Raja

Ctein's article might be apropos here:

The Online Photographer: 'Expose to the Right' is a Bunch of Bull

Ignore the title, and most of the article, the bit that might be relevant if the management of highlights around Figure 4 of the piece. Since the background of this picture is by design a blown-out highlight, managing the transition from cat to background is essentially this problem, and is the basic failure mode in the land of digital.

I say "might be" deliberately, it might be apropros. It might not be. Ctein's always worth a read on the technical front, though, so I drag this one out pretty often.
 
Nice hi-key on the first one. Here's a caption for the vintage one "Why do you want to do this to me?" joking.

Both are nicely done.

Thanks Ron!
 
Ctein's article might be apropos here:

The Online Photographer: 'Expose to the Right' is a Bunch of Bull

Ignore the title, and most of the article, the bit that might be relevant if the management of highlights around Figure 4 of the piece. Since the background of this picture is by design a blown-out highlight, managing the transition from cat to background is essentially this problem, and is the basic failure mode in the land of digital.

I say "might be" deliberately, it might be apropros. It might not be. Ctein's always worth a read on the technical front, though, so I drag this one out pretty often.

I did underexpose the original shot about .5 stop. It was all ambient, and fairly dark. The door behind the cat is white, but was in shadow.

Like almost everything else in photography, You have to know when to use different techniques... that is what gets people in trouble.. not knowing when to use what! lol!

ETTR is one of those things that works when it is appropriate... just like underexposure works some of the time, but not all the time.
 
Ctein's discussion of ETTR itself isn't very interesting, it's his discussion of RAW conversion with a custom curve to manage transitions from the highest tones to the blown out parts smoothly that is, I feel, the interesting part of that piece. I actually find the use of curves in conversion to be a pain in the butt and use some different techniques, but they work out roughly the same in the higher tonalities.
 
Ctein's discussion of ETTR itself isn't very interesting

Nor is apparently aware of spot metering :/

The problem with ETTR is that it is not standardized, it always has these vague "slightly over expose" definitions. If you take the meter reading and bias everything 1/3 stop, then yeah, you can expect hilights to be difficult to expose because you have no idea if the exposure ceiling are within 1/3 stop of meter. Likewise, if this is your approach, you'll have no idea if 1/3 stop is enough ETTR to really make a difference.

ETTR is probably not the best approach, but rather what I do is "expose for the hilights, process for the shadows" (EHPS, perhaps?), which makes sense within photographic lexicon and accomplishes the same desired result as ETTR.

I actually find the use of curves in conversion to be a pain in the butt and use some different techniques, but they work out roughly the same in the higher tonalities.

I am not entirely certain where in the pipeline curve adjustments take place in RAW processing. While, IIRC, some GPL processors do explicitly permit curve adjustment before *and* after the data is converted, I get the impression that typically curve adjustments are applied AFTER conversion, not before. If this is the case (and I am not saying it is for certain), then applying a curve in the RAW processor does not have any advantage than to do it within your primary editor, provided that you're using a two-step workflow.
 
Last edited:
Ctein's discussion of ETTR itself isn't very interesting, it's his discussion of RAW conversion with a custom curve to manage transitions from the highest tones to the blown out parts smoothly that is, I feel, the interesting part of that piece. I actually find the use of curves in conversion to be a pain in the butt and use some different techniques, but they work out roughly the same in the higher tonalities.

Detail retention in highlights is always an issue, and transitions can be difficult to manage. I will read the article when I have more time (work gets in the way, sometimes). Thanks...
 
first one reminds me of like a pencil drawing by a talented artist. i like the edit.

Thanks... that was sort of the look I was going for... Going to print it and frame for a girlfriend gift! lol!
 
That look awesome. I cant say anything about it from a photography sense as Its better thany anything Ive tried like this.
 
Ctein's discussion of ETTR itself isn't very interesting, it's his discussion of RAW conversion with a custom curve to manage transitions from the highest tones to the blown out parts smoothly that is, I feel, the interesting part of that piece. I actually find the use of curves in conversion to be a pain in the butt and use some different techniques, but they work out roughly the same in the higher tonalities.

This is worth looking at Gorman-Holbert Conversion Method
 
I am not entirely certain where in the pipeline curve adjustments take place in RAW processing. While, IIRC, some GPL processors do explicitly permit curve adjustment before *and* after the data is converted, I get the impression that typically curve adjustments are applied AFTER conversion, not before. If this is the case (and I am not saying it is for certain), then applying a curve in the RAW processor does not have any advantage than to do it within your primary editor, provided that you're using a two-step workflow.

Both of the raw processors I use (DPP & ACR) have a curves adjustment tab, although I don't think it's called that. I prefer to do the drastic overall curves adjustments on the 14-bit raw file, export an 8-bit tiff and then only do minor or selective adjustments in PS. I know a lot of people work with 16-bit files in PS and do the heavy "bending" there, but I just don't like handling 16-bit files and find that this method works for me - no banding or other artifacts.
 
If I have a highlight issue, for me the simplest is to simply export two different 8 bit TIFFs and then blend to taste in the highlights. It's a light touch of HDR without the tonemapping, I guess.

This largely works best for me because of the particular mix of software I use.

Anyways. Ctein's approach of using a curve to ease that transition to the blown out highlights is a good one, but not the only one. I mentioned it here because I felt that Charlie's picture would be improved with a little love on exactly that transition in a couple of spots, but more importantly because I think it's a good issue to keep in front of people in this digital age.
 

Most reactions

Back
Top