KSP2 EA Grand Discussion Thread.

[cocoscacao]

17 minutes ago, Periple said:

They clearly decided, very early on

I'm very worried about their early decisions from the moment Nate officially addressed wobble as an issue, for the first time in one of the "upnates". I don't know if they are allowed to share such details, but even the recent dev vlog didn't mention if they have any potential solutions for this. Correct me if I'm wrong, I don't recall the video in its entirety. We don't even know if they started addressing it now (which would be crazy), or from the beginning of the development... whenever that beginning was. And if the latter is the case... then...

[PDCWolf]

21 minutes ago, cocoscacao said:

I'm very worried about their early decisions from the moment Nate officially addressed wobble as an issue, for the first time in one of the "upnates". I don't know if they are allowed to share such details, but even the recent dev vlog didn't mention if they have any potential solutions for this. Correct me if I'm wrong, I don't recall the video in its entirety. We don't even know if they started addressing it now (which would be crazy), or from the beginning of the development... whenever that beginning was. And if the latter is the case... then...

They mentioned having a tool to compare multiple solutions to see how each impacted the game (which suggests they've thought out and programmed or are programming more than one solution, at least in a very basic way).

[Guest]

38 minutes ago, MarcAbaddon said:

Another problem I have here is that I feel that it's something that should been tackled earlier in the Dev process. It's one of those central things you have to plan around early when you make another KSP using Unity physics. It'd be one thing if they tried a solution early on and it turned out a dead end, but it seems they are talking conceptually. Which again, seems a bit late.

That’s easy to say but there are loads of things competing for that “must be solved early” spot. Thrust under warp, terrain that isn’t blobby, trajectories up to interstellar scale, multiplayer, the list goes on. I think a lot of these are more important to solve early than joints!

41 minutes ago, MarcAbaddon said:

I agree with this take, but I also feel that if you decide to go with 1 from the start you shouldn't make claims about how you are going to or at least trying to slay the kraken.

Yeah I bet they regret that now!

[Kerbart]

58 minutes ago, cocoscacao said:

I'm very worried about their early decisions from the moment Nate officially addressed wobble as an issue, for the first time in one of the "upnates". I don't know if they are allowed to share such details, but even the recent dev vlog didn't mention if they have any potential solutions for this. Correct me if I'm wrong, I don't recall the video in its entirety. We don't even know if they started addressing it now (which would be crazy), or from the beginning of the development... whenever that beginning was. And if the latter is the case... then...

Well the very early decision to go with Unity's physics model instead of rolling their own... Look at the current state of the game and keep in mind that this kind of physics simulation is complex and really hard. Do you want Intercept to roll out their own code for that? Even if they manage to make it bug-free—and we all know how that goes—it also has to be high performance.

And as stated countless times before, the problem isn't just wobble. That's easy to fix. The problem is to address the wobble without trivializing the game. Anyone telling me that "in real life rockets don't flex and stretch" is doing a fantastic job of pretending not to know a lot about physics and engineering. Of course they do. That's the nature of the materials we use.  Saturn V almost pogo'd itself to pieces before they managed to reduce the issue. The problem is scale, and making it fit in Unity physics model, and finding the balance between the game being fun (as in frustration-free) and the game being challenging. It's not a simple problem to solve properly, and they likely tabled it because there were bigger fish to fry.

[cocoscacao]

29 minutes ago, Kerbart said:

this kind of physics simulation is complex and really hard

True, I'm not disputing that. I admire them for what they're trying to accomplish... without a single clue on exactly how they'll manage it... unless affordable home quantum PCs become a thing. But someone probably put a lot more thought into it than I did.

29 minutes ago, Kerbart said:

And as stated countless times before, the problem isn't just wobble. That's easy to fix. The problem is to address the wobble without trivializing the game

 Yet we still didn't get "partially procedural parts", which would help a lot.

Edited October 5, 2023 by cocoscacao
tie-poh

[Bej Kerman]

6 minutes ago, cocoscacao said:

Yet we still didn't get "partially procedural parts"

Sort of like the wings?

[The Aziz]

9 minutes ago, cocoscacao said:

Yet we still didn't get "partially procedural parts", which would help a lot.

Nah, at the moment, a simple stack of same sized parts doesn't move a lot. Your procedural parts wouldn't help where the joints, which are inevitable, are the weakest - size change, separators, fairings, radial attachments. Also - just how many longest tanks on top of each other do you actually need on your ships so that, say, 3 aren't enough?

[cocoscacao]

58 minutes ago, Bej Kerman said:

Sort of like the wings?

No. In discrete steps. Check structural tube from KSP 1. You have an option to set it's length from 1 to 6. Something like that.

54 minutes ago, The Aziz said:

Your procedural parts wouldn't help where the joints, which are inevitable, are the weakest - size change, separators, fairings, radial attachments.

That's ok, but with stacking 3 fuel tanks instead of one, we have 2 weak spots to worry about. I just think those joints are adding useless complexity for no reason.

[Kerbart]

54 minutes ago, cocoscacao said:

That's ok, but with stacking 3 fuel tanks instead of one, we have 2 weak spots to worry about. I just think those joints are adding useless complexity for no reason.

To play the devil's advocate here, a stack of 75m high needs, to have some flex in order to reflect reality. Or what about a 400m tall stack? Having a maximum tank size enforces those joints, making taller stacks less stable than shorter stacks. Just as in reality. In that sense the added complexity is not useless as it's a tool to provide structural challenges when building a ship. This might very well be the reason for the devs to not implement procedural tanks.

I'm not in charge of that, and you'll probably disagree with the validity of such reasoning but it would suggest that there is a reason for the complexity.

My personal view is that we should have procedural tanks for a variety of reasons but that internally they should be represented as a stack of smaller tanks, so we retain a noodle factor that punishes unrealistic tall and slender structures.

[Alexoff]

23 minutes ago, Kerbart said:

To play the devil's advocate here, a stack of 75m high needs, to have some flex in order to reflect reality. Or what about a 400m tall stack? Having a maximum tank size enforces those joints, making taller stacks less stable than shorter stacks. Just as in reality. In that sense the added complexity is not useless as it's a tool to provide structural challenges when building a ship. This might very well be the reason for the devs to not implement procedural tanks.

I'm not in charge of that, and you'll probably disagree with the validity of such reasoning but it would suggest that there is a reason for the complexity.

My personal view is that we should have procedural tanks for a variety of reasons but that internally they should be represented as a stack of smaller tanks, so we retain a noodle factor that punishes unrealistic tall and slender structures.

For realism, such long tanks should not fold like sausages, but explode in a sea of fire. The lack of durability in the game (more precisely, infinite durability) breaks realism. Probably the most unrealistic part of KSP1/2 is standing the rocket directly on the engine nozzles. It seems to me that no one does this. It should be possible to do this on Gilly, but definitely not on Kerbin or Dune.

[cocoscacao]

2 hours ago, Kerbart said:

To play the devil's advocate here, a stack of 75m high needs, to have some flex in order to reflect reality. Or what about a 400m tall stack?

It doesn't have to allow such values. You can limit the length to the sum of 3 current longest tanks. I think that's reasonable.

1 hour ago, Alexoff said:

Probably the most unrealistic part of KSP1/2 is standing the rocket directly on the engine nozzles

100% upvote on this. Please open a suggestion thread.

[Alexoff]

45 minutes ago, cocoscacao said:

100% upvote on this. Please open a suggestion thread.

I very much doubt that anyone will implement this. To do this, devs need to introduce strength into the game.

[The Aziz]

It feels like some of you play a different game than me. I just launched a rocket mostly made of different parts of the same size and there was no wobble at all. The only bit where the joints went a little squishy was at fairing base—0.625m decoupler - which, as we already know, is expected before it's fixed. The rest was rock stable.

[MarcAbaddon]

10 hours ago, cocoscacao said:

It doesn't have to allow such values. You can limit the length to the sum of 3 current longest tanks. I think that's reasonable.

100% upvote on this. Please open a suggestion thread.

I wouldn't want a hard limit on length that is simply artificially enforced by the game but not reflected by in-game physics. That'll place too many constraints on creativity. Doesn't mean there shouldn't be structural limitations on it in practice, just that should be ways around it with proper  construction.

Edited October 6, 2023 by MarcAbaddon

[cocoscacao]

8 minutes ago, MarcAbaddon said:

I wouldn't want a hard limit on length that is simply artificially enforced by the game but not reflected by in-game physics.

How does that differ from what we have now?

[PDCWolf]

3 hours ago, MarcAbaddon said:

I wouldn't want a hard limit on length that is simply artificially enforced by the game but not reflected by in-game physics. That'll place too many constraints on creativity. Doesn't mean there shouldn't be structural limitations on it in practice, just that should be ways around it with proper  construction.

It is sadly a very common denominator for people to talk about wobble forgetting you're supposed to use the same tanks, structural parts, and many other stuff to create things that aren't rockets.

Edited October 6, 2023 by PDCWolf

[Bej Kerman]

21 hours ago, cocoscacao said:

22 hours ago, Bej Kerman said:

Sort of like the wings?

No. In discrete steps. Check structural tube from KSP 1. You have an option to set it's length from 1 to 6. Something like that.

Would you care to quote whoever said we'd get "partially procedural parts" and explain why it must  be how you  interpreted it?

[cocoscacao]

3 minutes ago, Bej Kerman said:

Would you care to quote whoever said we'd get "partially procedural parts" and explain why it must  be how you  interpreted it?

Nobody. Nothing must be as I say. I'm just presenting an idea (already exist in mods), that feels to me like a natural progression. I'm curious what other people have to say. 

[Bej Kerman]

2 hours ago, cocoscacao said:

Nobody.

So you just remember someone saying partially procedural.

[cocoscacao]

1 hour ago, Bej Kerman said:

So you just remember someone saying partially procedural

Nope. Coined it myself, since I don't know how else to call it. The point is...? 

[Bej Kerman]

1 minute ago, cocoscacao said:

1 hour ago, Bej Kerman said:

So you just remember someone saying partially procedural

Nope. Coined it myself, since I don't know how else to call it. The point is...? 

Why say "we still didn't get partially procedural parts" as if the devs promised such, if you don't think the devs ever promised such?

[Royalswissarmyknife]

I agree with @cocoscacao Idea

I actually have had an Idea like that myself for a while now

The inspiration was from SAF instead though.

Spoiler

Unrelated but I figured out why the forums sometimes didnt let me type. 

 

Edited October 6, 2023 by Royalswissarmyknife

[cocoscacao]

2 hours ago, Bej Kerman said:

Why say "we still didn't get partially procedural parts"

Poor choice of words I guess. If this problem was put on my back, that would be my first idea. Real rockets don't bend, sure... but real rockets also don't carry silly payloads that are (and should be) possible in a game. Since material stress/bending calculations are absolutely out of the question, wobble is the next best thing. Allow some over the top designs, but discourage insanity. However, when things start getting big, lowering the part count is one of possible optimizations. As others mentioned above, there are some drawbacks, but in my mind, you can set some reasonable limitations to them. 

[Infinite Aerospace]

To be fair they need to do both, I think the window for Kerbal Space Program II is fast closing. We're reaching the point where there's double digits (that's right, not triple, not quadruple, LESS than one hundred) people actually on this game at points. That is shocking player numbers, even for an early access title, I think unless the both the lack of content and the damned rocket deformation is addressed quickly, I don't see the game surviving I really don't.

 

I know we have this idea that a publisher will indefinitely continue to back them, but will they? What are Take Two getting out of this right now except for losing money, and really, really negative feedback. The game has no future unless something about the overwhelming consensus regarding the game can be changed, and even that is a tall order. You're talking about a swathe of people who have seemingly already made up their minds on the game.

 

For the record, I'm not one of those people, I think there's still immense potential here but I'm not going to fool myself into believing that it can continue how it is, that and I think the time for perceived excuses has came to an end, it's do or die time for the developers. Deliver something or watch the dream die, that point in the road is fast approaching.

3 hours ago, regex said:

I would hate to have a poorly thought out interim solution become the permanent one later down the road because the playerbase couldn't let go of it or we had moved so far forward in development that changing it would significantly affect established gameplay/require an extensive refactor.

At this point I couldn't care less if a better solution is found down the road, that breaks my craft. There isn't a compelling reason to even play the game to the extent where I'd have any craft to even lose.

 

I think 99% of the dwindling player base would take an interim solution, like Auto Strut at this point, if even just as a stop gap. I mean a game in active development is going to have times where your game needs to be started from scratch, it is what it is and I think we all accepted that buying into an early access.

 

But, in-action is just a giant middle finger to the community if I'm honest and I fully have the teams back at this point.

[Alexoff]

53 minutes ago, Infinite Aerospace said:

That is shocking player numbers, even for an early access title

In 2k23 it's not shocking, check Redfall, Forspoken or Gollum. But of course, these are far from the best examples and hardly anyone is happy to be in the vicinity of such games. Although they were expected and hyped no less than KSP2.