I'll try to get the thread back on topic

I've been looking at Torgue for some time. As mentioned concidering any features not available is foolish. The safest and most sure way to pick an engine is to use the release that is stable and to not count on any features promised that is still in beta. What you have is what you have.
Though octree is nice to have it's not a must have. Octree has many disadvantages as BSP depending on what your doing, especially when working on underground dungeons. Here BSP fairs much better and gives good results much faster. It alsi has the advantage of rendering very fast on old systems and being much easier to line up seemless textures. Like wine, sometimes old is good

However if I was working on outdoor levels I'd want octree hands down. I could get results in BSP but it would take a myraid of tricks that would create a lot of work. In a perfect world therefore the engine really needs seemless use with both formats such as engines like irrlict. If an engine really wants to boost about culling this really is the the top dog. Since currently neither torgue nor TV3d can do that then your real pluses and minuses depends on what your doing.
As for torgue our team decided against it for other reasons. Cost. Bang for the buck is just not there. Like Dark Basic pro they get you on the plug-ins.$500 for the engine, $250 for the lights plug-in, $995 for multiplayer---it goes on and on. You can't pick the INDY picks as they forbid making any money-so you have to pick the comercial ones.
Aside from the obvious money issue torgue does have very good documentation. The book is well written and aimed at the older release and doesn't even touch on accessing the source code but on the lower level development it is done very good. Thousands of other tutorials skattered throughout the forums and it becomes very clear TV3d has a very long way to go to catch up in this dispartment.
In fact I believe it's the documentation and community base that continues to keep torgue the number one engine on devmasters, because technically there are a lot of better engines listed. Getting 'how to' information is paramount to any project. Fail in this area and people look elsewhere.
So if anyone is trying to pick between the two my advice is first ignore the feature list, don't count on anything not already out and tested, look at your skill level and your pocket book. And see which is better for what you want to do.