View Full Version : Scanning Film Speed Test/what's your favorite film testing method?

21-Jul-2010, 06:11
Hi all,

I'm getting interested in doing a film speed test for hp5 and my equipment.

From my limited knowledge of it, it seems like you generally do a film speed test related to the paper you are using in relation to your film. The problem is that I don't print at this point, and I'm fairly unlikely to anytime soon. Are there any resources for doing film testing related only to scanning?

And not to start a fight, but what is your favorite film testing method? I believe there are multiple levels of detail...my preference would be a pragmatic level of most information gained for least amount of materials/time. I'd like to be in the ballpark, preferably somewhere near third base.


neil poulsen
21-Jul-2010, 09:23
In standard film-speed tests, one doesn't determine the speed based on the paper being used. Film speed is, and should be independent of paper selection. Select the film speed such that four stops less light than what the meter reads puts a minimum level of detectable detail on the film. I find on a regular basis, that black and white film ASA tests out at about half that of the manufacturer's stated film speed. So, an ASA 400 film will often test out at an ASA of 200. It depends on the film, though.

Using a transmission densitometer, Ansel Adams found that 0.01 density units above film base plus fog was a good standard for this minimum detectable level of detail. This of course assumes that you have a densitometer. For myself, I wouldn't consider doing black and white photography without a densitometer. Sometimes, you can find them at reasonable prices on EBay. I've seen methods that use scanners in some way. But, I don't recall where I saw those. Maybe a friend might have a reliable scanner, or find a lab that will take readings for you. Maybe someone else on the forum knows about using a scanner for these tests?

So, the film-testing process would go like the following.

Photograph something that's evenly lit, like a 68% Kodak neutral gray card. Put eight sheets of film into holders and take photographs of that surface. Set the meter at ASA 100, set the aperture to about f8, and select the shutter speed so that four stops less light than that read by the meter actually reaches the film. You may need to adjust the aperture a little from f8 to achieve this.

Take a photograph. Focus as if you were taking a photo at infinity. Then, take six more photos, and each time, decrease the aperture diameter by 1/3rd stop. (So, the f-stop number increases by 1/3rd stop each time.) Develop these plus one unexposed sheet of film at the manufacturer's recommended development time and temperature. These seven sheets of film will correspond to ASA's of 100, 125, 160, 200, 250, 320, and 400 respectively.

Once developed and dried, read the sheets of film on the densitometer. The film's speed is that ASA such that the corresponding sheet film is 0.01 density units greater than the reading you get for the unexposed sheet of film.

As to illuminating the neutral gray card, I have much better luck using the blue, daylight corrected bulbs that one can maybe still find in a pro photo store. I believe that black and white panchromatic film is daylight corrected, so the daylight bulbs work better.

Jay DeFehr
21-Jul-2010, 10:32

Film speed test is something of a misnomer, because these tests almost always test something other than the film's speed. The speed of HP5+ is well known and documented by Ilford- it's ISO 400. What you're suggesting, and what Neil describes above, is testing for inaccuracies in your equipment using film as the known variable. An inaccurate shutter doesn't change the speed of the film, but testing can determine how you should compensate for it. Even when testing for a film/developer combination, it's the developer that's being tested against a known value, and not the other way around.

So, if you're concerned your equipment might be inaccurate, simply bracket a typical, daylight scene over a wide range of f stops, develop normally, scan and evaluate. Check your light meter by comparing its reading to the sunny 16 rule. Correlate the negative with adequate shadow detail to its exposure value, and that will indicate your deviation under the conditions of the test.

Paul, you should keep in mind that as far as concrete values go, the ISO # is as good as it gets. Exposure Indices are fluid and more subject to personal interpretation than the influence of equipment or chemistry. I often expose my portraits at 2X ISO speed, not because I've somehow managed to make my film more sensitive, but because I like the way the tones are arranged in the print. Shadow detail is neither sacred, nor mandatory, but an important reference point for evaluation. Don't get too hung up on numbers. I hate to see film wasted on testing.

Good luck, and have fun!

Peter De Smidt
21-Jul-2010, 10:46
Traditional reflection camera meters give readings such that the luminance that they're reading will produce a middle gray on the negative with normal development. That luminance is called Zone V. Every doubling or halving of the exposure gives a new Zone. So for example Zone VI is double the exposure of Zone V, and Zone IV is 1/2 the exposure of Zone V. Traditionally the range of Zones goes from I to VIII, with zone I being a tad bit darker than the reading through unexposed but developed film, referred to as film base plus fog. With traditional bw film and developing, Zone III is the lowest Zone that produces good shadow detail on the print and Zone VIII is the highest zone that produces good highlight detail. (Caveat: modern films keep detail with significantly higher exposures than Zone VIII, but you'll have to burn them in on the paper, or switch to a less contrasty paper to get them to display detail on the print.)

Ok. So now you're out in the filed with your camera. I'll assume that you have a spot meter to make things easier to explain. The same system will work with different types of meters, but it gets a little more complicated. You meter the scene and find that there's a 5 stop difference between the darkest and lightest areas where you want good detail. You will now place that lower value on Zone III and the higher value will fall on Zone VIII, 5 zones higher than Zone III. For example, suppose that the darkest area where you want good detail gives a meter reading of 1/50th of a second at F22. If you set your camera to those values, you'd be placing the shadowed area on Zone V. (Remember, that's what meters tell you.) But you want it on Zone III. Therefore you close down two stops, changing the aperture to F45, you increase your shutter speed to 1/200th of a second, or you close down to F32 and change your shutter speed to 1/100th of a second, any combination that'll put the luminance that you're reading on Zone III.

When you're film testing, you want to discover how much exposure will give you the Zone I density such that you'll get a Zone III that'll produce good detail on the print. Normally for darkroom printing, that's a Zone I that gives a transmission densitometer reading of at least .1 above film base plus fog. (I prefer a higher zone one density of around .15 to .2 for optical printing.) You will then find your development time such that Zone VIII gives a reading of 1.35 above fb+f for a diffusion enlarger and 1.25 above fb+f for a condensor enlarger. These values would then allow you to print a normally developed and exposed negative of a 5 stop scene on a grade 2 paper and have good detail at the points in the scene that you put on Zone III and Zone VIII. (Caveat: just because a paper is labeled grade 2 doesn't mean that it meets the standards for grade two.)

Note that the lower negative densities are affected mainly by exposure, whereas the higher densities are affected mainly by amount of development.

What I've just described is standard Zone System practice for darkroom printing. Scanning changes things a bit. In many people's case, they can get more detail out of the thinner areas of the negative with scanning than with optical printing. For example, with optical printing I want my Zone I density to be .15 to .2 above film base plus fog, as I said above, but with scanning, though, I get good detail down to .05 or even a bit lower. This means that I can expose my negatives that I plan on scanning less than those I plan to print optically. Less exposure generally produces finer grain and higher sharpness, and since scanning tends to emphasize grain in bw negatives, exposing less is an advantage. That said, if you don't know if you'll ever print the negative optically, you should expose and develop for optical printing, as that'll give a great negative for optical printing and a very good negative for scanning.

A good way to test how a scanner handles various densities is to scan a step wedge with known densities.

21-Jul-2010, 10:58
There are as many right answers to the question as wrong.

The best exposure is the one that yields the best prints.

Some studies show that print quality is maintained (plateau rather than peak) over a range of exposures.

The following method can be used to determine the minimum exposure for best prints.

It is a reasonable substitute for hours or days or years of trial and error. But you could still tweak it after making some prints.

Zone I technique:

Expose a uniform target to zone 1 at some ISO setting.
Process the film and if the density is 0.1 above fog your ISO setting for original exposure (and future exposures) is/was good enough to produce excellent prints.

(If you don't know what 0.1 density is, you can place the negative over your meter. If the frame in question cuts the reading by one-third of a stop compared to an unexposed but processed piece of film, you are set.)

21-Jul-2010, 11:42
Thanks for all the help and answers...I'll need to process little by little.

Some issues I have right now:
I only have an incident light meter, which I use for everything. It's a Minolta Autometer V F. I don't have the reflected light thing to put on it. It's unlikely that I'll be able to afford a spot meter any time soon.

This photograph got me thinking that I'd like to be able to dial in exposures a little better:

Coming from digital, I would have been worried that I would lose details in her face because she'd be on the low end of the scale, where there's more noise. So I metered for her face with the incident meter (i.e. the white bulb was in the shadow of her face).

Now I ended up with a good exposure for her face and I like how that came out. Just wondering, if I got my metering more accurate, would I be able to get shadow detail into the highlights, which are currently blown out?

There's a touch of additional contrast in the curve I put on the picture in photoshop, but that's more or less how it scanned.


Jay DeFehr
21-Jul-2010, 12:05

In my opinion, regarding the example above, there are no important shadow details, or highlight details- it's all in the middle. I don't think this kind of shot even benefits much by an exposure meter, so I don't think your metering technique is terribly important here. I agree your highlights are blown, but that's a lighting issue- specular highlights don't typically make a very good background for a portrait. If you decreased development to address those hot spots, your mid tones would go flat. In short, I think you're barking up the wrong tree with film speed testing.

Lovely subject!

Ken Lee
21-Jul-2010, 12:41
You might find this brief article helpful: Testing Black and White Film (http://www.kenleegallery.com/html/tech/testing.html)

Peter De Smidt
21-Jul-2010, 12:54
An incident meter is fine. If you want to get quite technical with one, then check out "Beyond the Zone System" by Phil Davis at your library. If you aren't going to vary development by subject, then don't bother with that book.

You're scene is, I expect, more contrasty than a standard 5-stop scene, with the main subject lit by open sky (not direct sun) and some ares lit by direct sun. You have given enough exposure, at least for scanning, as there's pretty good detail in the dark areas. The only issues are the super bright areas in the background around her head.

Here are some things you could do about it with comments:
1. You could expose less. This would darken the offending areas, but it would also darken everything else. You would probably lose shadow detail. I don't recommend this.

2. You could develop less. This would lower the contrast of the negative. As such, this would lessen the brightness difference between your subject and the background. The down side to this is that you'll lose some tonal separation in your subject, and without local controls (dodging and burning in a darkroom, or various analogues in Photoshop), those areas will still be brighter than your subject.

3. You could take two exposures of the subject, one with an incident reading at the subject, and one with 2 stops less exposure. Scan both and combine in Photoshop.

4. You could compose your subject such that there aren't distracting bright areas in the background. This is the easiest choice.

5. You use a reflector or flash to increase the exposure on your subject such that the bright areas in the background get darker relative to your subject. This is the best choice if you can't re-compose to eliminate the distracting areas.

6. You could clone out the offending areas in Photoshop. This is tedious and hard to do well.

I regularly use an incident meter. Set the iso so that you get good shadow detail, which it looks like you've done. Meter the light falling on the most important part of the subject. In this case I would've held the meter fairly close to her face and pointed at the camera. Set according to meter. If there are brighter areas of illumination that could be distracting, do 4. or 5. above, or if you don't have a better choice do 3.

Brian Ellis
21-Jul-2010, 14:53
My favorite method is to let The View Camera Store do it for me. I've done my own testing for film speeds and development times. The first time around it's educational, after that it's tedium. So given a choice I'd rather let someone else do it for me at a cost of $30 (last time I looked). I've spent more than that on film when doing the tests. Plus The View Camera Store will give you more information than you can get on your own unless you own the Expo/Dev program and a Palm Pilot with which to use it.

Henry Ambrose
21-Jul-2010, 18:34
I'm on my phone so this short--
What you want for scanning is a negative that conveys the scene information to your scanner via densities that fit within the capabilities of your scanner -- what your scanner can "see". If you send the scanner a neg that is too dense or one with range greater than the scanner you have lost that scene information.

The simple and easy version is to shoot HP5 at 250 or 320 in common developers and scan it.

The problem with your picture is a lighting problem, not a developing problem.

21-Jul-2010, 19:13
Thanks for all the advice. Peter...you know I used to be good at lighting. Just finishing nursing school and getting back into photography. I get it...to much contrast in the scene. I'm not sure what I was thinking about being able to hold everything together.

Personally I don't necessarily mind the highlights in the background, but I would have liked to have some detail in them so I could manipulate them a bit.

Brian...that sounds like a good idea. I'll keep it in mind. My plan is to only shoot HP5, at least for a year, so it might be worth the investment. Well...mostly only HP5.

Ken...I liked your article a lot. I'm going to give that a try.

Jay...yup. I get it.

Is there a used densitometer that's usually recomended to someone on a budget? Or is that a rich man's game? What about a step calibration thing for my scanner?


Peter De Smidt
21-Jul-2010, 19:24
I haven't looked in awhile, but good densitometers were available cheap on Ebay awhile ago. You're other option is to send test negatives to someone who'd read them for you. I could do that, as could many others. The View Camera Store offers this service for a small fee, as someone mentioned above.

Stouffer step wedges are available for a reasonable amount.

Peter De Smidt
22-Jul-2010, 06:37
How did you scan it? Are you sure you're not clipping highlight detail? Look at the negative on a light table with a loupe. Can you see any detail in the bright areas that you aren't getting in the scan?

22-Jul-2010, 07:03
Hi Peter,

I scanned it on an Epson 4870 with a better scanning rack.

I scanned it 2x with the same results. Right now I'm making it very grey to see if I can get something out of the highlights. Perhaps I can make a layer over it and burn the highlights in.

The negative is really thin in the specular highlights. There may be the tiniest hint of fog.

PS thanks for your offer regarding the negatives. If I get my act together I might take you up on that.


22-Jul-2010, 12:01
The negative should be dense in the highlights. I would bet the detail is there in the negative (depends on the film; I bet money it's there if you used TMY2.)

You just can't easily print or scan it. Look for Ken Lee's scanning tips. You could also do two scans, one of them of the highlights and use some HDR trickery to put the detail back in the highlights without badly messing up the tone of the image.

Your scene contains a high contrast woods scene with many bright spots and shadows. The subject is in the shadows and exposed properly. I think it's a well made photo. If you wanted to reduce the contrast at the time of exposure, you could wait for a cloud to go over, or use some creative lighting to weaken the depth of the shadows.

22-Jul-2010, 15:56
The negative should be dense in the highlights. I would bet the detail is there in the negative (depends on the film; I bet money it's there if you used TMY2.)

duh. yeah. not thin. dense. they are that. I'm working on doing two scans to see if I can pull in the highlights. I just want them to be a little less than pure white and fit in a little more organically with the rest of the picture.

Peter De Smidt
22-Jul-2010, 19:19
The real Achilles heel of consumer flatbeds is their ability to deal with high densities. Since development greatly effects the high densities, it's possible that you should develop a bit less in such a contrasty situation. A way to test this would be to get a 31 step 4x5" Stouffer step wedge and make some test scans. That way you could find out at what density your scanner loses good separation between tones. You could then set up your development and exposure such that you don't make areas where you want detail to dense for your scanner.

Another alternative would be to have someone with a more capable scanner scan the negative for you. I have a Cezanne scanner, and I'd be happy to scan your negative for the cost of postage. Drum scanners do even better than professional pre-press flatbeds with dense materials, but these scans can get expensive. Before springing for one, definitely check with a loupe to see if there's really detail in those areas.

Brian Ellis
23-Jul-2010, 08:14
Thanks for all the advice. Peter...you know I used to be good at lighting. Just finishing nursing school and getting back into photography. I get it...to much contrast in the scene. I'm not sure what I was thinking about being able to hold everything together.

Personally I don't necessarily mind the highlights in the background, but I would have liked to have some detail in them so I could manipulate them a bit.

Brian...that sounds like a good idea. I'll keep it in mind. My plan is to only shoot HP5, at least for a year, so it might be worth the investment. Well...mostly only HP5.

Ken...I liked your article a lot. I'm going to give that a try.

Jay...yup. I get it.

Is there a used densitometer that's usually recomended to someone on a budget? Or is that a rich man's game? What about a step calibration thing for my scanner?


I assume you're talking about a transmission densitometer since we're discussing zone system testing. I don't know about "usually recommended" but you can find plenty of used transmission densitometers on ebay for little money since the labs that formerly used them aren't around any more or have switched to digital.

I bought a Macbeth Somethingorother digital transmission densitometer for about $200 on ebay but it wasn't the smartest purchase I've ever made. I used it once for testing and never used it again. A combination transmission-reflection densitometer would be more versatile since the reflection side can be useful for things other than zone system testing, e.g. when printing digitally to measure the dMax of different ink/paper combinations or to build your own profiles. However, they cost more money than a straight transmission densitometer.

I don't know about a step calibration thing. I used a 21 step wedge in the darkroom but haven't found a need for one when scanning and printing digitally. Maybe someone who uses one could explain the benefit.

Peter De Smidt
23-Jul-2010, 09:08
The benefit is that you can figure out the response curve of your scanner, noise characteristics of the scanner at different densities, and you can also tell where specific settings cause clipping.

Bob McCarthy
23-Jul-2010, 09:35
With a scanner I don't use the same process to evaluate ISO and development times as I did with optical enlargement, thought the general principles hold true.

I do include some of the clear film edge in my cropping box with my scanning software. I adjust scanner density by using the eyedropper on the clear edge to set DR = 0.

That pins the lower end. If you use a piece of film shot at zone I it should read base + .1 if your software has a densitometer function.

As for development time, I shoot for log 2.5 as normal. If I have a ++ scene I found I am still inside the capability of this scanner (a pro level flatbed).

If I have a minus scene then off to photoshop, or tweek in the scanning software.

I owe Fred Picker my inspiriation for testing of film speed from 20-30 years ago. I think of my scanner as a funny looking enlarger,and proceed appropriately.


Bob McCarthy
23-Jul-2010, 17:23
I miswrote DR=0 and should say, Z=0 (Zone 0)

Sorry for any confusion


28-Jul-2010, 18:25
still working on all this stuff, thanks for your help everyone.

Peter, do you think there's much difference in use between a 21 step wedge and a 31 step wedge? The price difference is quite severe.


Peter De Smidt
28-Jul-2010, 21:34
Hi Paul,

The 21 step version should be fine, as long as the densities go up fairly high.

2-Aug-2010, 11:50
Hello all,

Still dubbing away on all of this.

So I have a sort of test thing in my head that I'm planning to do. Was going to run it over with you before I do it.

I was going to shoot 5 copies of two negatives:
1) A high contrast scene with a mix of bright highlights and dark shadows, that's sort of horizontally even in composition
2) a low contrast scene

But my trick idea was that I was going to take an incident light exposure with my meter, set the aperture at that reading. Set the shutter speed 2 stops down and pull the dark slide out 1/5 of the way. Take a shot. Pull it another 5th, take a shot. Another 5th, now I should be at correct exposure. Take a shot. Etc...for two more shots.

Then...develop at my current time, -1, -2, +1 and +2. Scan all the negatives and then I should have a good idea what development time will work for a good negative for scanning?

Any gigantic holes in this method? Will this pulling the darkslide thing work?


2-Aug-2010, 12:03
Is there a used densitometer that's usually recomended to someone on a budget?

A photographer needs a densitometer like a fish needs a bicycle.

The amount of creative control you have over exposure dwarfs any tiny film speed differences you might find out by doing elaborate film speed testing. Besides, the manufacturer already does elaborate film speed testing for you...it's written on the box.

Look, this is what I do: I take pictures. Sometimes I even use a meter to measure the light. Then I look at the pictures. As a photographer, I know what I want the pictures to look like. If they don't have enough shadow detail, and I can't get said detail out of the negative because it's too thin in the shadows, then I expose more. If I can.

It's not that complicated. Go take pictures. Make sure you expose your film enough, if you have enough light to do so.

Henry Ambrose
2-Aug-2010, 12:41
Hey Paul,

I, other sheet film users and especially the film makers really appreciate you keeping sheet film alive by shooting lots of tests. <grin>

Seriously, its become generally accepted practice to make negatives for scanning that are essentially normal or normal minus for scanning. And there is a good bit of lee way as long as you don't make a negative with a density range that your scanner can't see all at once. Do a search for this subject and you will find many, many pages of discussion on this subject.

You're kinda re-inventing the wheel but I'm sure that it'll help hone your personal process, which is a good thing. There's nothing like knowing something is right as opposed to thinking its right. Be sure to report back with your test results and add to the knowledge base here.