Sign in to follow this  
GoldForest

A change to how thrust works

Recommended Posts

Right now, thrust is pretty simplistic. You engage thrust, it pushes your craft forward. If any part of your craft blocks said thrust, it negates all thrust and stops it from working. 

So, my suggestion is this:  Thrust is physical
Reason I'm suggesting this is for the problem of blocked thrust due to a wing or control surface or part. We've all been there. Built a plane and placed a part behind the jet engine, go to launch, turn on the engines, full thrust... and... you go no where because the thrust is blocked. 

So what I suggest is to make thrust act more like physical particles than a static invisible force. 

This would allow control surfaces and wings to be placed behind the engine inside the jet stream. It would also allow control surfaces to act like thrust vectors, pushing the thrust in any direction. 

I can also see this being useful for hidden engine builds like replica B-2, in which the engines are hidden inside the wings and there's exhaust ports for the thrust to come out. Since the wings wouldn't block the thrust, and instead direct it, it wouldn't have any problems. 

Share this post


Link to post
Share on other sites
4 hours ago, GoldForest said:

So what I suggest is to make thrust act more like physical particles than a static invisible force

It sounds desirable, and given enough computing power it would certainly make things act more intuitive/realistically.

But KSP is already heavily constrained by how much physics it has to calculate every frame. Calculating thrust the way you ask would add a lot of workload to that, potentially tanking the game for many players. Physics simulations, especially in games, tend to use simplifications and shortcuts exactly because of that, so a game remains fun to play.

 

5 hours ago, GoldForest said:

I can also see this being useful for hidden engine builds like replica B-2, in which the engines are hidden inside the wings and there's exhaust ports for the thrust to come out.

A stock way to do this currently is to take advantage of the fact that exhaust, regardless of the width of the visible graphics used, is calculated as a razor-thin line. Place wing sections or other useful parts strategically, leaving just the tiniest gap open, and thrust is no longer blocked. So it is already possible to build 'stealthed' exhaust ports. Just mind the gimbal range.

A little more difficult to use in a practical manner: exhaust damage is only calculated up to a certain maximum distance. Anything placed beyond that distance has zero effect on thrust.

Finally, with a little more handywork, the engine part cfg (or the craft file) can be edited to disable exhaustDamage entirely (this is how RCS, Ant/Spider, and the Dawn ion engine are excepted from exhaust blocking).

Share this post


Link to post
Share on other sites
1 hour ago, swjr-swis said:

It sounds desirable, and given enough computing power it would certainly make things act more intuitive/realistically.

But KSP is already heavily constrained by how much physics it has to calculate every frame. Calculating thrust the way you ask would add a lot of workload to that, potentially tanking the game for many players. Physics simulations, especially in games, tend to use simplifications and shortcuts exactly because of that, so a game remains fun to play.

Some rockets and space planes have ~100+ engines so real particle calculations would really cripple  big launches, no fun and a game is to maintain game play.

Share this post


Link to post
Share on other sites
Posted (edited)
18 hours ago, swjr-swis said:

It sounds desirable, and given enough computing power it would certainly make things act more intuitive/realistically.

But KSP is already heavily constrained by how much physics it has to calculate every frame. Calculating thrust the way you ask would add a lot of workload to that, potentially tanking the game for many players. Physics simulations, especially in games, tend to use simplifications and shortcuts exactly because of that, so a game remains fun to play.

 

A stock way to do this currently is to take advantage of the fact that exhaust, regardless of the width of the visible graphics used, is calculated as a razor-thin line. Place wing sections or other useful parts strategically, leaving just the tiniest gap open, and thrust is no longer blocked. So it is already possible to build 'stealthed' exhaust ports. Just mind the gimbal range.

A little more difficult to use in a practical manner: exhaust damage is only calculated up to a certain maximum distance. Anything placed beyond that distance has zero effect on thrust.

Finally, with a little more handywork, the engine part cfg (or the craft file) can be edited to disable exhaustDamage entirely (this is how RCS, Ant/Spider, and the Dawn ion engine are excepted from exhaust blocking).

 

16 hours ago, Aeroboi said:

Some rockets and space planes have ~100+ engines so real particle calculations would really cripple  big launches, no fun and a game is to maintain game play.


I fully understand that a system like this would restrict part limits even further, so I was thinking maybe making it an optional thing in the menu settings, or one further, the save file settings, so that it doesn't effect all saves, just the one where you enabled, what I'm going to call, 'Physical Thrust'.

Or another thing could be an Engine check box, allowing for more controllability of Physical Thrust. I think this would be the most likely answer. As @Aeroboi said, some people make uber space planes with 50+ Rapiers, so having the ability to make only about 4 to 8 of those physical thrust and have the rest be non-physical thrust would be beneficial. I know the rapier has thrust vectoring, but it's miniscule compared to say the Panther. If you could put a control surface behind the rapier engine and have it redirect the thrust, it would give the space craft a lot more maneuverability.

The same could be said about rockets, though, I don't know why you would use flap thrust vectoring on a rocket engine. Maybe if you were making a stealth rocket and the engine was hidden away inside. 

Edited by GoldForest

Share this post


Link to post
Share on other sites
13 minutes ago, GoldForest said:

The same could be said about rockets, though, I don't know why you would use flap thrust vectoring on a rocket engine. Maybe if you were making a stealth rocket and the engine was hidden away inside. 

That's how a lot of early prototypes were steered before we figured out engine gimballing.

Sorry mate, can't have this one unless there's a really effective programming trick you already know to accomplish it. There's no point going to the effort of physically simulating thrust deflection around airfoils or other objects when an abstraction will do the same job for a minuscule fraction of the performance hit, at the cost of odd edge cases. FAR is the closest we have and even that isn't doing exactly what you want here. That's just the price we pay for trying to simulate the physical universe, it gets exponentially more complex and resource intensive the closer you try to model reality. When it comes to things of minor importance like this, we just have to settle for less. Particle simulation in particular is a big no-no if you want it to run in realtime, we really want as few of those as necessary in stock KSP.

Share this post


Link to post
Share on other sites
12 minutes ago, GoldForest said:

@Aeroboi

 

The same could be said about rockets, though, I don't know why you would use flap thrust vectoring on a rocket engine. Maybe if you were making a stealth rocket and the engine was hidden away inside. 

Historically, it's because tilting a hinged metal flap at the bottom that is utterly unrelated to the rest of the engine operation is a far simpler feat of mechanical engineering.  So it was easier to solve the problem that way at first.  Gimbaling the engine means needing to feed either cold fuel or hot exploding gas through some kind of pivoting point in the "plumbing", and yet keep the characteristics of that flow absolutely identical through all possible positions that pivoting point can be set to.  That adds complexity.

Share this post


Link to post
Share on other sites

You could probably simulate it with some fancy modeling and using traditional gimbal. If the fins were modeled directly into the engine part I don't think they'd block exhaust. You could probably even get them to glow just like engine bells do. You could make something that looks like a V2 or Redstone engine with the fins it would just be one part that you snap on.

Share this post


Link to post
Share on other sites

I don't really see this as a problem that needs solving. modders have already found ways of building engines that use thrust vanes (using additional invisible thrust transforms) that work perfectly fine. The only benefit I see to this is adding thrust vanes yourself to engines, which could be done way simpler than a full particle simulation. Ie: detect what control surfaces are directly in the path of the thrust using a sphere cast, and doing some math to rotate the thrust transform in response to them. 

Share this post


Link to post
Share on other sites

It's actually unnecessary for physical reasons - chiefly, Einstein's Equivalence Principle. Acceleration is always, always caused by a continues bombardment of faster particles upon a slower particle or substance - except in the case of gravity, which is an acceleration caused by... Nobody knows. Lots of theories, some great, some garbage, but nobody knows. But THRUST is an acceleration caused by the exploding particles of the fuel striking the front (top) of the combustion chamber. They then bounce back or down and exit the chamber, in various ways of course depending on the engine.

Their exhaust doesn't CAUSE thrust, it is the RESULT of thrust. The thrust isn't pointing backwards or down, it's pointing forwards (or up). That's the vector you have to follow. Now a wing or fin behind the exhaust can cause DRAG, but drag is not thrust. Vectoring the thrust isn't just vectoring the exhaust, it's vectoring the forward-moving exploding substance that is CAUSING the thrust.

To sum up, the angle of the exhaust isn't the same thing as thrust at all. Anything that happens outside the engine doesn't affect what's in front or above it - that's action-at-a-distance, and is not physics. There are no actions at a distance in physics; there are no attractions, only repulsions. Even magnetism is a local lesser repulsion; it's not actually an attraction. All perceived attractions come from fields of less repulsion, relative to the local environment. All motion and therefore energy is caused by direct collision between masses, gravity aside.

The ship doesn't feel what happens AFTER its engines, since that's behind it already. All thrust happens INSIDE the engines, not behind them.

Share this post


Link to post
Share on other sites
10 hours ago, JaredTheDragon said:

The ship doesn't feel what happens AFTER its engines, since that's behind it already. All thrust happens INSIDE the engines, not behind them.

Well, it does feel it if the exhaust hits a another part of the ship, because then it creates thrust on that part. Deflected thrust can even move the ship backwards

In general, though, a raycast or...cylinder-cast? would be more effective and simpler than modeling particles

Share this post


Link to post
Share on other sites
This thread is quite old. Please consider starting a new thread rather than reviving this one.

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Sign in to follow this