[Leica] M8 picture in magenta light

[davison_m at msn.com (MARK DAVISON)]

> >>Considering the variation of light conditions I'd be glad I had an
>exposure at all! A pox on the on going concerns of people about M8 and
>IR.<<
>
>...My concerns are begging to feel a little feverish! :-)
>

Certainly it is futile to worry too much about color rendition when shooting 
stage performances which are lit with colored gels.

However, it is very common to shoot color portraits of people sitting in 
windows, often with incandescent interior lights mixed with the window 
light.

>From our experience with film, we expect our digital cameras to be able to 
do a reasonable job of reproducing skin tones and fabric in this situation.

My experience with the M8 shows that if you want to do this reliably, and 
repeatedly, without having to resort to a lot of post-processing, you should 
always shoot with an IR blocking filter on the lens.

For an example, see: 
http://gallery.leica-users.org/v/MarkEDavison/M8+filter+study/

Note especially the difference in the color of the sweater with the Marine 
Corps logo in the areas which are being hit by incandescent, the change in 
the color of the seat upholstery, and the change in the flesh tones. Note 
also that there is very little change in the color of the Marine Corps logo, 
the hair, and the wood panelling.  IR contamination does not act like a 
uniform color shift--the color shift depends on the degree of IR in the 
illumination, and the degree to which the object reflects IR.  You can't 
take it out by changing white balance, and I have had mixed luck taking it 
out with custom color profiles.  The color profiles which take out the 
magenta in synthetic black do nothing to remove the shift in synthetic 
greens as in this example.

If you open the original files in Photoshop you can also see a noticeable 
difference in the Red channels--the Red channel in the portrait taken 
without a filter is visibly blurred, compared to the Green and Blue 
channels.

I have seen similar problems in almost all indoor color portraits I have 
taken with the M8--the example shown here is not an isolated or "rare" 
instance.  It's dim here in Seattle, so we often have the lights on inside.  
And it's cold and damp, so we wear our synthetic jackets.

In summary, the M8 IR contamination concerns are real if you are serious 
about shooting in color and you will notice it.  If you use your M8 to shoot 
black and white, you can pretty much ignore it, unless you intend to mix a 
lot of the Red channel into your final black and white image (in which case 
you must be aware of possible loss of resolution).

To simply declare a pox on those who point out the desirability of using IR 
blocking filters with the M8 is to give bad advice.  This is not a partisan 
issue--it's just physics.


Mark Davison