6. Long Exposures

The last stand of film has been the long-exposure domain, in which digital simply hasn't been competitive. The arrival of Nikon D2H and D2X made a serious dent in the film stronghold, and now D200 adds itself to the growing rank of long-exposure candidates.

Nikon D200 has the ability to create virtually noisefree, long-time exposures, and of very long duration (if you can provide battery support for, or hook the camera into an EH-6/6e AC/DC adapter). The camera tracks the exposure time even on a "B" setting with high accuracy, and it uses this information to run a second "blank" exposure internally to cancel out noise accrued within the first exposure. The second stage in case of the D200 does not last exactly as long as the first, on an average about half the time, and is only operative on exposures lasting more than approx. 8 seconds.

The D200, like the D2H, uses a 16-bit counter to count time steps of a "Bulb" exposure, and eventually this counter overflows. Since the noise-reduction circuitry probably depends on knowing the exact exposure time, this means the D200 would cease to do effective NR after 109 minutes 13.5 seconds (= 65,535 steps each of 1/10 sec duration). The EXIF data field is limited to 16 bits, so you likewise won't get recorded the true duration of any exposure beyond 6,553.5 seconds. Very unfortunately, this gives a poor performance for really long exposures, say 3-5 hours.

You might object to the idea of a camera spending up to half of a shooting session running its noise-reduction program. However, since unlike film there is no reciprocity error involved, the practical importance of the NR delay is less significant than you might think. This simply means that you will get twice the output for twice the exposure duration, something film cannot match at all for long-time exposures. In fact, reciprocity error will cut the effective speed of a 100-125 ISO film to a level below that of the D200 for any exposure longer than a few minutes. So, if you set a D200 and a film-based camera side by side and tripped the shutters simultaneously for a long exposure, you would finish first with D200, even with its NR delay.

I could not find night skies sufficiently dark for any exposure beyond a few minutes, so this time I ran the long exposure tests with body cap on the camera. Admittedly this is a less harsh test than exposing for dim subjects, but you have to do whatever is feasible for a given situation. I did this series of tests outdoors at ambient temperatures ranging between -12° C and -2° C, mainly during late evening or night. I feed the D200 from the AC mains to avoid depleting the battery pack while testing.

The perfect test result should be 100% perfectly black pixels, that is, with a zero value. This occurred for exposures up to nearly 20 minutes, a tremendous result indeed. However, because the data for the longer exposures got lost with my server crash, I cannot show the graph. I did recover the 4-hour exposure though, and it is shown below. The 4-hour exposure was run in the deep of the night, so conditions should be optimally stable for this test, but now noise increased significantly since the internal time counter overflowed (remember it's 16-bits only).

Black as Black Can Get, or The Proverbial Black Cat in a Coal Cellar? This is the output from Nikon's D200, set @ 100 "ISO" with long-term noise reduction on, after an exposure of 4 hours. The image looks perfect, but since the NR hasn't been effective throughout the entire exposure time (due to the overflow of the 16-bit counter, refer to the text), there is in fact some noise buried deep within the apparent darkness.

(Above) This is the same image from a 4-hour exposure, brutally enhanced to show the residual noise. (Below) Another long-term exposure, this is 73 minutes without NR on, ambient temperature -7° C Both examples indicate that the self-induced (largely thermal) noise affects all four sides of the CCD, quite unlike other Nikon DSLRs. We also see that NR is very effective if you have turned it on. My review camera hardly displayed any hot pixels at all, even for this exposure lasting more than 1 hour (without NR)

© Bjørn Rørslett/NN

Now, shooting black cats in coal cellars might not be everyone's idea of the perfect picture, so I went out in the middle of the night to get some more interesting long-term exposures. One of these is shown below,

Long exposures with street lights or similar, strong point light inside the frame, indicated that the D200 gracefully handled the overexposure bound to happen at these highlights. In fact, the inevitable clipping of the highlights seemed better and better controlled as exposure extended. I'll have to do more research in this field, but there seems to be a remarkable progress here compared to film (which will blow out highlights and the glare from them will diffuse into the surrounding shadows as well). See the example below to appreciate this,

However, the price to be paid for all this long-time wizardry is power, or rather battery, consumption. Lots of it in fact. Thus at least under a sub-zero Norwegian winter climate, 30- to 60-minute exposures seem to be an upper practical limit because anything longer will completely drain even a fully-charged battery pack. I have tried so should know. Do remember the need for conducting a noise-reduction stage afterwards. However, if you can provide stable mains power (with the EH-6 unit), up to 109 minutes will work just fine. For longer exposures than the 109 minute limit you'll need to engage the D2X instead of the D200, or go the proven silver-halide road. Film is still the better solution for doing extremely long exposures, but we are talking about a problem for only for a small minority of the users here. Virtually all exposures taken with any camera are shorter than 2-3 hours in duration.

Just for the fun of it, I investigated what would happen if you intentionally let the camera deplete the battery while doing a long exposure. I clamped down a MC-30 Remote Release with the camera set to "B" and just let it run. After some 46 minutes, the camera gave in and terminated the exposure, did not run any NR, and just wrote the image to the CF card. This "break" obviously enabled the battery to recalibrate, so the camera immediate commenced a new exposure, this time lasting approx. 15 minutes before it again recorded the file to the CF card. After that last attempt of getting a picture, the camera just was dead. So, Nikon evidently gives highest priority to the camera being able to transfer the image to the CF card, and this does make sense. Better to have an image without NR than no image at all.

Given the superb low-noise performance of the D200, I wonder whether this is just a simple dark-frame subtraction, or whether the D200 has more advanced and clever tricks up its sleeve. Since the NR part of the exposure is just about half as long as with the D2X, it's pretty obvious some new wizardry is going on here. A pity though that Nikon skimped on the internal counter to allow the return of the 16-bit limitation of the D2H.

Nikon D200 Reviewed