My album is definitely far from ready for use, it's in draft mode but good enough for viewing purposes.

The issues happen on both on my actual smartphone (Samsung) and a Chrome mobile simulator extension which is pretty faithful to the real thing and one can look several models of Android and iPhone.

• Theme images on core album and on linked video page are not showing up full - they get cropped on both sides. They are 150px high and quite wide, shouldn't they show up in-proportion on a smartphone, no matter how small that may look?*

• YouTube preview and play in J-Album is totally unwatchable - not centered and highly cropped (more like chopped). Also a width/height proportionality problem. Once on clicks through to YouTube, the videos show up and play as they should.

This main album page is linked below, then go to "Music and Videos" to see that page's theme image banner and click on the first video or any.

Album Homepage

Not sure what's going on there. The first time through the album, the poster image for a YouTube video is horribly pixelated - I don't know what image it's trying to use, but it looks like a thumbnail that's been blown up way beyond its actual resolution. Doing a forced cache refresh often cures the problem, but only temporarily. The problem returns.

But I'm not seeing any unusual cropping, nor do I see any aspect ratio distortion - it's purely a resolution/focus issue.

I'll have to leave that one to Laza - YouTube videos in my own skins don't seem to have that problem - screenshot of a quick test using Neptune.

Grab one side of the browser window and make it smaller (narrower). Laza does use some viewport breakpoints to do some quick resizing on a phone, but the degree of image cropping varies by a huge amount.

In my little test album, the one you've linked to, the variation isn't as noticeable, because the theme image has no content outside of the middle half - the logo is surrounded by blank space. So, you don't notice the cropping. But take a look at the attached screenshots, and pay attention to how much of the theme image you're seeing, including the blank space around the logo, as well as the top and bottom of it. The file names indicate the device.

It's purely accidental in this case - I didn't spend any time choosing a theme image - but this illustrates one of the cardinal rules for choosing a theme image: Make sure the important part of your theme image is in the middle, because the left and right sides are going to vanish at some point.

It's not just Neptune - all of my skins appear to handle this correctly, even the lightbox ones, where I'm using a third-party script whose internals are murky to me. But in every case, the underlying mechanism is the same - the YouTube "embed" code is being used within a very simple iframe, and the iframe is sized for the viewport. In short, the skin itself isn't choosing the poster image, nor is it determining how to display it. YouTube itself is taking care of it. That makes the Tiger problems with it all the more puzzling.

ETA: But if you want a skin that will never crop your theme image, you'll need one of AndreWolff's skins, like FancyBox.

At last count, the "most used skins" metric was based on the number of "likes" received in the last 365 days, not on how often the skin is used (which is almost impossible to measure). Of course, most users never "like" whatever skin they're using, and if they're not uploading to jalbum.net hosting, they're never even prompted for it. In fact, I haven't seen that prompt in a long time, which leads me to conclude that it didn't survive the move to cloud hosting.

There was some discussion about basing it upon the skin used for the most recently modified albums, but since anything that's not hosted on jalbum.net is usually invisible, that wouldn't be much better. I don't know whether that change was made, though I suspect it was. I stopped paying attention to it a couple of years ago.

In short, that list tells us nothing, and the list contains enough obvious anomalies to make one suspicious of the data. But no one has ever figured out a better way to measure "popularity."

Well, yeah, except that users keep asking, "What's the best skin?"