Anartia jatrophae, butterfly
Moderators: rjlittlefield, ChrisR, Chris S., Pau
- rjlittlefield
- Site Admin
- Posts: 23626
- Joined: Tue Aug 01, 2006 8:34 am
- Location: Richland, Washington State, USA
- Contact:
Agreed, the image is very nice.
Bottom line, I don't have a good idea what is causing the banding.
--Rik
Photoshop levels adjustment shows that there are clearly some bright/dark swirly bands (like contour lines on a map, showing hills). In a focus-stacked image, bands like that are often due to slight variations in exposure from one image to another. But in a single shot like this one, I'm not sure what would cause it. I notice that the image file length is significantly shorter than this forum allows -- only 142 Kbytes versus 300 Kbytes allowed. But I would expect JPEG compression artifacts to be blocky, not smooth like these bands appear to be.There's a little bit of banding in the green background, I am not sure if this is due to my monitor or the image.
Bottom line, I don't have a good idea what is causing the banding.
--Rik
Those kinds of bands in smooth transitions (often seen in skies, for examples) are usually caused by gamut problems. If the image was made in a very large color space like ProPhoto, then the 256 levels for each primary color have to cover a wide area in color space, and the gaps between successively higher steps (say, from Green = 157 to Green = 158) will correspond to big differences in perceived colors.
- rjlittlefield
- Site Admin
- Posts: 23626
- Joined: Tue Aug 01, 2006 8:34 am
- Location: Richland, Washington State, USA
- Contact:
Gamut problems can cause big steps in what should be a continuous ramp. The big steps appear when an 8-bit image in a wide gamut like ProPhoto gets converted to a narrower gamut like sRGB. For example, a ProPhoto ramp of a certain bright green color, when converted to sRGB, no longer has any pixels where R=64 or R=66; the values just go as R=63, 65, or 67.Lou Jost wrote:Those kinds of bands in smooth transitions (often seen in skies, for examples) are usually caused by gamut problems.
But the bands that I'm looking at here are different. These are alternating bright/dark bands, with cyclic variation and no big steps. For example in the upper right corner of the image, in the area around x=740,y=60, along a 45 degree transect from upper right to lower left, the G component varies cyclically over a range from 115 to 118, taking on all possible values in that range. I think this is not consistent with a gamut problem.
It would be interesting to know whether the bands appear in images that were uploaded to Flickr, or if they only appear in the images that Flickr produced for serving back.
--Rik
A wide-gamut monitor (like mine) can indeed be part of the problem.
For me, the wide gamut is worth the side-effects. I need accurate colors, especially in the green end of the spectrum, which is where ordinary monitors fall down. When analyzing aerial images of forests, my monitor really shines, while common laptop monitors are horrible. When my colleagues visit me they are always shocked at the difference between their laptops and my monitor, side by side with the same forest image.
For me, the wide gamut is worth the side-effects. I need accurate colors, especially in the green end of the spectrum, which is where ordinary monitors fall down. When analyzing aerial images of forests, my monitor really shines, while common laptop monitors are horrible. When my colleagues visit me they are always shocked at the difference between their laptops and my monitor, side by side with the same forest image.
- rjlittlefield
- Site Admin
- Posts: 23626
- Joined: Tue Aug 01, 2006 8:34 am
- Location: Richland, Washington State, USA
- Contact:
- rjlittlefield
- Site Admin
- Posts: 23626
- Joined: Tue Aug 01, 2006 8:34 am
- Location: Richland, Washington State, USA
- Contact:
Lou Jost wrote:If the image was made in a very large color space like ProPhoto, then the 256 levels for each primary color have to cover a wide area in color space, and the gaps between successively higher steps (say, from Green = 157 to Green = 158) will correspond to big differences in perceived colors.
I agree with both of these. But I think only the second one applies here. The image as posted is already sRGB (as reported by Photoshop CC when given the URL to download directly), and I don't see any of the characteristic big steps that I would expect if it were suffering from the first problem.A wide-gamut monitor (like mine) can indeed be part of the problem.
--Rik
I'm using an ASUS ProArt wide gamut monitor. Maybe it is the reason that the banding is obvious to me. But as far as I can recall, this is the first time I have noticed banding on an image in this forum, and I've looked at nearly every image that has been posted here since I joined. So I suspect there is also something about this image that exacerbates the problem. In any case the image is beautiful.
It is. Rui's stream is worth looking at.In any case the image is beautiful.
Do you see banding on this:?
http://photomacrography.net/forum/viewtopic.php?t=32588
Chris R