ksp areodynamic model

[asap1]

The current ksp aerodynamic model uses very basic and often unrealistic(note its still pretty good compared to most games)  

[pandaman]

Yes, but I think it will be more likely an improved version of KSP1 aero rather than a full aerodynamics sim.  

I doubt there will be two optional systems, just possibly a difficulty setting depending on what they do, but I very much doubt it.

[Guest]

KSP aero it's not simpler than realistic aero, it just has a different set of rules and kinks, +1 for a more realistic model.

On the other hand hard no for having 2 different and conflicting systems at once, It would just be a mess.

[Brofessional]

Two different models would be a mess in regards to sharing craft or learning the game.  Also one would likely end up neglected anyway as the community forms a consensus on which is better.

If they can make the aero model more realistic without hurting performance too much that's great, but I don't expect anything drastically different than KSP1.

[Dientus]

5 hours ago, Brofessional said:

Two different models would be a mess in regards to sharing craft or learning the game.  Also one would likely end up neglected anyway as the community forms a consensus on which is better.

If they can make the aero model more realistic without hurting performance too much that's great, but I don't expect anything drastically different than KSP1.

Add my two cents to the "no two models" column.

 

I do hope for more accurate representation and calculation of aerodynamics and shearing forces, but would like more accurate metal and materials strength and flexibility as well to balance it out.

[Bej Kerman]

A part of me will be really disappointed if the realism doesn't approach Ferram. Aerodynamics shouldn't be baby-proofed; I feel the same way about aerodynamics as those who were scared of orbital mechanics being simplified when KSP 2 was earlier in development. With spaceplanes and such being so prominent in KSP, the aero model should be held with almost as high of a regard as orbital mechanics.

[Guest]

Currently speaking. I don’t think it’s a priority. The current one seems to work just fine. 
 

if they had time I think it would be a good thing to do. But it shouldn’t be top priority 

[K^2]

Intercept's physics programmer is from FEM background, so I think improvements on aerodynamics are very likely. Also, kind of necessary, since they're going with procedural wings.

[Incarnation of Chaos]

They have already confirmed they're sticking with KSPs style of aero.

6 minutes ago, K^2 said:

Intercept's physics programmer is from FEM background, so I think improvements on aerodynamics are very likely. Also, kind of necessary, since they're going with procedural wings.

They could easily just set up the drag cube based on the dimensions of the wing, still more complex than KSP but I wouldn't take it to imply FAR style aero.

[K^2]

8 hours ago, Incarnation of Chaos said:

They could easily just set up the drag cube based on the dimensions of the wing, still more complex than KSP but I wouldn't take it to imply FAR style aero.

Look, I'm not even a mechanical engineer, and if someone was to suggest this hack to me, I'd give them a death stare that would instantly make them reconsider. I might be placing too much faith in someone whom I know only from their LinkedIn profile page, but I don't think an FEM engineer is going to accept something with such obvious flaws. Never mind the fact that CoP is not going to move correctly, but you can literally just create a frame of a wing and use it as a full wing in that setup. I know KSP wings do a lot of dumb stuff, but this is an opportunity to do better. And you don't have to do fine simulation to do better. All you need to do is break the wing into smaller sections, each with its own surface area, chord, and CoP. Then just apply a standard lift and drag curves. Bonus points if you increase drag and reduce lift of the end sections to fake vortices. But even without that, simply simulating a wing as multiple sections would be an improvement, giving you both a more faithful simulation and better control over the planes you're building. And given that this is how all but top-of-the-line flight sims are set up anyways, it seems like it'd be a natural path forward with procedural wings and, by extension, flight dynamics in general.

[mcwaffles2003]

9 hours ago, K^2 said:

I think improvements on aerodynamics are very likely. Also, kind of necessary, since they're going with procedural wings.

51 minutes ago, K^2 said:

Look, I'm not even a mechanical engineer, and if someone was to suggest this hack to me, I'd give them a death stare that would instantly make them reconsider. I might be placing too much faith in someone whom I know only from their LinkedIn profile page, but I don't think an FEM engineer is going to accept something with such obvious flaws. Never mind the fact that CoP is not going to move correctly, but you can literally just create a frame of a wing and use it as a full wing in that setup. I know KSP wings do a lot of dumb stuff, but this is an opportunity to do better. And you don't have to do fine simulation to do better. All you need to do is break the wing into smaller sections, each with its own surface area, chord, and CoP. Then just apply a standard lift and drag curves. Bonus points if you increase drag and reduce lift of the end sections to fake vortices. But even without that, simply simulating a wing as multiple sections would be an improvement, giving you both a more faithful simulation and better control over the planes you're building. And given that this is how all but top-of-the-line flight sims are set up anyways, it seems like it'd be a natural path forward with procedural wings and, by extension, flight dynamics in general.

Wondering why procedural wings make updates to aerodynamics necessary.

Is it due to something like all the wing parts in KSP 1 are standardized by being premade parts so they could all be given specific variables concerning their lift/drag/etc properties whereas with procedural wings since there are no preset sizes those figures will need to be made up on the fly anyway since the shape and size changes?

 

[K^2]

1 hour ago, mcwaffles2003 said:

Wondering why procedural wings make updates to aerodynamics necessary.

If a wing moves mostly at speeds and in directions it was designed for, it's aerodynamic properties stay fairly well confined. You can get a pretty good model going by specifying wing area, center of pressure, glide ratio, critical angles of attack, and slope and intercept for the lift coefficient between these extremes. It's pretty close to what KSP does, and if your wings aren't doing anything weird, it's good enough. Of course, KSP wings often do something weird.

If you have a procedural wing, you could bake out the same characteristics and still use the simple model, but it's actually more work. What you can do much easier is to treat each section as its own wing. Technically, it's nothing new compared to building out large wings from many, many sections, but such wings already behave better than larger parts. Procedural wing would also eliminate all the self-intersection and overlap problems when you are building compound wing, so you'd basically be getting the best version you could have hoped for with careful design automatically.

Now, KSP stock wings have other concessions. They have a floatier feel due to lower drag at high AoA, they generate same amount of lift regardless of position, etc. That's likely to stay as is, by the sounds of it. Which is fine, IMO. Having procedural wings represented as compound sections would give you a lot more in terms of proper feel then a more honest simulation of individual sections.

[Jack Mcslay]

I think this is needed. I want effective airplanes to reasonably  resemble what real ones look like KSP's aerodynamic model is too easily exploited and trying to optimize planes eventually lead to designs that make no physical sense

[Incarnation of Chaos]

8 hours ago, K^2 said:

Look, I'm not even a mechanical engineer, and if someone was to suggest this hack to me, I'd give them a death stare that would instantly make them reconsider. I might be placing too much faith in someone whom I know only from their LinkedIn profile page, but I don't think an FEM engineer is going to accept something with such obvious flaws. Never mind the fact that CoP is not going to move correctly, but you can literally just create a frame of a wing and use it as a full wing in that setup. I know KSP wings do a lot of dumb stuff, but this is an opportunity to do better. And you don't have to do fine simulation to do better. All you need to do is break the wing into smaller sections, each with its own surface area, chord, and CoP. Then just apply a standard lift and drag curves. Bonus points if you increase drag and reduce lift of the end sections to fake vortices. But even without that, simply simulating a wing as multiple sections would be an improvement, giving you both a more faithful simulation and better control over the planes you're building. And given that this is how all but top-of-the-line flight sims are set up anyways, it seems like it'd be a natural path forward with procedural wings and, by extension, flight dynamics in general.

I mean I don't want a aeronautical model like KSP, but I'm not going to get my hopes up too much for anything more involved.

Especially since there's a lot of people coming from KSP who see that "dumb stuff" as a feature.

 

[DeLacy]

I think the atmospheric part is a bit neglected in general, I don't play it at all simply because flying and tumbling around with afterburners gets old after a while.

[K^2]

1 hour ago, DeLacy said:

I think the atmospheric part is a bit neglected in general, I don't play it at all simply because flying and tumbling around with afterburners gets old after a while.

In terms of flight physics, I find it mostly salvageable. There are very specific ways in which you build aircraft to make them usable. But that's kind of true of real aircraft as well.

I think the bigger problem is that once you collected science and some contracts from areas near KSC there's just no reason to bother with flight. Occasionally, it makes sense to make a suborbital space plane for a contract, but this is just busy work at that point. You never use planes for anything fun after the initial stages of the game. And that's made worse by the fact that a lot of good airplane tech is locked pretty far up the research tree, so by the time you have access to it, you can't make good use of it.

Having airplanes as a starting point in career can make that all way better, IMO. And yeah, making aerodynamics a bit more intuitive and providing players with better tools to design functional planes would make it not feel like a waste of time.

[Frozen_Heart]

Really hope we get a more realistic model.  Would be nice for swept wings to actually have an effect.

[K^2]

4 minutes ago, Frozen_Heart said:

Really hope we get a more realistic model.  Would be nice for swept wings to actually have an effect.

To be fair, I don't believe swept wings do much until you get to transonic speeds, and I highly doubt the aerodynamics of the game will support that sort of nuance. For starters, it would turn designing functional aircraft into an absolute nightmare. Forgot to consider layer separation due to the fuselage profile? Well, you've just lost input from your tail control surfaces and are now in a supersonic dive into the terrain that you can't exit. Hope you didn't bring additional crew and can just eject the pilot!

Though, some simple effects, like a sharp increase in drag at transonic speeds requiring afterburners to "punch through" to supersonic might be a nice addition.

[DeLacy]

29 minutes ago, K^2 said:

I think the bigger problem is that once you collected science and some contracts from areas near KSC there's just no reason to bother with flight. Occasionally, it makes sense to make a suborbital space plane for a contract, but this is just busy work at that point. You never use planes for anything fun after the initial stages of the game. And that's made worse by the fact that a lot of good airplane tech is locked pretty far up the research tree, so by the time you have access to it, you can't make good use of it.

Having airplanes as a starting point in career can make that all way better, IMO. And yeah, making aerodynamics a bit more intuitive and providing players with better tools to design functional planes would make it not feel like a waste of time.

Totally agree. I personally find the career mode a hassle and like the science mode more but you simply progress too quickly through the tech tree imo so there's just not much impetus for exploring other outer or inner planets (except Duna and maybe Eve).

I really want them to rebalance the tech tree and science payouts so you actually have to explore a lot and make use of the MPL module, stations etc.

[K^2]

2 minutes ago, DeLacy said:

I really want them to rebalance the tech tree and science payouts so you actually have to explore a lot and make use of the MPL module, stations etc.

There should be a lot more to research, so that alone should give you an incentive to keep going to outer planets to unlock interstellar tech. But even so, early game needs a rebalance. I don't know if starting with aeronautics is the right decision, but I think that should be an option. Ditto probes before manned flight. There is honestly enough there that you can gives you flexibility in order in which you progress through the tech tree. Maybe first tier of unlocks should be a choice between probes, aircraft parts, or larger rockets? With probe branch containing enough smaller rocket parts to get a satellite into orbit? This on top of rebalancing how career and science work would make things a lot more interesting.

And to bring it a bit more on topic, I would really like to see props being improved and be viable as your first piece of airplane tech. The first plane you can build in KSP is a VLJ. The only reason to even consider props for anything practical is a tilt rotor. And even then, you're probably going to combine it with jet propulsion and hinges, using props only for VTOL. There's zero reason to build a conventional prop aircraft in the game, and it's kind of unfortunate.

[Jack Mcslay]

16 hours ago, K^2 said:

And to bring it a bit more on topic, I would really like to see props being improved and be viable as your first piece of airplane tech. The first plane you can build in KSP is a VLJ. The only reason to even consider props for anything practical is a tilt rotor. And even then, you're probably going to combine it with jet propulsion and hinges, using props only for VTOL. There's zero reason to build a conventional prop aircraft in the game, and it's kind of unfortunate.

Right now I have a propeller-driven UAV to be sent to Eve and take atmospheric measurements, so there's that. Also, we'll have more atmospheric bodies some of them probably without oxygen and there's a chance the devs could be inspired by space exploration in the past months and add a tiny double propeller that can be attached to a probe core and fly on Duna.

If we're to get a good propeller aircraft gameplay they better have good automated cruise controls, it can already get tedious to cross the ocean on Kerbin on a jet, let alone doing so in a much slower propeller aircraft.

Proper cockpits would be nice too, it's pretty silly to have cockpits that look like fighter jets for a propeller plane, not to mention unnecessary costly and heavy as they don't need to be pressurized.

[shdwlrd]

19 hours ago, K^2 said:

And to bring it a bit more on topic, I would really like to see props being improved and be viable as your first piece of airplane tech. The first plane you can build in KSP is a VLJ. The only reason to even consider props for anything practical is a tilt rotor. And even then, you're probably going to combine it with jet propulsion and hinges, using props only for VTOL. There's zero reason to build a conventional prop aircraft in the game, and it's kind of unfortunate.

I couldn't agree more. Unfortunately we don't know if any of the BG parts will make it into KSP2. Even if they initially do prop and rotor engines using the original method, (the same way as the turbine engines but with a ton of animations) would be worth while in the beginning stages of the game and possibly later on depending on what planets you're on.

2 hours ago, Jack Mcslay said:

If we're to get a good propeller aircraft gameplay they better have good automated cruise controls, it can already get tedious to cross the ocean on Kerbin on a jet, let alone doing so in a much slower propeller aircraft.

Proper cockpits would be nice too, it's pretty silly to have cockpits that look like fighter jets for a propeller plane, not to mention unnecessary costly and heavy as they don't need to be pressurized.

KSP1 has some good autopilots other than MJ. But I understand what you are getting at. It can be tedious flying a slow prop plane around, even worse with a helicopter since you have to keep on fussing with it. 

The lack of variation in fuselage profiles and cockpits has always drove me nuts. I do understand the reasons why they were never added. You don't need anymore redundant parts. The addition of switchable cockpit configurations would help with that some. It's not the end all solution, but it could help.

One thing KSP doesn't like is light craft with high lift. This will trigger the infini-glide issue which isn't ideal when learning to control planes in KSP. Since larger, heavier crafts will behave closer to how you expect them to.