Olympus-OM
[Top] [All Lists]

Re: [OM] ***SPAM*** Re: Digital BW filter tests

Subject: Re: [OM] ***SPAM*** Re: Digital BW filter tests
From: "philippe.amard" <philippe.amard@xxxxxx>
Date: Mon, 2 Mar 2015 21:13:09 +0100
My L1 is curently on its way to Taiwan where an ebayer wanted along with the mighty lens but I still don't understand why you folks insist on using in camera conversion when we all know that a good raw processor and a little attention and love of light would fare much further, with or without filters ...

OK, I'm out ;-)

Philippe

Le 2 mars 15 à 17:49, Ken Norton a écrit :

Monobrow Bob wrote:
Interesting. I tend to prefer the monochrome conversions from RAW files.
The in-camera stuff just doesn’t quite get there. For me. <g>

Where I slightly beg to differ would be in regards to the DMC-L1 and the E-3. I've always liked the in-camera E-3 B&Ws and the DMC-L1 was certainly influenced by both Olympus and Leica in the programming of the algorithm. There is just something "right" about those files--when it comes to getting
people pictures.

The thing that seems to make the biggest difference that I've found is
using a lens-filter to match the general color sensitivity of B&W film. Most B&W films tend to be more blue-sensitive that color films AND digital sensors. In otherwords, B&W films are more like tungsten color films in
that they are more blue sensitive and therefore look better in indoor
artificial lighting.

You see that with Tri-X film, which looks great indoors with no filters, but the moment you step outside, you want to put a yellow filter on the lens. This is to counteract that unnatural blue-sensitivity of the film.

If you change a digital camera's white-balance to tungsten you'll come a lot closer to how B&W films see the color spectrum. Should you do that with
a filter or with a conversion setting in Lightroom? Good question. It
depends a lot on the sensor and how the three sets of color sensels of the sensor respond to light and their spectral response and noise patterns.

Take the blue sensels as an example. They are usually the most noisy of the lot. You would think that adjusting white-balance in your raw converter diminishes the blue-sensel "volume" in the mix. It actually does not. The raw converter is typically doing a 3-pixel merge. Not four, but three--the two greens for every one blue and red have to do with how the matrix is merged, not any particular advantage to having twice the number of green sensels--except Olympus and Fuji will sometimes use two greens of different responses, which forces a 4-pixel merge which is not nearly as inherently
sharp or well behaved as a 3-pixel merge.

(some converters actually do let you deep-dive into altering the ratio of how the pixels are merged, but the Olympus and Adobe converters do not).

So, we have a situation where the blue sensels are overly noisy for a given picture. The noisy blue sensels are merged in with the green and red and then the white balance is applied. Unfortunately, the noise addition of the
blue sensels has already been added to the merged calculation and the
derived pixel value has this noise built-in.

What if, you can balance the blue, red and green pixels at time of
shooting? Instead of the scene being overly yellow cast (interior
lighting), you can put a tungston correction filter on the lens and adjust your exposure to compensate. Ten years ago, the sensors were still noisy enough that the increased exposure usually offset the improvement you got off of the blue sensels, but with this method of color correcting BEFORE
the light hits the sensor, you are essentially mapping the noise floor
evenly across all three color sensels. The primary benefit isn't so much with noise improvement, but with far better behavior in Zones I-II, IIV-IV. You don't have one color clipping before the rest and you can actually do
extensive shadow and highlight recovery without strange color casts or
out-of-gamut conditions.

So... To a picture destined for B&W conversion, either in-camera or in
Lightroom, if you put a cyan or blue filter on the lens you'll better match "Tri-X" in how it sees. If it's for outdoor shooting, you can usually just forgo the filter entirely as a bare lens will give you a basic Tri-X with yellow look. But indoor picturs will be way off and the look is just wrong. A tungsten correction filter or a light cyan or blue filter will generally
be all that is required to alter the spectral response curves enough
(without screwing up metering too much) to give more "proper" digital B&Ws.

The DMC-L1 and E-3 in-camera B&W converter is actually doing this to some
extent which is why they do better than the average bear at in-camera
indoor B&W pictures, but kinda stink up the show with outdoor shots.

So, for indoor color digital photography, if you do some lens-filter
color-correction, you absolutely will end up with a greater usable dynamic
range and better uniformity in color response across the brightness
spectrum.

Sorry for the mixed topics of color and B&W, but you really can't talk
about one without talking about the other.

--
Ken Norton
ken@xxxxxxxxxxx
http://www.zone-10.com
--
_________________________________________________________________
Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
Archives: http://lists.thomasclausen.net/mailman/private/olympus/
Themed Olympus Photo Exhibition: http://www.tope.nl/


One sees clearly only with the heart. What is essential is invisible to the eye. Antoine de Saint Exupéry in Le Petit Prince.
NO ARCHIVE




--
_________________________________________________________________
Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
Archives: http://lists.thomasclausen.net/mailman/private/olympus/
Themed Olympus Photo Exhibition: http://www.tope.nl/

<Prev in Thread] Current Thread [Next in Thread>
Sponsored by Tako
Impressum | Datenschutz