View Full Version : Loss of contrast when flattening Photoshop layers


Raymond Mayo
10-02-2007, 06:29 AM
Hi everyone

Why is it that when you flatten Photoshop layers the image contrast sometimes shifts a touch? I've noticed it a few times down the months, but only now has it bugged me enough to look into it more, but I can't find an answer anywhere!

Basically, the image I'm trying to flatten now is black and white, though it is in RGB mode, and I've identified that the culprit is a curves layer. Even if I just merge that one layer down into the background there is a slight, but quite noticeable in the final print, loss of contrast. I have the curves set to Luminosity blend mode, and have also tried it in Normal mode, but the result is the same.

Any ideas?

Cheers

Raymond

Doug Nelson
10-02-2007, 06:39 AM
Have you tried printing without flattening? In that case it is still flattened, but further down the chain. It might help you determine where the workflow is interfering with your image quality.

Raymond Mayo
10-02-2007, 07:02 AM
Yep, I've tried that, and the image comes out fine, and as it should be. It's only when it's flattened in Photoshop that the loss of contrast appears (and more specifically when the curves layer merges into the background layer). Trouble is I want to save the file as a flattened TIFF to send out to someone, and looking as it should do!

edgework
10-02-2007, 07:45 AM
Yep, I've tried that, and the image comes out fine, and as it should be. It's only when it's flattened in Photoshop that the loss of contrast appears (and more specifically when the curves layer merges into the background layer). Trouble is I want to save the file as a flattened TIFF to send out to someone, and looking as it should do!What happens if you do merge visible to a new layer, instead of flattening.

What is the screen scaling when you do this? I've seen exactly what you're talking about, but it only happens if my screen scaling is at 25% or 33%, something that forces Photoshop to do pixel-averaging. When I scale to 100%, there's no difference, and I've never had it affect a proof.

Could you post the image?

cricket1961
10-02-2007, 07:53 AM
There used to be a visual bug(might still be).
Try viewing at 100% and flattening then. IF it changes then it is the file. If it does not then it is the visual reference only on the monitor and you don't need to worry about it.

Chris

Raymond Mayo
10-02-2007, 10:54 AM
Ok, so it does appear to be just an effect on the monitor. I just printed the image unflattened and then flattened, and the proofs are identical, even though the image appeared to change on the screen. I must have made a mistake earlier on and not printed exactly the same image, both unflattened and flattened.

Thanks for responding everyone.

Raymond