[Hover on a thumbnail to see the caption; Click on a thumbnail for a larger picture]
I bought a Nikon E5700 digital camera on 2003-Jan-9. This camera uses CompactFlash memory cards, and I bought a 256MB such. It takes pictures of up to 2560x1920 pixels (5 megapixels); other sizes offered are 1600, 1280, 1024 or 640 pixel wide in 4:3 ratio, and 2560 wide in 3:2 ratio. Note that most digital cameras as well as most computer monitors use a 4:3 ratio of width to height, whereas most film cameras use a 3:2 ratio.
The E5700 stores the pictures in NEF, TIFF or in one of 3 grades of JPEG. NEF is Nikon's own "raw" format (NEF == Nikon Electronic Format); they provide software to support such on Mac or Windows; support for other (e.g. Unix) platforms is available from www.cybercom.net/~dcoffin/dcraw courtesy of Dave Coffin. Another useful site is www.aamot.org/ole/photography.html.
With a 256MB card, one can take exactly 31 full-size NEF's or approximately 104, 208 or 413 full-size JPEGs at FINE, NORMAL or BASIC compressions respectively, ... or approximately 3674 VGA-sized BASIC JPEGs. The JPEG estimates are provided by the camera firmware. My experience indicates that the camera, while bang on about the size of a NEF (no surprise), is conservative about its size-estimates for JPEGs; with my typical photos, a 256MB card has room for about 160, 250, 500 full-size JPEGs at FINE, NORMAL, BASIC respectively.
The manual describes FINE as roughly 4:1 compression, NORMAL as roughly 8:1, and BASIC as roughly 16:1. My experiments indicate that FINE, NORMAL and BASIC are at least as good as 95, 85, 75 in the usual Quality units, as used by other JPEG software.
Question: say I want to take 400 photos between visits to my computer, am I better off using 1280x960 at FINE, or 1600x1200 at NORMAL, or 2560x1920 at BASIC. I did some searching on the web to see if anyone had done some experimenting to answer such questions, but found mostly people saying the obvious: that smaller file sizes mean lower quality. (Beware: many say "image size" when speaking of file size; for me, an image consists of pixels, a file consists of bytes.) agfanet.com/en/cafe/photocourse/digicourse/lesson14/cont_chapter05.php3 uses terminology I understand, but provides no answers. cs.mtu.edu/~shene/DigiCam/User-Guide/950/resolution.html shows the sort of experiment I wanted, and indicates that a "big image at lower quality" is better than a small image at higher quality (for similar file sizes).