First off, compare the Mozilla screenshot of part II with screenshot from Opera 7. You will notice a that there is a small difference. At first, I thought this was due to gamma information being saved with the file, but after using PNGCrush to compress the images and remove the gamma infomation, the difference remained. It was later discovered that this is a bug that causes Opera 7, beta 1 and 2 to render the background twice if position: relative was used without using eiher float or z-index.
The MSIE rendering has, as you will notice all of the transparency information removed, but the page is still fully functional and viewable.
Right now, I'd appreciate it if anyone would care to send me screenshots of browsers that manage this test well, and those who cannot handle the test at all. Send it to:
blog@virtuelvis.com(If that address has any uppercase characters in it when you paste it into an email application, they are there just to hide the addresses from spam harvesters - just remove them.)
