Post History

Dots and pixels can be the same or different, depending on context.  In a computer file, an image is usually made up of a rectangular array of color (or gray) values.  We call those <i>pixels</i>, for "picture elements".

"Dots" is usually used when trying to realize an image on physical media.  Dots can be very much like pixels, or not, depending on the medium.

For example, a die-sublimation printer can produce an arbitrary color at each location.  That's just like a pixel.  The physical characteristic of the printer might dictate a fixed dot pitch, like 600 DPI (dots per inch).  In effect, this is the density of pixels used to show an image on the output medium.

Let's say, you have a 1024 x 768 image you want to print 8 inches wide on a 600 DPI die-sublimation printer.  The printing software will automatically resize your image to 2400 pixels across, then print that with pixels matching dots directly.  In this case, you can argue that dots and pixels are the same thing, since the 2400 pixel across internal image was printed.

It's not so straight forward with other output technologies.  Consider an injet printer that can only either squirt a little bit of cyan, magenta, and/or yellow ink in any one place or not (we'll ignore the special case of using a forth ink to make black directly).  This allows each dot to be one of only 8 colors, which is not what you want your photograph to look like.  This printer provides the illusion of smooth color scales by using lots and lots of little dots.  In that case, dots are much smaller than pixels, since it takes a number of dots to realize the information in a single full-color pixel.

It can get even more complicated.  Some inkjet printers have some control over the size of the ink dots they produce.  They generally can't vary them meaningfully to give continuous color scales on their own, but you do get more than 8 colors per dot.  It takes fewer dots, and therefore less area, to reproduce the information in a single original pixel.  The number of dots/pixel can be quite hard to pin down.

Specs are for monitors can be different.  Old CRT color monitors had fixed dots that were either red, green, or blue.  Their intensity could be arbitrarily controlled, but their locations and pitch were fixed.  These RGB dots were arranged in a triangular pattern so that any clump of three adjacent dots forming a triangle were each of a different color.  Such clumps were called "triads".  Triads ended up in a hexagonal pattern.  Neither the triads nor the individual color dots were pixels.

Modern LCD monitors have clumps of RGB splotches, but those are usually arranged in a rectangle.  One RGB clump then <i>can</i> map to a single pixel.  Such monitors inherently display a fixed-size image in terms of pixels, like 1920 x 1080 for HD TV.  

It gets complicated, and the terms aren't always used consistently.  The little rectangular array elements of an image in a computer file can pretty reliably be called "pixels", but everything else is less certain, and gets blurry in common usage.

The one thing you can rely on is that marketing specs will highlight whatever value sounds most impressive, regardless of how much real-world bearing that has on the resulting output quality.