The "When" of Processing

Back when I had to walk fifty miles to school through drifts of snow, raw processing was simple: Hand off the raw data file to the converter and wait for it to finish, often overnight.

Okay, none of the statements in the opening paragraph are true, but the basic premise behind it is a myth that has sustained for a long time: (1) take raw image; (2) post process raw image. Plus: #2 is handled by a single product. 

It's that #2 that's wrong, and has been wrong in some sense for decades. 

Early in the DSLR era, Nikon provided a plug-in that was essentially the black box of the camera's image processor (that created JPEGs and TIFFs) that you could use with Adobe Photoshop. This was one of the first of the "pre-process" conversions that have been possible; many more have followed. 

These days, there are three basic possibilities you have to consider when processing raw files:

  1. Preprocess. Adobe's AI noise reduction, DxO's PureRaw, and Topazlabs Photo AI all come to mind here. These add a step to your work flow by creating a DNG file from your raw file, purportedly to give you a benefit you wouldn't get otherwise. Here, you change the raw data.
  2. Convert. This is the center of the raw universe, where a demosaic routine (and more) is used to generate RGB pixels to deal with. ACR/Lightroom, CaptureOne, DxO PhotoLab, and many more all live in this space, including some free programs from camera makers that are, well, what you get when you pay nothing. 
  3. Filter/Plug-in/Effect. Once you have RGB pixels, you still may want to do something with them to "correct" them. Many of the early pioneers of digital photography software lived in this space, creating tools that, in the beginning, mostly plugged into Adobe Photoshop. These days, they may plug into pretty much anything.

You might have noticed something: both DxO and Topazlabs have had significant efforts for #3, but now have products that do similar things in #1. For example, DxO Nik Define does noise reduction on RGB pixels, but now DxO PureRaw can do it on raw data. Ditto Topazlabs DeNoise AI versus Photo AI. Moreover, both of these #1 products are growing with the number of things they want to fiddle with. Photo AI 2.0, for instance, now deals with noise, sharpening, exposure, and color (the latter two are "beta" in the shipping product). 

One of the things I'm now trying to figure out is what is really "optimal" in my photography when it comes to processing. Workflow is one of those things that keeps changing. I hope to have an article later this year that steps through that with a modern viewpoint, from field to output. Stay tuned.

Looking for gear-specific information? Check out our other Web sites:
DSLRS: | mirrorless: | Z System: | film SLR: all text and original images © 2023 Thom Hogan
portions Copyright 1999-2022 Thom Hogan
All Rights Reserved — the contents of this site, including but not limited to its text, illustrations, and concepts,
may not be utilized, directly or indirectly, to inform, train, or improve any artificial intelligence program or system.