[Leica] First IR photos with M8

[Howard Ritter]

So the filter came from B&H today, a B+W 093 filter with a lower cutoff at 830 nm. Wow, it’s really black! And I got the problem with modern SD cards sorted and now the M8 is humming along with a 32GB card.

Late autumn does not present the best landscapes for IR, with a paucity of green foliage, but here in NC there are still enough leaves on the trees to allow for testing the gear. I stepped out of the house and found enough to make for a good picture, with blue skies and some picturesque clouds.

Here are three: http://gallery.leica-users.org/v/hlritter/For+Gallery/ <http://gallery.leica-users.org/v/hlritter/For+Gallery/>

All were taken at ISO 640, hand held, with an Elmarit-M 24/2.8 lens wide open and set with the infinity symbol at the f/5.6 mark. Exposures were in the 1/20 range or longer. The files were stretched so the histograms filled the entire range rather than the original ~60%, improving the brightness and contrast. All were converted to greyscale.

At full screen, all three seem to lack the expected sharpness. I can think of three possible reasons, the obvious ones being uncertainty about the best focus point (I tried several focus marks, but I’ll have to experiment further) and hand holding with a slow shutter. A third reason occurs to me: The lens is achromatic across the visible part of the spectrum, but obviously not for the part that extends into the IR, or there would be no focus shift to compensate for. Can we assume that it IS achromatic across the full range of IR wavelengths that the sensor is sensitive to? I don’t think so. If it isn’t, and there’s enough energy in parts of the IR spectrum that are far enough away from the notional focusing wavelength, then there will be progressive blurring of images formed by wavelengths further and further from the wavelength that is exactly focused.

In fact, this latter effect MUST be present with IR filters whose passbands extend into the visible deep red: When the lens is focused on the mark, the achromatic property of the lens puts the visible red in focus, but not the dominant IR, or there’d be no need for focus compensation. So there must be some degree of blurring with such filters; why isn’t it an issue, visually speaking? That would be an argument for using the smallest practicable aperture, while I was using the lens wide open for shortest shutter speed. Also an argument for using “tighter” filters, with longer cutoff wavelengths, to limit the range of wavelengths. Which however call for longer exposures or faster shutters. But with “looser" filters, the additional light that enables shorter exposures is less in focus… Oy, my head hurts.

Tomorrow, experimenting with a range of apertures. On a tripod.

Any thoughts?

—howard