TIP: My B&W conversions

[Sebastian]

I'm by no means an expert, but people seem to like what I've been doing with B&W lately, so I thought I'd share the method.

It is incredibly simple, I just adjust levels to get the highlights and shadows set, convert to greyscale, and then apply an s-curve. Usually as an adjustment layer so if I'm unhappy with a print I can open the PSD file and make adjustments down the line.

This is the curve I used for the "hands" image.


And this is an example of the effect.

[OldSchool]

Nice lesson Sebastian.
Thanks,
Tim

[another view]

So that's how you did it - thanks for sharing. Looks great!

[gahspidy]

Sebastien,
Good of you to share. The effect is very impactful. Do you convert with the Channels mixer method? I would think also, that you would be editing the files in 16 bit depth especially with a steep S curve as what you have shown.
I have been playing with a method that I have stumbled upon through my experimentations.
I use it for some, but not all pictures. I have used it on my chinatown series conversions.
I convert ( after doing a basic levels adjustment) to B&W using the channel mixer method.
I then go to Image> Adjustment> Selective color. I select the black color and increase to about +10, then the nuetral to about -8, and then White to approx. -25. This gives me a tonal pop, and a perceived increase in sharpness. Then a bit of dodging a burning here and there if i want a certain effect. Also done on layers so that the end result can be reduced if desired. It's effective on certain images and not on others.

[ken1953]

Thanx for the lesson Sebastian. I just ordered PS CS2 premium and have been wanting to try some b/w conversions. I will definitely give this a try. This was really an inspiring post.
Ken

[another view]

Do you convert with the Channels mixer method? I would think also, that you would be editing the files in 16 bit depth especially with a steep S curve as what you have shown.

I got to thinking about this yesterday - but according to his post he uses greyscale mode first. I wonder how it would work with channel mixer, getting the results close first and then using the curve, but I guess I can experiment with that. I agree about 16-bit, but I think Sebastian shoots jpegs most often (not positive though...).

[Sebastian]

Yeah, I just open the file, convert to grayscale, and use the curve. If the image is RAW, I do all adjustments in 16 bit, if it is a JPG, there is no point but to do it all in 8 bit.

And AV is right, I mostly shoot JPG. The shots in this post were JPG.