Filmtest with contrastcurves and grainstucture.

[Mark Crabtree]

Interesting. The TMX curve in D76 certainly looks more like I would have expected overall, though it does still have that odd bump in the middle. I had no idea Rodinal was giving that much different highlight contrast. TMax films do seem more sensitive to developers and development than traditional films, but a similar thing showed up with some other films in the Rodinal test.

 

[Photo Engineer]

We used D76 straight for our tests. The use of 1:1 is not usual for product release.

PE

 

[Bill Burk]

In my family of TMAX 100 tests, there might be a 'bump' but if I changed a single reading by 0.01 it would smooth the curve. So I don't believe it's a characteristic of the film, more likely a test anomaly.

I appreciate seeing D-76 1+1, and it would be good to also see D-76 straight. In terms of speed, I have two curves at Contrast Index 0.72 which nearly coincide, one is 12 minutes in D-76 undiluted, the other is 17 minutes in D-76 diluted 1 to 1. So I don't believe there is a significant speed difference either way.

 

[Dag_Nystuen]

In my family of TMAX 100 tests, there might be a 'bump' but if I changed a single reading by 0.01 it would smooth the curve. So I don't believe it's a characteristic of the film, more likely a test anomaly.

I agree, I also think the very slight bumb in our curve is a test anomaly. We are in the prossess of getting new teststrips from Stouffer.

 

[Mark Crabtree]

I thought the same thing and was tempted to say it, but don't think the bump showed in any other film which seems a bit odd.

It really doesn't seem like a big deal to me anyway, but I mostly look at curves to get an idea of the overall response of the film. For my own photography I much prefer something with a slight shoulder to a film like Acros or TMY. Others prefer the opposite, but either way curves give you an initial guide. From there I think taking pictures is the most productive way to learn about a film; at least that is true for me.

 

[Mark Crabtree]

Yes, that is why I prefer TMX to TMY and Acros for my use. But it looks like I won't be using it in Rodinal, not that I really intended to anyway.

 

[Animalcito]

Hi Dag,

well I measured the Tmax 100 in Rodinal ca. 1 year ago at 1+50 dilution, 20°C and an agitation of 3x/60s.This results in a beta value of ~0.55, I'm using a condenser enlarger. Also under these circumstances this double S-structure is visible, less pronounced but it is there.

Ciao
Marcus

 

[Bill Burk]

If the double-s shape in Adonal occurred below the 1.0 density mark, I'd be inclined to experiment with possible highlight detail enhancement effects.

But since it's above the density I usually use for printing, the part I would use looks just like a bit of a soft shoulder. Not as significant a curve shape problem as it may seem on the graph.

(As for the slight bump in D-76 from my curves, my bump comes from a single reading, the rest of the family doesn't have a bump).

 

[mauro35]

I was curious to see the full resolution image of CMS20 film, but the link seems broken, has anybody else tried?
Here is the page and the full resolution image link is on the bottom right of the page.

http://www.fotoimport.no/filmtest/filmCMS20.html

 

[Photo Engineer]

Look at some of the other films in Adonal! It means that the film was not designed for those conditions.

PE

 

[Nicholas Lindan]

The data on the web site is remarkable - good job.

As to bumpy curves, well, they are a specialty of VC paper.

There is a paper on the Darkroom Automation web site that goes into the bumps in VC paper The Workings of Variable Contrast Papers and the Concept of Local Gamma.

Ideally the 'S' curve of an emulsion is the integral of the normal distribution - the famous 'bell curve.' A simple bell curve doesn't have a flat top - and a simple emulsion doesn't have much of a straight line to it's HD curve.

It seems all (or mostly all) 'modern' films and papers are made up of one or more emulsions with different sensitivities, the reason being to get a straight line up through the highlights and beyond. If the two emulsions don't cross-over well, and one emulsion's highlights don't blend with the next's shadows, there is either a hump or a flat spot.

It would be interesting to see the plots of 'local gamma' - the derivative of the HD curve - for these developer/film combinations. To get a good reading of local gamma you need to run the test exposure at 0.1 stop intervals or better - a problem with film.

 

[Photo Engineer]

It is my understanding that these are the curves of the films and not any paper. Thus the bumps in a VC paper do not enter into the data.

As for the films, they are all composed of a mix of several emulsions, sometimes 2 and sometimes 3. They use a fast, medium, and slow and if the "join" is not perfect, you get bumps. Often, a film is perfect in one developer and bad in another. Just peruse the data and you will see cases of that in virtually every set of films and developer combination. I think that Rodinal is one spectacular example.

PE

 

[Nicholas Lindan]

It is my understanding that these are the curves of the films and not any paper.

You are right.

The article I referenced is about paper because printing is the part of the photographic process that Darkroom Automation is involved with. But the physics of the thing are the same with paper or with film.

As for the films, they are all composed of a mix of several emulsions, sometimes 2 and sometimes 3. They use a fast, medium, and slow and if the "join" is not perfect, you get bumps.

And the same applies to VC paper: multiple emulsions with different sensitivities. The only difference is that with paper one of the emulsions is sensitized to green light and so it's sensitivity viz white light can be moved around. With VC paper the join can only be perfect at one point - that seems to be #3 1/2 for Ilford MGIV. Interestingly many photographers came to the conclusion soft negatives give the best prints on Ilford - without any knowledge of the paper's HD curves.

Often, a film is perfect in one developer and bad in another. Just peruse the data and you will see cases of that in virtually every set of films and developer combination. I think that Rodinal is one spectacular example.

Couldn't agree more. Thanks to Dag's work it will be much easier to pick good combinations of film and developer. It would be nice to have all the curves gathered together in a PDF file.

 

[Photo Engineer]

Agitation, time, temperature and the phase of the moon all have something to do with the results obtained by development and printing modern films on modern papers. It used to be that all curves were virtually the same giving upward bowed or downward "saggy" curves until mixed or blended emulsions came along. This was only achievable by using modern process control. Look where it got us!

It looks good in the plant, but all customers do not use D76!

Combine the film non-linearity with the same problems in paper and the prints can begin to look pretty strange.

Dickerson and Zawadski published an article on paper, and recently Ilford revised their paper formula. It would be interesting to see paper curves and curves of select film/developer combinations printed on paper.

PE

 

[Photo Engineer]

I had to do this many times over to "prove" the newest sensitivities and curve shapes didn't mess things up.

PE

 

[NJH]

Thanks ever so much Dag for this testing, it has helped me enormously. Looking at your curves and through all the test shots I am going to develop at home with FX-39 and FP4+ and Tmax 400 films, I posted a thread here recently that basically boiled down to me wanting to try and get the tonality I see and sharpness like neopan 400 in ID-11 (over dev'd by my lab though so very grainy) but with finer grain and resolution like the modern films.

 

[Dag_Nystuen]

We have now published an updated version of the D-76 test.

http://www.fotoimport.no/filmtest/fkd76.html

We have replaced our old stepwedges with new ones, we have crosschecked our densitometer and made a new test that we belive are more accurate.

We are thinking of making simular tests on photographic paper also, but that will have to wait a bit.

Thank you for your coments and thanks. It is inspiering to get this feedback.