First in a series; consider yourself warned!
One of the things that Dan James and I (and probably others) are always going on about is how people get caught up in the myth of “more MPs = better pictures”. Yes, I think we can call it a myth. One that’s sold by camera companies who rely on regular “improvements” in such areas to convince people to junk their old cameras and buy new ones. So let’s take a look at megapixels and how they effect what you see on the screen.
The screen is an important aspect of this argument. What is the resolution of the device you are using to look at these pictures? Mine is 1366 x 768, or about 1 MP. That’s one megapixel. My Canon’s sensor is 18 times that. Does it show under ordinary circumstances? No. In fact I shrink my pics down to 640 x 480 (or 427) most of the time because that’s all that’s needed for on-line viewing. Less than one megapixel.
Now let’s look at some shrunken shots. The Canon is capable of several settings from 18MP on down: 18, 8, 4.5, 2.5, and 0.3 MP. Here they are, all reduced to that lowest number at 720 x 480 (which is still larger than I usually use):
Can you tell the difference? No, of course not. They’re all the same size and resolution, right? So the advantage must be in larger renditions, right? Right. Now, how large do you need to see?
If you’re making really big prints higher resolution is preferable of course. Kodak says the minimum resolution for a 16″ x 20″ print is 1600 x 1200. That’s 1,920,000 pixels, or 1.9 MP. Yes, about 1/10 what the Canon is capable of. By that standard a “full size” print could be 56″ x 56″! How big are you going to make your prints, and how much detail do you need in them?
Here’s what happens when you chop a 720 x 480 segment out of the full 18 MP (5184 x 3456) image:
Thus we see that for ordinary viewing purposes the chief advantage of high resolution imaging is the ability to do some post-shoot zooming and still get an acceptable quality smaller image with good detail. By enlarging this segment just a little more we see that the image is not razor sharp, which alas is a failing of the lens it was taken with.
Here is the 18MP picture reduced to “full width” of 1366 pixels for my screen (although it’s longer than the computer can handle at 911 instead of 768):
In the next two segments on this issue we’ll explore lens sharpness vs. resolution and, if I can figure out a way to properly demonstrate it, how our eyesight works with all this (a particularly challenging objective given my eyesight).