Is it the (allegedly 100% sRGB compliant) monitor on my laptop or does the second image have a green cast?
The original looks OK on my calibrated iMac and MacBook; and both images were processed with the same workflow and look equally "black" there (for the lack of a better word). However, I looked again at them on flickr and the second (Wendy) looked to be in need of a bit more contrast which I applied directly using flickr's in-built editor. Here's the result:
Thanks for the feedback - it appears we have a flickr mystery, but since the originals are fine at my end and the flick edits are fine at yours, , in the words of the Beatles, I'll let it be
When I open the second file I get a message that there is a problem with the embedded ICC colour profile, so I'm not sure where that error is coming from. When I bring the file into Photoshop, the image is not 100% gray scale, so I suspect that something might have gone wrong on the upload.
That being said, both images are well done and have strong impact.
The first one looks a little "muddy" and the subject's dark hair fades into the dark background a bit too much. I use a hair light in those situation to provide that separation. I find that the hands in front of the face are a bit distracting. In my own portraiture, I try to avoid placing hands on the side of the face closest to the camera.
The lighting in the second one is strong and effective. If this were my image, I would burn down that bright earring just a touch as it pulls a bit too much of the viewer's attention toward it.
Regardless, these are minor points and you should be happy with those two strong images!
Thanks for these helpful suggestions Manfred.
Yes, you are spot on about "Jessica" looking a bit muddy - I felt the model suited a soft look in contrast to the treatment of "Wendy" (once we got around the upload problem) but a bit too soft .... Good points on the hand position and the extent to which the hair line fades into the background too; I'll take another look at these issues.
I have not had any warnings about the embedded profile but there is definitely something wrong somewhere.
My comments were based on using Firefox, my default browser. If I switch to Edge () then, in the first post, the second image doesn't have a green cast, instead it is washed out and has very low contrast. The second version is OK.
I never trust what my browsers show as the degree of colour management that they use seems to be somewhat variable and even the ones that have a history of good colour management (Firefox is one browser that comes to mind), they break this on occassion when the browser is updated.
I will always download and examine an image in a photo editing tool, before commenting on the colour. In my case, I open the file with Photoshop and this is where I get the error message on the problems with the colour profile.
NB my version of PS (is prehistoric and) does not have that SWOP profile as its default CYMK profile so what I get from PS is not actually an error but a warning of the profile mismatch.
Also, when I open the image in Affinity Photo 2, it is again reported to be a CYMK (US Web Coated) image. AFP uses the correct profile and displays it as expected, i.e. no green cast. If I try DxO PhotoLab, it won't open at all, "CYMK images are not supported by PhotoLab."
Whatever has gone wrong, the version I get on downloaded, and I've done that several times now, is always a CYMK file.
I've just gone back to the file copy of the first upload to flickr, and also to the version here, opened both in Affinity Photo 2 and in Preview, and both are sRGB. It looks like this is one for the "curiosity" folder; there's no rhyme or reason to what we are seeing and as it has not happened again let's move on.
Definitely seems to be a Flickr issue.
I downloaded the original version of #2, before you added contrast, and opened it in photoshop. In photoshop, it has lots of contrast, unlike the one in your OP. I didn't see a green color cast, but I used the info panel in photoshop to double check, and it's neutral: the R, G, and B values are usually close to identical. The only places I found any appreciable difference from neutral were a few areas that had higher values for R, but the differences were small, and I couldn't see it visually.
Thanks for taking the time to do this Dan - that information is more than sufficient for us to consider this as "case closed" .