Night photography - film, exposure, development

I have had good results with TMax 400 for 4x5, using 20% less development time than normal development. I have been using a handy exposure calculator which is available on the South Bristol Views website: Just select Graflex manuals and scroll down

Click here for a night exposure calculator

All films have some reciprocity failure, it's just that the newer emlusions can go to slower shutter speeds before it kicks in. Acros is, as far as I know, a form of the newer "t-grain" films such as Tmax and the Delta films.

I don't think there is a type of film that is ideal, esp. at the kinds of shutter speeds you're talking about. Of course, something like Tmax or Acros will give you a much shorter compensated time at a metered 1 minute exposure than TXT or some other traditional emulsion. You can easily get out into some seriously long exposure times with traditional emulsions. I have shot TXT, FP4, and Delta 100 at night, in very low light, and all give good results. It just depends on the look you want. I usually go for FP4 myself.

I find that my meter is usually accurate enough to meter _something_ in a scene in low light. If I can get something, I can usually guesstimate from there for my exposure time, then I factor in reciprocity failure.

I haven't tried stand development myself. I just cut development time by about 20% and it comes out alright. Night time scenes are contrasty, and I try to compensate but, at the same time, I want it to look like _night_, with dark shadows and bright highlights.

allan

I have used HP5 for night photography Night Shot

When you hit reciprocity curve on a film the result is higher relative contrast between highlight and shadow/black areas since the highlight portion of a scene is usually not into reciprocity whereas the shadow areas are and the result is under exposure in the shadows. (if you metered for the shadows, which is not usually possible, then the result would be very dense or blown out highlights). I always meter a highlight area and using normal zone system rules and reciprocity figures for Hp5 works fine without adjusting normal dev. You just need a subject with enough light in a highlight area to take a reading with a spot meter.

Acros 120 doesn't hit reciprocity until above 2 minutes metered exposure time. A 2 min exposure on HP5 is a reciprocity corrected 25 sec meter reading.

The example shot used the same dev as for normal daylight but the trick is to get a really good highlight reading and place on zone 7,8 or 9 and not worry too much about the shadow areas. It is night time after all and you can expect some areas to be really dark. The shadow areas do have good detail in the actual print but the scan has lost some of it.

What kind of film do you think is ideal for night photography?

I really like Fuji Acros for B&W and Provia or Velvia 100F (not Velvia 50) for color slides. I recently did some tests of color print films (Fuji Reala and Kodak 100UC) the results of which are available here. I haven't had a chance to do any real shooting with them, though.

Low ISO like Pan F?

Pan F wouldn't be my first choice because it's already a pretty high contrast, low latitude film. You want somethat that gives you a longer, flatter straight line region to capture the extreme range of light often present at night. But if you've used it and like the results, then go for it.

More generally, low ISO films work well at night because they tend to suffer from reciprocity failure less than 400 ISO or higher films, so often end up being faster for multiple minute exposures.

Is there a difference between t-grain and traditional grain films in terms of night photography?

The T-grain films tends to have less reciprocity failure. They also have less of a shoulder, which depending on your subject and the look you're aiming for may be either an advantage or a disadvantage.

And how about that Acros stuff?

It's great. By far my #1 choice for B&W night photography, and a close second to APX 100 for general purpose use.

If my normal exposure for a scene is five minutes, what would it be for Acros (since it doesn't have reciprocity failure)?

Acros does have reciprocity failure, it's just that it's very slight. You have to open up the aperture 1/2 stop above 2m.

You'll have to tell me what film you're using for your normal exposures for me to make a valid comparison with Acros. But lets look at a 5m exposure on Acros, which would be about a 8m exposure on Tri-X or a 42m exposure on APX 100, FP4+, or similar slow conventional film. It'd also be a bit sharper and a lot less grainy than any of these.

What kind of development is best for long exposure shots?

I usually just use regular development in Xtol 1:1. For Acros, I rate the film at EI 50 and develop for 9.5m, which is a approximately normal development. For conventional films (e.g. Tri-X), where reciprocity failure tends to amp up the contrast, I'll do a 1 stop pull: I rate TX at 200 and develop for 7:15 in Xtol 1:1. Acros doesn't seem to need the reduction in development, but it does need a bit more exposure than EI 100, even for daylight photography.

I've tried stand development with really dilute HC-110, development time of about 70 minutes without agitation, and liked the results.

I've read reports of good results with this, but I've always had problems with air bubbles when I've tried reduced agitation.

Is this a good way to develop night shots? Obviously I need to control for the heavy contrast in scenes with only one light source.

If you get results you like, then for sure its good way to develop. I'm not sure how well the newer t-grain emulusions respond to stand development, two-bath development, etc. compared to older conventional emulsions, but it's certainly worth a try.

Overall, it sounds like you have a pretty good handle on what needs to be done for night photography. Now it's just a matter of trying a variety of materials and techniques to see which combination gives you the results you like the best. And be sure to bracket a lot, since even the best meters yield good guesses at best in very low light.

-Jon

Generally, the most important factor with night photography, assuming you can set up a tripod and take long exposures, is reciprocity characteristics. Tech Pan has long been the preferred astronomical film because, with a gas hypering treatment, it has almost no reciprocity failure -- exposures that calculate as an hour might only require two or three hours (compared to multiple nights for most other films at the same calculated exposure time).

For good or ill, however, Tech Pan is effectively gone. The best reciprocity capacity I'm aware of in current B&W films is T-Max 100, which requires no correction at all under 10 seconds, and beyond a few minutes is the fastest film available short of gas-hypered Tech Pan. There are a few color films that are reported to have even better reciprocity -- Fuji transparency films, for instance, have been reported to need no correction out to several minutes -- but they're also slower than ISO 100 to begin with and not subject to development controls in the same way B&W films are.

Stand development or greatly reduced agitation, with a highly dilute developer, will tend to promote compensation (HC-110 Dilution G works very well for this, with agitation on three minute, five minute, half of deve time, or no agitation after the first minute), which means the highlights (typically overexposed by the time the shadows have anything in them, because highlights are less affected by reciprocity failure anyway) get less development and overall image contrast is reduced by compression of the highlights -- a strongly compensating formula can preserve detail in a printable range with up to a dozen stops variation in light over the scene (HC-110 isn't quite that good, but it's cheap and useful for normal development as well).

Wow, thanks for the detailed answers.

Andrew: That exposure scale sure is "jiffy," but I'm wondering how useful it is, considering updates in films since 1964. ISO 100 and ASA 100 would be the same, I would think, but reciprocity failure would be different... or is the scale pre-reciprocity calculation?

Mr. Perfect Exposure: Thanks for the tip about metering for the highlights; that makes a lot of sense to me.

Jon: Thanks for that long response. I'm not yet entirely sure of what toe/shoulder mean - really, what the implications of it are. You said: "The T-grain films tends to have less reciprocity failure. They also have less of a shoulder, which depending on your subject and the look you're aiming for may be either an advantage or a disadvantage." Does this mean they'd be less likely to blow out the highlights? Or phrased another way, they'd successfully recieve more exposure without blowing out the highlights, so that the shadows could recieve some detail?

About Acros, you said: "lets look at a 5m exposure on Acros, which would be about a 8m exposure on Tri-X or a 42m exposure on APX 100, FP4+, or similar slow conventional film." Are you serious? An 8 minute exposure on Tri-X is 42 minutes on APX 100? I didn't realize there was that huge a jump from conventional 400 to conventional 100 ISO (or did you mean TMX instead of tri-x?)

When you say that Acros has finer grain than the other films, is that a function of reciprocity failure, or simply the newer grain structure? As in, if I prefer daylight TXP grain over daylight Acros grain, would that be the same as night TXP grain and Acros grain?

Donald: I run a college darkroom, and have kept the film developer limited to the various dilutions of HC-110, for ease of use. There's nothing keeping me from having my own private stash of a different developer, though. What would be a better compensating developer than HC-110?

Thanks for your responses, everyone.

Terry

Terry, my suggestions may not be helpful because I tend to be a bit of a libertine in my approach to nighttime photography. Sometimes my exposures under a full moon depend on how long it takes me to finish a glass of wine.

Other times I just set my Nikon F3HP to AE mode and let it decide. It's been surprisingly accurate, especially considering that Nikon's literature mentions only an 8 second timed shutter speed. The first time the shutter went beyond 5 minutes I thought either the shutter was shot, the mirror was stuck or the battery died. However it's gone as long as an hour, depending on available light and aperture, and produced usable exposures. So sometimes I'll just set it to AE mode and bracket using the otherwise usually awkward exposure compensation dial.

However I do use Fred Parker's "Ultimate Exposure Computer" as a guideline. I consider it an essential reference for difficult exposure situations.

See: www.fredparker.com

I use Diafine and Microphen to develop most of my long exposure nighttime photos. Both help keep contrast under control so I guess they qualify as compensating developers. They seem more effective at that than Rodinal and HC-110, which I also use.

However I do like some of the results I've gotten with Tri-X in a very dilute solution of Rodinal, usually 1:200, for two hours of stand development. I agitate only for the first 30-60 seconds. So far, so good - no problems with uneven development. It's very grainy but that's fine with me for certain shots. Usually the results I see in very dilute Rodinal aren't all that much different from more concentrated solutions. A bit more grain in the more dilute solutions.

FP4+ has good reciprocity characteristics, is very forgiving of exposure error and turns out well in Diafine.

I shoot a lot of TMX but I can't remember ever having done so for long exposure nighttime photography. I should try to find an article Howard Bond wrote for a magazine about a year or so ago that analyzed the reciprocity characteristics of T-Max films (and Tri-X if I'm recalling correctly). Apparently TMX can be a good choice, which for some reason surprised me at the time.

I don't think there really is much difference between choosing a film for nighttime use and choosing one for any other use. Tri-X looks like Tri-X, FP4+ looks like it always does, etc. Judging from the nighttime photos taken by other photographers, most on Tri-X, it responds similarly for them.

There's really no reason you couldn't use Pan F+. Well, other than the much longer exposures needed, which cuts down on your opportunities to bracket or make other shots on the same night under fairly consistent conditions. Diafine really tames the contrast of Pan F+ and doesn't make it significantly grainier (if anything it just looks sharper).

I tend to mention Diafine (and Microphen) a lot. It's an extremely easy developer to work with, very forgiving - time and temperature are nowhere near as critical as with other developers - and a very good choice for developing film exposed under very contrasty conditions. Conversely, it's not a good choice for low contrast situations. Contrast cannot be boosted significantly with extended development.

Occasionally, tho', some interesting results can occur, but they're fairly unpredictable. I tried a roll of TMY exposed at around EI 500 for development in Diafine. The outdoor photos on a sunny day had blown highlights and little shadow detail, which is an unusual result from Diafine. But a couple of photos I took of a plant in the atrium of a museum under very diffuse lighting were very striking. There was a moody feeling in the tonality that I found appealing.

I tend to stick with Tri-X in 35mm and medium format for nighttime use simply because it's familiar. I've used it for 30+ years in all kinds of light with all kinds of exposures, pushing, pulling, whatever. I wouldn't say it's ideal for all conditions but like a good friend it's reliable.

There are several examples in my photo.net folders, along with some information about exposure and development.

I'm not yet entirely sure of what toe/shoulder mean - really, what the implications of it are. You said: "The T-grain films tends to have less reciprocity failure. They also have less of a shoulder, which depending on your subject and the look you're aiming for may be either an advantage or a disadvantage." Does this mean they'd be less likely to blow out the highlights? Or phrased another way, they'd successfully recieve more exposure without blowing out the highlights, so that the shadows could recieve some detail?

It's kind of hard to explain without a picture, but I'll try. If you look at a film's characteristic curve (from the datasheet, for example), it kind of looks like an S. At very low exposures, the film doesn't respond at all to the light -- there's simply too few photons to trigger the chemical reactions necessary to record the image. Then, the response starts to increase. This region is called the "toe" and is where the curve starts to transition from a horizontal line (no response) to a diagonal line on the graph. In this region, the slope of the response is increasing, so the same increase in exposure at the start of the toe produces less density than at the end of it.

After about a stop or two worth of exposure, the slope stops increasing and the film's response once again becomes a straight line, though now it is a diagonal line going up and to the right. In this "straight line" region, the film responds roughly linearly to increased exposure: X units of exposure produce Y units of density both in the shadows and in the highlights. After a couple of stops, the response starts to become nonlinear again and the slope starts to decrease. This region is called the "shoulder" and like the toe, it has a nonlinear, but now decreasing response to exposure. Eventually, the response again becomes a horizontal line indicating that the film has reached its maximum possible density.

From an image making point of view, the key thing to remember about the toe and the shoulder is that tones are compressed when they lie in these regions. The advantage of this is highlights blow out and shadows block up gradually. While there may only be 1 stop or so worth of density in the toe/shoulder, it may use it to record 2-3 stops worth of exposure. Contrast this to something like a digital sensor, which has no shoulder: the sensor records full detail up to the clipping point, after which you get nothing but white. Film will gradually record less and less detail as it gets closer to its maximum density.

To make a long story short, the advantage of a broad shoulder is that highlights will blow out gradually. This can be useful at night, where you'll sometimes have light sources that are 8-10 stops above your middle gray exposure. The disadvantage is that there isn't much detail recorded in the shoulder, so if you wanted to burn in these highlights they'd probably look very flat.

T-grain films, generally speaking, have longer straight line regions. Where a conventional film might give you 8-10 stops of linear response, a T-grain film might go 10-12. This means you can get more detail in your highlights, but you'll have to work harder in the darkroom or in photoshop burning them in.

If you're using stand development and liking the results, you might not be too thrilled with the newer films. Stand development, and other compensation techniques, work by broadening the shoulder of the film.

I personally like the long straight line region of the modern films, but I am about doing about 90% of my printing digitally now, where it's very easy to dial in whatever highlight compensation I desire. In the darkroom, they often required a little more attention.

Another thing to watch out for with the T-grain films is that in some developers, the response curve actually increases in slope in the highlights, meaning that highlights are rendered with higher contrast than the mids and shadows. These negatives are a royal pain to print conventionally. I know the Tmax datasheets show this type of response in Tmax developer (which I've never used) but surprisingly, I saw exactly this effect with Acros in Rodinal 1:50, which is supposed to have a slight compensation effect. I don't see this with my current developer of choice, Xtol 1:1. I couldn't guess how they'll react to HC-110, but it probably depends heavily on the dilution you use.

About Acros, you said: "lets look at a 5m exposure on Acros, which would be about a 8m exposure on Tri-X or a 42m exposure on APX 100, FP4+, or similar slow conventional film." Are you serious? An 8 minute exposure on Tri-X is 42 minutes on APX 100? I didn't realize there was that huge a jump from conventional 400 to conventional 100 ISO (or did you mean TMX instead of tri-x?)

Yep, this would be why I quickly stopped using APX 100 at night, even though its probably my favorite film in every other respect. The rule of thumb I learned for reciprocity correction of conventional B&W films (and it's so far worked pretty well for me) is that 2s becomes 4s (2x), 4s becomes 12s (3x), 8s becomes 30s (4x), etc. So if the scene meters as 4m at EI 100, that's 1m at EI 400, which turns into 8m (8x) using the above rule. But if I'm using ISO 100 film, that's 4m x 10 or 40m (don't know where the 42m came from, probably a typo).

In reality, the rule of thumb does break down with really long exposures, but it gets you close. If you look at Ilford's datasheet for FP4+, the longest exposure in their reciprocity chart is 35s, which becomes 200s, a 5.7x increase (and close to the 6x that the rule of thumb would use here). Even if the slope of the reciprocity curve stayed constant (it doesn't, it increases) you're still talking about a 4m exposure turning into 23m. Reciprocity failure is ugly.

With exposures this long, any rule or chart is only going to get you in the ballpark exposure-wise, and the contrast is going to get really pumped up at that degree of reciprocity failure, so I'd probably do a series of brackets at 10m, 20m, 40m, and 60m, and hope I don't forget my gloves. That's why I love Acros: 4m turning into 5-6m is a lot more liveable, and I usually only have to bracket +/- 1 stop.

When you say that Acros has finer grain than the other films, is that a function of reciprocity failure, or simply the newer grain structure? As in, if I prefer daylight TXP grain over daylight Acros grain, would that be the same as night TXP grain and Acros grain?

It's strictly a function of the new grain structure. If you prefer TXP grain over Acros in general, it's not going to change much with long exposures. That said, I find Acros's grain more appealing than Tmax's, but it's no Tri-X.

-Jon

Terry,

I mostly use Acros for that purpose. Lack of reciprocity failure is cool when you've forgotten to bring the appropriate reciprocity tables with you, for films such as Tri-x for example...
OTOH I too like Tri-x a lot for long exposures. You can see a sample below, exposure was 3mn or so (and yes, I haven't caught a lot of fireworks despite those 3mn...).