Which is faster / less resource intensive : Shadow Maps or Shadow Volumes? Thanks all.

Depends on a lot of things...

Hardware, native support for shadow maps, special stencil operations to make shadow volumes faster, scene characteristics, etc, etc, etc

Some hardware (i.e. NVidia) has special settings that duplicate(maybe even quadruple) the rendering throughput when rendering to a shadow map. There is also hardware (from both NVidia and ATI I believe) which has support for double sided stencil, which makes shadow volume rendering faster.

Sometimes speed does not matter as much, if you are working on a fighting game, you can have beautiful shadow volumes for both characters, but it is very unlikely that you can afford shadow volumes on large scenes.

Note also that it is easier to get soft shadows with shadow maps than with shadow volumes.

There has been endless debates about which technique is better, and IMHO it really depends on what you are trying to achieve. Shadow volumes can eat your fill rate really fast, leaving very little room for doing other effects (i.e. post processing effects), but this is fine if you are not computing them for a lot of objects. Shadow maps are easy to understand and implement, but you do not get the nice hard shadows you can get with shadow volumes. On the flip side you can implement nice soft shadows easier with shadow maps than with shadow volumes.

My 2 cents, Hope this helps

This looks like it could be a very interesting article, but I don't think the link is quite working properly?