This may not be the most appropriate place or forum for this question - but hopefully it's reasonably on-topic...

I've been looking into occlusion maps, and with these (and rendering in general) I often see options for hardware or software generated.
This difference between the two seems fairly straightforward with regards hardware rendering (which I assume is simply 'real time'). But when it comes to generating an AO map I can't find much distinction between the two methods other than the algorithm used.

Anyone explain to me the differences, and why/when one would be used over the other?

Thanks


J