Some would argue that that sounds more like a project management failure than a change in market strategy—that they shipped a beta product as done, that someone screwed up somewhere, etc.
I don't think so, though; I think Apple just made a correct call in shipping the "core" of the product early (in the same way most enterprise products get 'Early Access' releases) to allow actual professionals to start evaluating the architecture of the software and tailoring their pipelines to fit it. Professional-use software doesn't get used on day 1 of release; it gets played with by an internal pipeline infrastructure team for six months while the working craftspeople continue to use the stable pipeline.
Handing out something that would be "patched into usefulness" at right about the same time it would actually be put into use makes sense, in strictly consequentialist terms, for both Apple and the professional studios—although Apple latching onto that release to do a bunch of marketing of their product as if it was usable for indies already, could be seen as a bit scummy.
> since it only supported one CPU. For compressing feature films, dual CPUs are a must.
This might not be the case right now (I don't know much about video encoding), but encoded video has keyframes, and effectively you can "chunk out" a set of raw frames to become a keyframe plus its successor interstitial frames, and then parallelize on those chunks, right?
In which case, I'd expect Apple figured that even if CPU-based compression is currently en vogue for this use-case, it'd be obviated by GPU-based highly-parallel compression soon enough, making the dual GPUs a much more significant benefit and the single CPU a much less significant drawback.
Tell me if I'm getting that wrong; I honestly have no idea.