Can someone explain to me in layman's terms the difference between Viewshed and Observer Points?
Viewshed analysis calculates the number of observers that can see a given location. No identification is given as which observers can see that location. The result is simply a single numeric value. If you have 10 viewpoints, then your resultant raster will have values in the range 0 - 10.
Observer Points calculates visibility in the same way but the result is reported differently. Observer Points returns a look-up matrix that identifies specifically which observer (based on its id) or combination of observers can see a given location. Let's say you have 10 observers, your output raster will probably have a far greater range of values than 0-10 because each value is a code that identifies the combination of observers that can see the given location. This information is provided in the accompanying attribute table which has n-rows and 10 columns (in this case). Each column represents an observer and whether it can see a raster cell with a given code (1 = visible, 0= not visible). Given the huge increases in number of possible combinations as the number of observers rises, ESRI have limited the maximum number of observer points to 16.
So, your attribute table for Observer points in our example might start off something like the following (I have not made any attempt to include all the rows obviously):
In this case a raster cell with a value of 1 can only be seen by Observer 1. A raster cell with a value of 12 can be seen by observers 1 and 3 but none of the others. I made row 23 break the pattern just to give a different example and to show that not all combinations are necessarily represented in the result. In this fictitious example, a raster value of 23 would mean that observers 1,2,6 and 9 can all see the grid cell... and so on.