Aero Revamp: What do You Want to See?

[Red Iron Crown]

The link between MASS and DRAG needs to be broken. Center of drag, center of mass, and center of lift are separate things. Currently in KSP the Center of mass IS your center of drag. This is what can cause your pointy delta wing space planes to flip backwards on re-entry. right now a heavy nose is also the point of most drag, rather than the big flat wing surfaces at the rear.

Nitpick: Your center of mass and center of drag are not necessarily identical in stock aero. Spam a bunch of high or low Cd parts at one end of a vessel and it will move the CoD a different amount than the CoM.

I agree though that basing drag on mass is a poor first-order approximation. KSP deserves a better system.

[Capt Snuggler]

Nitpick: Your center of mass and center of drag are not necessarily identical in stock aero. Spam a bunch of high or low Cd parts at one end of a vessel and it will move the CoD a different amount than the CoM.

I agree though that basing drag on mass is a poor first-order approximation. KSP deserves a better system.

huh, interesting. so the CoD is measured separately then, but just hidden from the player.

maybe it could be easier to fix...

annoyingly removing the fuel mass from the drag equation exasperates the base mass=drag issue so it would need a whole going-over to fix it.

[Xaiier]

Although you seem to be of the view that realism is irrelevant, you must be aware that this is very much not a universally held opinion.

The relatively realistic nature of KSP is what drew me to the game; if it wasn't there, I wouldn't play. I don't play any computer games apart from KSP (been a committed gamer for 30+ years, but my preferences run mostly to tabletop gaming), and I have no interest in physics-free space-themed fantasy. I have no problem with continuing to include low-realism options in the game for those that want them, but if they were the only option available I would rapidly abandon the game.

I am of the view that KSP would be best served by a very simple (if perhaps slightly improved over the current one) aerodynamics model, and that each player can then choose if they so desire to upgrade that with mods.

I really don't feel super strongly about this either way, however what really gets me going is when people toss "realism" all over the place as if that alone is a valid argument and join aerodynamics elitist clubs to bash stock together, exclaiming in disgust together how stock is such a terrible thing because it isn't realistic. This is why I reacted so harshly to your original post, as it was the same kind of post I've seen hundreds of times over in countless numbers of threads and it never accomplishes anything by being restated. Like I said in my initial post, this needs to be a discussion about gameplay and how that can best be served, not realism for its own sake.

[Artophwar]

I just want something that would simply simulate proper shaped rockets. Requiring Nosecones and fairings and benefitting properly shaped craft. I also want to see reentry damage made default with the new aero model. It is something that could be toggled in the difficulty settings. They said they were going to add reentry damage a long time ago when they made the reentry effects and I think it is something that should be in the final game as it is a fun gameplay mechanic to create craft that can survive reentry or watching thing horribly fail.

[Val]

I'd very much like to see a aerodynamics model that doesn't punish using nose-cones, and does punish flat pancake rockets.

I'm a fairly new player. I started in July. In the beginning, I tried to build long sleek rockets, because I thought it'd make it easier to punch through the atmosphere.

I was very disappointed to go on the forums and find out that it worked better to build out radially, because it has no appreciable aerodynamic impact.

I my mind it should be a trade-off between a long low-drag, but structurally unstable rocket, or a flat high-drag, structurally stable one. Right now long rockets are both high-drag and structurally unstable, which I think, is why people learn to build those whacky looking things...

I'd also very much like to see fairings and cargo bays shield their contents.

[Ziff]

I am of the view that KSP would be best served by a very simple (if perhaps slightly improved over the current one) aerodynamics model, and that each player can then choose if they so desire to upgrade that with mods.

The name of this thread is "What would you like to see?" and all you've said is you don't want realism. You've made that quite clear that for some reason the idea of realistic aerodynamics is a bad thing but you have made no case as to why that is. You've made no attempt to add to the thread discussion itself of "What would you like to see?" you've only said what you don't want. Your argument "I don't want realism for the sake of realism" is of no importance in this thread. Tell us what changes you would make to the aerodynamics system that improves gameplay. If you can make a case for an unrealistic system that betters the current one than maybe some people here might agree with you.

...however what really gets me going is when people toss "realism" all over the place as if that alone is a valid argument and join aerodynamics elitist clubs to bash stock together, exclaiming in disgust together how stock is such a terrible thing because it isn't realistic.

What really gets me going is people who think the word "realistic" is somehow inherently bad and detrimental to gameplay. Stock aerodynamics is terrible and it disgusts me to even think that adding aerodynamically shaped nosecones INCREASES drag rather than reduces it. It goes against every intuitive instinct. I've been playing since version .12 or something like that (right before the demo was released) and I never built planes until sometime around .24.2. I like rockets, so I built rockets. I don't remember exactly when they were introduced, but at some point I started adding nosecones to everything. You know it wasn't until I posted some pictures on the forum that someone had to tell me that all the nosecones I was adding were just useless dead weight and that they did nothing for the aerodynamics of my craft.

I am making a case for realism on the basis that we all have some basic knowledge of aerodynamics and it is intuitive to use what we know and understand when playing KSP. The current system is broken in those regards.

So, that being said, I would like to see an aerodynamic system based upon the shape of the parts and not based on the mass of the parts. One in which things like nosecones lessens the drag on a craft. That longer thin shapes are better than short ones.

Edit: I would also like to see an aerodynamic model in which airflow doesn't hit every part. Because that's basically what happens right now. Anything inside a cargo bay or a fairing increases drag because it increases mass. This is just silly and ridiculous.

Like I said in my initial post, this needs to be a discussion about gameplay and how that can best be served, not realism for its own sake.

No one here wants for realism because realism. We want realistic aerodynamics because it's better and more easily understood than the terrible substitute that is stock aerodynamics. I mean nothing else about this conversation needs to be said except Mass has more effect on drag than Shape. So if you have a better idea, feel free to join the discussion, if you don't then go make a thread called "Why we don't need realistic aerodynamics in 0.91."

Edited December 22, 2014 by Ziff
A word

[TheSandDuna]

I think it would be nice if the nosecones and possibly fairings could actually help reduce drag, rather than add weight and drag as they currently do. I'm sure that's in the cards eventually or else they wouldn't have created the nosecones in the first place. The other aspect I noticed is that I can pretty much land anything with enough parachutes on Kerbin because of the re-entry atmosphere will slow me down. Even at ridiculously deadly angles and from straight up scary heights there is no penalty and you can land crafts right now. They could work on that.

[michaelsteele3]

There should be an option somewhere (perhaps the ALT+F12 menu?) to enable/disable aerodynamics.

[tomf]

My problem witht the idea of drag being related to the cros sectional are of a rocket is that we have limited ability to fold things and build things in space. There are lots of things that in the real world you might expect to bolt together or unfold in space e.g. the lunar rovers that aren't possible in KSP. With the current aero you can just stick it on top of a rocket large enough to get it into space in one piece but if you had to wrap it in a fairing you could end up with some rockets that won't go to space today.

[Red Iron Crown]

My problem witht the idea of drag being related to the cros sectional are of a rocket is that we have limited ability to fold things and build things in space. There are lots of things that in the real world you might expect to bolt together or unfold in space e.g. the lunar rovers that aren't possible in KSP. With the current aero you can just stick it on top of a rocket large enough to get it into space in one piece but if you had to wrap it in a fairing you could end up with some rockets that won't go to space today.

It would just be more difficult, but not impossible. People launch all manner of silly things with FAR/NEAR, even without fairings.

[Ziff]

There should be an option somewhere (perhaps the ALT+F12 menu?) to enable/disable aerodynamics.

Disable it meaning.. what exactly? Disabling aerodynamics would mean planes can't fly at all. There would be no lift generated by wings. I think if by disabling aerodynamics you mean revert aerodynamics to the system we have now you just don't understand how terrible the current system is.

[Laie]

Well, I want needle-shaped rockets to work better than pancakes. I want nosecones to be beneficial. I also want splashdowns to be safer than landing on the ground (I guess that fixing aero may also help with other fluids).

More important than the specifics of the new aerodynamics would be to have more rigid joints that actually allow for the construction of tall, single-stack rockets. My typical rocket is 10-20% struts by part count; I often use several columns not because they're necessary, but because they can be strutted against each other. When and if I go single-stack, I often find myself slapping on radial parts for better strutting. The result is reminiscent of a suspension bridge.

[Kegereneku]

I would like to attract attention on two ongoing exaggerations in this discussion.

When people say they don't want too much realism, or (like me) warn that realism =/= better. It doesn't mean those peoples (or me) don't want a definite improvement over KSP's -now inadequate- aerodynamic model. Just as nobody will argue that they want nose cone to increase drag. They just don't want to spend 5 more hours designing a single planes/rockets over meaningless details that have no place in a videogame and only please a vocal minority.

In the same vein, not everyone arguing with the phrasing "more realistic" necessarily wish the FAR / NEAR extreme or mind what they want (ex : reentry) being done is a resolutely more fun-driven than strictly realist way. An easy example being how rocket engines were made purposefully weaker so you need multi-stage rocket as if it was real without other inconvenience of reality.

I have seen more than a few players root for extreme implementation because they feared the game to underdeliver, then be shockingly pleased/challenged with a simplified game-logic that satisfied entirely the feeling of verisimilitude they actually wanted, not wishing more.

So it might be worth putting nuance where it's due more often. Myself I DO want an improvement over the actual model, I do want features such as "lessening drag through fairing/cargobay", engine shocking with speed (yes it's possible) or even "reentry heat" but there is such a thing as acceptable break from reality or too much realism.

Obsessing toward realism only however is wrong. This is not a bad word but not a solution-to-everything as some make it out to be, to the point you get to question one's understanding of the subject. The ones arguing for more realism don't necessarily get right what it entails.

I've said it before and I'll continue as needed : KSP does not make you a "real" spaceplane-engineers, so it is only adequate the problems you face are equivalent.

Although you seem to be of the view that realism is irrelevant, you must be aware that this is very much not a universally held opinion.

The relatively realistic nature of KSP is what drew me to the game

Excuse me to intrude, but the key word in your post is "relatively". I also wanted to point out that the idea of realism being the only thing that matter is an even less held opinion, else we would be playing Orbiter.

[SkyHook]

I think the biggest thing I want is a sense of logic and predictability. If if LOOKS like it should fly, then it should. If it looks like a crazy contraption, it should kill its creator, like crazy contraptions do IRL

[Red Iron Crown]

No one is saying those things, Kegereneku. Your statement that designing craft will take 5 more hours is not supported at all, nor is the idea that only a "vocal minority" wants more realistic aero (we have no real numbers to go by on where the player base stands on it). We all want KSP to be fun, no one is obsessing over wanting a hyper realistic Orbiter clone. We're discussing how the aero model can be made better, something even Squad acknowledges is desirable and planned.

I strongly doubt that there is anyone who thinks the new model will be less or equally realistic as the current model. The discussion is about how a more realistic model should work, and what will be fun.

[NathanKell]

In fact, the only poll conducted on the forum thus far showed a quite large majority wanting FAR implemented. That said, it's a majority of those who took the poll, but...that's the data we have.

[allmhuran]

There seem to be attacks directed at Xaiier from all sides, so I'm here to provide support.

As I understand it, Xaiier isn't suggesting that "more realistic" aero (drag in particular) is necessarily bad. The suggestion is simply that "realism for the sake of realism alone" is bad. He even said exactly this in a previous post.

I think those who take objection with that position need to consider things a little further, so let me break it down. Either you agree with the statement "realism for the sake of realism alone is not a good idea", or you do not.

If you agree then cool, you're not actually opposed to what Xaiier is saying.

If you do not agree, that is, if you are of the opinion that realism for the sake of realism is good, then you have some major problems, friends. If you believe this then you surely must apply your belief consistently. So, no more launch and build in a day for you, because realistically rockets take years to build. Don't like random part failures? Too bad, because realistically parts randomly fail. Ever land your plane somewhere other than on a runway? Well, realistically, your wheels would probably be digging into the grass or dirt and your plane would be kaput. Do you enjoy driving rovers around? Not for long! Realistically your electricity requirements would be vastly more than they are in KSP, and your electricity generation would be much slower. The list goes on and on. So it seems that the call for "realism because realism" is calling for a hell of a lot of things that would make KSP a much less enjoyable game.

I'd like pointy things to go faster, and I'd like wings to stall, because pointy things going faster means people have a goal that they can work their design towards, and wings stalling would make flight more dynamic and fun.

Edited December 23, 2014 by allmhuran

[Xaiier]

The name of this thread is "What would you like to see?" and all you've said is you don't want realism. You've made that quite clear that for some reason the idea of realistic aerodynamics is a bad thing but you have made no case as to why that is. You've made no attempt to add to the thread discussion itself of "What would you like to see?" you've only said what you don't want. Your argument "I don't want realism for the sake of realism" is of no importance in this thread. Tell us what changes you would make to the aerodynamics system that improves gameplay. If you can make a case for an unrealistic system that betters the current one than maybe some people here might agree with you.

I have already expressed my opinion and reasoning for such in my back and forth with RedIronCrown. I would like a system like our current one except where cargo bays properly shield things inside and nosecones actually work.

What really gets me going is people who think the word "realistic" is somehow inherently bad and detrimental to gameplay. Stock aerodynamics is terrible and it disgusts me to even think that adding aerodynamically shaped nosecones INCREASES drag rather than reduces it. It goes against every intuitive instinct. I've been playing since version .12 or something like that (right before the demo was released) and I never built planes until sometime around .24.2. I like rockets, so I built rockets. I don't remember exactly when they were introduced, but at some point I started adding nosecones to everything. You know it wasn't until I posted some pictures on the forum that someone had to tell me that all the nosecones I was adding were just useless dead weight and that they did nothing for the aerodynamics of my craft.

You say stock aerodynamics is terrible. Compared to what? Reality? This is exactly what I am talking about, the basis for your unjustified claims is entirely reliant upon comparing KSP to real life. That a nosecone increases drag is inherent in the mass->drag physics of the Kerbal universe, that may be a dumb design from a gameplay perspective, but there isn't anything inherently wrong with it.

I am making a case for realism on the basis that we all have some basic knowledge of aerodynamics and it is intuitive to use what we know and understand when playing KSP. The current system is broken in those regards.

As I stated previously, my experiences with KSP have not led me to believe that people have natural intuition about how all this works.

So, that being said, I would like to see an aerodynamic system based upon the shape of the parts and not based on the mass of the parts. One in which things like nosecones lessens the drag on a craft. That longer thin shapes are better than short ones.

Edit: I would also like to see an aerodynamic model in which airflow doesn't hit every part. Because that's basically what happens right now. Anything inside a cargo bay or a fairing increases drag because it increases mass. This is just silly and ridiculous.

Agreed.

No one here wants for realism because realism. We want realistic aerodynamics because it's better and more easily understood than the terrible substitute that is stock aerodynamics. I mean nothing else about this conversation needs to be said except Mass has more effect on drag than Shape. So if you have a better idea, feel free to join the discussion, if you don't then go make a thread called "Why we don't need realistic aerodynamics in 0.91."

If you grew up in the Kerbal universe where mass = drag and played a game called Human Space Program in which drag was based on physical shape you could be making the exact same arguments that those mechanics are ridiculous and should be changed. Because I disagree that "easily understood" is inherent in realistic aerodynamics, you are going to have to provide reasoning as to how it would improve the gameplay, and how the drawbacks from doing such would be outweighed by the benefits.

[Wanderfound]

You say stock aerodynamics is terrible. Compared to what? Reality? This is exactly what I am talking about, the basis for your unjustified claims is entirely reliant upon comparing KSP to real life. That a nosecone increases drag is inherent in the mass->drag physics of the Kerbal universe, that may be a dumb design from a gameplay perspective, but there isn't anything inherently wrong with it.

As I stated previously, my experiences with KSP have not led me to believe that people have natural intuition about how all this works.

Agreed.

If you grew up in the Kerbal universe where mass = drag and played a game called Human Space Program in which drag was based on physical shape you could be making the exact same arguments that those mechanics are ridiculous and should be changed. Because I disagree that "easily understood" is inherent in realistic aerodynamics, you are going to have to provide reasoning as to how it would improve the gameplay, and how the drawbacks from doing such would be outweighed by the benefits.

Stock aero is terrible (to me, in my personal subjective opinion) compared to FAR, and, yes, reality.

I know what real planes look like. I had a basic idea of why they had tailfins and ailerons and things (that basic idea is now a lot less basic, thanks to my experience with the modded game). So, when I started designing, I built things that looked like real planes.

They flew okay, because just about anything will fly in stock aero. But it didn't feel like flying a plane. They took forever to climb or get up to speed, glided like bricks and gripped the air as if were made of toffee. Air is fast, air is slippery; none of these sensations came through the game. Flying them to orbit and back just wasn't much fun.

And then I discovered that to make them perform better, or to get larger planes off the ground at all, I'd be best off turning it into a ridiculous looking Christmas tree of parts rather than try to learn how to construct something sensible.

And then I discovered FAR. Never flying stock again.

[Xaiier]

Stock aero is terrible (to me, in my personal subjective opinion) compared to FAR, and, yes, reality.

I know what real planes look like. I had a basic idea of why they had tailfins and ailerons and things (that basic idea is now a lot less basic, thanks to my experience with the modded game). So, when I started designing, I built things that looked like real planes.

They flew okay, because just about anything will fly in stock aero. But it didn't feel like flying a plane. They took forever to climb or get up to speed, glided like bricks and gripped the air as if were made of toffee. Air is fast, air is slippery; none of these sensations came through the game. Flying them to orbit and back just wasn't much fun.

And then I discovered that to make them perform better, or to get larger planes off the ground at all, I'd be best off turning it into a ridiculous looking Christmas tree of parts rather than try to learn how to construct something sensible.

And then I discovered FAR. Never flying stock again.

Thats fine, thats why we have FAR. The ability for each player to customize their experience is one of the great things about KSP. That being said, there are a few things I don't want:

-FAR being implemented directly into stock, thus removing the players choice (obviously someone could make a dumbed down aero mod, but that isn't optimal)

-stock implementing a half-baked version of FAR which ends up screwing up the current stock design possibilities but still leaves FAR supporters unhappy

-any significant change which would make FAR impossible, which is one of Ferram's main worries with the new aero. It would not take much for FAR to become technically unfeasible within the new system, and SQUAD hasn't been the best lately about making new features easily moddable.

[Boomerang]

As someone who's toyed around with FAR and NEAR both and ultimately stuck with the stock aero model, I'd like to see:

-Drag shielding on parts within cargo bays, perhaps on parts completely within the collision box of other parts as well

-For shape to matter when it comes to drag. Broad rockets don't need to be impossible, but nosecones should be beneficial and awkward pancakes should behave, well, much more awkwardly

-To have a reasonably calculated center of drag, as well as an indicator for it in the editors, a la the CoM/CoL/CoT

As far as atmospheric density goes, if it's a choice between the 'soup-o-sphere' of stock or the thinness of FAR, I'd rather they stick with the former. While not terribly realistic as-is, I feel like greatly reducing the atmospheric density would require a boat-load of balancing to be done to the game across the board, in order to keep stock about as challenging as it is now.

That means looking into nerfing all engines, probably reworking the scale height of the atmosphere, and seriously considering how practical we want spaceplanes to be. Most people are going to seriously struggle building SSTO spaceplanes if you take them out of the stock atmo and drop them into something like NEAR.

That being said, if there's an option to mildly reduce the atmospheric density without hugely changing flight behavior and without having to rebalance most aspects of the game, then I'd be alright with that as well.

Essentially, like many people, I'd like things to behave in more reasonable ways within atmospheres. However, this is meant to be a game, rather than a simulator. I don't think it's terribly reasonable to expect a very nearly realistic aero system when things like the sizes and densities of celestial bodies have been handwaved around so much make room for generally realistic behavior in a scaled-down system.

Make the stock aero system behave more reasonably than it does now. But leave a highly-realistic system to something like RSS and the RO suite.

[Ippo]

Whatever they choose to make, I hope they won't break FAR because I will be staying with it anyway - I'm not even going to wonder what aero model they'll make becaus I already know it's never going to be on par with FAR. I bought this game on the promise of realistic physics simulation, as advertised on their site, and that's what I want from the game.

l;Dr: Realism for realism sake is not that bad if that's what you thought you were buying.

[Jouni]

I think there are three main things that should be changed in stock aerodynamics:

1. Drag should depend on the size, the shape, and the attitude of the craft, instead of its mass. This would encourage building rockets that look like rockets, make atmospheric reentries more intresting, and get rid of the slow initial part of the ascent.
   
2. Reentry damage should be implemented. It's the only major part of the game that's still missing.
   
3. Lift should be proportional to the square of the airspeed. Combined with the reduced drag for aerodynamic shapes, this would make planes depend less on having ridiculous amounts of thrust. As a result, jet engines could be nerfed to the point that they would no longer be viable engines for rockets.
   

[Darnok]

As someone who's toyed around with FAR and NEAR both and ultimately stuck with the stock aero model, I'd like to see:

-Drag shielding on parts within cargo bays, perhaps on parts completely within the collision box of other parts as well

-For shape to matter when it comes to drag. Broad rockets don't need to be impossible, but nosecones should be beneficial and awkward pancakes should behave, well, much more awkwardly

-To have a reasonably calculated center of drag, as well as an indicator for it in the editors, a la the CoM/CoL/CoT

As far as atmospheric density goes, if it's a choice between the 'soup-o-sphere' of stock or the thinness of FAR, I'd rather they stick with the former. While not terribly realistic as-is, I feel like greatly reducing the atmospheric density would require a boat-load of balancing to be done to the game across the board, in order to keep stock about as challenging as it is now.

That means looking into nerfing all engines, probably reworking the scale height of the atmosphere, and seriously considering how practical we want spaceplanes to be. Most people are going to seriously struggle building SSTO spaceplanes if you take them out of the stock atmo and drop them into something like NEAR.

That being said, if there's an option to mildly reduce the atmospheric density without hugely changing flight behavior and without having to rebalance most aspects of the game, then I'd be alright with that as well.

Essentially, like many people, I'd like things to behave in more reasonable ways within atmospheres. However, this is meant to be a game, rather than a simulator. I don't think it's terribly reasonable to expect a very nearly realistic aero system when things like the sizes and densities of celestial bodies have been handwaved around so much make room for generally realistic behavior in a scaled-down system.

Make the stock aero system behave more reasonably than it does now. But leave a highly-realistic system to something like RSS and the RO suite.

We have options now, so both your idea and NEAR or FAR are possible in stock.

[eddiew]

FAR, but with a game option to turn off disintegration (default off for easy/normal). Stuff flies much better, but it's no fun watching your early rockets fall apart because you tried to turn just a little too fast, or having your first space plane disintegrate on re-entry after an otherwise successful mission...

Primary reason for calling FAR over NEAR is the sim window in the SPH. With NEAR, you're still pretty much guessing at stability, where FAR lets you know in advance if you're gonna die. Lol, it could be made a little more friendly though. The decimal point accuracy of the display is a bit overkill and scary