since I had to really "know" any of that stuff. I like the overall composition, and I'm viewing it on a laptop which leads me to shy away from commenting too much regarding resolution. I too think stopping down would help a lot. Were I to borrow your tripod holes so to speak, I would maybe shuffle myself over to the left a bit and see how that looked. I enjoy the arc of the bridge a lot though.

 

[Steven Lee]

@MTGseattle That's the lens you sold me, BTW Link to location. @Ian Grant thanks for the tip! To remove all possibility of misunderstanding, which element in the scene perfectly lands on "1/3 of the way in"? Thanks again.

 

[Alan Edward Klein]

Steven, Just a side reminder. Your camera's rear standard has base and asymmetrical tilts. Note that it would not change the recommendations necessarily.

 

[Steven Lee]

@Alan Edward Klein yep, I remember. My understanding is that it's mostly a focusing aid, as the rotation axis is well-known so you can focus on it without having to refocus later. TBH I am still not "feeling" the convenience, but probably because I'm so clunky that these optimizations aren't helpful to me yet

 

[btaylor]

Ha! I thought that looked like Fort Point. Burn through a hundred sheets and it will all become clear!

 

[grat]

Those old guys called themselves 'f/64' for a reason f/16 is pretty far open in large format terms.

To be fair, f/64 is more of an 8x10 setting. Diffraction is going to start kicking in around f/32 on 4x5. Sometimes it's worth it for the extra DoF, but it will start affecting sharpness.