PDA

View Full Version : Camera movements and sharpness



Draganski
22-Apr-2015, 13:03
I am new to LF photography and have just started using camera movements.

My question is: does sharpness decrease a lot if i use a lot of rise? I know that there can be problems with vignetting, and i check all four corners of the ground glass to ensure good coverage. my impression is that the sharpness is reduced when i use maximum rise.
I use a fujinon 150w lens on a chamonix 045-n1. The lens covers the maximum front rise of the camera as far as i can tell.

I have ordered betterscanning scan holders, and i will attach the camera bag to the tripod to ensure more stability, just to rule out some of the factors that come to my mind.

Also i got some skewed negatives, obviously the film was moving in the fidelity holders during transport, is that also a factor for sharpness?
Thanks, dragan

Preston
22-Apr-2015, 14:09
Dragan,

Sharpness should not be affected by using front rise. With the Chamonix, you have to be careful when using front rise so that you don't unintentionally add some lens tilt. After doing your rise, be sure to check that the amount of tilt has not changed.

Is the perceived lack of sharpness the same across the entire negative, or is it localized. Without that info, we'd be shooting in the dark.

--P

vinny
22-Apr-2015, 14:46
Maximum rise with that camera and that lens is putting the image circle to the test. Yes, sharpness will be decreased because nearly all optics are sharpest at the center. Just because you don't see vignetting, doesn't mean you've used too much movement. In most cases stopping down increases coverage so using lots of rise at a large aperture like f11 won't yield great results compared to f32.

fishbulb
22-Apr-2015, 16:24
Maximum rise with that camera and that lens is putting the image circle to the test. Yes, sharpness will be decreased because nearly all optics are sharpest at the center. Just because you don't see vignetting, doesn't mean you've used too much movement. In most cases stopping down increases coverage so using lots of rise at a large aperture like f11 won't yield great results compared to f32.

Yep.

A lens projects an image circle like this. Typically it covers more than just the film, but it depends on the lens and the size of the film.

http://oopoomoo.com/wp-content/uploads/2012/03/TS-E-Circle.jpg

Most lenses have sharpness distribution that is centered - the center tends to have fewer aberrations than the edges. The bigger the image circle, the more you can shift or rise/fall before you reach the edges or before you get into the less-sharp areas.

Here is a sharpness map of the Zeiss 85mm f/1.4 (the old, non-Otus version) at f/2. Obviously the center is much better than the borders.

http://i.imgur.com/6PXf5aO.png

And here is the same lens at f/2.8. The more you stop down, the bigger the image circle, and the better the performance (up until diffraction sets in anyway).

https://i.imgur.com/Nqvl2xU.png

Preston
22-Apr-2015, 16:35
Yes, sharpness will be decreased because nearly all optics are sharpest at the center.

Hmmm? I'm having trouble parsing this, Vinny. Are you saying that since the lens axis (assuming zero tilt) is above the center of the film plane that the light hitting the film is coming through closer to the edge of the lens? If so, is this what's causing the reduction in sharpness?

If there is reduction in sharpness with a large amount of front rise, I suppose I never noticed it since most of my exposures are in the f22-f32 range.

--P

Ken Lee
22-Apr-2015, 17:23
If you use a lot of shift you can inspect the image on the ground glass with a loupe and observe for yourself. Brightness also falls off as we move away from the center.

All things being equal, Large Format lenses are not as sharp as those for small formats, but in exchange they can provide comparatively enormous coverage.

Dan Fromm
22-Apr-2015, 18:36
Preston, sharpness in the center is better than sharpness at the edge of the circle covered because some aberrations get worse off-axis. When the lens is decentered, image quality in the edge of the frame opposite the direction the lens is moved (lens up, bottom of frame, Dragan's situation) can suffer.

Some aberrations are sensitive to aperture, so stopping down sharpens up the outer part of the circle covered (makes it grow, in fact) and improves overall image quality until diffraction's effect overwhelms that of the residual aberrations.

For a table showing dependence of 3d order aberrations on aperture and distance off-axis, see http://toothwalker.org/optics.html

DG 3313
22-Apr-2015, 18:47
Draganski,

If the film was moving in the Fidelity holders during transport, I wonder if the film was inserted into the holder correctly? Yes, that is a factor.......

mdarnton
22-Apr-2015, 19:01
If you use a lot of shift you can inspect the image on the ground glass. . ..

Yes. I just got a 250mm Ilex Paragon, and was curious how it covered 8x10. Tessars supposedly turn to crap before they go dark, so I got out the magnifier. The lens is surprisingly good out to as far as my GG's clipped corners. Then I cranked in a little rise, and it went bad realllllly fast. Really bad. I thought only tuna salad went that bad that fast.

Preston
22-Apr-2015, 22:16
Preston, sharpness in the center is better than sharpness at the edge of the circle covered because some aberrations get worse off-axis.

Thanks, Dan. That makes perfect sense.

--P

Draganski
22-Apr-2015, 22:21
I used f22 and f35. On one image i suspect i added some tilt unintentionally, because only the top of the image is soft. The others are ok regarding sharpness, but i know this lens can do much better, because i have got a nice sharp negative where i did not use movements.
As soon as i have time, i will post some examples.

Doremus Scudder
23-Apr-2015, 03:02
Draganski,

A couple of things may be happening, or several together, to affect sharpness when you use rise.

First, though, you should know that using rise correctly should result in very little loss of sharpness. The only thing affecting sharpness, when rise is done right, should be the less-sharp area at the edges of the lens' image circle (as described very well above). And, this loss of sharpness is only at the edges of the image. Even when using rise, part of the image (the bottom) will be from the center of the lens and be very sharp. Plus, your 150mm Fuji W should be pretty sharp even at the edges, especially when stopped down to f/22-32. So if you are seeing significant loss of sharpness overall, or a big difference between center and edge sharpness, it is probably not the lens, but something else.

Some other things that could account for image degradation when using rise are:

1. You're compressing the bellows a lot and it pushes on the front standard, tilting it a bit and moving the plane of sharp focus. This happens often when using short lenses. However, you should be able to see this on the ground glass. Make sure to check focus top and bottom again after applying the rise. If this is happening, you'll have to use the camera's tilt movement to bring the image back into correct focus.

2. If you are experiencing overall softness when using rise, you might be de-focusing the camera somehow when applying the movement. Again, check focus after applying the rise and before stopping down.

3. You are using a Fresnel screen and it is not focusing off-axis light as accurately as it should. This was an issue with the o45-n1 (corrected with the 045-n2 I believe). If your ground glass image is sharp under the magnifier, but edges of the film are still not rendered sharp, this could be your problem. The solution here would be to try just a plain ground glass and see if the problem disappears. If so, you might not want to go back to the Fresnel with this particular camera, or you can research the possible fixes (I think Chamonix has some info about this on their website).

Your scanning holders shouldn't make any difference; they would be equally sharp or unsharp for all your negs. As for the skewed film: you're likely not loading correctly, i.e., missing one of the hold-down rails when loading, allowing the film to twist a bit in the holder (this can affect sharpness a lot too!). Practice a bit in the light with some scrap film to get the feel of things before going back in the dark. Also, check to make sure the film is loaded correctly after it's in the holder by pulling up gently on both bottom tips.

If none of this helps, or you're confused about how to go about fixing things, you can PM me. I live in Vienna (Augarten/Karmeliterviertel) and could meet with you and give you some hands-on help if we can find a good time for us both.

Best,

Doremus

Jim Jones
23-Apr-2015, 07:21
There are at least two Fujinon W 150 lenses. According to one source, the f/5.6 has an image circle of 245mm and the little f/6.3 an image circle of 198mm. The validity of such specifications depend on how critical one is. Using them, a graphic solution for maximum rise for the f/5.6 is 60mm with landscape format and 46mm with portrait format. For the f/6.3, 32mm with landscape format and 26mm for portrait format. There may be errors in both the original image circle data and in my graphics, so approach these limits with caution.

Drew Wiley
23-Apr-2015, 10:05
In some architectural shots it's not terribly important if there's some minor image degredation in the upper corners due to excess rise, because that area of the scene might just be blank sky or soft clouds anyway. But that also means that you're about to enter falloff territory, esp with certain wide-angle lenses, so it just depends on you subject and what you are trying to achieve with it. It also depends on your scale of reproduction and whether or not you can get away with very
small f-stops to improve the problem, if necessary. Published image circles are not cut-and-dried. Some mfgs define the usable circle much more liberally than
others. And for certain applications, like precise graphics reproduction, the published circle might be much smaller than what is realistically usable for general
photography. It always helps to mention the exact model lens you intend to use, and not just the nominal focal length, when seeking hints.

Lenny Eiger
23-Apr-2015, 12:04
I used f22 and f35. On one image i suspect i added some tilt unintentionally, because only the top of the image is soft. The others are ok regarding sharpness, but i know this lens can do much better, because i have got a nice sharp negative where i did not use movements.
As soon as i have time, i will post some examples.

On your next test, add f45 and f64. Fujinon lenses are very sharp, you should see where yours falls off. I bet its not until 64, or higher. It will also give you some more latitude...

Lenny

Drew Wiley
23-Apr-2015, 12:19
Any 150 lens is going to distinctly suffer from diffraction anything below f/32. Your nitpicky "sharpness" about this versus that lens is basically gone after a certain point. Tiny f-stops are the great equalizer, in a negative sense, that is. Whether this matters or not obviously depends on the degree of enlargement and what takes priority - optimal sharpness of maximum utilization of the image circle. These are obviously technical decisions. Sometimes we decide to break the rules for aesthetic reasons. I might actually want a distant feature on a tall building actually a tiny bit less resolved because it fools the eye into subconsciously ascribing a greater sense of scale. I do this with great mtn peaks sometimes too. Detail can become a sophisticated tool, just like perspective itself, just so you use these tools judiciously in a subtle way and not in some over-the-top conspicuous way. A good illusionist never shows his hand.

Lenny Eiger
23-Apr-2015, 12:36
The reason I said test it with your lens is because I have. I routinely shoot at f45 with my lenses and I am very happy with the sharpness.

Apparently, some people think that diffraction is a bigger deal than I do. I've been presented with all sorts of numbers, charts and diagrams. None of this actually makes a difference in my setup, for whatever reason. So I have simply acknowledged that there is a disagreement.

I say waste an extra sheet of paper or two, and decide for yourself. I love my depth of field.... I don't want to shoot at f22....

Lenny

Drew Wiley
23-Apr-2015, 13:02
You didn't print 30x40 Cibachromes from 4x5 like I did, Lenny. It's blatantly apparent in an application like that. Sure, I'd rather print those things from 8x10,
and generally do. But even with 8x10 film, which gets enlarged half as much apples to apples, I'd prefer not to make a glossy print larger than 20x24 taken at f/64. This all depends on what someone wants. The aesthetic question is the overriding one. But technically, there is a point where it's a real factor. Depth of field is it's own ballgame, and the strategy is a little different between different formats. Since I am comfortable with multiple formats, I have no problem thinking how to optimize each, though I'll obviously try to decide in advance of any outing which kit is best for that particular day or location. Guess I'm one of those guys who preaches the rules, then turns around and deliberately breaks them for known aesthetic reasons.

vinny
23-Apr-2015, 13:10
Lenny vs. Drew.
Who will win?

Drew Wiley
23-Apr-2015, 15:33
I already won three decades ago. All you gotta do is look at the prints. That was obvious to everyone once Ciba took over from dye transfer and you could really pack a lot of detail into a big print. Wasn't just me printing these things. Some were doing it way bigger; and if they had a consistent plane, like tabletop studio
sets, wouldn't even stop down below f/32 on 8x10 film. Not that detail is the whole name of the game; and if your film isn't flat in the holder, you waste the performance of those slightly wider stops anyway. But this is a thread related to sharpness per se. Inkjet is somewhere in between - not really the best test of
sheer lens performance.

Draganski
25-Apr-2015, 05:22
The problem might be related to the fresnel lens, as Doremus Scudder pointed out. I have upgraded to a yanke fresnel/ground glass combination a while ago. I will try the old ground glass with the fuji and see if that helps.

Lenny Eiger
27-Apr-2015, 14:13
Drew lost 30 years ago. Anyone that would print in that garish Cibrachrome is already nuts.... ;-)

I have zero interest in tabletop... it's an entirely different story...

Don't listen to either of us. Test it yourself on what you do, and see what happens.

Lenny

Draganski
28-Apr-2015, 13:48
I can definitely confirm after a bit of testing today, that the chamonix fresnel lens and especially the yanke fresnel/ground glass i own, are not suitable for using the fuji w 150 with movements like rise. The light falloff is immense. I guess the angle at which the light hits the fresnel lens when you shift it, is just not appropriate for that fresnel design. Focussing gets more difficult and maybe even the focus point shifts.
When i use just the chamonix ground glass everything looks much brighter and evenly lit. The only advantage i see with the fresnel is low- light without movements. I will use just the plain ground glass for my next shots, more to follow.
Thanks to all for your feedback and the entertainment :)

rfesk
28-Apr-2015, 17:00
Fresnels have their own focal length. Most are optimized for focusing in the middle range (normal) lenses for a given format size. Ask Bill Maxwell who makes
some of the best.

Doremus Scudder
29-Apr-2015, 02:30
I can definitely confirm after a bit of testing today, that the chamonix fresnel lens and especially the yanke fresnel/ground glass i own, are not suitable for using the fuji w 150 with movements like rise. The light falloff is immense. I guess the angle at which the light hits the fresnel lens when you shift it, is just not appropriate for that fresnel design. Focussing gets more difficult and maybe even the focus point shifts.
When i use just the chamonix ground glass everything looks much brighter and evenly lit. The only advantage i see with the fresnel is low- light without movements. I will use just the plain ground glass for my next shots, more to follow.
Thanks to all for your feedback and the entertainment :)

Dragan,

A Fresnel screen will indeed be more difficult to use off-axis. The problem is that the light is not refracted straight back at a right angle to the plane of the Fresnel screen. What you are seeing is not technically fall off, but the result is similar. I use a loupe that I can tilt to find the best viewing angle. That helps a lot. What I'm trying to say is that this is typical of Fresnel screens and may not have anything to do with your focusing.

Back to your focusing problem, because there may be another component to it. If your Chamonix 45-n1 came factory-equipped with a Fresnel screen (you mention it above), then replacing that with another Fresnel combination (the Yanke) or even a plain ground glass may result in a misalignment of focus and film planes.

If I remember correctly, the Chamonix Fresnel screen was positioned between the lens and the ground glass, and there were problems with focusing certain focal lengths of lenses and using lenses off-axis. I'm not really sure of all this, but it would be worth reading the information on the Chamonix website about this problem. You may have to shim your ground glass to achieve proper focus. Maybe other Chamonix 45n1 owners can help out here.

There's an easy test to see if your focus plane is the same as your film plane; set up your camera pointing downward at about 45 at a newspaper lying flat on tabletop or floor. The closer you can set this shot up, the more accurate it will be. Focus carefully on a line of newsprint with the lens wide open and make a note of the text. Then, without stopping down (or moving anything!), make an exposure. Develop and see if the line of text you focused on is indeed in focus. If so, no problems. However, if not, you will need to adjust the position of your ground glass.

Best,

Doremus

Draganski
29-Apr-2015, 05:49
Hello Doremus,

I have read a long forum post from 2009 regarding Cahmonix 045 and focusing issues, as well as the info on the chamonix homepage.
http://www.largeformatphotography.info/forum/showthread.php?56287-Chamonix-camera-45N-1-focusing-error
http://www.chamonixviewcamera.com/Fresnel_lens_problem_of_45N.pdf

The 045-N1 was designed without calculations for a fresnel lens, so shimming the ground glass is not necessary. i will nevertheless run a focus test as you suggested.

i had the same problem with my pentax k10 digital camera a while ago - the viewfinder screen was not aligned properly and i had to shim it. it was complicated and took me a while to figure it out. i couldn't imagine that i would have the same problems with a large format camera :)
thanks, dragan

Drew Wiley
29-Apr-2015, 13:00
The best use for fresnels is playing frisbee with your dog. Otherwise....

prendt
29-Apr-2015, 13:09
Hello Doremus,

I have read a long forum post from 2009 regarding Cahmonix 045 and focusing issues, as well as the info on the chamonix homepage.
http://www.largeformatphotography.info/forum/showthread.php?56287-Chamonix-camera-45N-1-focusing-error
http://www.chamonixviewcamera.com/Fresnel_lens_problem_of_45N.pdf



Interesting reading, thanks for the links to the threads. GPS was right about Chamonix cameras, I give him a medal for it. A very interesting intervention on his side.