No one is going to pay $15 a month for a game with cruddy graphics and very low population most of the day.
No one? Many players have done so for many, many years because the flight modelling is the best and you're free to pitch whatever aircraft and tactics you wish - arena disposition permitting - against potentially massive amounts of or at least hugely-varied human players. That's unique.
I notice all the other growing games are A V A and without Icons...
I hardly think a sensible stratergy is to try to force AH into a poorer replication of more recent games. They have their strengths and weaknesses the same as AH. Nor do I think it's a particularly good idea to chase the younger and more fickle ego-centric gamers. AH has appeal to other people. A couple of points which might not occur for various reasons:-
NONE of the present flight sim offerings have yet, nor likely will last as long as AH has. Doubly so if you include AH1 and 2's predecessor. There is clearly a special sauce in there somewhere, the longevity is nothing short of remarkable and seldom is acknowledged.
By now this game should fairly have a cult and legendary status. Yet you see hardly a mention because it's age and design-decisions are not in alignment with recent offerings. Why must everything pander. Some things are what they are.
At the time of the Steam launch I volunteered to help HiTech with some design work and was shared certain confidences which were pertinant to that work (and not mine to share further). Suffice to say many forum members seem to have been confusing causation with corellation for years now and have been taking that failure to revamp the player-base as some kind of evidence there's something wrong with the core of the game. The primary reason prospective players did not stick around had nothing to do at all with the core game nor the subscription model. Couldn't possible be any meaningful inference made about that from this release. Let's just say it was a sort of highly off-putting user-interface issue for those who had not given flight-sims a good go.
If I were chief advisor in charge of game architecture, future policy, whiskey-rationing and that little gravelley patch behind the outside toilet (which I'm not), I'd suggest a half year revamp, a consequential launch of AH4 on Steam with revisions to the user interface and certainly an some HTC-hosted human-populated alternative arena to the MA. The MA is a very peculiar place to those who have not spent a long time here. That's hard to empathise with by all of us who have, well, spent a long time here. It compromises all of our suggestions about revitalising the game. We all also speak from our preferences from years of gameplay. Maybe a rolling AvA (amongst other things) also offered. But not ONLY offered.
Regarding graphics, there are still people playing Counterstrike (the old one) and other similar games. The question is not whether the graphics are fantastic eye-candy replete with wow-factor but rather are they fit for purpose which IMHO is to give you sufficient immesrion to enjoy the actual activity of the game. For me it's perfectly good enough. Also there is advantage in that, wihout 72-layers of ray-traced sundogs and reflections your PC requirements also let you reach a more general, non-gamer and likely more mature demographic.
Finaly a word about the Dale God Entity. I don't have it, but by all accounts he has knocked the ball out of the park with VR compared to other well-staffed snd well-funded games. And it's not the first time, his patented net-lag code is the backbone to which this truly only MMOG flight sim hangs. I think he's a tiny bit autistic-savant (sorry Dale if I'm wrong or offend, to me there's no offence in such a gift), and I think he works best with people who have an alternate set of skills for balance - shall we say. If employees are out of reach then if I were him I'd be asking people like Lusche to help in addition to some user-interface specialists and so on. If nothing else a periodical reality-check with others about design propositions would really be beneficial.
Parallel arenas should definately be run and experimented with. Fiddling with the MA drastically will turn the lights off right-quick IMHO. Flight simming is a tiny subset withon a tiny subset, with obvious limits because of that. Driving simulators have a much wider base of interest but now you need thousands of $ to get competitive equipment. Not so here and that's largely thanks to Dale and his flight model.
Just my 20-cents