Color-Contrast-Exposure Equation

[Ron Kruger]

What I first noticed years ago with film seems to be consistent with digital: that color saturation and contrast sort of relate to each other, and that exposure affects both.
In other words, for most shots under adequate light, I got better color saturation and contrast (or the appearance of it) by underexposing -1/3 stop or so. It gave shots more pop.
With digital, however, instead of exposure compensation, I have bumped the saturation and contrast one notch in my camera and seldom need to underexpose a shot to get this effect.
I do this because I shoot for publication and have learned that the steps of the printing process rob images slightly of color and contrast.
This practice has come through years of trial and error, but I’ve never really given the relationship of color-contrast-exposure any serious thought as to what is going on (on a physics level), and was wondering if some here might like to discuss it?

[another view]

I'm in agreement, but really don't know anything about the technical aspect of it.

The main point I noticed in your post was that it seems that you're giving your client what they want, whether or not they're asking for it specifically (theory vs. reality).

KR was much better at ISO80 to my eyes, and this was underexposing 1/3 stop from 64. Velvia 50 always looked like it had this effect built-in, especially if you ever tried it at 40.

With digital, I prefer contrast to be a little softer than I need at time of capture which keeps the options open later in PP.

[Ron Kruger]

Which options do you mean, View?
Even though I'm as happy with the digital results I'm getting as I ever was with Kodachrome 64, I still miss the richness of Kodachrome 25.
I told one of my editors (who probably uses more of my images than anyone) that I was bumping the saturation and contrast slightly to compensate for printing, and asked him if that still applied the way it did when I was editor of a quality, four-color magazine, and he just said: "Whatever you're doing, don't change anything." (Theory vs. reality)

[another view]

Which options do you mean, View?

It's really easy (well, usually) to add contrast, but very difficult to reduce it and still get a good looking final image. Same is true for saturation, but to less of that extent. This is all based on the experiences of someone who knows the basics and is far from an expert...

But I don't want to be an expert in PP. The way I approach it, a certain amount is required but I don't want to spend a lot of time with it. As always, YMMV. I see a lot of shots I really like that are heavily stylized, but I know that's not me, that's all. I don't think you're going for that either, so you may want to experiment with lower contrast and possibly saturation at time of capture.

[Ron Kruger]

Hmmmmm. My approach is to put as much emphisis upon getting what I want when the shutter button is pressed and to do as little as necessary in PP.
All the editors I've talked with suggest that, and each has cited cases of good shots (composition wise) sent to them that have been ruined by PP "creativity." They say it is best to leave that to the real pros on their staffs.
Your suggestion of expermenting with lower contrast and possibly saturation at time of capture is one I'm going to take seriously.
But that brings up another queston: Is the saturation and/or contrast adjusted during PP of the same quality as that at time of capture? I suppose there is more than one way to skin a cat, but which leave you with a cleaner pelt?

[another view]

Is the saturation and/or contrast adjusted during PP of the same quality as that at time of capture? I suppose there is more than one way to skin a cat, but which leave you with a cleaner pelt?

Not exactly (underexposed shots that have been lightened look different in the highlights) but that's a good way to put it. I try to get it "pretty close" at time of capture but if it were perfect that would almost be a matter of luck. By perfect, I mean simply that I don't think the shot needs anything.

I think the over-done PP thing is a result of two things - an uncalibrated monitor and the use of a new toy (the software). Just because grass can be made neon green doesn't make it better, and if the monitor is uncalibrated how do we know what the image actually looks like anyway? This is why I say the first step to PP has to be evaluating the shot to see what it needs. If nothing, great - on to the next.

You've already got a great path to your final image so don't stop doing it. I would recommend playing around with different ideas just to see what they do and if they could help you in a situation, such as:

--Overexposing (but not to the point of blowing highlights) and darkening in PP; a.k.a. 'Expose to the Right'.
--Underexpose and lighten in PP (will increase noise but might make it possible to get a low light shot where you normally couldn't since you might gain a couple of stops).
--Try shooting with low contrast and increasing in PP
--Try lowering the contrast in a shot in PP and see what you think of the results.

[freygr]

Overexposing - Underexposing ---- contrast and saturation -- this is where shooting in RAW will give you the edge. The 4 extra bits per color channel gives you a few more stops of range to play with with out losing quality.

[SmartWombat]

Underexposure will add noise in the shadows.
I would stick to increasing colour saturation in software - it's really easy in Lightroom, just slide the Clarity slider to the right a bit.

[Sushigaijin]

Exposure is relative. The metering mode and camera firmware decide what "0" is, there is no reason that the camera knows better than we do. I usually expose for the histogram; I find a full histogram is a rich, saturated and contrasty photo, whether it is "0" or +0.3, -0.3. The number is irrelevant. If your camera looks better at -0.3, so be it!

And for printing, I've found that a slight over processing is desirable also. I bump contrast, saturation and even a bit of brightness for print - so often that I do it to every frame I keep now. And, I edit on a calibrated monitor so it isn't just a calibration problem. I think the issue is that paper has to absorb some of the ink (contrast) just to fix the image...a lossy reproduction, if you will.

[Ron Kruger]

I'm in full agreement, Erik. Auto doesn't always know best.
In fact, spot metering with my 200mm is about the only auto metering I trust, and then only when I'm shooting larger animals, such as deer, which are generally darker than their surroundings. In these cases, it's similar to walking up to the subject and putting a hand-held meter on it.
I believe auto metering systems, because they are an averaging system (with the possible exception of spot metering), tend to err on the bright side.
A "loosy reproduction" is a great way to put it. This is a fact, even on photo paper, and it becomes even more true as paper quality decreases through magazines, tabloids and newspapers.

[Medley]

I think you're very close Erik. I attribute the difference to different lighting methods. The computer illuminates the image from behind it (forgive me for forgetting the technical term), while a print uses reflected ambient lighting.

In terms of contrast, some softening occurs in converting the image from square pixels to round dots of ink. The type of paper and ink can also greatly influence the "spread" of the ink, causing more softening. The latter can be accounted for by using a calibrated monitor in conjunction with accurate color and printer profiles, and the former can be compensated for using sharpening techniques.Digital sharpening is really nothing more than increasing contrast in very specific areas of the image.

As to your question of wether contrast boosted in pp is of an equal quality to in-camera adjustments, it certainly can. In fact, it can be better. I get better sharpness out of my prints than my camera is capable of producing. The main problem with both saturation and contrast adjustments is that in most software (Photoshop included) the adjustments are counterintuitive. They produce the greatest boost in the areas that need it least, and vice versa. It's possible to compensate for this, but it requires more than a passing knowledge of the software.

I would say that if your editor is happy with your images, then leave these things to the techies. They'll appreciate the job security anyway.

- Joe U.

[Ron Kruger]

I find something Medley said very interesting: "Digital sharpening is really nothing more than increasing contrast in very specific areas of the image."
I shy away from sharpening images, because I've suffered some bad results, but I often hit "auto contrast" in PS and have noticed that while it boosts the contrast, the image also gets sharper, while the color saturation gets deeper at the same time.
That's the main reason I started this thread in the first place.

[freygr]

I find something Medley said very interesting: "Digital sharpening is really nothing more than increasing contrast in very specific areas of the image."
I shy away from sharpening images, because I've suffered some bad results, but I often hit "auto contrast" in PS and have noticed that while it boosts the contrast, the image also gets sharper, while the color saturation gets deeper at the same time.
That's the main reason I started this thread in the first place.

Unluckily your eyes are playing tricks on you. A sharp image will be highly defined but increasing the contrast will unmask details hidden by low contrast. The sharpening program does a lot more data manipulation than adjusting the curves.

[SmartWombat]

I edit on a calibrated monitor so it isn't just a calibration problem.

If you want the on screen display to accurately represent what will be printed you also need to be able to tell the program how the output treats the colour.
So that means profiling your printer, or using a standard profile for the lab.
Part of the reason I use Costco is that their lab profiles are on the web (let's see Snappy Snaps or Boots do that!) and I can have a better idea of the output.

[Medley]

Agreed Paul. The other reason I use Costco's lab is that I work there (though not in Photo), and I'm there 5 days a week. Couldn't possibly be more convenient.

Yes, Ron you are correct. The difference between sharpening and auto contrast is that sharpening applies the contrast only to the edges, and auto contrast affects the image globally.

freygr, I agree with you to a point. But fortunately, having your eyes play tricks on you is part of what digital sharpening is all about. Technically, sharpness is defined as the amount of detail at the high-contrast edges of an image. Since there is no way to add detail to an image, there is no way to really "sharpen" an image.

But carefully increasing the contrast at the edges of an image will fool the eye into believing that there is more detail than is actually there, providing the illusion of sharpening. This is true even in areas where increasing the contrast provides no additional detail information. In reality, the image doesn't really get sharper, but the eye perceives it as being sharper. Any low contrast area where more detail is actually brought out only serve to further the cause.

- Joe U.

[dbutler]

And if you are shooting in raw, Paul is right. Underexposing will result in noise. And, digitally, you can capture more cleanly when you move more toward overexposure than underexposure. You will recover a whole lot more detail, barring clipping. It is why I use my histogram view over just my LCD image, and keep it to the right.