Film Speed

[Trius]

In another thread initiated by Wintoid regarding development times for pushing Neopan in Rodinal, the discussion took a few different turns (on RFF? Go figure...), including the subject of "true film speed". Drewbarb and I were starting to head down the path of discussing whether a film's "true" speed is affected by the selection of developer. (And, by extension, dilution.)

There are at least two schools of thought:

* Speed is determined by the film emulsion, period.
* Speed is at least somewhat determined by the developer.

So, I'd be interested in a discussion of this. However, there are two assumptions/positions that I make/take:

1. I consider film speed to be determined by the minimum exposure required to register a Zone I detail. I.e., just visible shadow detail above film base+fog.

2. I am interested in empirical evidence under at least moderately controlled conditions. References to two or more photos under entirely different conditions as "evidence" of one's position, while perhaps interesting, don't interest me personally.

You may differ with my assumptions, and that's cool. But for me personally, I have been trained as a Zone system worker, though of course it's a modified approach based on situation, materials, objectives, etc. So that's where I'm coming from.

Has anyone done any controlled tests to "prove" one position or the other?

Edit: Perhaps obvious, but I'm referencing conventional b&w film here.

 

[Xmas]

Trius

The ASA & DIN standards have a (the same) standard for development, and for the point above the curve knee (your Zone 1), then they add a (the same) safety factor.

About '60 they both altered the safety factor from 2.5 to 1.25 stops, but now I'm really reliant on memory.

Some of the Ilford films are not to the ASA/DIN standard, instead they have picked a practical point, the curve may be strange...

Pre WWII received wisdom was it was a mad house....

Your obvious caveat is appropriate colour film is a more practical mechanism...

Noel

 

[drewbarb]

I will come out here and say that I don't buy the idea of "true" film speed, since different developers will give different results. Your given definition of film speed as the minimum exposure required to render Zone 1 detail is accurate, of course, but to a point. This has to be in relation to the variable of development- unless you are talking purely about latent image. Since different developers will yeild this ISO level of detail at different E.I.'s and with different developing regimens, this will effectively reveal the "true" film speed with the given developer.

When manufacturers rate films with ISO speeds, they always do so in relation to their assumed standard development, which is a time and temperature combination with what they specify as the standard developer for that film. A good, if extreme, example of this is the old Kodak Tech-Pan, which was to be developed in Technidol developer.

In anycase, I am eager to read other's ideas and thoughts on this thread,and thanks for kicking it off, Trius.

 

[markinlondon]

I find film speed to be dependent on many variables and would never commit the great interweb sin of declaring the "true speed" of any emulsion. For the last year I've mostly been shooting Ilford films and developing in ID-11. My preference for giving one extra stop of exposure and cutting development by 20% was an aesthetic decision but based on testing. I can only agree with all the comments made so far only to add that the "true speed" applies solely to you and your technique and environment.

 

[Xmas]

The ASA/DIN speed on the white or yellow box does not rely on witches brews...

I can feel the flames licking at my feet

Noel

 

[kaiyen]

First, it's not ASA, and it certainly isn't DIN. It's ISO. 🙂

Second, I completely disagree with the first point, and feel that "speed," as defined by shadow detail (which is a more vague version of Trius' perfectly valid Zone I definition), is entirely the result of film _and_ developer. Period.

My process for determining EI is kind of loose, but works for me since I am not a test-a-holic. it is available at http://photos.kaiyen.com/pmwiki/pmwiki.php?n=Main.Filmspeed

allan

allan

 

[Xmas]

The standard is ISO thanks, but the number on the box is ASA or DIN.

Ilford don't use it for delta (I think), and pre WWII they did use witches brews because few people used real photo meters, and they were in to marketing. It was the more general use of meters which allowed the 2.5 margin to be reduced to 1.2.

Nicely warm already, they burn withches dont they, Salem, Death of salesman, ...

Noel

 

[Trius]

OT: Mark, I really like your new avatar.

 

[drewbarb]

Allan- I just checked out the page you linked to your testing method for finding your own E.I. Well done! This is a very sensible way to find good, repeatable results for real-world film work. Bravo.

 

[kaiyen]

Actually..it is ISO. ASA and DIN are outdated. Anyway.

allan

 

[kaiyen]

Drew,
Glad you found the page helpful. I find that it's just scientific enough to be valid (to me), but loose enough that I am not spending all my time with a densitometer.

allan

 

[markinlondon]

OT, thanks, Earl. It wasn't a waste of a frame after all then 🙂

On topic, I was thinking about this after posting and my conclusion was that you wouldn't get much of an argument on this subject as this forum seems to be mercifully free of the "true speed" zealots.

 

[vicmortelmans]

My process for determining EI is kind of loose, but works for me since I am not a test-a-holic. it is available at http://photos.kaiyen.com/pmwiki/pmwiki.php?n=Main.Filmspeed
allan

Quite interesting, Kaiyen, I posted a procedure the other day that is similar to yours. Difference is that I propose not to use a real subject, but a flat field (even surface) and stretch the scientific part into the drawing of the actual film response curve.

See HERE.

In general, the 'ideal ISO' should be related to the whole process that's applied:
1- choice of film
2- choice of developer
3- choice of development process (dilution, time, agitation)

I think I'll go ahead with this test once my (first) batch of ADOX 100ART arrives.

These are the processes I'll try to find the 'ideal ISO' for:

1. 'normal process' Rodinal 1+50 10' 3x5sec agitation
2. 'push process' Rodinal 1+100 40' stand development

According to the massive development chart this should give about ISO100 for 1. and about ISO400 for 2. (extrapolated).

The results of the test will also learn me which process is suitable for typical contrast situations (although the answer will be quite obvious).


One aspect of your test is puzzling me: you only look at shadow detail, while my test suggests to look at the whole detail (zone) scale and match the exposure so there's both maximal shadow and highlight detail. In Adam's time, exposure (for the shadows) and development (for the highlights) could be decoupled, not using roll film, but this thread is about standardizing a (roll) film process, so I'd think shadow and highlight detail are in it together.


Groeten,

Vic

 

[Trius]

One aspect of your test is puzzling me: you only look at shadow detail, while my test suggests to look at the whole detail (zone) scale and match the exposure so there's both maximal shadow and highlight detail.

This is where I strongly disagree. First, I did not specify roll film vs. sheet film.

Second, "in Adams' time"? Did all sheet film manufacturing end in 1984? Someone should have told me that my 4x5 gear is totally useless. [insert smiley of your choice here]

Third, whether you use sheet film or roll film, development STILL controls highlight/contrast, not shadow detail. In the first post of this thread, I set the "ground rules" for my thinking. I don't expect everyone, or even anyone to adhere to that; this is RFF after all! But I don't give a rodent's arse about development time with respect to film speed. IT JUST DOESN'T MATTER (within the constraints of minimal time to develop an image in shadow areas.)

Now, I DO acknowledge that sometimes, perhaps often given certain subject and lighting conditions, shadow detail (at least Zone I or so) just doesn't matter, and "true" film speed doesn't have any relevance.

As for the position that there is no such thing as "true" film speed... Why then is film speed subject to the International Standards Organization?

 

[vicmortelmans]

This is where I strongly disagree. First, I did not specify roll film vs. sheet film.
Second, "in Adams' time"? Did all sheet film manufacturing end in 1984?

Sorry, my misunderstanding... I only assumed it was about roll film.

Third, whether you use sheet film or roll film, development STILL controls highlight/contrast, not shadow detail. In the first post of this thread, I set the "ground rules" for my thinking. I don't expect everyone, or even anyone to adhere to that; this is RFF after all! But I don't give a rodent's arse about development time with respect to film speed. IT JUST DOESN'T MATTER (within the constraints of minimal time to develop an image in shadow areas.)

I get what you mean! Each film indeed has a characteristic response at the shadow area, which does not depend on development time (may depend on developer? I don't know). What I meant to say was that depending on the contrast (which does depend on the development time), you may choose a different ISO. If you have a very wide contrast range (10 stops or more), you may want to increase your exposure (higher lower ISO) to get even more shadow detail, while if you have a very narrow contrast range (5 stops or less), you may want to decrease exposure (lower higher ISO) to save some highlight detail as well...

As for the position that there is no such thing as "true" film speed... Why then is film speed subject to the International Standards Organization?

I guess a good reason may be because 95% of film is developed in standardized development machines? For the way these machines are set up, the indicated ISO speed is optimal.

 

[markinlondon]

I guess a good reason may be because 95% of film is developed in standardized development machines? For the way these machines are set up, the indicated ISO speed is optimal.

But only if exposure is optimal as well. Also, surely increased exposure = lower ISO and vice versa and thus not as you stated above, Vic.

 

[vicmortelmans]

Hi,

reporting on the test I ran on the ADOX (EFKE) 100. I took a range of exposures of an even surface, each 1 stop more than the previous and the 'nominal' exposure indicated by the camera meter @ ISO100 in the middle.

Development = Rodinal 1+50, 8', 2times 5sec agitation

The graph shows the exposures on the X-axis and the base 2 logarithm of the resulting 16 bit pixel file (after scanning the negative, that is) on the Y-axis.

At the shadows-end of the curve (left), the first 'reproducable' exposure is -4. The deeper shadows are compressed in the shoulder of the curve.

At the highlight-end of the curve (right), the highest 'reproducable' exposure is 6. That's not because of film characteristics, but because beyond that density, scanner noise is becoming a problem.

So the total sensitivity range is 10 exposures. Is this a common sensitivity range? Seems to map to the 10 zones of the Zone System, or is it coincidence?

My conclusion is that the ISO100 rating is quite OK, although for scenes that have interesting, but small shadow details (so my average meter won't pay too much attention to them), ISO50 (or +1 exposure compensation) is better.

It's nice how this visualizes to "expose for the shadows". Exposure should be set 4 or 5 stops lower than a spot metering for the shadows reads. And then there's still plenty of room for the highlights.

What would be the effect of changing development? Increasing the time would drop off the highlights in the scanner's noise zone earlier, so that's not good. I wonder if pushing is a good idea at all. Reducing development time should be investigated further, because the curve doesn't show the highlight toe. Maybe it will give an even wider dynamic range (at lower ISO).

Hmm. all that trouble for quite obvious conclusions.... 😕 But doing it was fun 😱

Groeten,

VIc

 

[Trius]

Vic: My point exactly. Thanks.

 

[40oz]

It occurs to me that what is really being tested if you use, for example kaiyen's method, is the speed of the film at your chosen development time. One would think the "true" or "best" film speed in your developer would be reached only by repeating the same test with multiple rolls developed in the same developer with varying times. Then you would select the best frames from each roll, and compare them to reach the best time.

I'm thinking that, if you scan or print, you would want to try each "best" frame to see which actually scanned/printed the best, because that would actually be your end result, not looking at negatives through a loupe.

Granted, this coming from someone who is still really doing everything by the seat of his pants 🙂 It's been a while since I shot a film at the rated speed, is all I'm saying lol.

I have learned a great deal on how to expose scenes from this discussion. Thank you all for sharing your experience.

 

[vicmortelmans]

I bumped into this guy's 'journal', which is describing the same procedure.
=> Characteristic Curve

He has some tests on the effect of agitation as well!
=> Agitation

His test is on FP4, which is modern emulsion. This is the main conclusion: "The minimal agitation technique appears to give better results because it affords higher contrast in the shadow detail, and lower contrast in the highlights with no distinct shoulder in the curve."

Look at the curves on his chart: the low-agitation curve has *higher* densities than the high-agitation curve. Isn't this strange?

From earlier discussions, I learnt that low agitation causes the active developer concentration to drop in the direct surrounding of high-exposure film surface, thus slowing down development, while high-agitation causes a continuous flow of fresh developer, so development keeps going on steadily. The difference would then be that low-agitation smoothens the highlights shoulder.

So I would expect the low-agitation curve to have lower densities on the overall curve, especially at the highlights area. Contrary to what's expected, the low-agitation curve has higher densities on the overall curve, especially at the midtones area!?!?

Still, the effect is advantagous, because (1) there's more contrast in the shadows and (2) the highlight curve is smoother.

Can someone explain?

I could repeat my test with different agitation. The first test already has quite low agitation (2 times 5 secs for 8minutes, that's even less than the 'low-agitation' in the above test). But then again, I'm working on a completely different emulsion and a different developer as well, so maybe it responds fundamentally different to agitation?

When time allows, I'll do another test, go into higher exposures and do 7 minutes stand development... I wonder what that will do. I expect a wider dynamic range when rating the film at lower ISO.

Groeten,

Vic