Think I have the fluctuating orbits thing figured out

[tseitsei89]

8 minutes ago, JJE64 said:

Jesus Smoking Christmas, guys.  I made a simple observation.  @NathanKell was good enough to come on here and address it specifically and in great detail (thank you, Nathan!), without making excuses or obfuscating.  Can we possibly have one thread that doesn't involve us collectively keel-hauling the devs over everything?  I'm the last person to defend Squad's business practices, and I've been very vocal in my opinions about them, but even I have to admit the rancor is reaching the point of ridiculousness.  

If you continue to chastise these guys every time they speak, even when they're giving answers that they don't have to give, in levels of detail that they're not obliged to provide, eventually they're going to stop communicating with the community.  He could have just given a simple, "gee, thanks, that's interesting" and let it go, but he didn't he tried to be forthright with us, but we just can't resist the urge to punch down at every opportunity.  Give the guys a break, already.

Thank you, @NathanKell for addressing my OP.  I appreciate it.

This times million!

Calm down guys. It's just a game after all...

And NathanKell has been super nice and helpful while explaining the situation in this topic. I know I wouldn't have remained so calm if people threw that kind of s**tstorm at me when I'm just trying to explain the current situation...

[Waxing_Kibbous]

This is a stupid question, but is there a reason that once a player establishes an orbit or rendezvous that the path/s can't get locked into place, and then the COM would move along the paths (at least in vacuum situations)? I imagine calculating the velocity along each part would be pretty straightforward, mathematically generate an animation curve. Rotations wouldn't affect anything, but thrust would, re-establishing the new path that would get locked in again (on rails?). Even going through an atmosphere would act as thrust, and coming back out into a vacuum would put the player back on the new locked-in path. Maybe I'm thinking of it too simply (or maybe Unity just doesn't work that way), but I used to enjoy rigging in Blender, and these are the kinds of problems one needed to solve.

[Fearless Son]

34 minutes ago, Majorjim said:

How and why would I have a single part craft in orbit? I guess that is why I never noticed it. So it happens now under a different guise for multi-part craft now am I right?

Orbital rescue contracts.  They typically involve a single part that can have crew inside it (a cockpit, passenger section, or science module) floating in orbit with no other pieces.  Take one of those contracts, leave it alone long enough, and watch what happens.  

Granted, few people would leave helpless Kerbals stranded that long after taking a contract specifically to rescue them, but it could happen.

[Majorjim!]

3 minutes ago, Fearless Son said:

Orbital rescue contracts.  They typically involve a single part that can have crew inside it (a cockpit, passenger section, or science module) floating in orbit with no other pieces.  Take one of those contracts, leave it alone long enough, and watch what happens.  

Granted, few people would leave helpless Kerbals stranded that long after taking a contract specifically to rescue them, but it could happen.

Riight, I never play career so I would never have seen it.

[Greenfire32]

2 hours ago, The_Rocketeer said:

So @NathanKell, a dev, comes on the forum and tells us:

• they've identified 2 separate bugs
• they know the causes of the bugs
• they already have a partial fix
• it is already in next patch
• the problem has been around for years and nobody ever really noticed it
• they have ideas to pursue for how to reduce it further
• that he, personally, took time out of his vacation to look into this

and your response is basically to say how little you appreciate being patronised?

My problem isn't with NathanKell, it's with how bad KSP has been treated by the Devs since we hit "1.0" status. Before if something was significantly wrong with an update, Squad simply didn't release it. Yeah it caused some people to get angry with the delay, but hey, at least they still had a working version of the game.

Now after 1.0 hits, it seems we're just releasing updates for the sake of meeting deadlines. Stability be damned.

My issue isn't with NathanKell, he's merely the messenger in this case and as such is on the receiving end of the "crap flinging." Unfair, I know, but also how it works. I am grateful that he's doing work, but I am ungrateful that the game is in the state that it's in, in the first place.

So I'll compromise. @NathanKell, sorry for taking it out on you as an individual,  but I am absolutely fed up with Squad releasing game breaking updates and calling them the "next version" of KSP.

I've been with KSP since before the mun was in the game and it irks me that exploding landing legs, immovable wheels, and exponential decay in orbits is considered "acceptable losses" for an update. These are things that make the game. We have to be able to use these things.

Now if the random name generator for our Kerbals breaks in the next update, I don't think anyone would be too hot and bothered about that. But not being able to land somewhere, use an airplane or rover, or have a satellite in orbit? Those are game-breaking bugs. We shouldn't have had an "update" until those were fixed.

And what makes this all worse is that we're now in "consumer release" since going 1.0. These kinds of bugs are twice as unacceptable because now it's affecting the general population of consumers instead of just a small community of what is essentially beta testers.

This is not a pattern I am willing to support. Release stable versions, Squad. Not hotfix after hotfix after hotfix after hotfix...

2 hours ago, The_Rocketeer said:

I think some people round here could probably use a break from KSP - they take this stuff way, way too seriously.

I cant play KSP due to all the aforementioned bugs plus the random crashing this last update has seemed to have added, so no break is necessary. Let's not forget that the reason why some of us are voicing our rather angry opinions is because we love KSP and want to see it become an awesome game. We don't scream and yell about this or that because we hate the game.

Edited May 27, 2016 by Greenfire32

[NathanKell]

@JJE64 et al, thanks for the kind words. Much appreciated.

Looks like I need to do one more bit of clarification though. Note that I said "masked". I did not say "overrode" or anything like that. The issue that you can perceive in prior updates with single part craft is present for multi-part craft as well, just the visual feedback is masked by a separate bug (the flickering). Once again, it is not a question of that bug (orbital drift) being new. It is not. You have lived with it for three years. It just hasn't been shoved in your face due to a second bug.

Further, note that the drift stops when you are on rails, both before and after 1.1.1. That has not changed.

[Mako]

52 minutes ago, Greenfire32 said:

And what makes this all worse is that we're now in "consumer release" since going 1.0.

 

I've been seeing this sentiment a lot regarding the 1.1.x releases. People seem to be getting hung up on the word release, and I'm not sure why. It's just a word; it doesn't mean that much without context. And in this case the context is that KSP is still an ongoing project with all the good new improvements and features and all the bugs big and small. "Consumer release"  are not usually still under active development and don't usually receive new features without DLC or expansions (which usually, but not always, require additional purchases). KSP is still being actively developed with new features and major improvements continuing since "release," so I find it hard to compare KSP to traditional game development.

So maybe reevaluate how you look at KSP. It is unfinished as defined by the development team's own actions. Why not just think of it as you would if the magic word "release" was never used? The fact that it was used hasn't changed the reality of what KSP continues to be. The word release and the 1.0 version were only ever really used to allow the game to begin console development; those things never changed how the development team has treated the project. In fact, you could argue development has been ramped up even more after 1.0 with the addition of several new devs, NathanKell included.

This game has not been "released" by traditional standards, so why get worked up over a word that doesn't really apply. Play 1.0.5 as the current stable version; play 1.1.2 as the beta, bleeding edge version; or wait a while and play the last version Squad puts out (whenever that comes).

I do understand the frustration. We all want things to just work, but when software is still under active and major development and you're getting your hands on the latest version there will be bugs and broken mods. Because of this I've taken to waiting for final or close to final release for early access games. I've discovered I both enjoy and dislike the roller coaster ride early access comes with thanks to KSP, and while I'm glad I got in on it when I did, I've decided to be patient when it comes to other titles. This is just my perspective on the matter.

 

I'm loving the deep development discussion we're getting from @NathanKell. I really miss the old dev posts about development.

Edited May 28, 2016 by Mako

[Geschosskopf]

5 hours ago, Kerbonaut257 said:

But this is a literally gameplay breaking bug. Many of us expect our orbits to, ya know, remain stable. Especially for things like remote-tech. I literally have to resort to cheating the orbital periods to equal because as soon as I focus the craft, it begins getting a shorter orbital period. Then, if I want to re-aim the dishes on said satellites, then I have to go back in and cheat the orbit back again. This is ridiculous.

It has been ever thus.  You're only noticing it now because it's happening faster.  The KSP interface lacks the precision required to stick an RT satellite exactly where you want it, and even if you use various mods to improve your accuracy, you'll still never get it quite right.  Why?  Well, all orbital calculations involve the irrational pi, so there are ALWAYS inherent rounding errors.  Then there are the additional rounding errors due to using floating point values, all before you get to any flaws in the calculation logic.  The result has always been, and always will be, that you can never get an RT satellite exactly where you want it, and it will never stay where you leave it.  Even pre-1.0, you couldn't go a month without having to adjust your satellites.  If you let it go longer than that, your whole network would fall apart.

This is one of the 2 main reasons (besides the whole concept just being too 1960s for my thoroughly modern space program) why I quit using RT well before even 1.0 came out.  It's just too high-maintenance, even when orbits didn't fall apart as quickly as they do now.  Once you get to a certain size of network, you have to spend all your time constantly correcting things, so can't do anything else.  Sure, you can put kOS on every satellite so they can correct themselves, but that also reaches the point of diminishing returns fairly quickly.

Basically, the whole RT thing is doomed due to inherent limitations of digital computers trying to approximate analog math.  And that's before you consider anything involving Unity or Squad.  And this has me utterly and completely terrified of the impending, possibly game-breaking bomb on the horizon in the form of some sort of mandatory communications network becoming stock.  I really, really wish that whole idea would go away.  If folks want to mess with such things, let them use mods.  But don't force players to need something that the inherent limitations of the game prevent from ever having long-term stability.

 

5 hours ago, Kerbonaut257 said:

I can't imagine the fix is THAT difficult. And would have been done in mere days if the patch hadn't been released before vacation. I doubt that was squads idea. But seriously. Who posts a big patch RIGHT before vacation without time to QA & bugfix? I am here to complain directly to whoever made that decision.

OK, go fix it yourself then.  I have no doubt you could fix it by tomorrow, given that you consider this a trivial problem.

[Andem]

4 hours ago, Mako said:

 

I've been seeing this sentiment a lot regarding the 1.1.x releases. People seem to be getting hung up on the word release, and I'm not sure why. It's just a word; it doesn't mean that much without context. And in this case the context is that KSP is still an ongoing project with all the good new improvements and features and all the bugs big and small. "Consumer release"  are not usually still under active development and don't usually receive new features without DLC or expansions (which usually, but not always, require additional purchases). KSP is still being actively developed with new features and major improvements continuing since "release," so I find it hard to compare KSP to traditional game development.

So maybe reevaluate how you look at KSP. It is unfinished as defined by the development team's own actions. Why not just think of it as you would if the magic word "release" was never used? The fact that it was used hasn't changed the reality of what KSP continues to be. The word release and the 1.0 version were only ever really used to allow the game to begin console development; those things never changed how the development team has treated the project. In fact, you could argue development has been ramped up even more after 1.0 with the addition of several new devs, NathanKell included.

This game has not been "released" by traditional standards, so why get worked up over a word that doesn't really apply. Play 1.0.5 as the current stable version; play 1.1.2 as the beta, bleeding edge version; or wait a while and play the last version Squad puts out (whenever that comes).

I do understand the frustration. We all want things to just work, but when software is still under active and major development and you're getting your hands on the latest version there will be bugs and broken mods. Because of this I've taken to waiting for final or close to final release for early access games. I've discovered I both enjoy and dislike the roller coaster ride early access comes with thanks to KSP, and while I'm glad I got in on it when I did, I've decided to be patient when it comes to other titles. This is just my perspective on the matter.

 

I'm loving the deep development discussion we're getting from @NathanKell. I really miss the old dev posts about development.

I just have to say, people didn't decide to release KSP. SQAUD did. They made the decision that they were leaving early access, but don't seem to understand that there are consequences to taking that label. If they don't wnat the game to be viewed that way, they shouldn't have brought it out of early access. The reason I cannot  honestly ignore the term release is because SQAUD made the active decision to put that monicker onto KSP. We don't need to reevaluate how we look at KSP, SQAUD needs to reevaluate how they percieve the difference between earl access and released. The word, in context or not, has meaning too it. Meaning that I am unsure KSP has earned at this point.

[gchristopher]

On 5/26/2016 at 10:51 PM, NathanKell said:

Thanks, that's very interesting! I think, though, that that might be an artifact of other weird stuff. What's happening under the hood is that KSP uses Unity PhysX with a timestep of 20ms, and that's really bad for orbits. Prior to 1.1.1, (i.e. 1.1, 1.0.5, etc) KSP just applied the force of gravity (as calculated at the vessel's CoM) over that entire timestep (i.e. told PhysX that for that timestep there was a force of x m/s gravity * craftmass in the direction of planet's-core) To lower the error from that, I implemented a naive version of RK2 where the final force applied was the average of that force, and the force of gravity you'd have after that frame (i.e. after the gravitic acceleration was accounted for in velocity, velocity * 20ms was applied to position, and grav force recalculated for new position).

However, over break (I am bad at vacations) I was talking with ferram, and he poked me into looking at it again saying "no, obviously that will lead to continual decay". And it does. Because (if you, dear readers, haven't already seen this) the force of gravity at the end of the frame will always be stronger, so we're averaging a correct-at-that-point grav force with one that is stronger. (That's for reasonably circular orbits; highly elliptical ones can get weird, AIUI).

So there's three takeaways here:

1. Euler sucks.

2. Floats suck (We do all rails orbital calcs in doubles, and we calc the force of gravity using all doubles, but the force applied is a float because PhysX)

3. Badly implemented smarter integrators are not necessarily helpful.

Yep, naive Euler integration does not conserve energy. I've been on a team worked with RK  integration and they also had a heck of a time with implementation details. Is KSP orbit calculation simple enough to fix the orbital decay with leapfrog integration? I think if it's restricted to a 2-body problem, that might be all you need, and it's simple to implement. (This does not solve the floating point precision issue at all, though.)

[Mako]

@Andem

I probably should not have post that in this thread since it's only tangentially related, but in response I want to say that I completely agree with you and anyone else who says that KSP's current state does not match the traditional use of the word "Released." If it were my decision to make I would have chosen a different way to market the game since people do have expectations tied to that word. I also want to say that Squad or anyone else could use the word "released" or "Version 1.0" or they could have called it "extended beta" and Version 0.95 and so on (or to be extra dramatic and hyperbolic since that's the reason for the internet they could have called it "Popsicle" and Version Eleventy.One), and it wouldn't change a thing about the actual software or Squad's intentions. It might change your perception of the software, but really you're the only one who has any control over your perception. What we see right now we'd see no matter how they referred to it. Ongoing development means major changes for good AND bad (gotta take the good with the bad), and completed development, traditionally "released," means no more major changes or features.

Obviously we can see that Squad labelled it "released" which traditionally means no more major changes, but they are not limiting themselves to the traditions of the label. I'm glad for that. And in this case the only reason I can see for even sticking that label there, since they obviously had no intentions of keeping with the tradition of that label (again, I'm thankful), was so they could begin console development. The early access release model is only now becoming available on consoles, and I'm sure it's a lot more difficult to deal with than with the PC environment. It would likely be far simpler to say 1.0 and then provide updates, and it appears that Squad chose that route. I don't know if they had many options if they wanted to pursue a console release.

To try to be as on topic as possible I'll just say that whatever the case may be, I'm glad for continued development, I'm glad for bugs being so thoroughly investigated and fixed, and I'm glad for the discussion and insight into this bug and others. Anyone could have said "well, it wasn't broken before, and now it is," but that wouldn't be accurate. It was broken before, for a long time, and the quick fix turned out to be a fix that broke things in a new and interesting way that happened to be more noticeable and problematic than before. And the quick fix shall soon be properly fixed. So I really have no complaints.

Edited May 28, 2016 by Mako

[Xavven]

All software has bugs. It's unavoidable. Let's get some perspective on life and realize that KSP's imperfect orbital calculations are not worth getting upset over. Go outside and take a break.

[radonek]

On Friday May 27, 2016 at 11:40 PM, Waxing_Kibbous said:

This is a stupid question, but is there a reason that once a player establishes an orbit or rendezvous that the path/s can't get locked into place, and then the COM would move along the paths (at least in vacuum situations)?…

I had exactly same "bright" idea and somebody from Squad was kind enough (they probably hear this often) to answer and point out why it would not work. In short, instead of slow drift craft would "jump" when going off rails. This was in fact exhibited in older versions of KSP during rendezvous when other craft would jump when entering physics engine distance.

On Friday May 27, 2016 at 1:55 AM, NathanKell said:

@JJE64 et al, thanks for the kind words. Much appreciated.

Looks like I need to do one more bit of clarification though. Note that I said "masked". I did not say "overrode" or anything like that. The issue that you can perceive in prior updates with single part craft is present for multi-part craft as well, just the visual feedback is masked by a separate bug (the flickering). Once again, it is not a question of that bug (orbital drift) being new. It is not. You have lived with it for three years. It just hasn't been shoved in your face due to a second bug.

I can attest to that. I learned the hard way to watch carefully low munar orbits long before 1.1. They were never very stable and in one particular case, I've seen craft to spontaneously deorbit.
 

[Renegrade]

...I'm a lot more concerned about the phantom forces that landing struts seem to create than any tiny orbital decay.  I've been making reusable landers with cubic octags for legs for a while now, and it's getting a bit silly.

On 2016-05-28 at 10:26 PM, Geschosskopf said:

Basically, the whole RT thing is doomed due to inherent limitations of digital computers trying to approximate analog math.  And that's before you consider anything involving Unity or Squad.  And this has me utterly and completely terrified of the impending, possibly game-breaking bomb on the horizon in the form of some sort of mandatory communications network becoming stock.  I really, really wish that whole idea would go away.  If folks want to mess with such things, let them use mods.  But don't force players to need something that the inherent limitations of the game prevent from ever having long-term stability.

I suspect the "Stock RT" thing will be more like "Stock AntennaRange", FYI...no need for GPS-like constellations of satellites, just something with a long range antenna nearby.   I'm cool with that (and otherwise agree with your rant about RT )

[Majorjim!]

On 05/28/2016 at 8:15 AM, Xavven said:

All software has bugs. It's unavoidable. Let's get some perspective on life and realize that KSP's imperfect orbital calculations are not worth getting upset over. Go outside and take a break.

 I hate to break this to you but the purpose of this website is to discuss KSP. If it has bugs those will be discussed here.

Feel free to be shocked by this revelation.

Edited May 29, 2016 by Majorjim

[The_Rocketeer]

4 minutes ago, Majorjim said:

 I hate to break this to you but the purpose of this website is to discuss KSP. If it has bugs those will be discussed here.

Feel free to be shocked by this revelation.

My turn to obliterate your preconceptions:

There's a BIG difference between 'getting upset about' and 'discussing' bugs. If 'getting upset' is happening, going outside is a really, really good idea.

[Majorjim!]

Just now, The_Rocketeer said:

My turn to obliterate your preconceptions:

There's a BIG difference between 'getting upset about' and 'discussing' bugs. If 'getting upset' is happening, going outside is a really, really good idea.

What preconceptions would they be exactly?

 Nothing good ever happened 'outside'. 

[Xavven]

1 hour ago, Majorjim said:

 I hate to break this to you but the purpose of this website is to discuss KSP. If it has bugs those will be discussed here.

Feel free to be shocked by this revelation.

Hey now, there's no point to starting a fight we me over the internet. Obviously this forum contains criticism over one of my favorite games. I'm not personally offended by that.

But it irks me that some people would beat up a developer who calmly and professionally explains how the bug made it into version 1.1.2. The guy even took time during his vacation to work on it and some people in the community proceed to act all entitled about KSP. You weren't even one of the people I was directing my comment at, but apparently that warranted rude sarcasm.

[Majorjim!]

Just now, Xavven said:

Hey now, there's no point to starting a fight we me over the internet. Obviously this forum contains criticism over one of my favorite games. I'm not personally offended by that.

But it irks me that some people would beat up a developer who calmly and professionally explains how the bug made it into version 1.1.2. The guy even took time during his vacation to work on it and some people in the community proceed to act all entitled about KSP.

This I agree with.