Ship handling in Atmosphere can be balanced this way. Yes.
Even with the game handling aerodynamics realistically the ships can be balanced by changing their shape and power output in atmosphere. Explaining it away with different efficiencies of engines and different aerodynamic influence of shields.
Pretty much +1.
If the shield helps aerodynamics by reducing friction / damage, then the ship will start to behave like in space, which defeats the original purpose of having ships behave in a different way in atmospheres than in space. So we have to walk a thin line…
Well there was always that idea about having the shields shape themselves to be a big fat saucer to work as a lifting body and make banked turns viable in atmosphere. 
The idea of using shields to explain aerodynamics seems like a dodgy one to me - shields are (probably) invisible. Therefore, I would feel a complete disconnect with the ship if it wasn’t flying in a way that it’s shape and design suggest it should.
In other words, if it looks like a brick, it should fly like a brick! And vice versa…
Maybe it’s alright. Let’s see what people think about this balance once Flavien drops the juggling balls.
The shield idea is a way to decouple the ships from realism. Allowing the devs a point where they can balance that does only influence ship behaviour per ship and only in atmosphere. Yes it kind of defeats the whole point of having realism. It is a thin line, yes. And I also want the ships to behave diffrently. Fun is also important though.
I also suggested there to be some visual feedback. Like thin visible layer of the shield when hit
by projectile or atmosphere.
Then perhaps shields should not be a passive layer of defense like they are in so many other games but an active module that consumes power at a certain rate… so pilots of larger ships who want to fly smoothly for a while would have to sacrifice the ship’s other capabilities somewhat. Also larger shields would consume more power, and unless they’re some sort of magic semi-permeable membrane, your own weapons and engine exhaust won’t be able to pass through. (I’m assuming it’s a bubble shield or a slim conical shield, not a skin-tapering shield as that wouldn’t help with aerodynamics)
True.
One possibility to balance that is to make the ship much more detectable, via any mechanics in place.
Also what @Naiba suggested. Simply make the shields a slightly more aerodynamic shape than the ship they are protecting.
You could add a mechanic to allow players to make shields slightly more ‘extended’ to make for a more aerodynamic shape, at the expense of a larger hitbox and lower strength. It would make the ship slower, but more maneuverable. Perhaps have a toggle-able on/off system that allows for the ship to automatically change the values based on what the pilot is trying to do or have a micromanaging system.
Not a huge advantage, but enough so that an experienced player will know how to exploit these to make a corvette slightly more effective than a brick.
I would, however, wait until adding an effective anti-fighter/bomber weapon systems before fleshing the system out. Because the only ship that would really benefit the most from having more maneuverability in an atmosphere would be the corvette, and that only really would be useful when combating interceptors and bombers.
(four people posted while I was typing lol)
This isn’t about making the shields more aerodynamic than the ship to ignore the atmosphere, this is about making a big flat ‘stopping’ surface to allow for much tighter turns when pitching than you’d see with space combat. Totally changing up the flight model to where rolling and pitching rather than direct yawing is the default choice with atmospheric flight.
( And then having ships spin out when the shields drop under fire 
)
*You might even get some interesting consequences for this, like people choosing to shut off their shields temporarily in the hope of rapid turns from such an unstable flight.
I do just want to derail back to this for a moment. Does this mean we’ll see a greater emphasis on the Newtonian motion on capital ships? They have increased mass and less directional thrust (compared to the little fighters) - an effect we can already see in the corvette.
I’m starting to envision flying controls where the pilot must think ahead for the massive ships, perhaps even turning retrograde to decelerate. This could really differentiate between large and small ships and add extra challenge to the biggest ones.
Also, won’t this mean capital ships would accelerate in a single direction faster than any small ship could (disregarding warp)? Big ships in Battlescape would probably have to avert the trope of WWII naval battles in space, except when cruisers line up in orbit to bombard a land base.
More force != more acceleration. F =ma. Increase force and mass proportionally and you keep the same acceleration.
^^This
Lowering friction will only really allow you to fly faster in atmosphere, but not turn differently. The idea of using shields is to present a larger surface area to generate greater lift/drag and decreasing the turn radius. This might also create an aspected shield target where you can shoot at a large area from a certain angle and still hit the shields.
Mechanically in the game, you could present the atmospheric model a different geometry than the ship’s current cube, though it may end up requiring a super large surface area in order to get the performance we want (without actually trying to run numbers). If that’s the case, I would introduce a shield layering concept. Multiple layers would act similar to bi-planes in that each wing presents the same surface area, but there are just more of them. You can cheese the mutual interference issues and just consider each layer to contribute a multiplicative effect for each one still holding.
Other methods might include actually letting players have some sort of control over the size of their shield’s surface area.
You’re right, but capital engines’ force probably doesn’t increase proportionally with mass depending on how the engines actually work. Larger chemical rocket engines historically produce more acceleration than smaller ones. Engines in I:B aren’t chemical rocket based, so the devs can do whatever they want to balance capital ship acceleration.
Still, if capital ship acceleration is merely the same as small ships, most battles would end up being quite distinct from Star Wars or many other space operas.
I’m confused. How can you tweak things to make it feel good when half your aerodynamic simulation is missing? Surely when you add lift that will change how ships handle?
Also surely the only things you should really change are the properties of the ships in which case adjusting one to make it feel good wouldn’t effect the others.
My assumption was that you’d implement drag and lift reasonably realistically, have simplified ship models to make the maths eaiser (eg main surfaces without small details) and then tweak stuff like the exact shape of these simplified models, the properties of the materials the ships are made of to adjust surface friction, and the amount of thrust produced by the engines at various atmospheric pressures.
If an interceptor is gliding through an atmosphere in a stable regime, then adjusts AoA, it sounds like in your current model it will just experience drag while the direction of it’s velocity would not change. If lift were implemented the direction of the ships velocity would change towards (depending on ship geometry) the direction the ship was now pointing - that would change the feeling of flight enormously.
Of course ships aren’t going to be gliding around unpowered, but at velocities greater than terminal velocity the effects would be even more pronounced.
@INovaeFlavien And this is my main issue with how atmospheric flight currently works in the prototype. If pitch up, I want to feel the air pushing the underside of my ship upwards.
At the moment, it’s too similar to space flight with airspeed resistance.
I hope the new atmospheric model will improve this!
Yeah, I totally understand. But you won’t achieve much of that with a realistic ( hell, even semi-realistic ) flight model. Ships are bricks and have no lift surfaces / wings. Even the interceptor, realistically, has little lift surfaces. So in order to achieve the behavior you want, I need to cheat. I need to tweak and introduce non-realistic variables to the equations. That’s fine with me, but I want to make sure everybody is on board with that 
It does!
You have to have very high angle of attack. Such that you don’t see you vector indicator anymore. But the deflection works in the current model!
I support first trying out how bricks fly themselves in the new models before “cheating”.
Ah okay, I haven’t actually tested for a while! But it will still be interesting to see if the changes improve it.
I just feel like it’s too easy to turn and face in any direction during atmospheric flight, like I’m in space.
Should ships try to pull back to the travel vector in atmosphere?
Indeed it does. If you want to think of it in simple terms, it’s really just newton’s 2nd & 3rd law in action.
Air molecules striking a surface will impart momentum to it based on their mass and relative velocity. Like billard balls, if they strike the surface off-center, the molecules will deflect in one direction, while the surface deflects in the other (lift and drag vectors)…I believe this is how Flavien’s current model works by drawing ray-traces at the aerodynamic model for the object.
Since air molecules only have so much mass, and there’s a specific density for the atmosphere, they’re only going to be able to impart a certain amount of momentum (and hence force) on a body moving through them. The more massive the object moving through it, the less effect the atmosphere is going to have on it. The more dense the atmosphere, the more effect it has on the object (try flying through water). Also, the faster the object/air molecules, the more momentum is imparted (try flying through water extremely fast!).
If an object has a very large surface for air molecules to strike, then you get more force simply by having a larger volume of air that will hit your object. Changing the angle of attack does the same effective thing by presenting a larger surface area towards the free stream of air molecules.
In any case, if your object too massive compared to the number of molecules it’s hitting and the speed it’s going at, then it will just plow right through the atmosphere, and turning will have minimal effect (which is how it should work if you’re going waaay too fast anyway)
Our current suggestion with using shields for the flight model is to artificially change the surface area of the flight model i.e. make it to where more molecules hit the surface. I also suggested a layering shield model to multiplicatively increase the surface area without actually creating gigantic shield surface areas. This also has the effect of gradually lowering your maneuverability as your shields get weaker as opposed to a binary on/off effect with losing shields. The actual forward shape of the shields should help with drag models as well.
Newton’s 3rd is only responsible for a miniscule fraction of ‘lift’ in all regimes of flight except high speed low density:
https://www.grc.nasa.gov/www/k-12/airplane/wrong2.html
Let’s take this discussion over to the Atmospheric flight model thread.