Did the developers take the wrong approach to KSP2?

[Monger]

Imho the development process is too intransparent yet to judge anything. We will see if they can deliver in the next few weeks. 

From a business perspective, imho they made the right calls: Make the game more appealing to new customers. KSP 1 was surprisingly successful, but due to its long development time didn't make that much money. There were several points in time where Squad was close to shutting everything down. You have to lure in more players, and for that you need to flatten the learning curve, make the graphics and sound quality a little nicer, provide tutorials, toss in a few features that you would expect from an AAA title. The hardcore gamers will grunt and moan, but they will stay until you can deal with them.

It's a sound strategy, but in order for it to work, they have to deliver regular, meaningful updates. We will see if that works out. 

Edited February 26, 2023 by Monger

[mattihase]

56 minutes ago, Opus_723 said:

I don't get why everyone is acting like the number of *vessels* is a big hurdle to the use of N-body physics. The very obvious approximation to make would be to have the planets on rails through a long pre-calculated N-body trajectory and vessels just responding to the resulting time-dependent field. It's not like our spaceships are gonna move the planets (or each other) appreciably anyway.

I mean in some sense that's partially true. Vessels don't need to be particles with gravity. Still, polling the positions of vessels is, at best, going to be a O(n) complexity calculation based on the number of vessels, so having the calculation be 1) simple and 2) not dependent on the results of itself from the last step is quite frankly a no brainer, especially if you have something like timewarp in your game which would necessitate either more calculations, more error, or both, and especially especially if you're planning for multiplayer in the future in which people will (presumably) need to be able to sync up between different points of time in the simulation.

 

As far as pre-rendered "on-rails" n-body trajectories go, people have timewarped centuries to line up good gravity assists for individual missions, on an interplanetary scale. There's no way the playerbase wouldn't quickly end up making it to the end of any reasonably sized prerendered trajectory set in their saves after a couple grand tour missions on an interstellar scale. Even if that wasn't likely, imagine how soul crushing it'd be to be the developer who has to take their hands off the keyboard and say "well, the game's just gonna have to stop working after a while".

[omega2003]

1 hour ago, dragonart777 said:

 i thank it is more the advertising that failed more then all.   

Yes, marketing made a big mistake: they made believe that everything was going well while the game is one of the most buggy I've seen. It's early access, but according to their marketing, it should have been much more stable. After all, aren't the developers supposed to play it all the time because it's so much better than KSP1? Given the gap between their marketing and the games delivered  yesterday, it's not surprising that they find themselves so poorly rated on steam.

I also think that the developers have nothing to do with it. They probably used the latest dev version without a debugging session because the release date was decided and their bosses didn't want to postpone it anymore.

I am one, to my great regret, of those who asked for a refund. Given all the current bugs and all the features they want to add, I don't see how they're going to get out of it. I would wait 6 months to a year to see how it evolves. I hope it will finally be good enough to buy it and enjoy playing it.

[Missingno200]

2 hours ago, Yakuzi said:

Please let this be the KSP themed origin story of what Cities: Skylines was to Sim City 5

Oh so we get a marginally okay and a somewhat bad space simulator, instead of just one somewhat bad space sim, and the marginally okay one is only "complete" with 331 dollars worth of DLC?

Yaaay... Looking forward to that one...

Edited February 26, 2023 by Missingno200
Taht.

[TheOtherDave]

2 hours ago, GregA said:

 I think it comes down to limited developer time.  Would you rather the developers be spending their time making a good space ship simulator, or a solar system simulator?  For a truly dynamic system like you are describing the developers would spend a bunch of time tuning it and getting it right, and those are resources that would be better spent making a good space ship simulator.  For the time being I think most folks reading and posting would rather the developers be working on their spaceship simulator, if forum thread titles are to be believed. 

"Developer" is a pretty broad category, though. It's not like every single developer is working on one problem, then every single developer pivots to another problem — they're almost certainly split up into teams. Are the people from one team going to drop what they're working on to help other teams? Maybe, but in a project this size you've gotta consider the on-boarding time and any other inefficiencies that might come from moving everyone around. Of course, I'm not saying it'd literally add 0 days to the schedule to do n-body, but I'm not sure the choices are binary as (my interpretation of) what you're saying makes them sound, either. Of course, even if it did literally add 0 days to the schedule, that still doesn't make it the right thing to do for the game as a whole — it'd definitely complicate the maneuver planning and that might turn off a lot of people — it's just something that I, personally, would like to see.

[Frostiken]

13 hours ago, garwel said:

I know what you mean. I have also complained elsewhere about KSP2's lack of interesting new mechanics, although I was talking more about things like life support, reliability, radiation, heat management, signal delay etc. They might also for example make kerbals more individual or autonomous (add some sort of AI). Planetary weather could be a very interesting addition.

I'm also disappointed that the developers basically took the easy road of just adding more celestial bodies and parts, something this community could have done very well on its own (and perhaps even better). There is still hope that some new features will be introduced later, but then the devs should think outside the box and try to not just create a remake + visual uplift.

I confess I'm thus far highly unimpressed with KSP 2. I also will say I followed literally nothing, I saw the teaser trailer however long ago and that was it.

KSP 2 is like... KSP 1.5.

Honestly the most shocking is that it looks like they straight up copied the same parts from KSP 1 and gave them a new coat of paint. I was at least expecting a lot of vanilla inspiration and implementation of established KSP 1 mod features, like having modular service bays vis-a-vis Universal Storage 2, or even do something completely different from the 'snap stack parts'.

Also I agree that Unity was a mistake. Unity is very easy and cheap to work with, very moddable too, but I've literally never seen a Unity game that grew bigger than 'indie game side project' and didn't completely fall apart. KSP 1 already felt like it was pushing the absolute limits of what Unity could do, and it didn't take all that many mods to suddenly run into major problems with the engine.

Edited February 26, 2023 by Frostiken

[Missingno200]

3 minutes ago, Frostiken said:

 was at least expecting a lot of vanilla inspiration and implementation of established KSP 1 mod features, like having modular service bays vis-a-vis Universal Storage 2, or even do something completely different from the 'snap stack parts'.

Well, the former... I'm not sure what that's about, but the latter I can give an answer to:They didn't want to change that because that's what made KSP the game that it is.
They were pretty upfront about making KSP2 a lot like KSP1, but with better onboarding, performance, and more features integrated in a very well balanced manner. So far, 2/3 are out, and the first one is still a bit of a stretch. Hoping that improves.

Edited February 26, 2023 by Missingno200
Making lots of typos today.

[Frostiken]

4 minutes ago, Missingno200 said:

Well, the former... I'm not sure what that's about, but the latter I can give an answer to:They didn't want to change that because that's what made KSP the game that it is.
They were pretty upfront about making KSP2 a lot like KSP1, but with better onboarding, performance, and more features integrated in a very well balanced manner. So far, 2/3 are out, and the first one is still a bit of a stretch. Hoping that impropves

Well, I don't know, the point of the thread is that they were far too attached to 'how KSP 1 did it'. Nobody who is into KSP 1 is going to NOT play KSP 2 if they came up with new ways of using parts beyond the 'lego' model. There's other options that don't have to limit creativity. Being able to procedurally shape and design rockets is an option. I don't know, but there's always new things you can do with technology to enhance experiences. Who among us is seriously going to say that four fuel tank length options was somehow a superior system to a procedurally-sized one? Given the popularity of the procedural tanks mod, apparently not many.

I dunno how long you've been following KSP, but do not forget that something like the Delta-V calculator was literally opposed to by the original dev team because they thought running out of fuel en-route was "Kerbal".

Edited February 26, 2023 by Frostiken

[omega2003]

Despite the opinion of the ksp1 developers, the deltaV map has always been very popular and many people used a deltaV calculator, often installed directly in game by a mod.

According original creators running out of fuel was in the category "failure is fun" but in practice everyone was trying to avoid it.

[Missingno200]

25 minutes ago, Frostiken said:

Well, I don't know, the point of the thread is that they were far too attached to 'how KSP 1 did it'.

Actually, this thread opened with "KSP2 is too much like modded KSP1."

21 minutes ago, Frostiken said:

There's other options that don't have to limit creativity. Being able to procedurally shape and design rockets is an option.

You're asking for RP or Juno. Lego stacking was a huge part of the appeal to KSP1. Furthermore, "limit creativity?" I disagree. Limitations breed creativity. Why would this suddenly limit creativity?

23 minutes ago, Frostiken said:

Who among us is seriously going to say that four fuel tank length options was somehow a superior system to a procedurally-sized one? Given the popularity of the procedural tanks mod, apparently not many.

We had this debate during the hype thread. It turns out, this was a HIGHLY controversial option and we couldn't come to a consensus. Someone brought up(at some point) that it's even ridiculous to suggest that lego stacking was a bad system, as it would emulate real life flexing, which happened quite a bit on old rockets.

21 minutes ago, Frostiken said:

I dunno how long you've been following KSP, but do not forget that something like the Delta-V calculator was literally opposed to by the original dev team because they thought running out of fuel en-route was "Kerbal".

I don't recall that ever happening, but then again I admittedly only played KSP1 once in a blue moon back in the day.

[Frostiken]

5 minutes ago, Missingno200 said:

You're asking for RP or Juno. Lego stacking was a huge part of the appeal to KSP1. Furthermore, "limit creativity?" I disagree. Limitations breed creativity. Why would this suddenly limit creativity?

Probably right around the part where we've seen KSP 1 utterly disintegrate when you add lots of parts and thus, the 'lego' style of design and implementation into Unity rapidly brought the entire game to the brink of self-destruction? Lego parts has enormous consequences for more than just design.

Quote

Actually, this thread opened with "KSP2 is too much like modded KSP1."

OP literally complained instead of making KSP 2 they just remade KSP 1. You are severely twisting words here. The part of the complaint about "Modded KSP 1" is the KSP 1 part. 

Also I don't even consider it 'modded KSP 1' given the volume and complexity of the mods I use.

Edited February 26, 2023 by Frostiken

[Missingno200]

1 minute ago, Frostiken said:

Probably right around the part where we've seen KSP 1 utterly disintegrate when you add lots of parts and thus, the 'lego' style of design and implementation into Unity rapidly brought the entire game to the brink of self-destruction? Lego parts has enormous consequences for more than just design.

Firstly, single core performance is a pretty big issue. That's why KSP2 is multithreaded. 
Secondly, that's not even the fault of lego design. That's the fault of using exclusively flexible connectors. They could take the Juno approach and then make all tanks merge into one(or have rigid connectors between them.) That would remove the flex and wobble if that was REALLY your problem.
Actually, your argument doesn't seem to understand the problem isn't "lego parts", but rather unusual physics implementation, combined with a bottleneck.

[Frostiken]

2 minutes ago, Missingno200 said:

Firstly, single core performance is a pretty big issue. That's why KSP2 is multithreaded. 
Secondly, that's not even the fault of lego design. That's the fault of using exclusively flexible connectors. They could take the Juno approach and then make all tanks merge into one(or have rigid connectors between them.) That would remove the flex and wobble if that was REALLY your problem.
Actually, your argument doesn't seem to understand the problem isn't "lego parts", but rather unusual physics implementation, combined with a bottleneck.

My point is that they didn't do anything new, and as near I can tell, from what you said, the entire decision was based on the demands of the extreme core fanbase (who were going to buy the game anyway), not because they wanted to make a new, interesting game with new, interesting design as a new, interesting installment. 

I'm willing to bet one of the real answers is that by copying the same lego design, it allowed them recycle the same art assets (which they CLEARLY did).

Edited February 26, 2023 by Frostiken

[Missingno200]

22 minutes ago, Frostiken said:

My point is that they didn't do anything new, and as near I can tell, from what you said, the entire decision was based on the demands of the extreme core fanbase (who were going to buy the game anyway), not because they wanted to make a new, interesting game with new, interesting design as a new, interesting installment. 

There's really no proof that the "extreme core fanbase" would've bought the game with or without the lego parts. I don't recall the extreme core fanbase even asking for the parts, actually, it was more the casual audience who jumped to the defense of the building style. The so called fanboys were more in favor of procedural EVERYTHING, including solar panels, structural parts, and of course tanks.

Your second part's also a mess. It sounds like you're complaining about the reuse of old mechanics that were proven reliable from the previous game. I mean honestly, "the entire decision to bring back no unit limits in Command and Conquer Tiberian Sun was based on the demands of the extreme core fanbase, not because they wanted to make a new, interesting game with new, interesting design as a new, interesting installment." Or hell, how about "the decision to bring back the fire flower in Super Mario Bros 3 was based on the demands of the extreme core fanbase, not because they wanted to make a new, interesting game with new, interesting design as a new, interesting installment."
You really sound like you just want to complain.

22 minutes ago, Frostiken said:

It allowed them recycle the same art assets (which they CLEARLY did).

You've made the claim! The burden of proof has fallen upon you! You're gonna need evidence to prove that they did this, because so far, it seems incredibly unlikely that they did.

Edited February 26, 2023 by Missingno200
The previous comparisons were... inadequate and more so referenced the genre. Made them more reference the games instead. More equal comparison as a result.

[whatsEJstandfor]

On 2/24/2023 at 3:05 PM, J.Random said:

My understanding is that it's impossible to have both in a consistent, reliable way. They don't transform into one another, so you either integrate all the time, or you have patched conics. Which is why I'm very suspicious of devs' claims that the 3-body will be implemented for the binary (and only there and nowhere else), and that it won't be done through barycenter hack.

They said they're using a bespoke solution, and you could treat Rask/Rusk/your spaceship as a restricted 3-body problem, for which exact solutions exist

EDIT: I just had to look it up and I might be wrong. Exact solutions exist for the case where two bodies are in circular orbits around a common point, and the small third body orbits them in the same plane, but, if that's the solution they used, you wouldn't be able to approach Rask or Rusk from the normal/antinormal directions I guess. So, yeah, I guess they still must be approximating, but, still, being a special case of the 3-body problem, it sure does seem like it'd be pretty trivial for numerical solutions to be computed by modern hardware.

Edited February 26, 2023 by whatsEJstandfor

[Chilkoot]

14 hours ago, dazoe said:

KSP2 was NOT a re-write. Looking at the disassembled code it is clear that KSP1 was used as a starting point and they just added more on top.

Don't know where you heard this, but a developer on Reddit ran the decompile and said there was essentially no similarities in the code base: https://old.reddit.com/r/KerbalSpaceProgram/comments/11b8s6f/i_looked_into_ksp2_code_here_is_what_ive_found/j9wm3dd/

[saxappeal89129]

14 hours ago, dazoe said:

KSP2 was NOT a re-write. Looking at the disassembled code it is clear that KSP1 was used as a starting point and they just added more on top. I believe that is why you can see the early KSP1 issuses popping up again. There is even KSP1 code that's not even used by the game, old dead code. At least i'm like 85% sure of that, my opinion is based off of just a quick browse through so I could be wrong.

Also reverse-engineering/disassembling is against the eula, which will hamper modding the game in the way KSP1 was modded. Funny thing is I decompiled the code before even launching the game and being presented with the EULA and I've decided to refund the game as I consider it a major step backwards from KSP1

 

This is frightening...

Would be interesting to get hands on meeting notes back back to Day 1... to find out at what level decisions were made. 

[Missingno200]

4 minutes ago, saxappeal89129 said:

This is frightening...

And also completely false. 2 fold, actually. EULAs aren't legally binding(as of US law,) and also we literally have counter-evidence it wasn't built on top of KSP1.
https://old.reddit.com/r/KerbalSpaceProgram/comments/11b8s6f/i_looked_into_ksp2_code_here_is_what_ive_found/j9wm3dd/

Whether or not you believe it, the fact is that at least this Reddit post SPECIFIES what is different, which is more than dazoe ever said.

[saxappeal89129]

25 minutes ago, Missingno200 said:

There's really no proof that the "extreme core fanbase" would've bought the game with or without the lego parts. I don't recall the extreme core fanbase even asking for the parts, actually, it was more the casual audience who jumped to the defense of the building style. The so called fanboys were more in favor of procedural EVERYTHING, including solar panels, structural parts, and of course tanks.

Your second part's also a mess. It sounds like you're complaining about the reuse of old mechanics that were proven reliable from the previous game. I mean honestly, "the entire decision to bring back no unit limits in Command and Conquer Tiberian Sun was based on the demands of the extreme core fanbase, not because they wanted to make a new, interesting game with new, interesting design as a new, interesting installment." Or hell, how about "the decision to bring back the fire flower in Super Mario Bros 3 was based on the demands of the extreme core fanbase, not because they wanted to make a new, interesting game with new, interesting design as a new, interesting installment."
You really sound like you just want to complain.

You've made the claim! The burden of proof has fallen upon you! You're gonna need evidence to prove that they did this, because so far, it seems incredibly unlikely that they did.

I dunno the parts DO look SIMILAR if not identical in many instances... whenever you're dealing with the senses i.e. vision, smell, touch it's all subjective particularly since not everyone has the same level of acuity. Just sayin', I think you're wrong! :-)

1 minute ago, Missingno200 said:

And also completely false. 2 fold, actually. EULAs aren't legally binding(as of US law,) and also we literally have counter-evidence it wasn't built on top of KSP1.
https://old.reddit.com/r/KerbalSpaceProgram/comments/11b8s6f/i_looked_into_ksp2_code_here_is_what_ive_found/j9wm3dd/

Whether or not you believe it, the fact is that at least this Reddit post SPECIFIES what is different, which is more than dazoe ever said.

Wow, incredibly defensive... makes me think that there's even more here.

[Missingno200]

1 minute ago, saxappeal89129 said:

I dunno the parts DO look SIMILAR if not identical in many instances...

Not proof enough. Similarity and reuse of assets are two totally different accusations.
(Why are you even here? You're not Frostiken.)

2 minutes ago, saxappeal89129 said:

it's all subjective particularly since not everyone has the same level of acuity. Just sayin', I think you're wrong! :-)

Then please, provide evidence. A part name that I can compare in both KSP1 and KSP2! There's plenty of parts that KSP2 does share with KSP1, but I can guarantee that I can show that the models and textures are totally different, thus sinking this asset reuse accusation once and for all.

[TheOtherDave]

3 hours ago, Missingno200 said:

We had this debate during the hype thread. It turns out, this was a HIGHLY controversial option and we couldn't come to a consensus. Someone brought up(at some point) that it's even ridiculous to suggest that lego stacking was a bad system, as it would emulate real life flexing, which happened quite a bit on old rockets.

I thought that was an odd argument… I played with Lego a lot as a kid, and for obvious reasons I’d always use one longer piece instead of multiple shorter pieces whenever possible.

[J.Random]

3 hours ago, whatsEJstandfor said:

They said they're using a bespoke solution, and you could treat Rask/Rusk/your spaceship as a restricted 3-body problem, for which exact solutions exist

EDIT: I just had to look it up and I might be wrong. Exact solutions exist for the case where two bodies are in circular orbits around a common point, and the small third body orbits them in the same plane, but, if that's the solution they used, you wouldn't be able to approach Rask or Rusk from the normal/antinormal directions I guess. So, yeah, I guess they still must be approximating, but, still, being a special case of the 3-body problem, it sure does seem like it'd be pretty trivial for numerical solutions to be computed by modern hardware.

I'm not talking about computing. It's doable, the awesome Principia mod is the proof of that. But, AFAIK (I'm not going to pretend I know all or even a lot, so I may be wrong too), you can't really use these two systems together. N-body can probably take over from conic patches ("here's the position and current velocity vector, here are the bodies applying force to the craft, go on, integrate") but I can't imagine the reverse, because this extra m/s accumulated from pull by 3rd body may lead to weirdness when you try to describe it as a single SoI orbit. Like, imagine you're in a halo orbit around L1 and switch from N-body to conics. Suddenly, you have radial velocity of the Mun at what, 3/4 or 4/5th of a distance from Kerbin, so suddenly you're on a hyperbolic trajectory (didn't calculate, but seems like it). They may have made the barycenter's SoI large enough so that there's negligible difference between calculation results, but high warp may still lead to glitches.

 

We'll wait and see, of course, but I have my doubts.

Edited February 26, 2023 by J.Random

[whatsEJstandfor]

2 minutes ago, J.Random said:

I'm not talking about computing. It's doable, the awesome Principia mod is the proof of that. But, AFAIK (I'm not going to pretend I know all or even a lot, so I may be wrong too), you can't really use these two systems together. N-body can probably take over from conic patches ("here's the position and current velocity vector, here are the bodies applying force to the craft, go on, integrate") but I can't imagine the reverse, because this extra m/s accumulated from pull by 3rd body may lead to weirdness when you try to describe it as a single SoI orbit. Like, imagine you're in a halo orbit around L1 and switch from N-body to conics. Suddenly, you have radial velocity of the Mun at what, 3/4 or 4/5th of a distance from Kerbin, so suddenly you're on a hyperbolic trajectory (didn't calculate, but seems like it). They may have made the barycenter's SoI large enough so that there's negligible difference between calculation results, but high warp may still lead to glitches.

 

We'll wait and see, of course, but I have my doubts.

Ahhhhh, I see what you're saying; I hadn't even considered that. I'm assuming, because they said so, that whatever solution they're using is able to handle this gracefully, but I'm definitely curious now to know how they're doing it.

[CMDRennie]

I think someone should start an n-body physics thread (if they haven't already). Clearly people have allot to say on this topic, myself included.

In fact, here:

Edited February 26, 2023 by CMDRennie
Added link to other thread

[CMDRennie]

6 hours ago, Missingno200 said:

Actually, this thread opened with "KSP2 is too much like modded KSP1."

While I don’t disagree, the point I wanted to get across was a dissatisfaction of core gameplay improvements and a lack of ambition. Of course I expect KSP2 to embody some of the most popular mods from KSP1, just as KSP did throughout its development.