Overhaul the drag model
SimplePlanes is a simulation game, which is great because you can build whatever you want and then expect it to fly exactly like it would in real life!
One massive thing is stopping any aircraft's flight dynamics dead in its tracks: drag.
SimplePlanes drag model in its current state is so bad in fact, that I recently had to cut my aircraft's drag by almost 98% and that only got it close to flying properly. Engines need to be more powerful to make adequate thrust (which really messes with takeoff performance) and when you send your power to idle at 10,000 feet it feels like you just ran into a wall with how fast your plane decelerates, and you need to enter a vertical nosedive to gain speed again.
THIS IS NOT REALISTIC!!!
This is why the drag model needs a total overhaul. Airplanes simply need to make less drag than they currently do in SimplePlanes. As in, 98% less drag. This will allow for some awesome stuff like better gliders and more efficient engines, as well as much better thrust physics.
Also, instead of focusing on individual parts, I suggest that the drag model looks at the plane as a whole and figures out how much air resistance the entire shape will create. Do away with the drag points system and instead use a Drag Coefficient to determine how much drag your airplane will create.
Implementing both of these things would help the flying part of the game a lot, and would help builders like me improve realism as much as possible on our builds. Thank you for your consideration.
C S commented
I'm burning 3 of my Updoots on this one, because I hate that we can "Mirror" our plane, and yet SOMEHOW the left side's parts create more drag than the right side's... inducing the annoying auto-roll! >_<
Yes, we can fix it with modding, but the idea of "Simple Planes" is that it's "Simple" to make an airplane that "just works". Right now, it lets us make an airplane that "mostly works", and we are left to dick around in the files (which I imagine is highly annoying for mobile users to do) in order to address what's wrong.
At the end of the day, the 'simple' fact remains: the drag system has bugs, bugs that are able to be encountered through vanilla creations that are unmodified.
Lastly, if reluctance to incorporate a better drag modeling system comes down to the new system being too process intensive, the obvious solution to me is to have both, and then only compute the new drag when the user requests it be updated.
If a plane is using the new modeling system and is modified in the designer, prompt the player upon playing whether they'd like to calculate and updated drag model.
--If Yes: Let it crunch the numbers and hold the game from spawning for those 5-30sec
--If No: Set a flag that marks it to remind them next time, and default back to current flawed system, so that they can still try it out. (after all, not ever time we load a plane to test something, are we doing so to actually fly it, so having to recalc every time isn't necessary)
Just my two cents. And I know the reason why SP isn't able to be as great as it could be is due to Unity sucking a giant egg when it comes to proper multithreading, with the core of the engine still being serial, but that also doesn't excuse leaving a flawed system in play. :P
I'd say cut it by 25% and see how that goes, also it would mess with previously made builds therefore it will never happen
while this is a big problem, I don't really care too much if it's fixed or not, I can just go into XML and take away some drag and usually it works (it might need a bit of fine tuning but it works). I will go out of my way to fix the drag, some people won't. you might have some really nice looking plane on the highest rated but it would accelerate to mach 1 in about 5 seconds, this person did not bother to make the flight characteristics realistic. but not like anyone cares about the planes flight, it's all about how much parts it has and how good it looks. I think people going out of their way to make something realistic shows the effort that has been put into it flying well. it's pretty easy to do manually, so I don't see much reason for it but it'd be nice.