🚀 go-pugleaf

Rags,

You set up a straw man which you then proceed to demolish, but the Tones and Zones piece contains numerous factual errors.

In digital imaging, there is no *fixed* relationship between the integer values and the photon count (and hence tonal value) they represent. The relationship depends on the gamma curve that's applied�that's what a gamma curve does. In the case of digital raw capture, it's linear gamma.

You're mistaken as to Photoshop's 16-bit implementation. The possible values are 0-32,768 (for a total of 32,769 possible values).

In your zone experiment, you miss the fact that you're looking at the (correctly) converted raw image, not the raw image itself (those are hard to look at) to get a midtone 127.

In your gamma discussion, you overlook the fact that brightness is a nonlinear human response to photons: you also confuse display gamma, CRT gamma (the former is the combined transfer function of monitor + video card), capture device gamma and image encoding gamma. Other than the fact that they all contain the word "gamma" they're separate things. "Light" doesn't have a gamma. The things we use to detect and quantify light do, and the root of this whole issue is that eyeballs have a very different gamma to silicon.

When you shoot JPEG, gamma correction is indeed applied in the capture device, as you say. However, when you shoot raw, no gamma correction is applied until you process the raw through a raw converter. That's one of the reasons you can do so much more with a raw than with a JPEG.

I'm glad that you've realized the importance of exposing for the highlights when you shoot digital, but you've confused a lot of things, and most of the things you refer to as properties of light are in fact properties of the eye, including the one that leads you to the extraordinary conclusion that light has a gamma of 2.

Encoding gamma IS the property that relates digital counts to tonal value. Your statement that "there's no such thing as digital gamma" is wildly incorrect, and distinctly unhelpful to anyone trying to make sense of this.

It's not a religious concern for me, but it's distressing to see so much misinformation being bandied about on what is fundamentally a fairly simple concept: silicon counts photons linearly while eyeballs do not, resulting in a discrepancy between the way each renders tonal values, necessitating correction to silicon capture for viewing by eyeballs.