HDR Film / integrated exposure mask

[keithwms]

Alright, this is an idea I had a long time ago but I just haven't had time to work on it so I thought I'd throw it out into the ether and see what ideas people have.

The goal is to increase the dynamic range. Yeah yeah I know it is already very large and better than digital etc. Set that aside, this has nothing to do with digital. This idea is more about making GND filters obsolete.

The idea is to put a mask onto the film which becomes opaque with exposure. Initially, I was thinking of a photochromic material such as one finds on the "transitions" lenses. But now I wonder if there is something less expensive....

Also, for colour film, this exposure mask would need to be as neutral as possible.

Thoughts? Ron?

 

[markbarendt]

How about stacking two film's in one holder for example, 1st film (main) developed as negative 2nd film (mask) developed (less) as positive.

Reverse the stack in the enlarger.

Might this also work as an unsharp mask?

 

[keithwms]

I thought of that once, Mark- to slip a pre-shot and developed neg over the new neg and thus make a light mask. That could work in principle, but registry would be a pain. And this idea is rather to have somehting that works on the fly, that you don't have to shoot twice. So perhaps if you had some sort of almost transparent layer with incorporated developer that then greys up slowly upon exposure...

Whatever the strategy, the mask would need to be much less expensive than the photochromic materials. Perhaps there is some polaroid magic that could be invoked- a material that polarizes upon exposure or such? Where is Land when we need him!

The advantage would be virtually limiteless dynamic range with no external filters required and no need to shoot multiple frames or do fancy processing.

 

[BetterSense]

Didn't they have reusable Cibachrome masks that worked like this? I've never seen one but I always thought it was a cool idea.

 

[keithwms]

Oh I didn't know that. Hmm...

 

[markbarendt]

My thought was only shooting once just having "two sheets in the holder" so to speak. Lower tech than your thought.

One thought is to use say HP5 for the main image and Delta 100 for the mask. The mask would underexpose and be aligned perfectly because it's exposed at the same time in the same holder.

 

[richard ide]

Interesting idea. the second sheet would be IIRC underexposed by 5 stops because of the AH layer on the front sheet plus whatever density the emulsion on the front sheet would add.

 

[keithwms]

I don't follow your idea, Mark- the exposed neg in the front wouldn't be cleared. Or am I missing something?

But if there were some sort of transparent material with an integrated developer...?

 

[BetterSense]

Well, if you could work in a studio, you could try soaking a negative in developer, squeeging it off, and exposing it right away. The developer would exhaust in heavily exposed areas, causing massive tone compression. It might require a hole in the darkroom wall/camera obscura type setup to be workable.

But then, if you are working in a studio, you can control the lighting....

 

[Lee L]

Contrast masking has a long history with varied uses in the wet darkroom. A quick overview is here:

http://www.freestylephoto.biz/traditional-processes-contrast-masking

As BetterSense recalls, it was nearly always used in the best Cibachrome prints shot in normal and higher contrast light.

Lee

 

[keithwms]

Yes, I am aware of contrast masks, but my point is that with something like a photochromic layer, there would be no muss, no fuss. I don't believe the photochromics are all that complex to coat...

Okay, I have the solution!!! A sheet of nice high-quality glass (or sapphire) coated with photochromics, and you just place that over the neg, with the photochromic pressed against the emulsion. Voila. This has to be tried

http://www.ppg.com/optical/opticalproducts/opticalfilm/Pages/default.aspx

Probably not neutral enough for colour film without some adjustment, but might be just fine for b&w.

 

[David A. Goldfarb]

For monochrome, printing out processes are generally self-masking. Shadows look good, but the highlights aren't there yet? Just keep exposing until the highlights are good, and the shadows will mask themselves, so they don't block up. This is a big attraction of albumen for me.

It's not without its limitations. You still have an ideal density range for the negative, but it offers some flexibility and interesting possibilities that aren't otherwise available without contrast masks and careful registration.

Color? Not sure if there is a self-masking tricolor printing out process, but of course if you're looking for something easier than contrast masking, a multilayer hand coated process with color separation negs ain't it.

 

[Lee L]

Yes, I am aware of contrast masks, but my point is that with something like a photochromic layer, there would be no muss, no fuss.

Apologies. I misread. There was a reusable masking material as BetterSense mentions, but I've forgotten what materials it used.

Lee

 

[markbarendt]

I don't follow your idea, Mark- the exposed neg in the front wouldn't be cleared. Or am I missing something?

But if there were some sort of transparent material with an integrated developer...?

I'm saying, as an example, stack two sheets of film HP5 on top close to the lens and Delta 100 behind/away from the lens in a modified sheet holder.

Set up and take a shot, the HP5 gets the normal exposure and the Delta gets the left overs after the light filters through the HP5.

Separate the sheets.

Develop the HP5 normally, develop the Delta reversed.

Put the two sheets back together for printing.

I have no clue if that combo would actually do the job and it would take some serious experimentation but theoretically the HP5 could be developed first and judged then the Delta could be developed as needed to get the total contrast into a printable range.

 

[markbarendt]

The problem I see with the photochromic stuff I've seen is the time it takes to change.

I don't know what it is called but for welding there is glass that instantly goes dark when an arc is struck.

 

[Athiril]

You do not need more dynamic range. That isn't the problem.

What you want to do is actually reduce the dynamic range.

For a simple example, a sunset, the sky is very bright, and foreground not so bright.

A high contrast shot becomes a low contrast dull image, a low contrast shot, becomes a high contrast vibrant image. When the whole range of the image is used from black to white.


For a vibrant image, what you want in the simple sunset example is to have the dMin of the sky at X, and the dMax of the sky at Y, while in the foreground you also want' to have the dMin at X and dMax at Y, instead of X-Z and Y-Z etc.


I've discussed a transition lens type setup between lens and film with several people, you can make one for a LF camera simply from a sheet of film, develop the sheet out, then weak bleach and then fix, and repeat, you want a few stops left, perhaps 1.0d or 1.3d, then you want a rehal bleach with chloride, to make a chloride emulsion in the dark, wash well, fix another sheet without processing, and wash well, stick that sheet against the emulsion side of the other sheet, in the dark, dry and take out in the light.

In theory it should act like a silver chloride transition lens, and 'reset itself' in the dark, so you can insert it into a slot in a LF camera, and start pre-exposing it. Though you may not want to pre-expose it at the same focus as you're going to shoot. Depends... testing required!

Of course the mask will be formed upon mostly from the UV and blue spectrum which may be a problem in several scenarios..


edit: forgot to mention, you will need a copper source to go along with that new emulsion, or it wont work iirc. You could wash it in very dilute copper sulphate, or copper chloride perhaps before sealing it. Or use a copper bleach to begin with.



Placing a light-formed mask right against the film will make a very defined mask, I do not think that will work how you want it to. What you need is a more general mask.

For example, if you shoot a test chart.. black and white line pairs, a defined mask will just make the entire thing grey. You'll just lower adjacent contrast, not normalise large continguous areas. It needs to be more out of focus for that. Placing it further up the path might do that.

 

[polyglot]

For it to work in-camera, you'd need a really-fast printing-out effect. Which I think it pushing it a bit, to get anywhere near the densities you'd need with normal exposures, you need a development step.

OTOH I have considered an auto-GND filter (applicable only to digital+tripod though); take a shot and have it rendered blurred in greyscale on a low-res LCD filter (bonus polarising effect) in front of your lens, then take the real shot.

 

[markbarendt]

I didnt explain my reasoning for the stacking and order well above.

My thought is that you need a material with differing sensitivity thresholds to do what you want Keith.

That was the thinking behind using a slower film behind a faster film.

The Delta 100 in my example would theoretically be somewhere around three to five (or 7, I don't know) stops behind the HP5 (two for the rating plus whatever amount the HP5 acts like a neutral density filter).

Once the threshold was reached the reversed Delta would in the end mask just the mid to low end of the HP5 curve and do it at a predictable rate.

It would not be unlimited and require longer enlarger times.

 

[BetterSense]

I don't know what it is called but for welding there is glass that instantly goes dark when an arc is struck.

It's not a type of glass. It's an LCD panel.

 

[markbarendt]

I just figured out that I'm wrong in suggesting reversal of the one sheet. I think it should both should be negatives.

A positive would just lower the contrast rate, a negative would maintain contrast rate and just add more density to the highlights.

 

[keithwms]

You do not need more dynamic range. That isn't the problem.

What you want to do is actually reduce the dynamic range.

We have a difference of definitions.

In the manner that I use the terms, the 'dynamic range' of the film (and associated development process) is the number of zones that can be recorded, from toe through knee in the response curve. The brightness range of the scene is the 'scene brightness range' or SBR. Two very different things, of course.

Anyway, I allege that a photochromic layer could fit a lot more SBR between the knee and toe. Again, the idea is to make graduated neutral density filters obsolete.

I agree with the other comments that the response time of a photochromic layer could be problematic, but I can think of several straightforward ways to overcome this:

(1) simply adjust the ISO of the film accordingly, through appropriate development i.e. pull processing; or,

(2) place an appropriate neutral density layer between the photochromic layer and the film, so that the film receives less intensity per time; or,

(3) place a shutter (or darkslide?) between the photochromic layer and the film and pull that after the photochromic layer has had time to respond...

 

[keithwms]

I just figured out that I'm wrong in suggesting reversal of the one sheet. I think it should both should be negatives.

A positive would just lower the contrast rate, a negative would maintain contrast rate and just add more density to the highlights.

Yep.

But I suppose that there is some merit to the idea of a multilayer emulsion, if the top one has some sort of incorporated developer that makes it grey up in the high intensity areas.

 

[Athiril]

I don't have my own definition, as I use the actual pre-existing definition of dynamic range. Which applies to everything.


Recording the entire dynamic range like that will not help you, everything will simply be flat, dull and low contrast. The intensity of two separate areas will remain distinctly different, the dynamic range of the sky for example will map to light grey to medium grey, the dynamic range of the foreground will map to medium grey to dark grey.

Instead of both being near white to near black, which is high contrast.

Masking reduces your dynamic range, and balances these different parts closer together and thus increases their contrast when reproducing their entire dynamic range. Pull developing does not. It lowers your contrast.


Basically, what you want, is for local contrast to exceed area to area contrast, which isn't there unless it's masked (like a grad ND) and the area to area contrast is reduced below local contrast.


It may be possible to achieve that in developing.

 

[mr rusty]

Interesting idea. I have no way enough knowledge to really comment on this, but what I think you are getting at is a film that controls the light in such a way that the image it produces when developed without special techniques looks like those strange HDR images that litter the net that look unnatural but in the same way when done well somewhat intriguing where the dark foreground and bright background are all perfectly exposed.

 

[Athiril]

Speaking of strange HDR, I've had halo'ing with Rollei ATP.