Problem with merging layers.

  • 1
  • Problem
  • Updated 3 years ago
after emerging of layers the quality is often very poor. It seems to dependent to some degree on the picture contents, as it does ́t happen always to such a heavy degree, but it happens very ofte. Photoshop CC 2015 latest version with Mac Mini latest version OS
Photo of Jan Peter Vos

Jan Peter Vos

  • 4 Posts
  • 0 Reply Likes

Posted 3 years ago

  • 1
Photo of Chris Cox

Chris Cox

  • 20280 Posts
  • 767 Reply Likes
What do you mean?

Merged layers look just like the layers before merging.
Photo of Jan Peter Vos

Jan Peter Vos

  • 4 Posts
  • 0 Reply Likes
No, they loose color and details. In some cases quite substantially. The picture becomes useless.
Photo of Chris Cox

Chris Cox

  • 20280 Posts
  • 765 Reply Likes
Zoom in to 100%. Odds are you're dealing with a noisy image, and the down sampled preview isn't the same as the full resolution result.

But literally - the merged layers must look exactly like the layer before merging (same math, same code, same data).
Photo of Jan Peter Vos

Jan Peter Vos

  • 4 Posts
  • 0 Reply Likes
That could be the case. The picture has very many, very small colorful details, which could be interpreted as noise. But it is the way I want it. The less detailed the different layers the better the merging, but I keep loosing details, fast always.
It is a nuisance, because it may take quite some time to achieve the effect I am looking for. I am active in photo art. You can see my pictures at www.vosart.eu.
Photo of Chris Cox

Chris Cox

  • 20280 Posts
  • 765 Reply Likes
Always zoom in to 100% to judge the actual appearance. Noisy or binary (like halftoned) images will show the most difference in the downsampled preview.

Again, you can't actually lose anything when merging. The merged image is the same as the preview, assuming you are at 100% zoom or greater. When zoomed out, you are seeing a downsampled preview which shows the average values (because you are seeing far fewer pixels) and can be misleading on noisy or binary images.
Photo of Jan Peter Vos

Jan Peter Vos

  • 4 Posts
  • 0 Reply Likes
Thanks Chris.