How did this asinine obfuscated word get into the vernacular? What was wrong with "X by Y"? I'm sorry, I'm one of those poor, pathetic people who is not good at doing fucking square roots in his head.
Square roots? Even nicer! 4:3 ratio! Gotta love it.
Agreed. Down with megapixels.
They're not even all 4:3, so it gets nastier still.
The problem with a horizontal * vertical number, however, is that a "pixel" in the sense used by most camera manufacturers is actually a red, green or blue point. Some are arranged in triads, many in a more unusual format with more greens than reds or blues. Most cameras actually unbake the data before storing.
Good breakdown on How Stuff Works. This page describes the above -- scroll down to the Bayer filter for a quick visual.
It's pretty hard to come up with a measure of picture information capacity that is
Megapixels, despite the dorky ring of the word, only fails on the last item.
At least with TV screens, you can measure diagonally, which nicely captures the two dimensions. (But now there are TVs with aspect ratios other than 4:3, but we'll ignore that.)
But then you still know what you're talking about when you say something like "22" 16:9" :D
The problem with a horizontal * vertical number, however, is that a "pixel" in the sense used by most camera manufacturers is actually a red, green or blue point.
The same is true of CRT monitors, and it doesn't matter. The bottom line is that the camera emits a JPEG or TIFF or something, and that's a simple grid of colored squares, which are easy to count.
The cynical explanation: Because some marketer realized that quoting a number that grows by N-squared for each generation of technology encourages upgrades more effectively than quoting a number that grows by N.
The more cynical explanation: Because quoting the X-by-Y resolution would call attention to the fact that the vast majority of users have no use for images anywhere near that large.
Even more cynical explanation: Comparing 3 and 5 is about the intellectual limit of most consumers - you think their eyes won't glaze over at figures like 2048x1536 and 2560x1920?
The cynical explanation: ...
Actually it was mostly a rhetorical question. But I guess we're just lucky they didn't re-use "DPI" to talk about cameras in some arbitrary, braindead Apple-like way. Actually that might have been easier to understand.
Okay, maybe I'm being an idiot here, 'cuz gott im himmel knows I'm totally confused - but - it's important to me to know "at the highest setting, I can get a 5x7 at 300 dpi" or "an 8x10 at 300 dpi."Case in point: the magazine XLR8R did in fact end up using illustrations based on my photos instead of my actual photos, because they weren't high enough resolution for their full-page spread.
And that, my friend, SUCKS ASS.
You can figure that out from "megapixels", but it's a pain in the ass. If a "3 megapixel" camera takes a 4:3 picture, then the image will be somewhere around 2048x1536 (equals 3,145,728). At 300 pixels per inch, 2048 pixels is 6.8 inches (2048/300). So with that camera, you can get an image that can't be printed larger than 6.8"x5" if you require 300dpi.
It's easy to get from "X by Y pixels" to "physical size at N dpi". It's a pain in the ass to get from megapixels to, well, anything, because finding factors of large numbers is hard.
Because it allows small increases in resolution to look hyooj.
I mean, moving from 100x100 (0.01 MP) to 1000x1000 (1MP) isn't actually a huge jump, but it sounds bloody impressive.