Ex7 - London, My two penny worth analysis

[mikemeister_admin]

Well this is the first time Mike has said "do anything" to improve the image.

I find it extremely interesting that we can all make such different renditions/interpretation of the same image - some of us added extra steps and/or multiple passes of CM.

I assume, like me, that you normally follow some sort of work-flow when tackling an image for real.  My first step is to crop - but none of us did that or straighten the verticals - but then we were all concentrating on CM.  I'm not sure the (bad?) composition could be improved much - the photo should have been wider and higher for us to do so.

Apart from the cast, our other main problem (I assume) was to get the skin tone 'correct'.  His left cheek started off at Lab 23,15,-3
We got
34,4,23
48,9,19
41,8,14
28,7,20
41,21,20
- quite a variation, but then our overall colouring and tonal interpretation were quite different.

The other area I found interesting was how we pulled/pushed shade detail in the window (from very light to quite dark) and also the toning on the stone work - perhaps how we interpreted London/City cleanness/grime.

I include 4 images
- our final results
- skin clocks
- stone clocks
- difference of our image compared to the original, with adjustment level layer (255 down to 90) to emphasis the change

The right black vertical window frame variation was quite a lot between us, orignal was Lab 7,2,-9 and we got
6,-6,6
10,-9,11
12,-1,-6
12,-9,12
9,-1,-1

Hope you find it all interesting - if Mike gives us another "do anything" next week, I do the same thing if you think it worth it.
Chris

[mikemeister_admin]

PS I assume Barry's blue difference is due to the fact that the image was not jpg - anybody know if that is right?

[mikemeister_admin]

Thanks for this - it's definitely worth it - even if I don't fully understand all of it.

[mikemeister_admin]

Hi Manda,

I assume you mean the last image.

I just take the original image and put a copy of the changed image on a layer above it and change the blend mode to difference.
I then add a Levels adjustment layer and bring down the White end from 255 to 90, to make the difference more noticeable.

Because I find that when I adjust an image, and concentrate on certain parts of it, I sometimes miss what I had done to other areas - so I use this view to double check I've not muck up things that I should not have!

If you like the idea, then I've put it as an action on my web site www.broadhurst-family.co.uk/lefteye/ that automatically does it for you in a new window.

Glad you found my comparisons interesting.
Chris

[mikemeister_admin]

I have walked through your pages!
Your whole website is interesting to read it.
So , I have saved it in Clipmates Collection under Curvemeister 3.0
There comes a time , I  read it.

Frits Egelie

[mikemeister_admin]

Very interesting comparison, Like Manda, I am not sure I understand it all.  About my example, you mention it was not JPG. What I sent in and what you are showing indicates JPG. I am not sure what you mean by "not JPG"
    Thanks
          Barry

[mikemeister_admin]

My mistake, I thought it was a .php file, what ever that is!

[mikemeister_admin]

Thanks for the link - I love your website (remember I was your number one fan in the dgrinners forum).

[mikemeister_admin]

I'm old and forgetful - I did not make the connection as I was thinking of Singapore and I misread your initial postings and took Asia for Africa! 
Silly me - so Hi again!
How are you getting on with PP5, assuming you got it for Christmas?  I've just got through Chap7 (&8) and although I get the ideas, I think it is the worse writings I've ever read - just a whole lot of I did this and that with no background of why - guess it will take another of couple of readings to try and see how Dan saw what to do.
Chris

[mikemeister_admin]

I find it rather interesting that (best I can tell) my readings for skin tone do not match yours. I believe my image is second in the top row and shows skin tones past 1 o'clock. Attached is a copy of the image with three skin tone readings and none are past 1 o'clock. Can anyone explain?
Thanks Joe S