Nikon D2H Reviewed. Long Exposures

The last stand of film has been the long-exposure domain, in which digital simply hasn't been competitive. So far, that is.

Nikon D2H has the ability to create virtually noisefree, long-time exposures, and of as long duration as you care to have (or can provide battery support for), but do observe the caveat described later. The camera tracks the exposure time even on a "B" setting with high accuracy, and uses this information to run a second "blank" exposure internally to cancel out noise accrued within the first exposure. The second stage should last exactly as long as the first, actual image-taking stage, to the 1/10 second in fact, and the duration is duly logged in the EXIF header.

You might object to the idea of a camera spending half of the time in a shooting session running its noise reduction program. However, considering D2H should be set to 200 "ISO" to get best quality and you traditionally would use a 100 ISO film for the identical shot, you really end up using equal time to complete the job anyway. Another, not intuitively obvious but nevertheless increasingly important issue in favour of long exposures in the digital domain, is that the LBCAST chip has no reciprocity error unlike film. This simply means that you will get twice the output for twice the exposure duration, something film cannot match at all at long-time exposures.


Night in the Woods Nikon D2H,@ 200 "ISO" with noise reduction on, AFS 300/2.8 ED-IF Nikkor lens, 20 minutes exposure at f/5.6	© Bjørn Rørslett/NN

The incident light levels for the wood scene above was far below the metering range of my Lunasix, which reads down to -6 EV and has no problem on its own to give sensible readings for a moonlit landscape. This equals saying it was really dark and in fact I didn't see a thing in the viewfinder, hence the rather casual composition of the image above. Neither did autofocus work, of course, until I coaxed AF to spring into action by using the AF assist beam of my SB28 DX flash. That stage completed, I simply went indoors and enjoyed a hot cup of coffee, and let the camera and the noise reduction circuits do their magic. After about 40 minutes of enjoyable home entertainment, I returned to find a well-exposed, virtually noiseless night shot safely tucked away on the CF-card. Brave new world.

Crop at 100% magnification

Some hot pixels can occur within the deep shadows, but please note how discreet their appearance is, thanks to the noise reduction algorithm employed by D2H. Had I added another round of coffee to my own schedule and let the camera run for another 10 minutes or so, these pixels might have been enlighted sufficiently to become ordinary ones. Nikon D2H tends to act like film in this respect, by accumulating light intensity onto its LBCAST chip in a similar way film does build up its latent image. At least this is my current appreciation of the matter.

Long exposures with street lights or similar strong point light sources inside the frame, indicated that D2H 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).

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 minutes 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), only your patience sets the limit. Or, so I thought, but further testing proved otherwise.

I spent several cold wintery nights trying to learn more about the inherent noise of the LBCAST sensor, and did this by running doubled exposures commencing from 1 sec. Each exposure setting was duplicated to give one image with and one without noise reduction switched on. I initially ran these tests by hooking the D2H up to my trustworthy MT-2 intervalometer. Since the exposure duration available with the MT-2 maxes out at 90 minutes, the longer exposures were run with the camera turned on manually and not through the MT-2. Up to 20 minutes of exposure (ambient temperature -12 °C) resulted in virtually noiseless images even with the NR function switched off. Longer exposure durations, from 30 minutes, started to show 3 hot spot areas in the lower part of the frame, and by 60 minutes (without NR) these spots were very prominent. Even so, NR handled the situation satisfactorily and largely covered up for the deficiency, although image dynamic range tended to suffer in the impacted areas of the frame.

However, my 180 minute exposure with NR on was very noisy, quite unlike the 90 minute result. Further testing documented that the NR function of the D2H ceases to function after 109 minutes 13.5 secs. Why is this? Simply because D2H has a 16-bit CPU, and the register tallying the actual exposure duration, counting 100 millisec units, eventually overflows after counting 65,535 units. Thus, the second stage of noise reduction will stop after 6553.5 seconds no matter how long the first stage lasts. You can't always win, when the bits are stacked in your disfavour. You can of course make a manual dark-frame exposure if you wish, but the neatness of having the camera do all the wizardry on its own is lost. Given the superb low-noise performance of D2H, I wonder whether this is just a simple dark-frame subtraction, or whether D2H has more advanced and clever tricks up its sleeve.