Heh. I’ve been fiddling with GIMP for years (see Links on my site for where I’ve found good info). Just never really settled down to it. I created the image in GraphicConverter in 8 easy steps.
step 1- create 3 new 50x50 pixel images
step 2 - create a rounded filled rectangle in the first image, set to have a 40px radius
step 3 - remove 25px from top and right sides (Picture -> Size -> Add/Remove Margins)
step 4 - select entire 1/4 image, copy/paste into upper-right corner of the second new image
step 5 - mirror the 1/4 image horizonally, copy/paste into upper-left corner of the second image
step 6 - remove 25 px from the bottom of the second image
step 7 - copy/paste the 1/2 image into the top half of the third new image
step 8 - mirror the 1/2 image vertically, copy/paste to the bottom of the third image
All done.
And that’s why I’ve not done much with GIMP; GraphicsConverter does almost anything I want as far as simple images go (well, and I’ve never had a good enough machine to make any serious graphics software anything but annoyingly slow...). I’ve managed to learn to do simple gradients and fades, and some semi-transparent layers with GIMP (see my Theo template), but that’s all.