Jump to content

Please Add Realistic Aerodynamics (FAR) and Perhaps Procedural Nozzles for Optimizing at Different Pressures


Recommended Posts

Realistic aerodynamics is not only an essential topic for aerospace engineering, it is absolutely FASCINATING.

I wanted to break the sound barrier with the Farrem Aerospace mod and the Realistic Progression mod tech tree. This led me to spend a whole 2 days researching aerodynamics, including watching the old Shell Oil film "High Speed Flight," which explained all the considerations of transonic and supersonic flight. This included elevator issues, thin wings, the area rule, sweeping wings behind the shock wave so they experience subsonic flow, etc.

I spent 2 days excitedly researching, but I had also taken both incompressible and compressible flow courses. Think of all the fun I would have had if I were to discover those aerodynamic principles for the first time!

If the developers want to make a game marketed as being a fun way to educate the next generation of engineers and scientists, they should go all the way and implement realistic aerodynamics so players can try to break the sound barrier and optimize the aerodynamic profile of their spaceplane for supersonic flight.

Procedural nozzles could even be added just to experiment with optimizing engine efficiency for different atmospheric pressures (future update?). Make the game the ultimate sandbox of aerospace engineering!

(The same goes for including realistic n-body gravitation like in the principia mod! But while it can and should be implemented I know it probably won't be...)

Edited by Spartan Hoplite
Link to comment
Share on other sites

14 minutes ago, Spartan Hoplite said:

If the developers want to make a game marketed as being a fun way to educate the next generation of engineers and scientists,

I don't really have an opinion on the whole topic. But that's not the point of Kerbal, it's not an educational game at its core, it's a fun game for people that are into space stuff, it does inspire a lot of people to pursue a career or an education into space related stuff, but it's not an educational game, with all the bells and whistles that that would bring.

That means that when deciding between 100% realism and a more fun and engaging gameplay experience, gameplay should win.

Link to comment
Share on other sites

8 hours ago, Master39 said:

...But that's not the point of Kerbal, it's not an educational game at its core, it's a fun game for people that are into space stuff...

That means that when deciding between 100% realism and a more fun and engaging gameplay experience, gameplay should win.

Can you think of any game designed as educational that is fun?

KSP is in a unique position where adding more realistic physics and engineering elements is an opportunity to make it more fun, not less. Besides, having  these features as toggleable difficulty settings makes it so people can choose whatever gameplay they want, giving the developers a lot of freedom with what they can add to the game. Players already do this with mods, but having features supported in the stock game removes stability issues and allows for more advanced modding.

The only issues with advanced features are the time required to implement and that the developers could mess it up by not utilizing the features to create compelling gameplay.  Turning physics and engineering concepts into a great game is an ambitious idea for ambitious developers.

Edited by Spartan Hoplite
Link to comment
Share on other sites

17 minutes ago, Spartan Hoplite said:

KSP is in a unique position where adding more realistic physics and engineering elements is an opportunity to make it more fun, not less

That adds complexity, which for many people would make it less fun.

18 minutes ago, Spartan Hoplite said:

only issues with advanced features are the time required to implement and that the developers could mess it up by not utilizing the features to create compelling gameplay.  Turning physics and engineering concepts into a great game is an ambitious idea for ambitious developers.

I can only suspect you aren't a game developer. Those issues are huge, which is why KSP didnt implement them initially. 

Link to comment
Share on other sites

9 hours ago, Spartan Hoplite said:

2 days researching aerodynamics, including watching the old Shell Oil film "High Speed Flight," which explained all the considerations of transonic and supersonic flight. This included elevator issues, thin wings, the area rule, sweeping wings behind the shock wave so they experience subsonic flow, etc.

I spent 2 days excitedly researching, but I had also taken both incompressible and compressible flow courses.

Procedural nozzles could even be added just to experiment with optimizing engine efficiency for different atmospheric pressures

That does not sound like fun to me. It may be to some, but if my Panther powered jet won't make 650m/s when I need it, I just won't bother.

Some people say that FAR is actually easier than stock aero model, after reading your post I highly doubt that.  Making stock propeller planes is hard enough with controlling their deploy angles for speed etc etc. Don't need that for everything else.

Link to comment
Share on other sites

9 hours ago, Spartan Hoplite said:

(The same goes for including realistic n-body gravitation like in the principia mod! But while it can and should be implemented I know it probably won't be...)

It won't and shouldn't be implemented in the core game, that's what mods are for. Making it part of the game would ruin gameplay for many players as orbiting stations and satellites would start drifting from their original orbits and would require to be adjusted constantly or could end up being accidentally slingshotted away of the orbiting body, which could be frustrating. In a mod this becomes an optional challenge.

 

I haven't used FAR, but from what has been said here, I agree it should stay as a mod too. 

Link to comment
Share on other sites

40 minutes ago, The Aziz said:

That does not sound like fun to me. It may be to some, but if my Panther powered jet won't make 650m/s when I need it, I just won't bother.

Everyone has had to do some research to play KSP, even if it was from watching Youtubers. In-game tutorials could make the experience less of a chore for those that don't like researching interesting topics like I do.

Its not like I had to do any math to understand the topic or to build a craft with it. FAR does all the math and makes it easy to see. This could be made with even better UI.

I'm sure lots of people feel satisfied when building something in KSP makes them feel smarter than they actually are need to be! (No one would argue engineering is as easy as KSP)

I think difficulty settings are key here. An option to model aerodynamics simply like KSP1 could surely be added, satisfying everyone.

Edited by Spartan Hoplite
Link to comment
Share on other sites

58 minutes ago, The Physics Engineer said:

It won't and shouldn't be implemented in the core game, that's what mods are for. Making it part of the game would ruin gameplay for many players as orbiting stations and satellites would start drifting from their original orbits and would require to be adjusted constantly or could end up being accidentally slingshotted away of the orbiting body, which could be frustrating.

Hypothetical: There will already be automatic resupply. If n-body is added (which I know it won't be), a feature for stations to automatically correct their orbits could also be implemented. The same would apply for if orbital drag were implemented, that way the player would only have to worry about maintaining orbits for craft being currently piloted

1 hour ago, The Physics Engineer said:

In a mod this becomes an optional challenge.

I believe the developers should take full advantage of difficulty settings to add optional challenges

Link to comment
Share on other sites

1 hour ago, Spartan Hoplite said:

A feature for stations to automatically correct their orbits could also be implemented. The same would apply for if orbital drag were implemented, that way the player would only have to worry about maintaining orbits for craft being currently piloted

So... A feature that changes literally nothing in player's experience because everything is automatic anyway.

Link to comment
Share on other sites

3 hours ago, The Aziz said:

That does not sound like fun to me. It may be to some, but if my Panther powered jet won't make 650m/s when I need it, I just won't bother.

Some people say that FAR is actually easier than stock aero model, after reading your post I highly doubt that.  Making stock propeller planes is hard enough with controlling their deploy angles for speed etc etc. Don't need that for everything else.

I will say in my experience with FAR it presents the right tools to go properly engineer a craft and is approachable with limited knowledge on the subject. Also, I looked up one of the videos @Spartan Hoplite mentioned and I find it is brief, enjoyable, and informative. If you would care to look it over:

I get that adding these nuances isn't for everyone and I get it if it doesn't make it into the stock game but I would be pleasantly surprised. I will say though that the stock aero model does at the very least need some very real improvement. I think many of us have gotten used to or were able to overlook its faults but it definitely has many of them. The cube drag model is over simplified and makes heatshields and fairings act cartoonishly in regards to their drag or lack there of it, drag along blunt surfaces can be nearly erased by attaching a part to a node then clipping it, etc... If I had to choose between the two models which I would prefer to step into then the  more realistic version would be my choice. I say this not because I think KSP should strive for realism but because I find more realistic model is more fun, has deeper nuance to explore, is less prone to glitchy exploitation, and is actually more intuitive in most regions.

Link to comment
Share on other sites

28 minutes ago, mcwaffles2003 said:

I will say in my experience with FAR it presents the right tools to go properly engineer a craft and is approachable with limited knowledge on the subject. Also, I looked up one of the videos @Spartan Hoplite mentioned and I find it is brief, enjoyable, and informative.

I'm glad you appreciated the video! I think something similar could be condensed and put into the game.

28 minutes ago, mcwaffles2003 said:

I will say though that the stock aero model does at the very least need some very real improvement. I think many of us have gotten used to or were able to overlook its faults but it definitely has many of them.

This reminds me that stock KSP doesn't even have airfoils, just flat wings! Many real wings are designed to generate lift with 0 angle of attack. Also flat wings just look ugly

Edited by Spartan Hoplite
Link to comment
Share on other sites

2 minutes ago, Spartan Hoplite said:

This reminds me that stock KSP doesn't even have airfoils, just flat wings! Many real wings are designed to generate lift with 0 angle of attack.

KSP flat wings till simulate airfoils in that they generate lift with no angle of attack, just they're not represented graphically. This bugs me though since the lift flips when you fly inverted.

Link to comment
Share on other sites

One thing that needs to be understood about lift and drag in KSP1 is that they are user defined values in the config files. The drag cube(s) can be calculated automatically by the model or defined by the user. The center of mass, lift, and pressure is calculated by the model or can be defined by the user. That's where the quirkiness for aerodynamics comes into play. All the parts are balanced based on their shape and intended roll. If you use them outside of their roll, you will get some interesting aerodynamic features from them.

In all reality, the aero simulation in KSP is fine. What needs to happen is the aerodynamics of the parts needs to be refined. 

Link to comment
Share on other sites

There are many choices between 100% realism and a design driven purely by game-play considerations,
and in many cases better realism helps with game-play.

KSP1 gradually evolved an aerodynamics model while the designers were learning what the game should be.   So it has rules that roughly simulate how lift and drag behave at trans-sonic speeds, and how the shapes of parts affect drag, but those rules are very strange sometimes

On the aircraft challenges, you see we spend some effort learning where KSP1 does things differently to real physics.  We can hope that KSP2 can simulate more realistic aerodynamics, so we are not induced to learn arbitrary rules, and so our competitive designs can be done more intuitively,

There are a few points where KSP1 did have a choice between realism and game-play:
+ larger lift and drag at low speeds, allowing supersonic aircraft to land on a short runway
+ a very gradual 'soft' stall, making it easier to land aircraft.
+ symmetric aerofoils, so they can go on rockets  without making the rocket spin (Real symmetric aerofoils are not that much less efficient than cambered aerofoils )

Some of these choices are stored as numbers and curves in *.cfg files (both in KSP1 and in FAR) so there is hope that it will be relatively simple to crank up the realism in KSP2.

Link to comment
Share on other sites

32 minutes ago, shdwlrd said:

In all reality, the aero simulation in KSP is fine. What needs to happen is the aerodynamics of the parts needs to be refined. 

this_is_fine.jpeg

 

_________________________________________________________________________________________________________________________________________________________________________________________________

 

This is one of the things I really like about FAR... why make all these different variables for each individual part hoping to guess the users intention when you can just map the surface of a vessel after its been completed and have it act as it should in air? Doing things piece by piece like this leaves large room for exploitation and you get heat shields acting better than wings and propellers at their specified job...

 

This is not fine... but it is hilarious

Edited by mcwaffles2003
Link to comment
Share on other sites

Well, FAR also describes wings with a set of variables (listed under FARWingAerodynamicModel in the  configuration files) but it still works pretty well.

The exploit with the heat shields is mostly due to the way KSP1 removes drag on surfaces where fitting parts are node-attached (since version 1.0,5).  The unusual lift of heat shields makes some contribution to the exploit, but zero drag is the big loophole.  

Link to comment
Share on other sites

@mcwaffles2003 I guess you missed one major part of my statement. Let me quote myself.

2 hours ago, shdwlrd said:

All the parts are balanced based on their shape and intended roll. If you use them outside of their roll, you will get some interesting aerodynamic features from them.

In the clip, they are using heat shields as props. That is using a part not as intended. That's why you are seeing that effect.

Link to comment
Share on other sites

19 minutes ago, shdwlrd said:

@mcwaffles2003 I guess you missed one major part of my statement. Let me quote myself.

In the clip, they are using heat shields as props. That is using a part not as intended. That's why you are seeing that effect.

I didn't miss it, that's the problem. The role of a part is determined by the maker using it despite what is intended by those who created the part. The piece by piece system is far more convoluted, susceptible to misuse, and prone to bugs than FAR. Also, just because FAR is more realistic btw doesn't inevitably mean it is harder to use, there's just more depth to which a player can refine a craft. The general rule of CoL goes right behind the CoM still applies and most reasonable KSP craft would work fine with for most of the regions a typical player would fly a plane. Not to mention isn't this kind of what the game is about? Building craft in a world of unyielding physics and learning through the art of engineering and reiteration to build really cool stuff.

In short, I would argue your position to keep the aero model as is comes with the caveats of intended use and you haven't given any reason why introducing and aero model similar to that of FAR would be detrimental to the player experience.

Link to comment
Share on other sites

17 hours ago, Spartan Hoplite said:

If the developers want to make a game marketed as being a fun way to educate the next generation of engineers and scientists, they should go all the way and implement realistic aerodynamics so players can try to break the sound barrier and optimize the aerodynamic profile of their spaceplane for supersonic flight.

Procedural nozzles could even be added just to experiment with optimizing engine efficiency for different atmospheric pressures (future update?). Make the game the ultimate sandbox of aerospace engineering!

KSP 2 is an expansive game, I consider aerodynamics to be important, but not the forefront problem the game wants to present and challenge the player with. 

I'm all for a revamp/improvements to the underlying aerodynamics model for KSP 2, but I would not expect it to be aiming for complete realism, the same way its orbital mechanics also doesn't aim for complete n-body simulations. As long as the updates don't take away from the core gameplay KSP 2 should be focusing on, I'm all for it as it is part of the overall challenge. 

that said:

KSP 2 can be a great game without being 100% realistic in all aspects. 

 

 

 

 

Link to comment
Share on other sites

17 minutes ago, mcwaffles2003 said:

In short, I would argue your position to keep the aero model as is comes with the caveats of intended use and you haven't given any reason why introducing and aero model similar to that of FAR would be detrimental to the player experience.

I was letting you know how the parts are setup in the stock aerodynamic system and how they are balanced. That's all. I can't be objective about FAR because I've had horrible experiences using it. Planes that are properly balanced and constructed are either uncontrollable or don't fly. Rockets that can't make orbit even though everything is over built. (4200dv and 6.9twr, no flat leading edges) Planes that explode with the slightest g-load spike. (Less than 1g) Or once you reach above 3gs. (Really! Even commercial airliners can survive 3gs without breaking apart. Granted they will need inspection afterwards before flying again. Fighters can handle 8+gs within their normal flight regime.) Unreasonably high takeoff, landing, and stall speeds. No effective glider like effects because of the high stall speeds. Basically making the game unplayable for me. :mad:

The only good thing FAR did for me was tell me about KE (KER).

Link to comment
Share on other sites

8 hours ago, Spartan Hoplite said:

I believe the developers should take full advantage of difficulty settings to add optional challenges

It could be nice to have it as some sort of hardcore or legendary mode, but I don't think they will direct their efforts towards it. It's probably better for them to provide an easy way to install mods so that the players who want it don't have major problems getting those mods.

5 hours ago, Spartan Hoplite said:

This reminds me that stock KSP doesn't even have airfoils, just flat wings! Many real wings are designed to generate lift with 0 angle of attack. Also flat wings just look ugly

This would definitely be a nice addition!

Link to comment
Share on other sites

1 hour ago, shdwlrd said:

I was letting you know how the parts are setup in the stock aerodynamic system and how they are balanced. That's all.

Didn't mean to sound like I was confrontational if that is how I came off.

1 hour ago, shdwlrd said:

Planes that explode with the slightest g-load spike. (Less than 1g) Or once you reach above 3gs. (Really! Even commercial airliners can survive 3gs without breaking apart. Granted they will need inspection afterwards before flying again. Fighters can handle 8+gs within their normal flight regime.) Unreasonably high takeoff, landing, and stall speeds. No effective glider like effects because of the high stall speeds. Basically making the game unplayable for me. :mad:

The only good thing FAR did for me was tell me about KE (KER).

Did you by chance play with FAR as part of the realism overhaul package? To me that is what it sounds like. If you ever care to try it again right click on any lifting surfaces and note the mass/strength multiplier slider. Sliding it up should be of great help.

 

Edited by mcwaffles2003
Link to comment
Share on other sites

17 hours ago, mcwaffles2003 said:

Did you by chance play with FAR as part of the realism overhaul package? To me that is what it sounds like. If you ever care to try it again right click on any lifting surfaces and note the mass/strength multiplier slider. Sliding it up should be of great help.

Last time I played with FAR was in .90. After 1.0 I've lost interest because the aero model was serviceable.

Link to comment
Share on other sites

3 minutes ago, shdwlrd said:

Last time I played with FAR was in .90. After 1.0 I've lost interest because the aero model was serviceable.

Idk what FAR was like back then but I think it's fair to say a lot has changed in the past 5+ years. It's worth a look at.

Link to comment
Share on other sites

×
×
  • Create New...