Quote - It's going to take more than dual core support to cut the render time down to something sensible - a complete overhaul of the render engine probably

I don't expect TG2 per-core render times to get better significantly. In software, order-of-mangnitude improvements come from better data structures and/or better algoritms. TG2 is a program which has been in the making for more than 10 years (I am sure that much of TG2 code is at least based upon 0.x one) and therefore I expect that by now the innards are as good as theoretically possible. Code optimization, tweaking, fragments written in assembler and the whole enchilada of low level techniques never give more than a 2-5 speed factor boost (at a dreadful development effort cost).

All this taken into account, I don't think that a 100 hours render with corrent build will get much better than 4 hours using an optimized code on an 8-core chip (8*3 speed boost). Given that serious renders are said (at least of Planetside forum) to lay in the multi-hunderd hours range, I do not think that TG2 will be able to render a professional image in less than 1 day per computer per day. This means that a 1 minute animation (25 * 60 = 1,500 frames) won't be rendered in less than one week on a large render farm.

As I wrote before, I think that TG2 place is to create skyboxes for HDRI (in which case a 1 day per skybox time is perfectly appropriate because you do not need 100's of them and, if so, you are a professional user who can put aside a few render boxes whose task is just creating skyboxes in the background).

Bye!!

GIMP 2.7.4, Inkscape 0.48, Genetica 3.6 Basic, FilterForge 3 Professional, Blender 2.61, SketchUp 8, PoserPro 2012, Vue 10 Infinite, World Machine 2.3, GeoControl 2