KSP 1 exploits (useful bugs) that you do OR don't want to be in KSP 2

[Vl3d]

I've been thinking about some bugs that are actually useful for players building exotics. The thing is.. I really don't like this type of exploit-based way of playing the game. I feel like the game should aim for rather grounded engineering principles and physics that (if possible) are closer to the realistic baseline (while keeping the fun in the game). Of course, some people love them and want them in the game. I will focus on the negatives while keeping a positive attitude first.

I have a few examples we can discuss - things that are in KSP 1 - but that I wish were not or will not be present in KSP 2:

• Part clipping that does not realistically calculate combined walls mass / area or combined internal volume. I hate parts clipping one inside another but keeping the functionality of both. I feel like this is not possible in real life - we can only clip by cutting / combining walls to make a bigger container with a certain internal volume, but we cannot duplicate that volume.
• So this means I don't like vessels that look small but have N fuel tanks or batteries or wings or solid rocket boosters inside.
• I don't like how we can move parts away from the root so they look like they're floating in the air. There should be visible struts / supports for any floating part.
• I don't like the high lift - low drag exploit using the heat shields. https://www.youtube.com/watch?v=wOlMlRf9qvo
• I don't like the decouplers exploit and other kraken drives that unrealistically allow for compounding force.
• Parts clipping in the ground when vessels reload.
• Being able to visibly pull apart parts (like docking ports) that are actually still stuck together.
• Unrealistically limp rubber rockets or extra flexible joints and the connected Ant engines rope exploit (wobbly should have its limits).
• Floppy robotic parts if we don't lock the joints every time (I made some bay doors that were flopping all over the place).
• Spinning ladders bug.
• Parts supporting incredible weight just because they are 200+ meters away and the physics does not load for them.

And other stuff we can find on the amazing channels of

• https://www.youtube.com/c/Danny2462
• https://www.youtube.com/c/Stratzenblitz75

Yeah, it's fun to see one time - like a SpeedRun, but in the end these are bugs. Mr. Tom Vinita said the team will defeat the Kraken. I feel like all this stuff should be removed especially if there's going to be multiplayer. It would be fair that we should all be able to build incredible things through hard work and smart engineering, not using game bugs and exploits.

Edited June 2, 2022 by Vl3d
Fixed the title and body

[Pthigrivi]

For the most part Im fine with losing unrealistic physics and kracken drives, but Id be a little annoyed if they eliminated part clipping. I do it for aesthetic reasons and I think players can draw their own fuzzy line over whats realistic. There are a lot of times when you’re clipping into something that’s obviously a bulkhead or space between truss elements, or just pushing a light in a bit so it looks married to the surface. I also sometimes float parts, usually temporarily while Im working but occasionally floated off one part for node/alignment reasons but visibly touching another so it doesn’t look magic. If players feel like abusing that thats just up to them. 
 

Ditto on a few unrealistic things like infinite engine restarts, saturable reaction wheels, and  RTG decay. Its just a simplified convenience that makes the game more playable. The one classic exploit Id be fine with losing is tapping time-warp to halt rotation. Im actually pretty happy to see things continue to rotate under warp. 

[Vl3d]

Infinite engine restarts, saturable reaction wheels, RTG and solar panels decay - should be stock difficulty options.

I'm totally fine with part clipping for aesthetic reasons and to move things around. I use this all the time. And in a single player scenario I agree that it's up to the player how he wants to LEGO things around. But when we are in a space race or comparing vessels to one another or we have some multiplayer interaction or if we build in a persistent world, I want it to be an even playing field without weirdness.

A while back I thought about some guide lines for part clipping - this would allow clipping but make things more realistic in a good way.

On 4/19/2022 at 4:36 PM, Vl3d said:

1. I've mentioned it in the list, but we really need a better way to interact with parts that are clipped inside other larger parts.

It's so frustrating to have to zoom inside and find the right angle in normal view so that I can right click a part (on a small probe core inside a tank, for example).

2. Total volume of hollow parts clipped together should be calculated using the formula:

vA clipped with vB = vA + vB - (vA intersected with vB)

3. A similar approach to calculate hollow structure mass. Meaning that if X is a hollow parts or part with living space, the total mass should depends on the exterior walls, not on the part walls that are now inside (and in practice would not exist anymore).

mX clipped with mY = mX + mY - (mX and mY mass of walls that are now inside the structure)

4. We desperately need a way to group merge fuel tanks together so they become a single tank. Also applies to hollow parts with living space that are clipped together.

[Death Engineering]

5 hours ago, Vl3d said:

1. I've mentioned it in the list, but we really need a better way to interact with parts that are clipped inside other larger parts.

It's so frustrating to have to zoom inside and find the right angle in normal view so that I can right click a part (on a small probe core inside a tank, for example).

[snip] THIS THIS THIS!!!

Edited June 2, 2022 by Vanamonde

[shdwlrd]

There's always going to be physics bugs in games. It's how the game deals with it that matters. KSP1 is notorious for flat out breaking when the physics got out of wack. (Danny2462 showcases them very well.) If KSP2 can deal with that level of breaking of the physics without crashing or deleting the Kerbal universe, I would take that as a good sign. 

Physics exploits themselves, if they are truly game breaking, (kraken drives) they should be corrected. If they are more of a nausance than anything else, they still need to be corrected, but a much lower priority. (Super buoyant parts, small part vibration that doesn't cascade into a bigger problem.) 

Some exploitation would be expected with part clipping. It would be really hard to police that one. Some use cases are obviously exploitative, but some are necessary to gain reasonable performance without completely changing your design. (Power production and storage comes to mind.)

Under normal circumstances, crafts with floating parts shouldn't be allowed to launch. But with all the Kerbal contraptions I've seen, floating parts are necessary to setup "stock" bearings and to free them. So there's another case where something that shouldn't be allowed is actually necessary in some cases.

[intelliCom]

On 6/3/2022 at 1:45 AM, Pthigrivi said:

 There are a lot of times when you’re clipping into something that’s obviously a bulkhead or space between truss elements, or just pushing a light in a bit so it looks married to the surface.

With part clipping, I often use it to embed batteries into large truss segments, especially if I'm building rovers. Makes for a strong part to use on the main body.

That being said, part clipping should have its limits. Perhaps they could fix the exploit of compressing thousands of tanks into a small space by only allowing a small bit of part clipping? You know, enough to be "married to the surface", as you described it. Many solar panels in KSP1 already visually do this.

Just a matter of finding what the limits should be.

Edited June 4, 2022 by intelliCom
missing word

[PDCWolf]

9 hours ago, Vl3d said:

 

in the end these are bugs.

• I hate part clipping with a passion, but since they denied us advanced tweakables there's no other way to get the parts you want unless you mod the game and make the part yourself. There should be a middle ground here, or rather a proper implementation of "there's X volume inside this part and you can fill it with whatever you want".
• Yes, at that point it is just plain exploiting.
• This extends building capabilities without being cheating per se, the only advantage is based on the dumb way they manage drag (empty nodes + bounding box).
• High lift low drag describes all the wings, they're absolutely broken like the whole aerodynamic system. I have no idea why they gave such aggressive stats to heat shields, considering they don't offset the CoM to allow controlled descents.
• That's entirely on the devs, they made parts that magically attract, along with a whole hacked-in system of force dispersion.
• The whole physics management during unpacking on all scenarios is royally broken, and there's a myriad of mods that attempt to fix that and expand functionality, I'd be really disappointed if they made the same mistakes again.
• Unity's crappy default joint system.
• Wobbliness should not be a feature, rockets would explode before visually bending anyways.
• They're weak, sloppy, buggy, and left unfinished. I regret paying for them every single day.
• Didn't know this one lmao.
• The whole part resistance system is completely out of balance and was never looked at again. Again, would be really disappointed if they made the same mistake.

Most of the things on your list are attributable to Unity's crappy physics, or stuff the devs overlooked or outright didn't bother fixing/polishing, but hey, they had to prove us wrong that the game was ready for release :^)

[Pthigrivi]

On the part clipping point I guess Ive just never concerned myself with how others play or choose to enjoy the game and I don’t personally need a finicky mechanic to tell me what’s realistic so leaving this to players personal preference seems fine. I also don’t see competitive gaming being the central focus of multiplayer so folks can just decide among themselves if part clipping is allowed or not in the instance. Making some arcane set of rules and invisible geometries for each part seems like a long run for a short slide to me. 
 

And Ive also never really had a problem with bendy rockets. If something I’ve built starts bending it lets me know Im either designing or flying it wrong, and seeing it wobble and flex communicates the nature of the problem better than a spontaneous explosion. There’s a limit of course but this just seems like a fine-tuning problem rather than something to be eliminated entirely. 

Edited June 3, 2022 by Pthigrivi

[t_v]

To add to the part clipping messing up competitions thing, take challenges like the Jool - V which don’t allow functional part clipping. If people are clipping parts, then they just don’t fulfill the criteria for whatever challenge is being attempted and don’t get to participate/get the credit. 

[Vl3d]

15 hours ago, Vl3d said:

Spinning ladders bug.

5 hours ago, PDCWolf said:

Didn't know this one lmao.

 

[PDCWolf]

10 hours ago, Pthigrivi said:

And Ive also never really had a problem with bendy rockets. If something I’ve built starts bending it lets me know Im either designing or flying it wrong, and seeing it wobble and flex communicates the nature of the problem better than a spontaneous explosion. There’s a limit of course but this just seems like a fine-tuning problem rather than something to be eliminated entirely. 

They implemented a hacked fix to sidestep their own gameplay fix, that's enough of a clue that it's not really a programmed mechanic. Nowadays if your rocket is bendy you can just right click > autostrut to X. Further on, most bending is caused by how dumb the joint system is (weakness being a result of different sized joints interacting). It is not really a gameplay feature, shouldn't be, and shouldn't be considered as one either. 

On this same hand, rockets don't bend in real life, just check Proton's most popular failure, where it even went completely lateral against the airstream and disintegrated first before visually bending.

[mcwaffles2003]

2 hours ago, PDCWolf said:

On this same hand, rockets don't bend in real life, just check Proton's most popular failure, where it even went completely lateral against the airstream and disintegrated first before visually bending.

But they do

Just because proton flexed beyond the point of failure doesn't mean that it didn't bend along the way to failure.

Flexibility is part of why challenger likely exploded as well, the solid rocket motor flexed at a joint allowing internal fire that passed between a grain boundary of the propellant to infiltrate the joint and burn away the O-ring seal. (Perhaps Im remembering this incorrectly, but I'm fairly sure)

Also... literally everything bends, nothing is perfectly rigid. Anyone who has been on top of a tall tower on a breezy day would attest to this.

[Pthigrivi]

2 hours ago, PDCWolf said:

They implemented a hacked fix to sidestep their own gameplay fix, that's enough of a clue that it's not really a programmed mechanic. Nowadays if your rocket is bendy you can just right click > autostrut to X. Further on, most bending is caused by how dumb the joint system is (weakness being a result of different sized joints interacting). It is not really a gameplay feature, shouldn't be, and shouldn't be considered as one either. 

On this same hand, rockets don't bend in real life, just check Proton's most popular failure, where it even went completely lateral against the airstream and disintegrated first before visually bending.

What I mean is if a rocket tumbles over I know I don’t have the aerodynamics right, insufficient stabilization or pushing too fast too low in the atmosphere. If it suddenly explodes on staging I know I need to adjust sepratrons or decoupling force or manage some other conflict. Wobble and bending is usually structural: the rocket is too long or there are payload problems or Im just pushing too hard at max q. I don’t personally care that the visuals are exaggerated compared to real life if its communicating important information about a flawed design. There’s a limit of course and at some point it looks ridiculous, but a subtle wobble and flex before RUD is a helpful visual cue. 

[Vl3d]

... the limp phenomenon.

Edited June 3, 2022 by Vl3d

[SciMan]

You know what would get rid of most of the problems of part clipping? Being able to weld parts together, like what the UbioZor Part Welding Mod did (does? IDK if it still works, it would be real nice if it did).

Of course, this would only solve part clipping to your satisfaction if the part welding was the ONLY way to get parts to clip into each other, and they wouldn't clip normally.

That way you could say clip parts to your hearts content, (and if it was minor things like putting girders in a fuel tank it would give it a higher impact tolerance at the cost of more dry mass and slightly less fuel capacity).
But the part welding would be SMART.
It would detect things like two fuel tanks clipped together, and instead of letting you have "two fuel tanks worth of fuel in the space of one fuel tank", if they were 100% overlapped it would simply give you double the dry mass for no gain in fuel capacity. If the parts were not clipped together as severely, the fuel capacity would proportionally increase, until they're fully separated and you once again get the full capacity of both fuel tanks.

And the only way you could violate that would be a check-box in the cheat menu, which would exist for the purpose of debugging, similarly to KSP 1 "hack gravity" type cheats.

 

Additionally, these welded parts could have code in them that detects a failure condition (structural overload, collision, aerodynamic stress, heating, etc.) and proceed to un-weld and weld the parts in flight (depending on if the stresses reduce or not, the part might end up permanently bent or outright fully broken apart into its separate parts).

If that system was further extended to the whole craft, you could replace the entire Unity physics joints system with something much more robust that actually "only bends when it's already too late" instead of wobbling on the launch pad like in KSP 1 with no auto-strut.

[PDCWolf]

11 hours ago, mcwaffles2003 said:

But they do

Under very specific circumstances and only in very specific ways, not as a wet noodle which is what the game shows. Further on, real life structural flexibility has nothing to do with in-game bending either. You sold me two false equivalences for the price of one. I refuse to relate the in-game wobbliness with any real life event, they're not the same, they do not represent the same, and they do not happen for the same reasons.

11 hours ago, Pthigrivi said:

What I mean is if a rocket tumbles over I know I don’t have the aerodynamics right, insufficient stabilization or pushing too fast too low in the atmosphere. If it suddenly explodes on staging I know I need to adjust sepratrons or decoupling force or manage some other conflict. Wobble and bending is usually structural: the rocket is too long or there are payload problems or Im just pushing too hard at max q. I don’t personally care that the visuals are exaggerated compared to real life if its communicating important information about a flawed design. There’s a limit of course and at some point it looks ridiculous, but a subtle wobble and flex before RUD is a helpful visual cue. 

Wobble can happen on a short, balanced rocket so long as you mismatch node sizes (which is why it tends to happen on the payload side), it can be (and most times is) completely unrelated to real life events, this is not at all equivalent to staging related complications which work pretty well as a real-life abstraction and the same goes for aerodynamic tumbling, they're well simulated and communicated, wobble isn't.

Edited June 4, 2022 by PDCWolf

[mcwaffles2003]

37 minutes ago, PDCWolf said:

Under very specific circumstances and only in very specific ways, not as a wet noodle which is what the game shows. Further on, real life structural flexibility has nothing to do with in-game bending either. You sold me two false equivalences for the price of one. I refuse to relate the in-game wobbliness with any real life event, they're not the same, they do not represent the same, and they do not happen for the same reasons.

Hey cool, I can do that too

 

[Jarin]

I want my docking-port-based reactionless thruster.

[SciMan]

And I want surface bases above a certain size that are constructed of docked-together parts to not randomly catapult themselves into the sky for no observable reason upon the game loading them into physics range (yes I know they "fixed" it, but the fix didn't really fix all of it, yes it still happens and I've had it happen to me, so it apparently still needs fixing).

If that means we have to lose kraken drives, so be it.

We can also do away with the weirdness of aircraft landing gear "only partially" interacting with the runway whenever you put a craft on the runway from the SPH. As well as the problem with aircraft landing gear suspension default settings just not being something that will ever work (constantly bouncing for no good reason unless you override the settings).

[Davi SDF]

Many of these are easily avoidable by "just don't do it".

Although some like "Parts clipping in the ground when loading" and "Spinning ladders" really should be fixed, but that is assuming that those bugs even exist in KSP2 in the first place.

Also, i'll be honest, i don't know about you, but going around the system with a Kraken drive to troll new players sounds really funny to me.

[Anth]

The biggest thing I will miss is being able to stop a craft from spinning in space by using timewarp.

Also stabilizing SAS or wobbling crafts using timewarp is another one, but hopefully I wont have to do that in KSP2 (which I wont be able to because its part of the above)

Another which is interesting that I doubt will be in KSP2 is being able to phase through another craft if using timewarp. But the only time I might use this is in 'simulation mode'

 

Edited June 9, 2022 by Anth12

[shdwlrd]

46 minutes ago, Anth12 said:

Another which is interesting that I doubt will be in KSP2 is being able to phase through another craft if using timewarp. But the only time I might use this is in 'simulation mode'

They did a dev blog about that. You will have to worry about crossing orbits with time warp. If there's going to be a collision, it's going to happen, time warp or not.

[Bej Kerman]

On 6/10/2022 at 1:22 AM, shdwlrd said:

On 6/10/2022 at 12:30 AM, Anth12 said:

Another which is interesting that I doubt will be in KSP2 is being able to phase through another craft if using timewarp. But the only time I might use this is in 'simulation mode'

They did a dev blog about that. You will have to worry about crossing orbits with time warp. If there's going to be a collision, it's going to happen, time warp or not.

Link or it didn't happen

[shdwlrd]

27 minutes ago, Bej Kerman said:

Link or it didn't happen

Edited June 16, 2022 by shdwlrd

[Bej Kerman]

11 minutes ago, shdwlrd said:

30 minutes ago, Bej Kerman said:

Link or it didn't happen

Nothing here said you'll need to worry about vessels colliding during timewarp - only that you'll have to worry about the vessel you are controlling right now colliding with another vessel in orbit while you're busy using it.