Bracketing for long time exposures ?greenspun.com : LUSENET : Large format photography : One Thread
I've been shooting night shots a few times and I was wondering whether lightmeters or spotmeters would be of any help.
Since I've found out that an average time of 10 / 30 sec would be fine for specific night scenes and sometimes 45 / 60 / 90 sec. for other situations, I was wondering if this "bracketing" way of shooting night photo would be "easier" than spotmetering the scene and going into complex calculations...
Unless the spotmeter would save me a lot of time, do you think exposing 3 or 5 frames (10 sec difference each) would be the best way of having a wider panel of exposure ranges ?
Thank you for all your helpful contributions.
-- Bruce Barelly (email@example.com), May 08, 2001
The obvious answer is to test on Polaroids
-- Garry Edwards (bestsnapp;firstname.lastname@example.org), May 08, 2001.
Those are not terribly long times, so a sensitive light meter (Pentax digital for instance) would most likely be useful, namely give useful readings to base exposures on. I use mine routinely for exposures at EV2 or EV3 with excellent results. Give it a try. OTOH if you are happy with the exposure times you have already worked out, what else is there to do.
As for polaroids, I would expect the reciprocity characteristics to be significantly different from any film for multi-minute exposures, so as to render them useless except after careful calibration beforehand. Correct me if I this is wrong please.
-- Richard Ross (email@example.com), May 08, 2001.
The problem with bracketing long exposures is that reciprocity failure will quickly throw you into very long times. VERY long. Someone said that your times aren't very long, and he's right. Some shots I've taken in low light (not too low, mind you) come in around a minute or so, but compensating for reciprocity failure pushes the times into the 10 or 20 minute range. I trust my meter in some low- light situations, but oftentimes intuition is all one has...
-- Chad Jarvis (firstname.lastname@example.org), May 08, 2001.
I would agree based upon my own experience that Polaroids won't help much due to reciprocity failure. Color Polaroid will shift in color and I'm not sure about the Black and White but you'll get some sort of reciprocity failure at those exposures. In any case, it won't match what your film will do. Plus Polaroids tend to be more contrasty and don't usually pick up shadow detail well like b+w print film can. Of course you can use the Polaroid Type 55 Pos/neg film which is a lower contrast film and picks up shadow details quite well.
I am assuming that you are asking about b+w film. To help answer your question, I think that 10 second internals is relative. If you exposure calls for 30 seconds then you would bracket by exposing at 15 secs (-1 stop) 30 seconds (normal) and 60 seconds (+1 stop). At a 60 second exposure, you have to bracket in 30 and 60 second intervals (ie- 30 seconds to get -1 stop, 60 seconds at normal and 120 seconds at +1 stop). Technically, you also have to figure into this reciprocity compensation (in both exposure and development) for certain b+w films like Tri-X. I think that you want to get at least a 1 stop difference in your bracketing on b+w film.
I hope this helps.
-- Yoichi Kawamura (email@example.com), May 08, 2001.
Unfortunately, the Polaroid reciprocity failure schedule for these lengths of exposures is nowhere near that of film: it is much, much worse. I base this on experience with P'roid Pro 100 B&W material. I'd bracket, make careful notes and start building up your own "database". As a start, anything that says it needs a 15 sec exposure I'd start at 20 sec, and go out to 60 sec in 10 sec increments. Also the atmospheric (clear/cloudy) conditions will influence how much light bounces back on to the buildings.
-- Ellis Vener Photography (firstname.lastname@example.org), May 08, 2001.
this is a subject i have a great deal of experience with-- i've been shooting in downtown Seattle at night for several years, with exposures up to 5 hours (if you're interested, check out www.chrisjordanphoto.com-- all my work is up there).
there are a couple of key things i need to know to help you, most importantly, what film are you using? the films all have different reciprocity characteristics; some of the B&W films are quite horrendous, with ever-stepening curves that require multiplying your exposure by ten, and some transparency films have very little reciprocity failure even at extremely long exposures. Velvia, for example, turns a bit green and the exposure approximately doubles, but that tends to level out, so that even if your exposure time is an hour, you can expose for 2 hours safely. Provia is also excellent, even better than Velvia that way, and with more latitude.
i've had great luck with my Pentax 1-degree spotmeter, and have discovered that it is actually accurate down to a full stop below its lowest EV number-- there is no mark down where the "zero" would be, but if you watch the meter drop down below 1, and imagine where the zero would be, you can still get accurate readings down in that range. that's an extremely low light level that requires looooong exposures even when the lens is opened up, and when it's stopped down then you'd better plan on spending the night next to your camera (and forget bracketting)!
another important factor (which i think eliminates your easy formula) is whether there are any lights in the picture. that makes a HUGE difference to the exposure. if there are no lights, then you can meter normally for the illuminated subject, however dark that subject is, and expose for the midtones. but, if there are lights in the picture, and you want them not to be totally blown out, then you have to take the into account when metering. usually that means sacrificing a lot of shadow detail and exposing for the lights.
a few more thoughts: when shooting at night, lens flare can be a big problem-- a distant streetlamp, even if it's not in the picture, can throw flare in there. i usually stand next to my lens while the exposure is happening, shading the lens with my body. and for the worst problems there's always Photoshop...
last comment: before inserting the film holder, tap it firmly on your knee to make sure the film is seated in the rails, and then insert the holder, pull out the dark slide, and WAIT for several minutes before starting the exposure-- the film needs to get used to the air inside your camera, and it will frequently "pop" into a different shape depending on temperature and humidity conditions. if the pop happens while you are exposing, you'll get a wierd ghost image that will drive you nuts.
good luck, and write back with any additional questions, etc.
-- chris jordan (email@example.com), May 08, 2001.
one other comment on bracketing: because of reciprocity failure, your bracketing times will not be the same as normally. i.e., bracketing in half stops, usually you'd ad 50% to the time each time. but, with long exposures and increasing reciprocity failure, you need to add more and more each time. for example, using transparency film, if your meter says the exposure should be 1 minute, then shoot one at a minute (which will be underexposed because of reciprocity failure), then another at 1:40, which will probably be just about right on, and another at about 3 minutes, which will probably be only about 2/3 of a stop over. if your meter says the exposure should be 5 minutes, then shoot one at 5, one at 8, and one at about 12-14. interestingly, it doesn't matter whether you use 12 or 14 minutes-- taking reciprocity failure into account, those two extra would only make for about 1/8 of a stop difference. i've made many REALLY long exposures where i wasn't sure within an hour what it should be and it doesn't matter anyway because an extra hour only adds 1/6 of a stop more light!
-- chris jordan (firstname.lastname@example.org), May 08, 2001.
Bruce: I don't know which film you are using, but if it is Kodak, you can download a PDF publication from their website on reciprocity failure, see publication E-31, "Reciprocity and special filter data." (Kodak doesn't like to call it reciprocity failure since failure implies their stuff doesn't work, just my opinion) The document is filled with charts and graphs for their products. I have just started using my camera well after the sun goes down, TRI-X in very dim light, exposures up to 15 minutes or so. I find the charts very helpful and they produce useful negatives, especially if you back off development time as suggested in the publication. At least in black and white, this seems hard to screw up. You might find this publication useful and it's free. Other manufacturers presumably can give you similar information. This is an inexact science, at best, but the chemistry of it (physics of it?) give much leeway. If you expose a 13 minutes exposure starting 45 minutes after the sun goes down, using your spot meter to get you the initial placement that you use on the chart to give you the 13 minute exposure (this would be a metered 75 seconds, which works out to 13 minutes per this publication)consider how much dimmer the light will be at minute 12 than it was at minute one. The light intensity as dusk becomes night can fall off pretty fast, it is obviously not a constant, though the chart probably assumes it is. You can lose one stop in 5 minutes when you're deep into dusk. The publication is a starting point, and with experience I think (at least in black and white) you can get printable negatives using it. It is, by the way, very interesting to check the TMAX data; for TMAX 100, that 75 seconds as indicated by the meter works out to only 2 and 1/2 minutes, much less than TRI-X which is the faster film for shorter exposures. If it were up to me, I'd use a spotmeter and get it right versus lots of bracketing, though with roll film (I assume this is what you're using)cost may not be much of an issue.
-- Kevin Crisp (KRCrisp@aol.com), May 08, 2001.
Re: My earlier suggestion that you test on Polaroids: I don't agree that polaroids will be unhelpful in this situation. Whilst colour polaroids have very strange recipocity characteristics compared with film, (and very strange everything else too!) B&W Polapan 100 should give you a reliable guide, especially as your exposures are not very long. You do not indicate the type of film you are using, but I think that you should take on board the fact that modern films have very different recipocity characteristics from those of only a few years ago and that recipocity failure is now much less of a problem than it used to be. If you are shooting trannies at night then presumably you will be shooting on tungsten balanced film, which is designed for longer exposures than daylight film.
-- Garry Edwards (email@example.com), May 09, 2001.
I've used Ilford films for ages and they print a reciprocity graph in their film brochures. It's the same graph for each of the films - I don't know why as I'd have thought different speed films would have had differing reciprocity characteristics. It's possible to either make out a table of figures from this rather than having to carry round the graph (for the most common speeds you use) or even to do a bit of measuring and put the figures onto a spreadsheet so that you can calculate the equation of the curve and then work out whatever correction time you want. I have to agree with the other posts that, at longer times, it's not an exact science and plus or minus a reasonable amount doesn't make much of a difference.
I tend to measure the exposure I want, add on the reciprocity correction and then bracket from that figure - it means you get something on the negative although the times can get very long on the third bracket at +2 stops (no point bracketing below the time for b&w negs as you start losing shadow detail). e.g. measured time = 30 sec, add on correction to give 2 min 40 sec, expose at 2min 40sec, 5min 20sec, 10min 40sec. The problem with this is highlights in the picture will be way overexposed if you're trying to get in a lot of shadow detail and you have todevelop & print the negs very flat to kill the contrast.
-- Andrew Pell (firstname.lastname@example.org), May 13, 2001.