View Single Post

Old 03-25-2019   #72
markjwyatt
Registered User
 
markjwyatt's Avatar
 
markjwyatt is offline
Join Date: Nov 2018
Location: Southern California
Posts: 618
Quote:
Originally Posted by willie_901 View Post
This is a fundamental, important point.

Internet image viewing is a nightmare. The effective MTF50 is significantly degraded by compression requirements. But this is trivial compared to the unavoidable, unknowable variations in viewing circumstances.

Eliminating compression effects and optimized screen viewing cannot eliminate the core problem.

There are inherent flaws in all mathematical models that transform continuous phenomena into a binary framework. The impacts of this issue can be negligible. But they are not zero. Often digital images are unintentionally undermined by improper technique.[1] Still, anytime we digitize film media we have abandoned the analog domain. The result is not a copy. It is a only a model for the original. The compromises are usually acceptable compared to the advantages of digital image rendering, storage and distribution.

Avoiding analog media altogether does not eliminate the fundamental problems of modeling human visual perception using binary methods. It use to bother me it's possible to take a digital image with a very high system MTF50 and simulate film image properties such as grain, development induced contrast effects, base-plus fogging and handling blemishes. Simulation may be derivative and therefore inferior, but it is not pointless.

Digital printing is another step where the impossible is attempted. Binary information must be transformed to represent continuous information. Fortunately, viewing distance is a powerful equalizer. Advances in printer technologies, media and paper formulations are significant. But it is not a simple, convenient process to make an excellent digital print.

These issues are exacerbated by the relatively low IQ of 35 mm film. If large negative photography became the standard after WW-II, preferences for analog photography could be different because visual perception of large-negative images are more similar to contemporary digital images.


1. Here's a study that "...demonstrate[s] methodologies, preferably analytical ones, for establishing and verifying digital imaging requirements for scanning resolution of photographic negatives and slides."
translation: it ain't easy, but it kind of works.
__________________
Flickr: https://www.flickr.com/photos/markjwyatt/
  Reply With Quote