The Megapixel Myth

This guy is on math crack. Square megapixels are indeed the proper way to compare resolution.

But he's right that megapixels aren't the whole story, and beyond a certain level they are irrelevant to typical users.

A nice page on this subject: http://www.wyofoto.com/Digital_Image_Quality/Image_Quality.h... details a plausible way to calculate perceived image quality.

There might have to be some adjustments made because the human eye isn't equally sensitive to all color differences, but this is getting closer to the mark.

I am surprised to see people here criticize the article -- I though it was obvious that the man has got a valid point. For all "normal photographers" -- shooting vacation photos etc. -- it makes no difference to have 3 vs 10 MP. What makes a huge difference is good optics (where generally tiny = bad) and how well the automatic programs behave.

I think people should be aware of this.

All these words. Show us pictures of different solutions side by side and let us compare on our own.

This quote "One needs about a doubling of linear resolution or film size to make an obvious improvement. This is the same as a quadrupling of megapixels." is not only subjective, it's flat out wrong. Theres a remarkable difference between a DVD (480p) and HD at 720p, and that's only a 50% increase in linear resolution in one direction (and only something like 15% in the other).

I've read almost everything on Ken Rockwell's site. He's very consistent, level, and fair, and you could get a pretty decent photography education just by reading it. He also makes the same point over and over and over: talking about cameras isn't as fun as taking pictures. Turn off your computer and go shoot!