Yeah that old school 'divide and conquer' approach is the way to go if your renderer doesn't do network rendering on single frames. It would require scripting to divide the main camera into sub cameras that you then save into separate files. Send these files to the slaves, reassemble the render(s) you get and it would be a fair approximation of how it works in other apps. Maybe one of you poser gurus would like to take a crack at it ?