Developer Insights #23 - Black Hole Sun

[mattihase]

1 hour ago, Oak7603 said:

My apologies, by wobbly rockets, I meant the rockets that start wobbling on the launch pad and then explode and then a relaunch fixes it. @herbal space program in the post above is proof taht this is still happening.

I feel like this is more an issue of lack of physics easing in the current game than the structural integrity being borked. If you can get past the first couple seconds (such as using cheats temporarily or launch clamps) it usually sorts itself out soon after.

 

Engine wise it's definitely among the more important features left that need to make their way into the game.

[cocoscacao]

2 hours ago, Oak7603 said:

the studio should be working towards fixing the small things that have a big impact, not this waste of time.

Those small things with big impact, aren't small things at all. To be fair, what @herbal space program mentioned, I feel for the guy behind this post. These miniscule additions add to the overall experience, and are well worth the effort. Also, "The studio" isn't just a bunch of dudes and dudesses(?) you can throw at one specific problem, and make it go away faster...

[The Aziz]

People complained that clouds were crappy. Something that adds nothing to gameplay. But they hired blackrack anyway. And behold, the stream of pictures of craft flying through pretty clouds. Nothing added to gameplay, yet people are happy. Because people like pretty things.

[steveman0]

10 hours ago, PDCWolf said:

It is not fixed, in the sense that the "fix" only accounts for a single configuration. In a game where you build land vehicles, aircraft, spacecraft, space stations, land bases, interstellar ships, and so much more, a solution that only ever works for vertically stacked tanks on the same central axis of the root part is useless. That's not even going into the performance tradeoff (huge) when you build big stuff with autostruts on.

Maybe you need to clarify what actually isn't fixed about it.  My experiences with it are that craft now much more realistically respond with an appropriate magnitude of bending forces to be realistic to materials that would be used in real life rockets.  I believe the only notable issue that even relates is as @mattihase has pointed out - that there aren't appropriate dampening forces in all situations that can allow a phantom force to sometimes feedback on itself in increasing magnitude to the extent that it can cause craft to rip themselves apart.  This is not fundamentally the same issue as wobbly rockets, it's fundamental to the modeling of the physics forces and not the joint stiffness.  This issue would overwhelm the rocket regardless of how stiff it is because it results in runaway forces.

Long thin metal craft with metric tons of mass at either end subjected to metric tons of forces are going to experience some degree of bending.  It isn't realistic to expect to build toothpick craft that don't bend under extreme load.  When you push the physical parts to the limits of the geometric construction, you should expect a craft to bend and fail catastrophically.  It's a fundamental failure mode for a craft of a given shape.  Where we are at in terms of the magnitude of craft design where this failure point is reached is just perfect in my view.  The noodly rockets we had were silly and making some of the things I've seen some people try to fly more rigid would be equally so.

Edited April 10 by steveman0

[PDCWolf]

4 hours ago, The Aziz said:

People complained that clouds were crappy. Something that adds nothing to gameplay. But they hired blackrack anyway. And behold, the stream of pictures of craft flying through pretty clouds. Nothing added to gameplay, yet people are happy. Because people like pretty things.

The atmosphere and clouds were unjustifiably bad for a 2023 game. They weren't cartoony, they were bad, buggy, ugly and killed performance. This was compounded by the atmosphere properties and rendering also being more or less the same, and them being unable to fix it for over 8 months. THAT is why they hired blackrack, because they were in over their head over something so basic, to the point it was cheaper and faster to hire an specialist than to teach themselves how to get it working.

Clouds and atmosphere weren't "a pretty thing", they are both the first thing new players see after the horrible UI, and were also making it hard to play the game because the starter planet has both clouds and atmosphere.

1 hour ago, steveman0 said:

Maybe you need to clarify what actually isn't fixed about it.  My experiences with it are that craft now much more realistically respond with an appropriate magnitude of bending forces to be realistic to materials that would be used in real life rockets.

And you made the same mistake. The game is mainly about rockets, sure, but it's far from the only thing that the game needs to portray. You don't see the ISS needing to strut the panels to the main hab for it to not spaghetti unto itself. You don't see the curiosity rover needing to strut the camera mast to the body because it's a surface attachment. Rockets work if you don't try anything too wild, anything outside rockets that happens to be big is still pretty much unaffected by the fix and grinds the game to a halt with auto struts.

"Realistically" is clearly a word you use without knowing what it means, unless we both get down into aerodynamic calculations to test bending forces of whatever material KSP is supposed to make its stuff of, and the configuration of attachment points (why are radial decouplers made of wet spaghetti?). The word you want is "intuitively", because it works to the idea you think you have about how a real rocket experiences forces and bending.

There's a whole discussion on wobble already so I don't feel like reviving the issue. What we have now is a bandaid for the most experienced (but just one), use case: single stack rocket with, at most, smallish radially attached boosters. Anything you build outside of that still flexes some, even the stuff that shouldn't.

[Meecrob]

16 hours ago, The Aziz said:

This thread has derailed long ago.

Oh heavens to Betsy, we forgot all about the thread where they joked about putting imperial units in KSP. Quick everyone, we must finish that riveting debate!

Seriously though, the thread isn't derailed so much as the topic is now something YOU do not wan't to talk about.

 

15 hours ago, Scarecrow71 said:

Stop harping.  Please.  You are literally just looking for something to be wrong where nothing is.

No, they are telling you that you think something is fixed because you have not seen the cracks yet. They have been very patient trying to explain it to people in this thread. It is not a difficult concept to grasp.

 

3 hours ago, steveman0 said:

Maybe you need to clarify what actually isn't fixed about it.

Why do people think the burden of proof is on the poster who went and did reserach on the topic, and not the poster who reads the post and goes "I don't believe you. I haven't looked, but you sound wrong."

3 hours ago, steveman0 said:

metric tons

What about imperial tons? Back to topic and all that. (I know it wasn't you who mentioned getting back to topic, I'm just making a bad joke)

Edited April 10 by Meecrob

[Scarecrow71]

9 minutes ago, Meecrob said:

No, they are telling you that you think something is fixed because you have not seen the cracks yet. They have been very patient trying to explain it to people in this thread. It is not a difficult concept to grasp.

And I'm going to die on this hill.  The original comment, without any clarification or qualifiers, was that rockets were still wobbly.  You can go look through this thread to see that for yourself.  And the truth is that the initial wobbliness was, is, and continues to be fixed.  All of this stuff about orbital colonies and ships with 1000s of parts came after the fact in an attempt to discredit my position that the wobbliness upon launch is fixed.  I even went so far as to state that it wasn't the best fix, and that if someone is building something with a bunch of stuff hanging off the side then yes, you are going to wobble.  But the central issue of wobbliness upon launch is fixed, and nothing anyone says can change that.  Doesn't matter what the fix is, or how it was implemented, or even if it's temporary and will be looked at later.  It's fixed.  Period.

Now please, stop harshing on me over this.  I'm not the enemy here.  If people want to discuss the original comment and my position, fine.  But don't tell me I'm wrong for answering the original comment by stating I didn't take anything else into account, or that someone qualified it after the fact.  I'm not a freaking mind reader.

Edited April 10 by Scarecrow71

[MechBFP]

8 minutes ago, Meecrob said:

Why do people think the burden of proof is on the poster who went and did reserach on the topic

Because that is how burden of proof works champ. The person making the claim has to prove it. No one here has proved that the issue isn’t fixed, all I have seen is some vague hand waving of “what about this theoretical situation that no one knows anything about?”  

[Flush Foot]

15 minutes ago, MechBFP said:

Because that is how burden of proof works champ. The person making the claim has to prove it. No one here has proved that the issue isn’t fixed, all I have seen is some vague hand waving of “what about this theoretical situation that no one knows anything about?”  

Indeed!

Happily we live in the former, not the latter 

[PDCWolf]

44 minutes ago, Flush Foot said:

Indeed!

Happily we live in the former, not the latter 

The original claim, the one we actually have proof from MechBFP, is that they fixed "wobbly rockets", by virtue of showing a like 100 meter tall single tower of tanks. We all agree that such a particular case is fixed. My claim, and that others like cocos echoed, is that such a fix is both a bandaid, and a performance hog, and also fails to account for a lot of cases, so the problem kinda really isn't fixed. Someone else (I think cocos too) claimed that rockets might still disassemble on launch on certain configurations.

Now, if you wanna blame me of something, sure, I'll spell it out for you: you can absolutely blame me for my lack of will to even bother opening KSP2, let alone designing anything in it, going through the trouble of launching, docking, and so on, just to grab a screenshot, or metrics of how performance is affected, and what not. So yeah, my claim, on my behalf, shall remain unproven if what you want is photographic evidence, on account of me having games I actually want to spend my free time with. Sorry I can't help with that.

 

Edited April 10 by PDCWolf

[NH4Cl Enthusiast]

Absolutely loving the shifting goalposts and retconning the wobble discussion. Back when they did the dev chat about it, some people were drumming very strongly that it's ok they took well over half a year after launch to even start talking about how to implement a fix because it needs to be solid and up to snuff, something that will be performant and not cause bugs or other issues with all the glorious features that will come in the future to allow building colonies and interstellar crafts and whatnot. As long as they do it right.

[Spicat]

Wobble is fixed, kraken attack are not (like the disassemble on launch people are talking about)

[Scarecrow71]

Just now, Spicat said:

Wobble is fixed, kraken attack are not (like the disassemble on launch people are talking about)

Thank you.  This is the difference between the original statement (Wobble is not fixed) and what cocos is actually complaining about (kraken attacks).  You just said it far more eloquently than I was trying to.

[cocoscacao]

1 hour ago, Scarecrow71 said:

And I'm going to die on this hill.  The original comment, without any clarification or qualifiers, was that rockets were still wobbly.  You can go look through this thread to see that for yourself.  And the truth is that the initial wobbliness was, is, and continues to be fixed.  All of this stuff about orbital colonies and ships with 1000s of parts came after the fact in an attempt to discredit my position that the wobbliness upon launch is fixed.

Nope. My original post here was that there are many unresolved core issues still present. Just one  of them happens to be wobbliness. Not wobbliness in the terms of now and then, but in terms of, it's a core problem if they intend to have much bigger scope now. Even devs stated this. Other core problems are mainly computational... You can find them in my posts above.

[Flush Foot]

1 minute ago, Spicat said:

kraken attack are not

#ThisHasAgedWell

[Presto200]

This is a whole big conversation, maybe make a new thread in Suggestion and Development for it? I personally disagree and would say it is fixed but I will say maybe they will later reiterate how they fixed it to handle your concerns, but for right now it is fixed as good as we need it to be

[The Aziz]

4 minutes ago, Flush Foot said:

#ThisHasAgedWell

Oh this again. Alright let's beat the dead horse: AS EVERYONE CAN SEE, Tom realized how big this claim was, backed from it and called it an ultimate goal. A goal that hasn't been reached yet, just like the game that ain't finished. 

That's all that needs to be said about it and I refuse to answer to any replies, goodnight.

[Scarecrow71]

5 minutes ago, cocoscacao said:

Nope. My original post here was that there are many unresolved core issues still present. Just one  of them happens to be wobbliness. Not wobbliness in the terms of now and then, but in terms of, it's a core problem if they intend to have much bigger scope now. Even devs stated this. Other core problems are mainly computational... You can find them in my posts above.

And my original response was to the original comment:  

On 4/9/2024 at 6:12 AM, Oak7603 said:

What about wobbly rockets.

After I responded to this directly, you came in and started making comments about specific situations that are, in fact, kraken attacks.  I stand by what I said:  As far as the original comment goes, the issue is resolved.  Temporarily or not, bandaid or not, autostrut or not, whether you agree with it or not.  I responded to the original comment, and I don't appreciate you harshing on me for that.

[Presto200]

Also I'm always a fan of communication and understanding through (civil) conflict. I just think that if its a conversation that has taken multiple pages on this April Fools post, it should probably have its own thread

[Flush Foot]

<deleteme.mods>

Edited April 10 by Flush Foot
Deleting comment as it relates to stuff that got purged

[cocoscacao]

6 minutes ago, Scarecrow71 said:

you came in and started making comments about specific situations that are, in fact, kraken attacks

Call them kraken attacks if you wish (very often attacks, I might add)... But that should give you a clue what it really is...

 

6 minutes ago, Scarecrow71 said:

Temporarily or not, bandaid or not, autostrut or not, whether you agree with it or not.

I bow to your infinite wisdom, and put my opinions in the trash can. It is fixed.

Edited April 10 by cocoscacao

[herbal space program]

3 hours ago, steveman0 said:

  I believe the only notable issue that even relates is as @mattihase has pointed out - that there aren't appropriate dampening forces in all situations that can allow a phantom force to sometimes feedback on itself in increasing magnitude to the extent that it can cause craft to rip themselves apart.  This is not fundamentally the same issue as wobbly rockets, it's fundamental to the modeling of the physics forces and not the joint stiffness.  This issue would overwhelm the rocket regardless of how stiff it is because it results in runaway forces.

What I was talking about with my Eve 10 stack was definitely this and not any inherent lack of rigidity causing my rockets to self-disassemble on the launchpad. With a few struts placed in the obvious places, I haven't had any problems at launch  with any of my craft since the most recent update. What I was talking about happens when the craft is in orbit, with no non-uniform forces acting on it, and although having SAS on seems to make it happen more often, it happens sometimes with SAS off as well.  It also only seems to happen  when I have a craft made from two or more  assemblies that were docked together in orbit. And it always starts very small and gets steadily more violent until the craft flies apart, which very much supports your model of incorrectly implemented dampening of small motions causing feedback loops in certain cases. FWIW,  it also seems to happen a whole lot less now than it did before, so they seemed to have fixed it in most cases, but it still does happen occasionally. 

[Spicat]

I think the conversation has spiraled enough and I think it would be more productive to upvote the associated bug report which is not very high (to be honest there is probably other ones but that’s the main bug I can think about):

 

[Flush Foot]

4 minutes ago, Spicat said:

upvote the associated bug report

I went to do just that, only to discover that I'd already upvoted it

[NH4Cl Enthusiast]

On the topic of the actual dev insight post, I think it encapsulates perfectly the state of the game and how the studio presents itself. This is how it reads to me:

 

Heyyyyy NASA collaboration yeeee we're real space science people here! Did you see it? NASA! We're like legit and know all the current science astronomy stuff! 

Eclipses are cool and we have them in KSP2! Well almost. They're not really eclipses and don't really work all too well but we can call them that cause it kinda is!? Here's math to throw you off! But in the future we'll have something awesome, maybe perhaps possibly for sure when the game gets good. In the future. For sure. Stay tuned for more insights! 

HEY HEY HEY community challengeeeeee! Meme time! Photos, memes, posts, we're totes vibing with the space community here!

 

 

No offense to the actual post though, it was ok but you must realize that if someone sees Kerbol randomly dimming for a moment, they're absolutely not going to look up in awe and admire the simulated eclipse? It's more likely nobody will even make the connection and understand what's happening.