if the engine is too much designed for the way directx and especially diretc3d acts, then you don't will see a support for other plattform that soon...
the option is to rewrite the core (maybe not the whole one but seperate really rendering from other things) but its a huge work and sometimes is better to start a new engine core than to rewrite a directx based one only to fit for multiplattform problem.

but to rewrite or write a new engine core it takes so much time, that is not realistic if they work now on this game project. so after or during this project maybe they see what could be a lot better in workflow etc. so that they recreate new editors etc. its still easier to make a new editor than an engine. The engine is more or less the same and you can use a lot of code twice for the new editor in the background, only the visual part of the editors must be changed. It's not a few days job but soon or later something must happen, and I think the engine itself is not that bad, tweak a little bit more the tools and people easily get more ideas of what great things they can create with 3dgs.
I used and use several engines ( closed source, propetary and open source) and for me 3dgs is still the best package even if I only use a6 extra and lite-c free. But I've also to admit that I'm very impressed about what I'm able to do with the other engine of the competitor who after only a few hours on the first day I used it. I feel familiar with everything and I have the feeling that I work professionally with it. But for rapid prototyping or rapid game creation I still consider 3dgs the best one, and the option for knowing that we can extend the acknex engine is and will remain hoepfully a great option