Converted People...

[Chrisclick]

It's been a while. I think my last started thread was back in November last year. Spent some time hospitalised since, back on my feet as of a few weeks ago. My first outing since, was to a local craft market held in February.

I was trying my hand at B&W conversions at the end of last year and thought that I would pick-up where I left off.

The difference is, that now I have gained enough confidence to photograph people. These folk were kind enough to respond positively when I asked if I could photograph them:

Clothes Guy 1/320 f/5.6




Guitar Guy 1/320 f/7.1



Bearded Guy 1/250 f/5.6




CC always welcome.

[Shadowman]

Nice conversions. Welcome back.

[Marie Hass]

Hi Chris,

I think, when we spend time hospitalized, it changes our perspective on life and what is important.

Fears seem less so, and we wonder what kept us back before.

Nice images. Welcome back.

[mugge]

Love them, that's all.
Cheers Ole

[John 2]

Welcome back Chris. Hope you have shaken off whatever took you to hospital. Three very nice portraits. My Wife is a potter and I enjoy going to craft events with her for the same reasons. Might be tempted to selectively increase the contrast by adding a slight vignette and lightening the facial areas of each. It might also be worth taking down the white BG in the first if you have a curves tool. Nice work.

[Manfred M]

Chris - it's good to read that you are on the mend. Hospital stays are no fun at all and I try to avoid them as much as possible.

While the framing of the shots works for me, your image processing is off a touch. The first one is close but the other two are rather flat and muddy looking. As an old-school B&W photographer (I started out with film in the wet darkroom), the one thing I was taught is that every B&W image must have a full tonal range from pure black to pure white, and this is missing in these images. With jpeg, you only have 256 shades in the image and you need them all. I've gone back and fixed up the black point, white point and mid-point (gamma in these shots to demonstrate what I mean); nothing else was touched. If you don't like what I have done, I have no issue at all deleting my edits.

Regardless; blacks should be black and not gray, whites should be white, and not gray. The mid-point should be set to make sure that the image brightness is correct.

[chauncey]

Manfred, can you remind us again in general terms of RGB numbers, where you set your
Black/White Gray points?

[Manfred M]

Manfred, can you remind us again in general terms of RGB numbers, where you set your
Black/White Gray points?

Black = 0,0,0

White = 255, 255, 255

Midpoint = whatever makes the image look right.

This answer while 100% correct is also 100% useless. I'll post something useful when I get back to my computer.

[Manfred M]

Okay - back and time for a more appropriate answer...

1. This is the original image showing the histogram. The black point is at 0 the white point at 255 and the image is rather flat and muddy looking. Note that there is no substantial data on the left before a value of 15 and the right at around 192.




2. Move the black point to 15 and the white point to 192. This remaps all the points and gives us data from 0 (pure black) to 255 (pure white). The image will now have an appropriate range of shades from pure black to pure white, which is quite important in a B&W image. Note also how the image is still a touch on the dark side though.

Notice how I use the word "substantial data", there could be some data and in fact there is some on the right hand side past where I moved the marker, but there is so little data in the image, it really doesn't affect it. Anything to the left of the black point marker will be assigned a value of "0" and anything to the right of it will be assigned a value of "255" and we will get a tiny bit of clipping on the white side, but this is okay.




3. Adjust the mid-point (gamma) to ensure that the image looks correct. Moving it to the left brightens things up and moving it to the right darkens the overall image. I happened to pick a value of 115 In my eyes, this image looks better now.



While I used Photoshop, one can also do similar adjustments, but not necessarily using a single function, in other editors.

[Chrisclick]

Many thanks for the welcome back and kind comments, John, Marie, Ole, and John 2.

Yes, Marie, I'm certain that you are absolutely correct. That was the case for me; a reality check - a shift in priorities, time to think.

Thanks again.

[bnnrcn]

Welcome back Chris, I'm glad that you are back to normal life As some technical comments have already been made, I just want to say that I like the expressions and framing .

[Chrisclick]

Hi Manfred, thanks for your input. While I like and appreciate your original edit of the Bearded Guy shot, I'm not real sure about your changes to the other two.

I downloaded your versions of all the shots and, while the Bearded Guy looked much improved, I was surprised to see that the other two shots had some blown out areas.

Opening them in PS and looking at the histogram for both, revealed:






Turning on warnings showed:





You would know yourself, that the elbow of the guitarist is completely blown out in your edit and reveals no detail at all.

While I understand all that you are saying and, do appreciate the pointers you provide, with regard to your edit of these two photos, I feel that you went a bit too far both into the blacks and the whites.

For me, it's a bit tough trying to learn about B&W. Knowing whose opinion to trust is problematic - everybody has one.

The last gem that I gleaned was from the first B&W shot that I converted back in November:



Go figure.......maybe I'll just stick with colour/color/kula.

[Chrisclick]

Many thanks, Binnur.

[Manfred M]

In B&W "blown out" does not have the same negative impact as in colour work. The name "Black and White" says it all. White (a.k.a. as "blown out") is in fact white. Anything else (other than pure black) is in fact a shade of gray. We don't call it "dark gray and light gray".

In the "old days" a publisher would not accept a print for publication unless it contained pure black tones and pure white tones as the image would not have met publication standards. As I mentioned before, I was trained back in the B&W film / darkroom days and that was one thing that was drilled into me as a young photographer. Those standards still exist today in the print publications business.

Let's also examine what "blown out" means in digital photography. It means that we have exceeded what the camera sensor was capable of recording, i.e. we have exceeded the dynamic range of what our equipment can capture. Frankly, there are some instances where blowing things out cannot be avoided. Take a shot of a bright light in a scene or the reflections of the sun off water (a.k.a. specular highlights). These are unavoidable and are okay to have in images.

The "blinkies" in your camera or the indicators of areas that are blown out in PP tools are just there to warn you that this is happening. It is up to the photographer to make a judgement call as to whether or not this is acceptable in an image. There is no rule of photography that says "pure white is forbidden" in an image. I'll take pure white if it improves the image and I will also suggest that small areas of it in an image are just fine. Large areas, perhaps not. Just as an aside, in Photoshop for sure, the "blown out" indicators don't come on the moment even a single pixel shows pure white, there is a threshold (a percentage) that has to be surpassed before this happens, so it is a warning, nothing more and nothing less. It is up to the photographer to determine how much he / she is willing to live with.

The only time I actually worry about having blown out areas in an image is when I print digitally (as this is not an issue with traditional silver halide paper prints), and there this will be a problem as in printing the white comes from no ink being deposited on the paper and this will produce a strange "bald" area in the print, so it is typical to process the image in post, but then to preprocess the output to ensure that rather than a value of 255 being pure white, we pull that back to a value of just over 240. Blacks are handled very much the same way and we pull black values up to around 15 to preserve shadow detail. These steps are unique to the printing process.

By the way, when you read my comments on colour work that looks muddy, I recommend exactly the same process, if the image contains blacks and whites. A bit more restraint is in order when they don't, but having an image that has a wide colour range virtually always looks better than one that doesn't.

[Chrisclick]

So you're happy with the guitarist elbow in your edit?

[Manfred M]

So you're happy with the guitarist elbow in your edit?

I would not have posted if I were not, but then the edit took me all of 10 seconds. But then, I'm working from a downloaded / screen captured jpeg, which is limiting what I can do too.

If it were my image and I were working from the raw file I would definitely do a bit of dodging and burning to tone down the highlights and bring out the shadow detail. This is something I do in a lot of my own work.

Doing it off a downloaded jpeg is a lot more difficult as the data in the file is simply not there (your camera probably captures in 14-bit and a jpeg is 8-bit; add to that the data loss from the B&W conversion, if you do the math, you will find you have thrown away over 99% of the data your camera has captured). I'm fairly certain that there would be even less clipping (I wouldn't say none) that way.

It is a tricky shot to edit because of the way the scene has been lit. The background is also a bit problematic and is causing some of the edit issues. The camera left arm is definitely "hot", but neither of us can do much about how the lighting was set up. White clothing is always tricky to handle; just ask any wedding photographer what he / she things about shooting a wedding with a sailor in white dress uniform and the bride in white as well.

Stage lighting is also generally fairly harsh, it tends to come from the top. The hat the musician is wearing is going to give areas of blocked shadow detail.

I guess in conclusion, I'd have to say that your edit should result in a pleasing image and if you are happy with the way it looks, then who am I to say? I edited them the way I would have, had they been my images.

[IzzieK]

Is using the Levels in Photoshop better than using the Threshold with Curves in the adjustment layers? Just curious...as I am more used to the latter than the former.

Anyway, Chris -- welcome back and hope you are feeling much better. I thought these images looked better in Lytebox but without it, it is just another shot. Manfred's edit looks much better because the image is now more pronounced, the eyes all have those catchlights too which is crucial to a better looking image in portrait photography.

[Sergio M.]

Very good set of images. I especially like #3.

Let this be a springboard to many more.

Sergio

[Manfred M]

Is using the Levels in Photoshop better than using the Threshold with Curves in the adjustment layers? Just curious...as I am more used to the latter than the former.

Izzie - both methods work and I also use curves for this type of adjustment from time to time.

With various levels of understanding by the members, I will often opt for the simpler tool to get the result. If (for a colour image) I demonstrate using a curves adjustment layer, by channel, as the "best" way of doing this, I think I would scare away 95% of the people.

That being said, the reason I would do this in curves is if I had a both a black point / white point issue combined with a contrast issue. I generally don't like adjusting the mid-point in curves.