He was talking about the background, not the foreground.

The difference is in color around the edges of the picture in the background change noticeably on a non-fullscreen image on my Android 12 device.

> The examples are just bad. If you want to show something, screenshot and enlarge it to show the artifacts.

Yes! Where's the red underlines and diffs? I can see the background banding, but the foreground looks the same at a glance except that some of them look ambiguously "off" in ways that could just be placebo.

You'd think a visual artist would be more interested in visual communication and not just a wall of text with un-annotated photos.

The article is about the background banding.

I think he was complaining specifically about the background banding.

I downloaded the images and then compared them via Beyond Compare.

After that it was pretty obvious what the author is talking about.

It's hard to see in the first set of images, but the second set is much clearer. In the WebP example, look to the right of the subject, about 1/6th of the image's width from the right edge. There's a hard transition between shades of grey. The JPEG version directly above it also has banding but each band is narrower so the difference at the edges is more subtle.

This seems to be highly subjective. I had absolutely no problem seeing those artifacts without any pixel peeping, they're that obvious.

WebP image gradients just looked broken (posterized) except the lossless one, which was (obviously) perfect.

> enlarge it to show the artifacts.

One might argue that if you need to enlarge it to see the artifacts, then the artifacts aren't perceptible enough and the codec is already good enough for the use case.

But we are philistines not pro photographers