Saturday, February 9, 2013

Answer to the Kodachrome / Digital puzzle

Here's the answer to yesterday's puzzle challenge about "Digital vs. Kodachrome." Even though the votes were 126 to 74 in favor of the one on the right being Kodachrome, actually the one on the left was Kodachrome. (Though, technically, as many of you pointed out, they were both digital in the sense that they were both presented on your computer or mobile screen.)

Photographer Walter Wick processed the digital capture to try to match the color and the grain to the unprocessed drum scan of the Kodachrome.

Walter says: "Your commenters are making some wise comments, even if their guesses are wrong. When I made the "color noise" version, I purposely held back a bit on noise and softening of the digital image on the hunch that some friends I planned to show this to would harbor a bias in favor of the legendary film. This seems to be borne out in your poll, but of course, there could be other reasons the vote is going that way.

"I would have loved to have tested Kodachrome against full-frame digital on tripod-mounted cameras with the same lens, but alas I was not able to pull that together before the December 30, 2010 deadline - the last day Kodachrome was accepted for the final processing run at Dwane's Photo.

"What you're seeing here is something quite different: a full frame film camera with a pocket sized digital - shot casually, hand-held on a beach. What surprised me was the resolving power of the small-sensor Leica compared to the full frame Contax. In the digital version, the man is slightly enlarged relative to the frame compared to the that of the film version. But this advantage pales by comparison to the relatively huge image area of the full frame film camera. This does not bode well for Kodachrome comparing favorably with the resolving power of full frame DSLR. That, combined with the huge amount of shadow detail in Raw files and extreme high ISO sensitivity of digital cameras pretty much explains why most photographers abandoned Kodachrome before Kodak did (there are other reasons too, such as vast improvements of E-6 film stocks). However, if you must have that "Kodachrome look", well, you still can!

"The graphic above shows the relative size of the two cameras, and separately, the size of the film frame relative to the size of the digital sensor.

"The film was professionally drum scanned at the resolution shown, matched to the transparency, and not further altered. Also shown is the native resolution of the Raw file from the Leica/Panasonic. It was that file that was color adjusted and had noise and blur filters applied to imitate the Kodachrome (as some of the commenters have correctly surmised).
Thanks, Walter!


Anonymous said...

I voted for the one on the left being Kodachrome, looking at the votes stats on the assumption that most people would be wrong, and it turned on to be right.

Keith Parker said...

Wait, I guessed correct?

Mair said...

I simply LOVE being right!

Rich said...

I think that's OK, as long as it won't be Always;-)

Roberto said...

I still think the guy is a Zombie Actor. -RQ

Jean At Home said...

LOL. I thought the Kodachrome was the one on the right, even though I didn't like the overblown color. It seemed to have less noise. What can I say? I was wrong in my guess.

a chris said...

I wonder how the comparison would be in lower light. But I guess a comparison would be less meaningful as soon as light budgeting becomes an issue, especially with the film and sensor being different sizes. I find my cameras (D90 and Panasonic LX3 - much the same hardware as the D-LUX4, I understand) really aren't up to something like indoor shots of moving people in ambient lighting, which is one of the things I think the majority of non-professional (unprofessional? heh) photographers would really find useful for most shots. Still, there's much to love about the digital technology.

Tom Hart said...

I guessed that the one on the right was digital, based on the higher definition of some of the beads. That was pretty obvious when comparing the side-by-side images enlarged. I was surprised that the vote was so lopsided in the other direction. Not that I would stake my life on my logic being correct, but it seemed like a good objective difference on which to base a guess - right or wrong.