How to Test a Film?

Considering that my (there was a url link here which no longer exists) was pretty successful in my opinion, I was wondering how everyone "tests" their film and developers. Right now, I'm shooting a bunch of lower-to-mid-ASA films (Efke 25, PanF+, Plus-X, FP4+, Delta 100, T-Max 100) at box speed and just seeing if I like the look of the results. Please spare me the condemnation, because I know this is not how you're supposed to test and evaluate film.* I know you're supposed to pick one emulsion and work with it a million times until you master it.

I was once told to start off shooting at the rated ASA, then the same scene a stop above and a stop below. And then, I should shoot at box speed but develop at +/- 30% of the development time. I tried the first part (varying my EI) and didn't really see a difference, so I put organized film testing on the back burner for a while.

Then I saw (there was a url link here which no longer exists), where the first post makes me feel like I should abandon photography altogether due to how little I know. I sort of understand the basics of the zone system, but nowhere near comprehending how this person is testing his film.

I'm wondering how the pro film users here go about evaluating a film (and possibly its response to different developers). Do you shoot a standard target, like an 18% grey card, a number of times at different exposure values? Or maybe some sort of sharpness-determining test pattern? Or if you go out shooting scenes, then how are you evaluating such subtle differences? And what's the end result - are you gradually plotting your own density curves?

And perhaps most importantly, can I evaluate the response of a film without a densitometer? They're really expensive and I can't imagine ever justifying the purchase of one.

* - after going through these films and seeing if I like the results, I will probably take the advice I find in this thread and go back to my standard HP5+ and try to work on perfecting my understanding of it...

I found this article from View Camera magazine extremey useful. Its very simple and non-technical, and for me it made the difference between sucking terribly and making some prints that I'm very pleased with. It's easier to do with sheet film but you can adapt it to roll film, you'll just have to shoot several rolls. But it saves a lot of guessing and hoping and wasting of film and time.

http://www.viewcamera.com/pdf/2006/VC_Getting Started.pdf

yeknom02 said:

I was wondering how everyone "tests" their film and developers.

Click to expand...

Some do extensive testing and others use what the film manufacturers print on their film boxes.

I looked into this a few years ago and decided on my lazy person's method:

Most articles I read about finding personal film speed came to the same conclusion. That was to use half the manufacturers film speed (i.e. a stop more exposure) and reduce development time by about 25%

I tried this and liked it so it became my standard. Ilford HP5+ at EI 200.

http://www.freewebs.com/stevesmithphoto/personal.html


Steve.

If you want to go a middle route between highly technical/expensive and non-technical/inexpensive you might consider the following. I haven't tried this, but I think it would work. I won't make a full description here, but just an incomplete outline of the idea.

In addition to your camera and film you will need 1) a light meter 2) a blank featureless wall of neutral color 3) a transmissive-type step tablet 4) your film development station 5) a light table, or at least some way to visually evaluate the density of the film.

Do the following:

1) Meter the wall at a number of iso settings.
2) Shoot the wall with exposures corresponding to each of the iso settings.
3) Develop the film
4) Take film to a light table for visual evaluation
5) While film is on the light table, place step tablet next to film
6) Visually match up each frame of the film with the step tablet, noting which step gives the best match
7) Use the documentation that came with the step tablet to determine the estimated density of the film
8) Make a plot of log(exposure) on x-axis and film density on y-axis
9) Evaluate the plot based on methods described elsewhere

Basically what I am suggesting is to use a step tablet as an inexpensive substitute for a densitometer. (I have a densitometer, so I haven't tried the method described above, but as I said earlier, I think it would work.)

I'm not knocking people who do meticulous testing of films, because it is very useful information that trickles down to many people, including myself. However, I think you may need to ask yourself, before embarking on this mission, how far you want to take it, and; what will be the net usefulness for you ? For example, if you are determined to really settle on a film or two and stick with the products that give you the 'look' you are after for some time, (after 10 or 20 rolls of tests), and stick with your test equipment and be able to control your processing regime rigidly, then go ahead with the meticulous testing: it could pay dividends to your technique. If, on the other hand, you are looking to use different cameras and lenses, shoot multiple subjects on the same roll, or not be able to get the chemistry you want consistently or change enlargers, for example, then I would say go easy and get to know a few parameters and general expectations of film/developer combinations: meticulous film testing isn't necessarily going to guarantee you better pictures.
If you do embark on testing procedure then what you suggested in the first part of your post would be the best place to start. Take your HP5+, buy a couple of different developers - say, D-76 and Xtol - and test rolls at 200, 320, 400, 640 and 800 using different dilutions and agitation or partial stand development. Make sure you keep detailed notes (including changes to temperature and agitation) and clearly mark the negative sleeves or it will become a mess. I suggest that you shoot 'real world' subjects (or those normal to you), because that is what you want to know about from the film information.
If the results aren't to your liking try other developers - did I say this was going to be cheap ? If someone says that 'w+x+y=z', don't be surprised that in your case (as in most other peoples') it doesn't. You create your own regime rather than try to fit into someone elses: however, take advantage of the guidelines and collective experience of the forums, this will make it cheaper.
Sorry to go on, good luck.
Mark Walker.
P.S. If you fancy yourself as an amateur scientist, get a densitometer, if you are like 99% of other serious photographers, evaluate your prints with your eyeballs.

Wayne said:

I found this article from View Camera magazine extremey useful.

Click to expand...

Thanks, great reading!

alanrockwood said:

8) Make a plot of log(exposure) on x-axis and film density on y-axis

Click to expand...

So what formula gives you exposure, and how do I calculate it from my camera settings and/or light meter recording? All I know is that the units of exposure are lux-seconds.

alanrockwood said:

Basically what I am suggesting is to use a step tablet as an inexpensive substitute for a densitometer. (I have a densitometer, so I haven't tried the method described above, but as I said earlier, I think it would work.)

Click to expand...

I did a Google search to find out what a step tablet is, and WOW, there are a lot of choices. How do I know what to get for the testing method you're talking about?

Several books treat this subject in some detail. Probably the most familiar is Adams, "The Negative." White, Zakia, and Lorenz, "The New Zone System Manual" and Davis, "Beyond the Zone System" also give good methods. Hutchings, "The Book of Pyro" has some interesting and useful methodology. The process is similar for all the methods discussed, although the details of each of the methods differ a bit. It is not difficult, but it is tedious, takes time, and uses a fair amount of film. The idea is to determine the threshold exposure (and thus speed) and dynamic range of the film for your film, equipment and technique, and then to determine the response to changes in development (thus finding the proper development times). The threshold exposure is not affected a lot by development, as long as you give sufficient development to get a reasonable image. That makes it easy to get in the ballpark for determining EI. (Actually, the manufacturer's recommendations are quite accurate for most work.) Depending on the contrast you get, you can then adjust development (and exposure, to some degree) to fine tune things. I like to use a step tablet in a special camera back to test sheet film, but that is quite impossible for 35mm. The Kodak Master Photoguide used to have a gray card and reflection gray scale that is wonderful for this sort of thing. Gray cards (18 percent reflectance) are still available. The reflection scale is a very useful addition to your target. Be sure lighting is very even. Remember that most meters are calibrated for 12 percent average reflectance rather than 18 percent. Be sure to check your calibration work against real life scenes by shooting some test scenes before you fully accept it.

I suspect you want to test your "system" and get an exposure index and development time. If you really wanted to test only the film you need a calibrated senstometer. There are only a handful of people that have that capacity here. For me, the utility of a film test is to compare less-well-known film to well established Kodak, Ilford and Fuji products and report the results on APUG.

Follow the already-posted advice and find a test you can do easily with your equipment at hand. But realize you are testing more than the film. You are testing your whole system including your meter, shutter, aperture setting dial, etc.

ic-racer, you raise an interesting point that reminds me of a question I had... Since there are so many films that have been rigorously tested and are very well-known, is there any resource that collects the exposure curves of these well-known films for selective comparison?

For example, is there anywhere I can turn to see the curves of Tri-X, HP5+ and, say, Neopan Acros plotted on the same graph so I can view their performance relative to each other? Or perhaps a list of data points so that I can create my own graph in a program of choice, such as Excel or Matlab?

I should test my film exposure/develop/print process to get it down. I should normalize my procedure. I understand why. I know how to do it. I can explain it for anyone who would be willing to learn how and why to do it.

But, admittedly, I don't.

I have done it many ways. Here is what I am doing now, mostly for incident metering. First, I do a speed test and normal contact sheet test, using a middle gray card as a guide. I rate and develop as the manufacturer recommends, and I follow the film development/agitation procedures outlined by Ilford. I use a lens hood. I meter with an incident meter in direct daylight near the middle of the day. I angle the card to avoid glare. I bracket in whatever increments my aperture ring will allow for a stop or two each way. Then I finish off the roll on other pictures. I then develop and make a "normal" contact sheet, in which I print the unexposed film edges right to the first point at which maximum black is reached (when the film edges first match the adjacent areas of the proofsheet that did not have any film on top of them). Then I see (with my eyeballs) which middle grey card in the proofsheet matches the actual grey card best, and use the corresponding EI in the future.

Then I test for contrast, so I can see what sort of brightness range the film can capture. I use the EI that I determined in the previous test. I do a series of exposures that will turn the contact sheet into a greyscale. I stop down a little bit into pure black (six stops down from the incident reading) and then step my way up the scale quite a way into pure white. Then I develop as the manufacturer recommends. I make a proofsheet so that my one exposure that was made at the meter reading matches the grey card, which also matches the frame that I called middle grey in the first test. (It is usually within a second or two of the "normal" proofsheet. I think different developer batches and temps cause the variation.) Then I take a look at how contrasty the film is, so I know what to do when shooting in certain light. If the film seems very far out of whack in terms of contrast, I will use a different development time and start everything over. I've had to retest to find new development times only with Delta 3200 (a very flat film) and Efke 50 (very contrasty).

That's pretty much it.

FWIW, I pretty much always get box speed or within one EI rating doing this. Doing Zone System film testing, you will probably not find this to be too common.

I use my spot meter to measure luminance range when I have the time. I use the incident meter to establish the normal exposure. I know from my contrast test how much range the film can handle. So I measure it with my spot meter. I then will tweak exposure and development if the range is excessively narrow or broad for tat particular film/developer combo.

Here is a data set to play with. The First column represents log inverse exposure. Put this on the x-axis reversed high to low. The second column is the y-axis log density values for Ilford delta 100. The third set of values is for delta 400. You can graph the two together and see how it looks. I post the data so you have a 'real' dataset and can see how to make an H&D curve in Excel or Matlab. The data are not ISO standard data; just results from some control strips to check development.

[x-axis 'log exposure']
0
0.13
0.27
0.42
0.59
0.73
0.88
1.02
1.14
1.3
1.44
1.59
1.72
1.85
2.01
2.12
2.27
2.4
2.55
2.70

[y-axis 'Delta 100']
2.18
2.02
1.88
1.75
1.6
1.46
1.33
1.19
1.04
0.87
0.73
0.58
0.45
0.33
0.21
0.12
0.06
0.03
0.01
0

[y-axis "Delta 400"]
1.57
1.48
1.4
1.32
1.23
1.15
1.07
1
0.92
0.83
0.73
0.63
0.53
0.45
0.35
0.27
0.19
0.12
0.06
0.03

yeknom2

just take a roll of film
and bracket your exposures.
shot 1 at box speed and what your meter says
and 1 above and 1 below ( fstops )
look at your film after you process it,
and then shoot a roll at the "speed" that you liked best.

you don't really need to do much more than that ...
people always try to make simple thing more complicated.

ic-racer:
I think your log density values should be negative...? Anyway, this is exactly the kind of data I'm looking for - is there a comprehensive repository for it somewhere on the web?

I'm really looking forward to performing these baseline tests on my system - finding a good film rating, developing time, etc. I haven't been able to determine them from simple variations on exposure.

yeknom02 said:

ic-racer:
I think your log density values should be negative...?

Click to expand...

Essentially yes, I just flip the X axis with a check box in the graphing program so I don't have to type in the negative signs. The actual y-axis numbers are really meaningless without any ISO calibration data so it does not make any difference either way. Again, each person's data is 'darkroom specific.' So sharing the info is not much use.
More specifically, the speed data in those numbers is meaningless because the y-axis has not been calibrated to an ISO standard (though I have a thread on how to do this). The slope and curve info is specific only to my conditions of agitation, temp, developer type, etc, the day the strip was processed.

Here is how Ic-Racer's data graphs with my excel spreadsheet. My interpretations are based off of a speed point of 0.1 negative density obtained at Zone I.

My calibrated step densities from Zone X down to Zone 0 are in 0.025 log exposure increments---so I plotted the negative densities as close to Ic-Racer's as I could, but I only count 20 steps.

Just my observations---obviously 400 is faster than the 100 as it takes less log exposure to reach a density of 0.1, but it is still near one stop too slow (i.e., the curve attains a 0.1 density at near Zone II instead of Zone I). So, by my graph its EI would be 200.

Delta 100 appears to be almost 2 full stops too slow to reach a threshhold of 0.1 neg density at Zone I----therefore, it's EI would need to be rated at 25 in order for the curve to reach the threshhold at Zone I. By my graph, the curve reaches 0.1 density at near Zone III!

I notice there is no data from Zone I to Zone 0, so perhaps these curves should be interpreted as being one stop faster than they actually appear, IDK. In that case, Delta 400 would be rated at 400 and Delta 100 would be rated at 50. When I graph my 21 step tablet, I get data from Zone 0 to Zone X.

Hmmmm

I don't believe Dale provided any exposure information with the data sets.

CPorter said:

Here is how Ic-Racer's data graphs with my excel spreadsheet. My interpretations are based off of a speed point of 0.1 negative density obtained at Zone I...

Click to expand...

I'm wondering - are the zones defined in terms of log exposure? I've only seen these graphs in terms of what toes and shoulders look like, never linking them to distinct values of density and log exposure. It would make more sense to me if Zone V were halfway between the maximum density and the minimum (base + fog).

jnanian said:

yeknom2

just take a roll of film
and bracket your exposures.
shot 1 at box speed and what your meter says
and 1 above and 1 below ( fstops )
look at your film after you process it,
and then shoot a roll at the "speed" that you liked best.

you don't really need to do much more than that ...
people always try to make simple thing more complicated.

Click to expand...

He didn't say he wanted to photograph, he said he wanted to test film.

Wayne said:

He didn't say he wanted to photograph, he said he wanted to test film.

Click to expand...

thanks wayne
i keep forgetting ...
some people endlessly test, others photograph ...
i am so glad when i started i pledged not to test

yeknom02 said:

...
I did a Google search to find out what a step tablet is, and WOW, there are a lot of choices. How do I know what to get for the testing method you're talking about?

Click to expand...

To answer this part of your question, I have a Kodak photographic step tablet No. 2, uncalibrated. BH sells this unit for about $109. However, you can get a MUCH less expensive step tablet from Stouffer Industries.

http://www.stouffer.net/TransPage.htm

Stouffer calls it a step wedge, but it is the same thing as a step tablet. I suggest the T2115 or the T3110. The T3110 has a finer step (1/3 of a f-stop equivalent step size, which is the same as 0.1 density unit step size) and is a little bigger than the T2115 (1/2 f-stop step size, which is equivalent to 0.15 density unit step size.) The price list link is here:

http://www.stouffer.net/Productlist.htm.

Calibrated versions of those same products are T2115C and T3110C, which cost a little more but are calibrated. You could probably get by with either calibrated or uncalibrated units.

Wayne said:

He didn't say he wanted to photograph, he said he wanted to test film.

Click to expand...

Well, now I'm willing to revise my statements. What I'm really after in the end is optimal printing. And optimal prints means knowing your process front and back, particularly how to rate each film and getting to know what that film is going to give you, especially at its limits. I did the bracketing thing once and I could hardly tell the differences frame to frame. I'll probably end up comparing bracketed exposures again, but after I've set the "boundaries" of my film(s), so to speak. I plan on filling a lab notebook very quickly...

My analogy* is that I want to tune my instrument, and that's best done one string at a time. If I try to judge tuning while playing songs - i.e., making pictures - it's a LOT harder to know what you want to fix.

(* note: analogy doesn't work for non-stringed instruments)

yeknom02 said:

Well, now I'm willing to revise my statements. What I'm really after in the end is optimal printing. And optimal prints means knowing your process front and back, particularly how to rate each film and getting to know what that film is going to give you, especially at its limits.

Click to expand...

This sounds like you want to read Way Beyond Monochrome, by Ralph Lambrecht. A wonderful book - I bought the second edition as soon as it became available, even though I already have the first. See http://www.waybeyondmonochrome.com/ for information about the book, including sample chapters in PDF form.

-tih

jnanian said:

yeknom2

just take a roll of film
and bracket your exposures.
shot 1 at box speed and what your meter says
and 1 above and 1 below ( fstops )
look at your film after you process it,
and then shoot a roll at the "speed" that you liked best.

you don't really need to do much more than that ...
people always try to make simple thing more complicated.

Click to expand...

There you go, John!

Frankly, if I had to test every film I use, for every possible combination of developer, camera, lens, meter, agitation variations, temperatures, moon cycles, etc, I'd abandon photography altogether.

I'm certainly not knocking anyone who wishes to conduct extensive and technical testing, but, at the end of the day, I really wonder if that translates to more successful images.

MaximusM3 said:

Frankly, if I had to test every film I use, for every possible combination of developer, camera, lens, meter, agitation variations, temperatures, moon cycles, etc, I'd abandon photography altogether.

Click to expand...

Me too.


Steve.