I don't disagree with what you're saying there. I tried to intercede as a level-headed middle-man. I think it was too little too late though unfortunately. I wish I'd seen the whole thing earlier.
I do still stand by my point that there are people using Arducopter for professional AP. They just aren't
here. MRF has a good collection of people, but it's by no means all of them.
It's true that Arducopter can be harder to set up than other systems. I don't think that it's because something is wrong with it. It's because we simply allow more adjust-ability... plus it has SO many more features. I think where that really hurts is that people who are trying to scratch together a living doing this, just don't have time to figure this stuff out. You only need a flying tripod, not a flying dolly track. I'm trying to position myself as a middle-man, that's where my strength lie. Education, training, setup, consultation, systems engineering, etc... One problem though is that, we've all been giving everything away for free for so long, people expect everything to continue to be free. This is not a statement about you, just a general statement about the program. I've had a number of contract talks, but they dry up as soon as it's time for money to actually change hands. Somewhat frustrating from my perspective.
As far as reliability of the system is concerned, I categorically reject the idea that DJI is better. Or really any system. Just because our flaws are exposed for all to see, does not mean that we have more of them. They're just more visible. A user helping to find a bug is not an indication of the problem with the program. It's an example of the strength of the program, and the benefits of open source. These exact same sorts of problems could exist on the closed source systems. It's just when you report them the response is "Sorry for your loss, please buy another" you get "Oh crap, that's a bug, we can fix that." And the problem gets FIXED, is the point. We have 10,000 people test flying the code, and submitting valuable data back that helps make it better. But, our failures are a little more visible.
And I really don't see anything about AP that has specific requirements that do not fit into the all-things-to-all-people goal of Arducopter. So your aircraft is kinda heavy and underpowered. There's really no magic there, DJI can't give your copter more power. It may leave you with more reserve power, but I think in a tricky situation like wind, or evasive manoevers, you may suddenly find you have the same problem. And you want your PID gains lower for smoothness. Ok, I actually developed code that gives you the best of both worlds. Really tight PIDS, with smoother control inputs. It got significantly changed when it went into trunk and I haven't flown it yet. I hope it came out OK. I was specifically targeting the AP crowd with that. Spline nav is going to be epic too.
Anyway, I guess I'm just rambling so I'll cut myself off.
