Saturday, November 1, 2003

Megapixels and digital storage media

There is one factor I feel strongly about, but you're probably not going to read about it in many other magazines: solidity. Figure B shows side shots of the three formats.

FIGURE B

You can see the relative flimsiness of each format. (click for larger image)

I (and this is strictly a personal opinion) really dislike Smart Media because the cards seem so incredibly flimsy. They're very thin and very large compared to the other formats. And they have a huge open area where the electrical contacts are made. Whenever I handle one of these cards, I'm always convinced I'm going to break it.

My personal favorite is CompactFlash. While it's considerably bigger than the postage stamp-sized SD cards, that's precisely why I like it. I have big hands and I feel more confident I'm not going to lose my CompactFlash cards. I'm always paranoid I'm going to lose the SD cards.

SD is a very nice format, however. If you're more secure in your ability to handle small things, SD is amazing. And, if you're going to be a Palm user as well, you might want to make SD your first thought.

Memory cards and the number of pictures you can take

Now it's time to harken back to math class, otherwise known as my article, "What the heck is a megapixel?" at http://www.computingunplugged.com/issues/issue200309/00001105001.html. In that article, we discussed the number of megapixels needed to take pictures that could be printed at various sizes. If you don't remember all the gory details, now would be a good time to go back for a refresher and re-read the article.

It turns out there's not a direct mathematical relationship between the number of megapixels in an image and the amount of storage space taken up by the resulting image file. That's because of compression. If you remember all the discussion about lossy and lossless compression from "Understanding how image compression works," at http://www.computingunplugged.com/issues/issue200310/00001115001.html, you'll recall that the compression algorithms work based on the image itself. This means that an image of a face and an image of a forest might wind up taking up different amounts of space on your memory card.

Further, because many cameras let you set the quality level of the compression, from the "compress-the-crap-out-of-it" low quality compression to "do-your-best-but-try-not-to-compromise-image-quality" of higher quality compression, you can make images take up more or less space as you save them.

Here's a rule of thumb: try to keep your compression set to high quality and just get a bigger memory card. It'll result in better images.

And now, another rule of thumb, this time about image sizes. In general, an image taken with a one megapixel camera and saved at high quality will take about a bit under half a megabyte. A similarly high quality image taken with a two megapixel camera will take about a megabyte. And a three megapixel camera image will take a bit under 1.5 megabytes.

So, for argument's sake, let's assume you have a 3 megapixel camera with a 64 megabyte CompactFlash card. If each high quality, three megapixel image takes about 1.5 megabytes, then you can divide 64 by 1.5 to get 42.666. This means you can hold approximately 42 (plus or minus a few) images on your card.