My thoughts exactly. I've spent the last couple of years poking around the different engines and have seen a lot of work, and talked to a lot of people who've worked them. W3D, Source, Cryengine, UDK, Unity, ARMA, etc. All of the modern engines have their own individual selling points. UDK is probably the best bet for a large community and a wealth of material, but now it is past its time. Cryengine is attractive for it's graphics and sheer power. Unity seemed to be one of the more user-friendly ones, kind of a mid-grade that found a good balance in capabilities. UDK will still see use for a while, but with UE4's release and continued development it is seeing the same community-driven atmosphere being pushed with everything they've got- the creators are well aware of what made UDK successful and are milking it for everything it's worth. When deciding what engine I'd be most likely to choose it came down to a choice between Unity, CryEngine, and UE4 and I ultimately decided on the latter because of the likelyhood that it will be extremely mutable in what it's able to accomplish (Blueprint system in particular looks excellent for projects like these with shared traits and mostly aesthetic variation). If you're going to make the switch then there's no point in not going with something that will define the next generation, not be made obsolete by it. That said, W3D is well invested into at this point. At some point there will be a time where everyone here can say they've done all they want to do with it and smash a bottle over her bow and push her out to sea, but the change beyond that isn't something to be taken lightly. The question is, what is the ultimate goal of each individual project and the intended gain by those making it. W3D won't necessarily be a high point on a resume, but as far as a intense hobby and something to pour your soul into... Well it's already proven that aspect beyond expectation at this point.