I have to say, it's beginning to seem like GS have once again missed the boat with regards to dissonance between AC13 and their clients' core needs and expectations of the program.
I may be (and truly hope I am) wrong, but from all the buzz and articles it would seem that AC13 is going to be all about Teamwork (or TW 2.0, as they put it) in much the same way AC12 was largely about the new CW tool and it's system-based interface. Which is all well, and good except for the fact and indication that GS seem to be catering to their big-ticket clients (read: Larger firms) rather than to the meat-and-potatoes issues of their bulk of their clientbase, which are small to mid-sized firms. Issues like a working Stair Tool (for once) that produces parametric stairs which are easy to create, document and modify and don't constrain one to a ludicrous fixed template of outdated stair-types, SEO issues, and the huge raft of other issues that require multiple work-arounds just to feign a pseudo-parametric process.
In ArchiCAD 12 the Curtain Wall tool was a great addition, but was it really that high up on the user wish lists. I mean outside of firms that focus on Commercial, healthcare and large-scale projects, it just isn't a useful (or even complete, or well-implemented) tool for architects and designers whose day-to-day projects range from residential to small-size civic projects. I'm not even including the multi-core support that was added to AC12 because that was well over-due (as is 64-Bit support), nor am I including 3D-documentation (because it's based on AC's outdated and clunky internal 3D engine, rather than an improved and more responsive OpenGL or DirectX platform) as legitimate major improvements to AC12.
And now we seem to be approaching AC13 where once again, the major upgrade seems to be in an area, that while beneficial to some, will not really be significant to a whole host of others who have got accustomed to working around Teamwork 1.0's perceived shortcomings or flaws (it's still far superior to Revit's team-based approach, so once again, was this really an area that direly needed urgent improvement?).
Meanwhile the nuts-and-bolts issues of the program (GDL alternative to parametric object creation, Improved versatile window/door library, Terrain/Massing/Site tool(s) 64-bit capability, etc etc etc) remain unaddressed. Perhaps they will prove me wrong and render this rant moot. But given their recent history, inability (or is it reluctance) to communicate with their own clients on their own moderated Talkforum, and their apparent objectives ( large-sized potential or future clients with null-BIM experience) I highly doubt it.