Any Examples of Upswept vs S-curve in prints?

[Athiril]

Hey,


Can anyone share examples of upswept curves on film enlarged in the dark room and s-shaped curves as well.

I'd like to see some side by side.

Images tend to disappear from old threads I find with google.

[Ken Lee]

You can easily make an upswept curve in Photoshop or any other tool which gives you a histogram and curves adjustment tool. What you see, is what you get. You can sweep up, as it were, as much or as little as you like.

[Drew Wiley]

Would be hard to objectively visualize this unless you were printing a graphics step
tablet in which the steps are evenly spaced. You would also have to expose, develop,
and print the film in such a manner that you go beyond the straight-line section of the
characteristic curve at both the high and low ends.

[Athiril]

Hi,

I'm not interested in playing with Photoshop on a random image.

I want to see images other people have printed in the darkroom from a negative with an upswept curve or S-shaped curve.

I didn't think that was too much to ask on a film forum.

[Frank Petronio]

An S-curve will increase contrast, the upswept curve will produce more open mid-tones and a longer range towards the highlights.

Of course someone could be nice enough to post a picture printed on Grade 2 and Grade 4 papers to illustrate, but how many people have jpgs of just such an example at their disposal? And why would they be willing to humor such a snippy twippet as yourself?

[D. Bryant]

I'm not interested in playing with Photoshop on a random image.


I didn't think that was too much to ask on a film forum.

Why not? You've indicated or implied in other posts that you are very familiar with and use PS.

Besides it will be more instructive to you if you do that sort of task yourself.

Don Bryant

[Ken Lee]

http://www.kennethleegallery.com/images/forum/24.jpg
Normal

http://www.kennethleegallery.com/images/forum/upsweep.jpg
Upswept Curve

http://www.kennethleegallery.com/images/forum/upswept.jpg
Upswept Image

[Ken Lee]

http://www.kennethleegallery.com/images/forum/24.jpg
Normal

http://www.kennethleegallery.com/images/forum/s-curve.jpg
S-Curve

http://www.kennethleegallery.com/images/forum/s-curved.jpg
S-Curved Image

[Ken Lee]

Tri-X in standard developers like D-76, HC-110, etc is said to have an upswept curve - and you can see why (in the simulation) we might want to give it an extra stop of exposure: to get the shadow values out of the "dumps" so to speak, where they will be more linear. We would want to give more exposure, but give less development to restrain the high values.

The S-Curve image looks like what we get when we strongly over-expose, but try to rescue the image. It also looks like what we often see when using a condenser light source, as opposed to a diffused source: We print long enough to keep some texture in the high values - but in doing so, the low values drop off the lower threshold. What we're left with is pretty... harsh.

[bob carnie]

Nice Examples Ken



Tri-X in standard developers like D-76, HC-110, etc is said to have an upswept curve - and you can see why (in the simulation) we might want to give it an extra stop of exposure: to get the shadow values out of the "dumps" so to speak, where they will be more linear. We would want to give more exposure, but give less development to restrain the high values.

The S-Curve image looks like what we get when we strongly over-expose, but try to resuce the image. It also looks a lot like what we often see when using a condenser light source, as opposed to a diffused source: We print long enough to keep some texture in the high values - but doing so, the low values drop off the lower threshold. What we're left with is pretty... harsh.

[Ken Lee]

- Thanks -

I guess we could do an S-Curve in the other orientation too.


http://www.kennethleegallery.com/images/forum/24.jpg
Normal

http://www.kennethleegallery.com/images/forum/s-curve-2.jpg
The other S-Curve

http://www.kennethleegallery.com/images/forum/s-curved-2.jpg
Result

[Mark Sampson]

Ken-
Thanks- that's an excellent demonstration. Pictures can indeed be worth a thousand words. Especially because I can't imagine how you could get that 'other' upswept curve just using film...

[Neal Chaves]

In the mid-80s, when T-Max first came out, I ran development tests, plotted the curves and made test prints developing in HC110B with agitation at 30 seconds. Curves were upswept and prints were awful. I tried some later when the special T-Max developer (which was too expensive anyway) came out and still didn't like the looks of it. It had very poor mid-tone contrast.

I've tried all the soups and films and keep coming back to Tri-X (now HP5 plus) and HC110. I have identical development schemes for them from E.I. 16 to 1600 (high value speed) with HC110 or Acufine (which just costs more). At a time of 5 minutes in B at 68 deg. Tri-X delivers a beautiful negative at E.I. 64.

If you look at the graphs in Davis' BTZS, you can pick off the speed of Tri-X as 64 at 5 min. in B. He never shares this with the reader or student, gurus never do, Adams says things like "I thought the shadows needed extra support so I placed them on Zone IV." This is the same as rating the film at 64. I also found no reciprocity failure when developed as above at exposures out to 1 minute (limit of my test).

[Harold_4074]

Especially because I can't imagine how you could get that 'other' upswept curve just using film...

Well, you could underexpose a print, let it develop until the midtones just start coming in, and then flash the white lights. If you yank it out of the developer at just the right time, you will have something very much like the example. If you think that it looks strange in the simulation, you should see an actual print.....please don't ask me why I know this :)

[Ken Lee]

Here's another series: a milder image, and milder curves applied:


http://www.kennethleegallery.com/images/forum/normal2.jpg
Original
http://www.kennethleegallery.com/images/forum/upswept2.jpg
Upswept Curve
http://www.kennethleegallery.com/images/forum/upswept2image.jpg
Result

[Ken Lee]

http://www.kennethleegallery.com/images/forum/normal2.jpg
original
http://www.kennethleegallery.com/images/forum/scurve2.jpg
S Curve
http://www.kennethleegallery.com/images/forum/scurve2image.jpg
Result

[Drew Wiley]

The problem with doing it in the darkroom is that the paper has a curve of its own, one
which is more restrictive than the film itself, so that objectifying the result involves
two separate things. I think it's easier to just go out and shoot and experiment, or else
visualize the way Ken is doing it. As much as possible I try to work with films with a
long straight line, or develop in such a manner to place the entire subject luminance
range on the straightest section of the curve. An exception would be portraiture, where I often prefer an upsweep to get more upper midtone and highlight separation.
But plenty of tricks are available to us.

[Eric Woodbury]

Ken, those are nice. I realize this wasn't the original question, but can you make the inverse of the upswept? Would that be an "over-swept"? Jumps up fast in the shadows and decays in the highlight. Thanks.

[Ken Lee]

http://www.kennethleegallery.com/images/forum/normal2.jpg
Original
http://www.kennethleegallery.com/images/forum/downsweptcurve.jpg
Downswept Curve
http://www.kennethleegallery.com/images/forum/downsweptimage.jpg
Result

[Oren Grad]

Pictures can indeed be worth a thousand words. Especially because I can't imagine how you could get that 'other' upswept curve just using film...

You can get something like Ken's backward S-curve using Phil Davis's DI#13 developer with TMX. Really. :) If I recall correctly, David Kachel also was able to generate a curve like that with some of his SLIMT experiments.

Quite a few years ago, Phil had a pair of excellent articles in Photo Techniques on the topic of gradation, in which he provided pictorial examples of the effects of different paper curves (with the negative held equal) and different film curves (with the paper held equal) on the tonal scale of the resulting print. Well worth seeking out for anyone who isn't yet comfortable on an intuitive level with the effects of film and paper curves. Ken's examples are excellent too, and the curve tool in a good photo editor like PS is a powerful tool for exploratory, interactive learning.

[Athiril]

In the mid-80s, when T-Max first came out, I ran development tests, plotted the curves and made test prints developing in HC110B with agitation at 30 seconds. Curves were upswept and prints were awful. I tried some later when the special T-Max developer (which was too expensive anyway) came out and still didn't like the looks of it. It had very poor mid-tone contrast.

I've tried all the soups and films and keep coming back to Tri-X (now HP5 plus) and HC110. I have identical development schemes for them from E.I. 16 to 1600 (high value speed) with HC110 or Acufine (which just costs more). At a time of 5 minutes in B at 68 deg. Tri-X delivers a beautiful negative at E.I. 64.

If you look at the graphs in Davis' BTZS, you can pick off the speed of Tri-X as 64 at 5 min. in B. He never shares this with the reader or student, gurus never do, Adams says things like "I thought the shadows needed extra support so I placed them on Zone IV." This is the same as rating the film at 64. I also found no reciprocity failure when developed as above at exposures out to 1 minute (limit of my test).


When you say Tri-X I assume you refer to Tri-X, but this is a LF forum and Tri-X isn't available in LF, only Tri-X Professional.

[Harold_4074]

You can get something like Ken's backward S-curve using Phil Davis's DI#13 developer with TMX. Really. If I recall correctly, David Kachel also was able to generate a curve like that with some of his SLIMT experiments.

Which, of course brings up the next question: other than as an educational exercise, why would you want to? I just can't visualize a subject that would benefit from this transform!

[Oren Grad]

Which, of course brings up the next question: other than as an educational exercise, why would you want to? I just can't visualize a subject that would benefit from this transform!

The classic example is a dark interior scene with a window or door open to a bright exterior scene. There's lots of information in the very lowest tonal values, a region in the middle of the scale where there's not much going on, and a lot of information in the highest tonal values.

[Harold_4074]

Of course. Now that I think of it, I've done that myself on technical images, to facilitate measurement between points that are in highlight and in shadow. I just couldn't think of a pictorial case where the "flat" midtones would work well.

[Michael Kadillak]

For those of us that do not use scanners or computer printing technology to make prints, we need to be capable of seeing the net effect of combinations of developers and sheet film within the wet print and understand the sometimes subtle changes that result.