Maneuver nodes are... really, really bad

[Gorby1]

+1 to all of these issues. It's bizarre to me that anyone thought this was an improvement over - or even as good as - KSP1 maneuver nodes.

[Draradech]

There are multiple bugs with the nodes, but I have found them to be accurate if:
- you switch to SAS hold directly before the burn
- you don't toggle map mode during the burn (resets the node)

Some discussion in here:  

 

[JoeSchmuckatelli]

39 minutes ago, Draradech said:

There are multiple bugs with the nodes, but I have found them to be accurate if:
- you switch to SAS hold directly before the burn
- you don't toggle map mode during the burn (resets the node)

Some discussion in here:  

 

I tried clicking on Maneuver after Prograde wasn't working.  That wasn't any better.  It's almost like you have to turn the SAS off?

Literally none of my planned maneuvers, timed with the burn timer, are coming out anywhere near where I hoped via the MN planning.  Thankfully I know how to manually correct these things; but new players would just be frustrated!

[Draradech]

Not SAS maneuver. That has its own bug. SAS hold works for me:

 

[Talka]

2 hours ago, JoeSchmuckatelli said:

I don't know if you caught Scott Manley's pre-launch ESA event video - but at one point he's pulling the nodes and it looks identical to the KSP node interface.  Skinny lines and all. 

On my phone so I can't time link to his video. 

Looks like the fat line thing was added just before launch. 

 

+1. This video illustrates perfectly how awful this system is.

[justspace103]

yeah the maneuver system is pretty frustrating. been encountering all these issues and more.  It seems to want to refresh when switching between map view and flight view

[Machinosaure]

Are you guys having issues also with the timing of the manoever? If my planned burn is longer than 10s, I can't follow the little countdown on the manoever, else i'll burn late and something will go wrong. 

[modus]

15 hours ago, JoeSchmuckatelli said:

I don't know if you caught Scott Manley's pre-launch ESA event video - but at one point he's pulling the nodes and it looks identical to the KSP node interface.  Skinny lines and all. 

Looks like the fat line thing was added just before launch. 

 

Yes! I just rewatched that. It looks so much better.

[Movado]

17 hours ago, JoeSchmuckatelli said:

I don't know if you caught Scott Manley's pre-launch ESA event video - but at one point he's pulling the nodes and it looks identical to the KSP node interface.  Skinny lines and all. 

On my phone so I can't time link to his video. 

Looks like the fat line thing was added just before launch. 

 

Here's the point in the video where Scott Manley is using the skinny maneuver node version...

So, yeah, the devs much have swapped it out for the fat boy version at release.

Edited February 27, 2023 by Movado
Forum bug duplicating the content of the post

[IbanLoL]

17 hours ago, JoeSchmuckatelli said:

I don't know if you caught Scott Manley's pre-launch ESA event video - but at one point he's pulling the nodes and it looks identical to the KSP node interface.  Skinny lines and all. 

On my phone so I can't time link to his video. 

Looks like the fat line thing was added just before launch. 

 

Yes, I did see some streams from that event and the graphics of the nodes have been changed indeed. The fat lines we have make the node slightly busier, but it does look a bit better.

The nodes are too busy anyways (even with the skinny lines) and what's missing is the ability to control the nodes even while you're not completely zoomed in on it, so you can see how the trajectory is affected by what you're doing.

At the moment, it's either you're fully zoomed onto the node, so you can pull the lines but you don't see your trajectory, or you're zoomed out so you see your trajectory but you can't properly adjust the node...

They'll improve this for sure, they'll have to. It's already pretty hard to plan inter planetary transfers, so I can't imagine how it would be like for interstellar transfers with this system...

[modus]

9 minutes ago, IbanLoL said:

The fat lines we have make the node slightly busier, but it does look a bit better.

You think the fat lines look better?  Imo they look terrible 

[IbanLoL]

1 hour ago, modus said:

You think the fat lines look better?  Imo they look terrible 

Well they may look a bit more "modern", or polished. I'm trying to find some positive aspects

[modus]

I admire that.  I'm having a hard time finding anything good about this maneuver system

[Tazooka]

1 minute ago, modus said:

I admire that.  I'm having a hard time finding anything good about this maneuver system

Same. I prefer the old system a lot more. Especially as it works, accurate and doesn't bug out

[Fluke]

In case anyone missed this... there is a mod to make the ap/pe sticky again. Works great, tested it and could adjust a Mun intercept by about 100m at a time. 

 

https://forum.kerbalspaceprogram.com/index.php?/topic/213117-sticky-orbit-markers-show-appe-during-maneuver-plotting/

[angelatthetomb]

What's needed is threefold:

 

-Ability to see orbits propagating into different SoIs as they are adjusted (I'm 100% sure they'll fix this).

-A way to observe orbital parameters changing live as you adjust the node, with numbers and not just visuals. A popup "orbital parameters after maneuver" window would be best, so that you don't have to look at the actual location of whatever parameter you're interested in. Bonus points if it includes parameters after an SoI change (i.e., shows you Mun periapsis even though the node is in Kerbin periapsis - KSP 1 never got this). Less confident this will happen.

-A method to adjust the maneuver node without looking at it. Best solution: a "maneuver node editor" that is not as ugly and finicky as KSP 1. It should appear near the lower-center of the screen (where there's some convenient dead space) and should have a large-scale maneuver node widget as part of the UI. This gives you a duplicate of the maneuver node widget that exists in the world, but is in a fixed location, easily accessible, and always available whenever you're in map mode with a maneuver node. The layout of the SAS hold axes widget provides a useful analogue. Above or to the right should be numerical input options.

[MarcAbaddon]

Honestly, I would love for them to just copy the Precise Maneuver editor mod, with a simple mode with only half the options for newcomers.

[Talka]

I still can't get past the fact that they've been internally playtesting this game and didn't change these maneuver nodes. Apart from bugs and a lack of optimization, it's the main constraint on this game being playable. Maneuver nodes are infuriating to use, and they should be one of the easier things to improve. They're also already a solved problem! KSP1 does them satisfactorily.

Anyways, I tried playing again for a couple hours today, and again gave up on an interstellar mission after frustrating myself fighting the maneuver nodes. It's such a shame. 

[Bej Kerman]

Just now, Talka said:

I still can't get past the fact that they've been internally playtesting this game and didn't change these maneuver nodes. Apart from bugs and a lack of optimization, it's the main constraint on this game being playable. Maneuver nodes are infuriating to use, and they should be one of the easier things to improve. They're also already a solved problem! KSP1 does them satisfactorily.

KSP 1 maneuver nodes are broken on a fundamental level, considering your maneuver as only an impulse. For burns longer than a few seconds, and especially for capture burns, the predicted orbit will be completely wrong. KSP 2 does it right and properly accounts for acceleration over the course of the burn.

1 minute ago, Talka said:

Anyways, I tried playing again for a couple hours today, and again gave up on an interstellar mission after frustrating myself fighting the maneuver nodes. It's such a shame. 

May I ask what kind of engine you're using to attempt this.... interstellar mission?

[cvusmo]

Agreed. Nodes are quite frustrating to make. There is a mod for fine tuning buy I just want what we have in KSP1.

 

The fine tune adjustment and how nodes work in KSP1 is awesome. 

 

I do like the new SOI. It makes it easier to get into orbit.

 

Warping on a manuever node path is a nightmare. If you misclick, you miss your entire node. Also the time is calculated wrong many many times for nodes.

[Kirarozu]

On 2/25/2023 at 7:18 AM, 机械主教71号 said:

Can't agree more. It's totally a disaster.

What's more, the node can't automatically pick the AP/PE, AD/DN, etc. 

We want the KSP1 type maneuver node.

Agreed. I don't know why they changed it to Arrows instead of the actual symbols. They are so tiny you can't really tell which one you're clicking on until you click on it. They should just use the KSP1 node.

[Talka]

1 hour ago, Bej Kerman said:

May I ask what kind of engine you're using to attempt this.... interstellar mission?

Whoops . Obviously meant interplanetary.

1 hour ago, Bej Kerman said:

KSP 1 maneuver nodes are broken on a fundamental level, considering your maneuver as only an impulse. For burns longer than a few seconds, and especially for capture burns, the predicted orbit will be completely wrong. KSP 2 does it right and properly accounts for acceleration over the course of the burn.

With KSP1 maneuvers, I can start a maneuver at half the predicted burn time to get a perfect orbit starting from the precise point I want (e.g., from Ap/Pe). With KSP2 maneuvers, I can't precisely time them around Ap/Pe because the maneuver nodes represent the start of a burn rather than the center of one.

[funkymove]

Please bring forward, the maneuver node editor functionality from KSP1. The current one in KSP2 is a kluge, at best.
Aside from the obvious floating point error problems, wobbly rockets, awkward VAB UI (because it attempts to be a common UI for both planes and rockets, despite their orthogonality), and awfully scripted smoke plumes... missing fidelity to the KSP1 maneuver node editor is a big loss for existing KSP fans, will make it harder to onboard new fans for KSP2, and reduces playability to reach other planets, much less not providing the requisite precision for interstellar travel (which was expected in KSP2).

After a couple hours of frustration with KSP2, I will NOT be playing it again, until it has, at least, reached parity with KSP1.
Especially after spending $49 for open alpha quality software, with very little of the release's planned functionality included, and who's predecessor both costs less, is more playable, and has a developed ecosystem.
After shelling out $49 for an effectively unplayable game, I'll check back in a year or two.
If you want the similary disappointed (people) to come back sooner, make sequels that are at least as playable as their predecessor.

Or, close the alpha, and discontinue the bait and switch tactics.

[Bej Kerman]

23 minutes ago, Talka said:

1 hour ago, Bej Kerman said:

KSP 1 maneuver nodes are broken on a fundamental level, considering your maneuver as only an impulse. For burns longer than a few seconds, and especially for capture burns, the predicted orbit will be completely wrong. KSP 2 does it right and properly accounts for acceleration over the course of the burn.

With KSP1 maneuvers, I can start a maneuver at half the predicted burn time to get a perfect orbit starting from the precise point I want (e.g., from Ap/Pe). With KSP2 maneuvers, I can't precisely time them around Ap/Pe because the maneuver nodes represent the start of a burn rather than the center of one.

Not really. As I said, they predicted orbits and did not at all take into account acceleration. If you have a low thrust craft and you want to achieve 0 inclination, you have to start your maneuvers well ahead of the 50% rule else you cut your engines at a much higher/lower latitude than the maneuver would anticipate. Only in KSP 2 can you properly do a low thrust inclination change using maneuver nodes (image credit: @Draradech)

The 50% rule was just a workaround for KSP 1's extremely flawed system that didn't always result in "perfect" orbits. It only correctly approximates no-gravity-whatsoever environments where the maneuvering object does not change in mass. I'm struggling to find the right phrasing, but I think @Immabed worded it perfectly:

On 2/26/2023 at 7:25 PM, Immabed said:

In KSP 1, the node assumed instantaneous change in velocity, so in practice it was essentially centred in the burn (though not completely, since decreasing mass as you use fuel mean the second half of the velocity change takes less time). This was simply a workaround for the node's lack of understanding of motion during the burn. In practice, if you start a burn ahead of the node, you never actually pass through the location of the node exactly, as your trajectory is altered ahead of the node, and depending on the burn this could be quite significant. I'm sure many of us can remember instances where our actual trajectory took us a bit into the atmosphere or otherwise did not match the presumed trajectory if you just look at the node itself, particularly when the burn itself was quite long in comparison to the period of the current orbit.

It didn't really deliver perfect orbits. Just an extremely vague approximation you can only approach with extremely high accelerations, unless you are dealing with extremely minor changes, like fine tuning intercepts. Again, KSP 2's system blows this issue out of the water. By starting the burn at the node, you can extend the burn without the start time creeping backwards, and the node remains anchored where your old orbit ends so it isn't part of the constantly changing mid-acceleration section of the orbit, which would complicate interacting with the node, especially when you make use of ion engines, and eventually, torch drives and interstellar engines, where the "middle of the burn" would creep outwards into interstellar space as you pull the prograde handle.

As demonstrated in Immabed's image, you can properly account for burn times when engineering orbits. The new maneuver system does exactly what you say the old one does, it puts you in the orbit you asked for, properly accounting for your acceleration (and probably change in acceleration, but I haven't tested that yet). You're allowed to complain that the maneuver isn't in the middle like you're used to it being, but in KSP 1 you couldn't do what's shown in the image and time the burn of a low-acceleration craft to land you precisely in a low-inclination orbit. This is just one of those changes you might need to get accustomed to when KSP 2 eventually (and hopefully) takes off, like procedural wings and workspaces. It's different but that doesn't make it bad.

[Lekiffer]

IF you focus on the body you're going to the blue circles will show where you will enter and exit SOI. Whilst focused at body click any part of your manoeuvre node (doesn't have to be the middle) then u can adjust it accordingly whilst being focused at the object ,sometimes you have to close the node for it to update your SOI markers so make a small change close manoeuvre node to see results, that combined with where you will enter the SOI gives you a clear path of where you will pass the body. You also get AP and PE height on the navball ball . Iv visited every planet and moon like this so it definitely works