Super User is a question and answer site for computer enthusiasts and power users. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Many times at work, someone will ask me if they can get an image (company logo, for example) in EPS or SVG format. They assume that since I put the image on the company site, that I could easily convert it to the format they need and send it to them.

Obviously, the problem is that the original image is a raster format and they're asking for a vector image. I often find myself having trouble explaining why I can't fill their request. Especially without giving a bunch of technical terms that I'd have to explain.

I'm not looking for an answer on how to convert JPG to SVG or similar but rather an easy-to-understand explanation of why the process isn't as straightforward as it seems.

Please help me straighten this out. Thanks!

share|improve this question
up vote 22 down vote accepted

I always say

Raster records the pixels in the picture.

Vector records the steps it took to draw the picture.

So if you enlarge Raster you get a big picture with big pixels. If you enlarge Vector the computer follows the steps to redraw at higher resolution.

If you have a chance to demo it for them, load up Word and put in 1 small (raster) image and 1 clip art. Then drag the corners to show what happens. You could even print out the before/after and hang it on your cube so you have something to point to.

share|improve this answer
+1. Nice and concise. – Joey Jul 31 '09 at 16:53
@arathorn, true; more added. – hyperslug Jul 31 '09 at 17:23
I really like the demo idea. Thanks! – Travis Jul 31 '09 at 17:30

In layman's terms.

Raster (I usually just say "bitmap" though) images record the colors in order. Blue, blue, blue, light blue, light blue, light blue... (At this point i'm pointing at a nearby object and reading colors on the top, left to right).

Vector image is a description about an image. "A light blue circle 12cm across. A solid blue background.

Then I explain how Rasters are better for photographs because that's how cameras and scanners see the image anyway, and Vectors are better for illustrations because you can make it any size without losing detail.

share|improve this answer

Vector graphics are to the etch-a-sketch, as raster graphics are to a mosaic.

share|improve this answer
Or old comic books when viewed though a magnifying glass. – Grant Jul 31 '09 at 18:22
Nice analogy and it makes sense, too. – Travis Jul 31 '09 at 18:57
+1 for the everyday references -- if the audience know those items! – Torben Gundtofte-Bruun Aug 12 '10 at 19:48

The main difference between a vector graphic versus a rasterized graphic is one is a coordinate based representation of what the object should look like. Flash animations are a good example of a vectorized graphics. A digital image is a perfect example of a rasterized image as it has a specific resolution. Scaling up the two is clearly illustrative of the differences. Vectorized graphics will scale without losing any detail as they are coordinate based. Rasterized images scale up and the pixels will become more appearant as the pixels become larger and larger.

True Type fonts are a good example because they scale to any size and look fine, when you use the fixed pitch system font you will noticed the pixilation as you increase the size.

share|improve this answer

Vector images consist of lines and curves. In general they are shapes with outline and fill and with that you compose everything else. Because you specify shapes in terms of mathematical equations they scale well and look the same in every resolution or size.

Raster images on the other hand consist of a regular grid of pixels each of which has a color. Changing the size of the image degrades its quality as you only have the original pixel grid as a starting point.

Converting from raster to vector basically means you'd have to try to find out what shapes can be used to approximate the look that was achieved with the raster image. For straight lines this is usually straightforward, but for curves and similar the process gets a little more complicated to get the look right.

Also it gets much harder if the raster image only has low resolution, such as on the web.

share|improve this answer

A raster image is like a paint-by-numbers picture you played with as a kid. Each pixel is assigned a number representing a color and the computer just fills in all the pixels to display it.

A vector image is like asking an artist to write down all the steps he takes to create a painting. He records every brush stroke by its location on the canvas, its shape, direction and size. Another artist can then follow the instructions to reproduce the same painting.

A computer is very good with numbers so it can reproduce either type of image with ease but turning a raster image into a vector image is like asking an artist to take a grid of numbers and a color pallet, determine what shapes the numbers represent and then record the brush strokes he would have used to create them.

Anyone can fill in a paint by numbers picture but what shapes and brush strokes it would take to create the same picture is open to interpretation. A good artist could get close but it would never be exact.

A computer can do no better than an artist. It can create smaller, more detailed brush strokes than even the most talented artist but it is actually far worse at recognizing complex shapes. So the best it can do is a close approximation.

share|improve this answer

Just explain to them that there are two different ways of storing images and yours is the other way.

share|improve this answer

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .