White point Black point...pointless?

[Abitconfused]

Having a white point and black point, I have read, has been preached for some time. While there is no law, I still believe this is sound advice generally.

This image has an almost white point and an almost black point.



We can find the whitest point using Threshold.



And mark it using Color Sampler Tool.



Then, using Curves white eyedropper, convert it to white.



Finding the blackest spot is easy too.



Experimentation is in order but, in this photo, a small shift toward black adds contrast without sacrificing critical detail.



Of course, add some sharpening.



Crop and you are done.

[george013]

Having a white point and black point, I have read, has been preached for some time. While there is no law, I still believe this is sound advice generally.

This image has an almost white point and an almost black point.



We can find the whitest point using Threshold.



And mark it using Color Sampler Tool.



Then, using Curves white eyedropper, convert it to white.



Finding the blackest spot is easy too.



Experimentation is in order but, in this photo, a small shift toward black adds contrast without sacrificing critical detail.



Of course, add some sharpening.



Crop and you are done.

Not sure, but the issue of the white and black point is one of exposure and dynamic range. What happens when you use the eye dropper? Aren't you changing the WB?

George

[JohnRostron]

I alway go for the simplest approach. How does yours differ differ from slecting a Levels adjustment and moving the black and white points there? Or even, horror of horrors, using auto-levels?

John

[Manfred M]

Ed - I usually use either curves or the levels adjustment to set the black and white point.

The reason I do so is that I can see the histogram and make my choice as to where to do the adjustment based on the amount of data that is to the left of where I set the black point or to the right of where I set the white point. The eye dropper and the auto levels don't give me that level of control. As George points out, the risk with the eyedropper is that one can introduce a colour shift.

[Abitconfused]

I alway go for the simplest approach. How does yours differ differ from slecting a Levels adjustment and moving the black and white points there? Or even, horror of horrors, using auto-levels?

John

I think you are correct John. Although with a number of layers, editing any one step may be more flexible. I think my main gripe is that black and white point are not considered axiomatic to best practices anymore.

[Manfred M]

I think my main gripe is that black and white point are not considered axiomatic to best practices anymore.

It depends on who you ask, Ed. My college profs certainly thought so and would take this into account when marking assignments. I would suggest that there are some exceptions here, but they are limited.

If you refer to the masses of self-trained photographers out there, then I would suggest you might be onto something.

[Donald]

You state that your main gripe is that black and white point are not considered axiomatic to best practices anymore. And Manfred argues that it depends on who you ask, suggesting that the masses of self-trained photographers are responsible for not adhering to having to have a black and white point.

It is certainly something that is well taught. But it can be well taught so that those who deviate from the mantra know when they're doing it and and why they're doing it. I don't think that anyone would describe as Paul Strand as being amongst the masses of self-trained photographers.

[Abitconfused]

Agree. I have believed that the color shift caused by using the Curves eyedroppers did more good in correcting color than bad in causing an undesirable color shift. I use a separate layer for each Curves eyedropper and most often use white, black or, less often, both. That way I can redo any eyedropper if I choose. Am I in need of a scolding?

[Manfred M]

You state that your main gripe is that black and white point are not considered axiomatic to best practices anymore. And Manfred argues that it depends on who you ask, suggesting that the masses of self-trained photographers are responsible for not adhering to having to have a black and white point.

It is certainly something that is well taught. But it can be well taught so that those who deviate from the mantra know when they're doing it and and why they're doing it. I don't think that anyone would describe as Paul Strand as being amongst the masses of self-trained photographers.

Donald - this is a tough call. Paul Strand, Alfred Stieglitz, Edward Steichen et al were instrumental in changing the perception of photography from a fairly basic tool to a modern art form. As with all art forms, there were trends and stylistic approaches, often dictated by the limitations of the technology of the time. Pictorialism is one photographic genre that was very much associated with these people, but to a large extent went out of fashion in the early 1900s.

The key driver of the black point / white point requirements was the publishing industry and the printing technology that was used to make photographic reproductions. In order to have an image accepted for publication, it had to show the full range of tones from pure black to pure white. Not meeting this requirement was the most common reason for the rejection of prints submitted for publication. The last time I checked (about 3 or 4 years ago), this was very much still the standard expected for the print industry, and this went for B&W and colour publications.

This is where my comment on training comes from. Most people have never submitted work for publications and are unaware of this requirement. Any trained photographer who submitted images for printing was and follows that standard. I personally feel that people that only view their images on screens can get away with a bit more as the contrast ratio with electronic displays is a lot more forgiving than the printing process.

[JenSCDC]

Agree. I have believed that the color shift caused by using the Curves eyedroppers did more good in correcting color than bad in causing an undesirable color shift. I use a separate layer for each Curves eyedropper and most often use white, black or, less often, both. That way I can redo any eyedropper if I choose. Am I in need of a scolding?

You could also just change the blending mode to Luminosity.

[DanK]

Agree. I have believed that the color shift caused by using the Curves eyedroppers did more good in correcting color than bad in causing an undesirable color shift. I use a separate layer for each Curves eyedropper and most often use white, black or, less often, both. That way I can redo any eyedropper if I choose. Am I in need of a scolding?

This seems like a complicated way to do things, and as George suggested, one that can causes unwanted problems. You are mixing two things in your procedure: expanding the tonal range (which is usually what people mean by "setting the white and black point" and adjusting color. There is no reason why the areas that are close to black and close to white should be even mixes of R,G,B, which I think your method imposes.

I just adjust the tonal range using levels, curves, or the black and white sliders in LR and deal with color mix separately.

[Abitconfused]

I agree with your assessment Dank. My illustrations were more in support of the idea of a definitive B&W point than an illustration of a procedure I most often use. My point was that a "looks OK to me" histogram may be an unskilled approach. I have found that experimenting with Curves B&W eyedroppers often produces (for me) surprisingly good results whereas B&W sliders are more predictable but less eyeopening or "fun" if you will.