KSP2 EA Grand Discussion Thread.

[psunavy03]

On 3/10/2023 at 5:56 PM, cocoscacao said:

Again if... I guess devs would know best.

Fun side note, now that you mentioned stressful maneuvers... I'm hoping that Kerbals will die from over the top G forces in this game, rather than just passing out.

I've flown fast jets and gone through centrifuge training . . . and this is not how this works.  The amount of G it takes to kill you is a whole lot more than it takes to have you G-LOC.  You don't (directly) die from G-LOC.  If you die in that kind of scenario, it's generally because when you lose consciousness, the jet happens to end up pointed at the ground, and the 15-30 seconds it takes you to regain useful consciousness is too long to keep you from turning into a smoking hole in the ground.

Edited March 14, 2023 by psunavy03

[psunavy03]

For you kiddos who are all shocked paying $50 of 2023 money for an EA title . . . here's an ad for SNES games in 1994.  Adjusted for inflation, that's a Street Fighter title for ~$141, Super Metroid for ~$111, and Illusion of Gaia for ~$151.  All for the finest 16-bit technology the Clinton administration could offer.

Edited March 14, 2023 by psunavy03

[Jacke]

1 hour ago, InterstellarDrifter said:

Does anyone else agree with my opinion that wobbly rockets, or anything else long and cylindrical in shape, in-game, are actually modeled this way on purpose because that's what happens in real life?

Real rockets do not wobble like that.  Sure, they flex.  But they're also designed to minimize that flex to the point where in KSP (either one) you wouldn't be able to see it.

It comes down to not setting up the joints correctly in Unity.  Unity joints can be made much firmer by adjusting parameters.  And with measures like what the Kerbal Joint Reinforcement mod does (turns a single joint in a trio of 3 in a ring) or later stock KSP1 (add partless autostruts to connect more parts together to reduce overall flex) things can be made much more realistic.

Edited March 14, 2023 by Jacke

[The Aziz]

4 hours ago, InterstellarDrifter said:

Does anyone else agree with my opinion that wobbly rockets, or anything else long and cylindrical in shape, in-game, are actually modeled this way on purpose because that's what happens in real life?

I think we think since rockets are made of metals and composites that they shouldn't bend, but this actually is far from the truth. Real rockets can bend. And if not designed correctly, they absolutely would wobble. 

Thought experiment: Take a 30m long tube of thin-walled steel at a 3.75m diameter and stick it out horizontally anchored at one end. How much do you think it would deflect. Answer: way more than you think.

Now, don't get all upset. I'm not saying KSP2 didn't get it wrong. I'm just pointing out that it is a simulator that is trying to simulate many physical behaviors, including extension, flexion, compression, and tension of materials and their construction, in order to deliver a more authentic engineering experience and aerodynamic behaviors of our rockets, in flight and when they crash.

Just something to think about, maybe ask a dev directly(if anyone can grab their attention) if the current rigidity model is what is intended?

P.S. KSP-1 rockets still wobble too if they're aspect ratio is not sufficient.

There's a subtle difference however. In KSP it's the joints where the rocket is held together, and the weakest joints are where the decouplers are. Sure it makes sense because they're supposed to be detachable, my rockets don't wobble in the middle of a fuel tank or between a tank and engine - it's always the separators. Logical, but there's more to that. Interstage fairings, be it from decouplers or fairings themselves don't provide any structural stability, at all. They're physicsless shells, not metal tubes. Your engine isn't held by the joints around the bottom of the tank above it, it's held by one point on the top of the decoupler. You can try building a Saturn V, or anything that has a smaller spacecraft between stages, hidden inside a fairing. That fairing may as well not be there because it does nothing for the stability, and your lander or whatever will poke through it, because the top of the fairing isn't physically connected to the command module.

And I doubt it's going to change because of craft tree structure, but gods this needs some work.

[Sturmalex]

My observation for KSP2 right now is that the gameplay is bug ridden, but most (not all, mind you) can be worked around. I am looking forward to the first patch and the ones following. I am optimistic that they will make the game a lot more stable/better rather quickly.

With all the big bugs out of the picture I can see the game becoming great already making the creation of crafts a lot less finicky and prone to breaking the game or save.

Then the devs can focus more on devloping the next big step which has to be science/career mode to give the player a purpose apart from exploration/craft building with no limits.

 

So overall I am very optimistic looking forward as long as this first patch delivers some good fixes. As a Dev myself (though not games but aerospace) I think Nate's decision to not pump out 20 Hotfixes to  fix one issue but create 30 new issues is a sound and  responsible decision. 

[cocoscacao]

4 hours ago, psunavy03 said:

I've flown fast jets and gone through centrifuge training . . . and this is not how this works.  The amount of G it takes to kill you is a whole lot more

I'm not thinking about jets. One time, I was returning from Eeloo, and when capsule reached thicker atmosphere, it blew up due to overstress (it had heat shields). Although, in this scenario, Kerbals are doomed anyway, there is probably a middle ground where capsule can survive, but it is fatal for Kerbals.  Though, it would be a rare occurrence.

Edited March 14, 2023 by cocoscacao

[LoSBoL]

45 minutes ago, Sturmalex said:

 

So overall I am very optimistic looking forward as long as this first patch delivers some good fixes. As a Dev myself (though not games but aerospace) I think Nate's decision to not pump out 20 Hotfixes to  fix one issue but create 30 new issues is a sound and  responsible decision. 

What I'm mostly eager to find out, is next to performance and bugs being stated to be worked on, is if we get QOL updates or other small features. 

There was an interview, either with Matt Lowne or Scott Manley, I don't know which of them it was anymore. But it was stated that it was prognosed we'd get two QOL updates before the next step in the roadmap would be taken bringing Science and Tree Progression. 

Maybe the patch will shed some light on that, so will it mostly be bug patch? Or are we going into the prognosed update cycle? 

[The Aziz]

I saw some UI improvements in the last dev diary, but there's no telling if it comes now or later.

[Wheehaw Kerman]

5 hours ago, Jacke said:

Real rockets do not wobble like that.  Sure, they flex.  But they're also designed to minimize that flex to the point where in KSP (either one) you wouldn't be able to see it.

It comes down to not setting up the joints correctly in Unity.  Unity joints can be made much firmer by adjusting parameters.  And with measures like what the Kerbal Joint Reinforcement mod does (turns a single joint in a trio of 3 in a ring) or later stock KSP1 (add partless autostruts to connect more parts together to reduce overall flex) things can be made much more realistic.

You know, I’m more than fine with wobbly rockets in the game, provided that a) you can build non-wobbly ones without messing around in settings and b) they come apart and RUD properly when they start tumbling.

 

[Sturmalex]

32 minutes ago, LoSBoL said:

What I'm mostly eager to find out, is next to performance and bugs being stated to be worked on, is if we get QOL updates or other small features. 

There was an interview, either with Matt Lowne or Scott Manley, I don't know which of them it was anymore. But it was stated that it was prognosed we'd get two QOL updates before the next step in the roadmap would be taken bringing Science and Tree Progression. 

Maybe the patch will shed some light on that, so will it mostly be bug patch? Or are we going into the prognosed update cycle? 

True. But just because there are said to be 2 QoL patches before new mechanics come into the game it doesnt mean there cant be 125 Bugfix patches aswell. :-) I am pretty sure this will be a bugfix patch and no QoL Patch. I dont see the first QoL-Patch being delivered for at least 2 months. But then again I have zero insight into their development schedule, their cycle etc. Its just guesstimating stuff.

[InterstellarDrifter]

6 hours ago, Wheehaw Kerman said:

You know, I’m more than fine with wobbly rockets in the game, provided that a) you can build non-wobbly ones without messing around in settings and b) they come apart and RUD properly when they start tumbling.

 

Right!! Agreed.  I really wish they wouldn't have gone more realistic with some RFNA exhaust and black smoke in the explosions, instead of the vanilla white puffy steam clouds. But, I do get that would not fit the art/design aesthetic, lighthearted feel of the game.

[Chilkoot]

6 hours ago, Wheehaw Kerman said:

You know, I’m more than fine with wobbly rockets in the game, provided that a) you can build non-wobbly ones without messing around in settings and b) they come apart and RUD properly when they start tumbling.

 

Pay attention, Kerbonauts!  This is what happens when you install your command module upside-down

(Edit:  literally - one of the directional sensors was installed 180 degrees to spec, making the guidance system think it was pointing downward at launch).

Edited March 14, 2023 by Chilkoot

[darthgently]

Meming...

https://drive.proton.me/urls/F3Y8JJK600#hjm9yhM6nwFq

 

[JoeSchmuckatelli]

7 hours ago, Wheehaw Kerman said:

) they come apart and RUD properly when they start tumbling

 

 

Sorry - you don't get to see that!  Pretentious Kerman strikes again!

[Wheehaw Kerman]

13 minutes ago, JoeSchmuckatelli said:

 

 

Sorry - you don't get to see that!  Pretentious Kerman strikes again!

I know, right?  Smack a lander and rover into a mountainside on Minmus and P.A.I.G.E. calls it problematic.  Sheesh.

[JoeSchmuckatelli]

15 minutes ago, Wheehaw Kerman said:

I know, right?  Smack a lander and rover into a mountainside on Minmus and P.A.I.G.E. calls it problematic.  Sheesh.

Watching epic failure and big booms was one of the promises of this game.  I hope it's the first thing they get rid of (once they start looking at player feedback)

[Stoup]

https://discord.com/channels/1039959585949237268/1039965578754007060/1085256095515283516

Enjoying getting in the habit of doing this, hon hon. In response to someone asking if they'd release the patch notes before the patch itself:

• the aim is the day before, but it could be day-of.

[i dont know how to forum]

11 hours ago, The Aziz said:

There's a subtle difference however. In KSP it's the joints where the rocket is held together, and the weakest joints are where the decouplers are. Sure it makes sense because they're supposed to be detachable, my rockets don't wobble in the middle of a fuel tank or between a tank and engine - it's always the separators. Logical, but there's more to that. Interstage fairings, be it from decouplers or fairings themselves don't provide any structural stability, at all. They're physicsless shells, not metal tubes. Your engine isn't held by the joints around the bottom of the tank above it, it's held by one point on the top of the decoupler. You can try building a Saturn V, or anything that has a smaller spacecraft between stages, hidden inside a fairing. That fairing may as well not be there because it does nothing for the stability, and your lander or whatever will poke through it, because the top of the fairing isn't physically connected to the command module.

And I doubt it's going to change because of craft tree structure, but gods this needs some work.

For what it's worth, this particular issue should be fixed or at least minimized with the next patch:

On 3/3/2023 at 11:40 AM, Nate Simpson said:

• Fixed: Engine plate floating node joints less rigid than other stack node joints (were not receiving multijoint reinforcement)

Engine plates are the intended way of doing interstages now, so this should help a lot. Personally I'd still like to be able to use normal fairings for interstages too (currently fairings cannot be closed against parts like the could in KSP1), and I'd like to see the tube parts be given interior nodes so they can be used as open-end interstages (the tube parts are incredibly useful but they would be so much more useful if you could place stuff inside them). More options is always good!

[The Aziz]

I hardly ever used the plates, despite their obvious usefulness, but it still doesn't help with the fairings which are still single joint, and they "close" around parts only visually.

The fact that tubes don't allow surface attachments is stupid, let's start with that.

[RocketRockington]

22 hours ago, InterstellarDrifter said:

Now, don't get all upset. I'm not saying KSP2 didn't get it wrong. I'm just pointing out that it is a simulator that is trying to simulate many physical behaviors, including extension, flexion, compression, and tension of materials and their construction, in order to deliver a more authentic engineering experience and aerodynamic behaviors of our rockets, in flight and when they crash.

I think you're misunderstanding something fundamental here.  You're.imagining they tried to do a more realistic simulation and got it wrong.  That's not what happened.  They only decided to stick with simulating things exactly as KSP1 did - with tuning that makes it even floppier.  If they wanted a more realistic simulation of the factors you mention, they would use a soft body simulator like beamNG.  Or they would have used KJRs solution- fuel tanks don't have 1 structural member that runs from a point in the middle of one tank to the middle of another tank and have all forces transmitted on that one beam, they're welded all around the wall, more similar to how KJR works with their multipoint attachments.

Fundamentally, KSP2 mostly tried to replicate KSP1's physics, either be a use they thought that was ideal, or because they didn't know or think they could pull off a better solution.

Edited March 15, 2023 by RocketRockington

[i dont know how to forum]

5 hours ago, The Aziz said:

I hardly ever used the plates, despite their obvious usefulness, but it still doesn't help with the fairings which are still single joint, and they "close" around parts only visually.

I'm not certain what you're asking for here then. You mentioned interstage fairings, that's what engine plates are intended for. You attach a decoupler at the floating node, it automatically creates a fairing between the engine and the decoupler. You then attach whatever you want inside the interstage (like the lunar module if you're making a Saturn V recreation) to the engine plate's interior node. Once engine plates are fixed in the next patch this should be structurally sound without struts.

Don't get me wrong, there are certain cases where I want to be able to use normal fairings for this so that I have more control over exactly where the fairing closes. I currently have a rocket that uses a combination of engine plates, tubes, and unclosed fairings to get exactly the shapes I want. I don't see how multijoint reinforcement could help here though since the fairing isn't actually attached to the part, unlike with engine plates.

5 hours ago, The Aziz said:

The fact that tubes don't allow surface attachments is stupid, let's start with that.

This would definitely also be nice, but it's worth mentioning that tubes are just another type of fairing. I think being able to attach parts to fairings could be a really helpful feature, but it would likely require them to significantly change the way fairings are set up behind the scenes.

[pandaman]

6 hours ago, i dont know how to forum said:

...but it's worth mentioning that tubes are just another type of fairing...

Is that so?  I thought the structural tube parts were more like empty fuel tanks, at least in the way they are implemented.

 

 

[The Aziz]

I haven't checked if things go through them but based on my short experience I think this might me the case.

Unlike KSP1 where they were solid, like they should.

[InterstellarDrifter]

17 hours ago, RocketRockington said:

 If they wanted a more realistic simulation of the factors you mention, they would use a soft body simulator like beamNG.  Or they would have used KJRs solution- fuel tanks don't have 1 structural member that runs from a point in the middle of one tank to the middle of another tank and have all forces transmitted on that one beam, they're welded all around the wall, more similar to how KJR works with their multipoint attachments.

Fundamentally, KSP2 mostly tried to replicate KSP1's physics, either be a use they thought that was ideal, or because they didn't know or think they could pull off a better solution.

Oh, I was under the impression that they were using a soft body algorithm, in KSP1, and therefore likely KSP2.

[RocketRockington]

6 minutes ago, InterstellarDrifter said:

Oh, I was under the impression that they were using a soft body algorithm, in KSP1, and therefore likely KSP2.

Nope.  It's rigid body physics and stock Unity (PhysX) physics.  None of the parts deform, which is what soft body physics would entail.  If we saw things like flexing wings (along the span, not just at the root) that would be soft body physics.

Vs reality, completely rigid rockets are a better approximation than the current wet noodle approach.  Rockets deform but not much.  While they are thin walled, they are held up by internal pressure and clever isogrid reinforcement.  Any rocket that bent like a KSP2 rocket would RUD, as the structure would be massively compromised.  Even if it didn't RUD, the structure would remain permanently bent, rather than continuing to flop around.  If you 'bend' a soda can it doesn't snap back into place later.

Edited March 15, 2023 by RocketRockington