Consumer video equipment is awash with different options and formats for storing, transmitting, generating and displaying video.  It should go without saying that they're not created equal, and that you'll get far better results with some than others.  This is a concept I've touched on a few times before (<a href="http://www.gamesx.com/rgbadd/caveatrgb.htm">once</a>, <a href="http://atarilabs.com/meat/2000/1201_videoprimer.shtml">twice</a>, <a href="http://atarilabs.com/meat/2001/1011_videoprimer.shtml">thrice</a>), but I simply can't leave it alone.  There are many more formats outside of the consumer universe, but for now the formats we're going to look at are RGB, Component, S-Video, Composite Video, and RF.
<br><Br>
The fundamental issue here is conservation of space, be it on the air on on a DVD or VHS tape.  An uncompressed video image is huge, it's gigantic, it's really quite large, and unless you can physically connect one device to another you will find compressing the video stream is almost mandatory.  When you want to transmit an image from, say, from a TV station to a TV in your home, the amount of space you use is a critical issue.  There are many TV stations sharing the same air and each is granted only a tiny amount of 'spectrum' to send a signal.  The same logic applies to storing video - if you don't compress it the amount of space required is staggering.  Without compression a DVD would hold mere minutes of video,  instead of hours.  By using different techniques to compress these images you can acheive remarkable savings in space with minimal quality loss.  
<br><Br>
<b>[ RGB ]</b> Computers, with the exception of some antiques, transmit images to the monitor using analogue RGB video and a physical connection.  As you might remember from high school, Red, Green + Blue are the three primary colours with which all other colours can be made.  If you combine Green + Red, you get Yellow.  If you use less Green, you will get Orange.  By varying each signal you can generate all the colours in the rainbow, and then some.  The problem is the space it takes - RGB is really only economical for images and video that are generated 'live' by a computer, camera or game system, and connected directly to the display.  Storing RGB signals is extremely wasteful as well.  RGB is not a rigidly defined standard, there's no specific RGB connector, resolution, refresh rate or anything of that sort.  Similarly there's no limits to these specs, beyond the equipment you're using.
<br><Br>
<b>[ Component ]</b> Component video is a mathematically complex system most commonly found on DVD players.  Here's a simplified version of how it works:  Three channels are used, one for brightness or Luminance (Y) and two for colours, one red, and one blue.  The display treats the brightness channel as green, and the colour channels are mathematically applied to produce a full-colour image.  There's no advantage to this system over RGB, except that manufacturers and movie studios can treat consumers like ravenous pirates and apply such 'features' as region-specific encoding and copy-prevention.  The quality difference between RGB and component is minimal, the quality of your source and your display play a larger role.  Component video is a standardized signal, using specific connectors, and generally sticking to a few specific resolutions.  Even though component video adheres to these standards, there's no real reason to.  This is an uncompressed signal, so the final quality depends on the source device;  if your video source is a compressed medium (like a DVD video) the image will have lower resolution in the colour channels.  If you use a 'live' video source like a game console, there's no significant disadvantage to component compared to RGB.
<br><br>
<b>[ S Video ]</b> S-Video is very much a defined standard, with a specific connector and resolution.  If you take the next signal and do everything possible to keep it clean and pure, you get S-video.  You can imagine the S stands for Super, tho it really means Separated because the Y and C channels are kept apart.  It's also much like component video with the Y channel and two colour channels combined into a C channel.  This format provides a very stable picture with only two wires, and you'll probably find if this is all you have then you're not really suffering much.
<br><br>
<b>[ Composite Video ]</b> Composite video was a real boon when it first started appearing on game consoles.  Sadly now it's looking very tired indeed, as there are many problems inherent to the format.  The colour resolution is very low, and there are problems with signal generation and subsequent decoding by the TV that all but the most expensive devices are powerless to overcome.  Artifacts such as indistinct and blurry colours, 'dot crawl' along high-contrast boundaries, succeptability to interference and 'crosstalk'...  It's a real mess.  Composite video uses only one wire, and blends the Y signal along with the C data in a process that might best be described as 'sloppy'.  It's a holdover from the early days of TV when the colour information was shoehorned onto the black and white signal.  A certain amount of 'bandwidth' was allocated, but due to the cost of trying to achieve the full spec with 1940-era components, the full spectrum isn't used.  I could go on and on, but suffice it to say that it's really ugly and you should upgrade as soon as you can.  Sell a kidney.