1) Open image.
2) Rotate if unlevel (usually is!)
3) Crop to 1024 x 680, I use 768 tall if the composition warrants it.
4) Adjust levels* by hand, each channel - Red, green, and blue. You can
use the 'master' or 'luminosity' channel, and it does work as well, but some images respond better to adjusting each channel at a time.
5) Adjust curves. In PS
elements this function is replaced by an applet that uses buttons and an incremental slider. Both are very, very powerful tools, and personally as important to me as....
6)...Sharpening**. Hardest of all to get used to. Canon (in one source) recommend 300% and a radius of 0.3 as a 'starting point'. Try this on a well focused, non-blurry 1024x680 image, and you will
get the jaggies, however at larger resolutions it is a great setting. Leave the radius as is, and play around with the percentage setting. Anything between 80 and 300 can work. Alternatively, if the image is good out of the camera, and you're at 1024 wide, try a radius of 0.2, and a percentage of 400-500. This results in very fine, surprisingly subtle detail sharpening and works very well.
None of this is set in stone. Alot of it has been acquired from experimentation, even more from friends, this forum, and even a screener or two have lent their expertise. At work I also sit opposite a graphic design class, and have picked up loads of little tricks.
*Learn the histogram. I won't go into detail here - use the help function in photoshop - it explains it well.
works by creating a duplicate image in memory (you won't see it), and applying a gaussian blur to that duplicate. How much
blurring is dictated by the 'radius' setting. It then compares the two images and, where it detects a difference in the brightness level of individual pixels at the same location in the images, PS
treats this as an 'edge'. The 'percentage' setting exaggurates this difference, thus giving the illusion of sharpness.
[Edited 2004-11-12 15:45:47]
It takes a big man to admit they are wrong, and I am not a big man.