There needs to be a better way to plan flights

[Dizzle]

So I have no idea if this will be addressed in the UI changes coming in 1.0 for real 1.1, but there has just GOT to be a better way to plan maneuver nodes. Having mostly kept my interplanetary missions to Eve and Duna, my experience with this has been present but limited until now. Right now I'm trying to get a probe out to Pol, and WOW it's difficult to see what I'm doing. I mean, what am I supposed to make of this?

 

 

I haven't been around the forums for a while so maybe I'm beating a dead horse, but seriously, what do I do here?

[RocketBlam]

I personally use MechJeb, because it addresses some of these shortcomings. I've said before that there needs to be some way to know when to transfer to another planet. If you're not using MechJeb, the only way to know when to burn is to consult some website that tells you the angles the planets need to be in, and with no way to measure angles in the game, it's a big guess as to whether you're in the right spot or not. 

I'm not sure if that's what you're referring to, but I agree it needs a better way to navigate to other planets.

[Goody1981]

For the further away planets if you fiddle around with the node until you get an intercept, then focus view on the planet/moon you're targeting. You can still pull the node handles, but now Pol (or wherever) is in the middle of your screen. Makes fine tuning the intercept easy

[The Destroyer]

11 minutes ago, RocketBlam said:

I personally use MechJeb, because it addresses some of these shortcomings. I've said before that there needs to be some way to know when to transfer to another planet. If you're not using MechJeb, the only way to know when to burn is to consult some website that tells you the angles the planets need to be in, and with no way to measure angles in the game, it's a big guess as to whether you're in the right spot or not. 

I'm not sure if that's what you're referring to, but I agree it needs a better way to navigate to other planets.

Or VOID/KER(I think)/protractor. I believe what he is reffering to is how hard it is to place manuvere nodes (especially when two lines are close together)

[Dizzle]

That's fine for getting to a planet, but as for getting Pol itself, focusing on it doesn't offer any new information unless there's already an intercept. The trouble is that with the jumbled mess in the screenshot, I have no way of figuring out how to align the orbits to actually GET an intercept.

[Alshain]

I agree (see signature)

[Venusgate]

22 minutes ago, RocketBlam said:

If you're not using MechJeb, the only way to...

Actually, KAC has transfer windows, not to mention Transfer Planner, but yeah, stock would be nice.

[Hannu2]

When you leave Kerbin's SOI you should plan a maneuver which brings you to about 0-200000 km periapsis. It is typically (not always) 100-300 days after escaping Kerbin (if you use low energy transfers). When you execute that maneuver you set a new one about 100 days before entering Jool's SOI. Wait until you are near that point. Then you focus to Jool and adjust orbit so that periapsis point is on Pol's orbit. If your midcourse correction was correct it should not take more than 20 m/s. Wait until ship enters into Jool's SOI.

Check periapsis and adjust if needed. Put maneuver to periapsis and adjust retrograde and normal velocities so that you have eccentric elliptical orbit somewhat larger than Pol's orbit on Pol's plane and periapsis at or very little inside Pol's orbit. Then put another node with zero dv on that orbit couple of days after periapsis. Map view should show target and your positions at closest encounter now. Use the first node and adjust orbit larger (decrease retrograde and keep inclination) so that you get encounter or near passing near periapsis. Wait and execute burn. Second node was because sometimes map view can not predict close encounters without it and you can now remove it.

After burn plan new midcourse correction which gives encounter with Pol, if burn was inaccurate. Pol's SOI is small and large burns are typically not very accurate. And few days before entering Pol's SOI adjust periapsis where you want.

These are not the most dv saving instructions but this is basic level which must be understood and learned to execute before more advanced gravity assists, distant plane change maneuvers or timing arrival so that you encounter moon at optimal position to brake. I recommend mods (I use MechJeb, Alarm Clock and Transfer Window Planner) because the stock map view is not very good and certainly not intended to this kind of accurate maneuvering. Squad's philosophy is moar boosters and harsh eyeballing and less nerdy stuff. I do not expect that it will change in any version. Fortunately the game is moddable to be fun for engineers too. But if you are purist it is possible in stock too. Just laborious and grindy (in my opinion) as every work without proper tools.

Edited December 16, 2015 by Hannu2

[Reactordrone]

It also helps to move the view so that you're looking down on the solar system rather than across it. Looking at that picture I have no idea what's going on.

[Dizzle]

9 minutes ago, Reactordrone said:

It also helps to move the view so that you're looking down on the solar system rather than across it. Looking at that picture I have no idea what's going on.

This only works when you're already on an even plane with the target.

[Plusck]

I disagree completely. I have always thought that the orbital path system is amazingly clear for the 3-D space and gravitational influences that it is representing.

There are glitches, sure, such as craft being offset from their pathways in interplanetary space, difficulties in placing nodes or whatever, that need to be addressed.

But the basic principle, the ability to switch focus, and the vast amount of information that you get just by zooming and moving around with the curser keys is just fantastic. IMO. It's basically what drew me into the game and had me playing the demo endlessly to start with.

[Boris-Barboris]

I always thought showing the target position phantom when hovering a cursor over your orbit patch would solve most of my problems: so when you point at your periapsis, map would always show where would your target be when you are at periapsis, regardless of wether you have intersection or not.

[CalculusWarrior]

10 minutes ago, Boris-Barboris said:

I always thought showing the target position phantom when hovering a cursor over your orbit patch would solve most of my problems: so when you point at your periapsis, map would always show where would your target be when you are at periapsis, regardless of wether you have intersection or not.

I agree, this should be included in stock and would improve rendezvousing by eye immensely. Of course, not to say I'm against improving the maneuver node planning system too, that needs work.

[Ehco Corrallo]

2 hours ago, Dizzle said:

So I have no idea if this will be addressed in the UI changes coming in 1.1, but there has just GOT to be a better way to plan maneuver nodes.

SQUAD has said that map view is a particular target for UI changes. They're definitely been working on the visual clarity. I'd assume that improvements to the map won't stop there. 

Edited December 16, 2015 by Ehco Corrallo
On the a visual clarity? I can write better than that, I promise you.

[WuphonsReach]

Personally, I refuse to use the little handles on the nodes and just use an addon like PreciseNode, which lets me use the number pad keys to make adjustments.  It's so much easier to use then trying to finely tune a maneuver node by dragging those fiddly little node handles.

For planning my transfer windows, I use the Transfer Window Planner add-on, which creates reminders in KAC (Kerbal Alarm Clock).  Execution of the nodes is either setup by hand, or by using MechJeb's porkchop plot tool.

[Red Iron Crown]

I use PreciseNode for this, for me being able to adjust a node when it's off camera is essential for interplanetary stuff. @Alshain's suggestion for an improvement to the stock node handling looks fantastic, I could probably do without PreciseNode if it was implemented.

[magnemoe]

6 hours ago, Dizzle said:

That's fine for getting to a planet, but as for getting Pol itself, focusing on it doesn't offer any new information unless there's already an intercept. The trouble is that with the jumbled mess in the screenshot, I have no way of figuring out how to align the orbits to actually GET an intercept.

You can put an node in interplanetary space to get your trajectory near Pol orbit.
You want this burn soon after exiting Kerbin SOI but getting to close is impossible because of inacuracy so the final adjustment has to be done later.
Now you also need to intercept Pol then its on the right side of Jool so you catch up with it.
first put one node some days out, then put another node inside Jool SOI and tweak it so you get very near Pol orbit. 
An final node to brake until you get an good intercept, if the braking is expensive you do the main part before entering Jool SOI. 

Mechjeb is nice to get the trajectory for intercept, you still have to do the braking nodes for hand. 

[CrashTestDanny]

The view in the picture you posted is not very useful.  If you right-click and pull your mouse down you will get an angle above the orbital plane of the planets, which will be easier to deal with.

When using the map view gets too frustrating, I do this - the earlier the better as the delta-V costs increase as you get closer to your target.:

1. set the ship to a prograde hold SAS or MechJeb - I prefer MechJeb.
2. wait for the ship to get to prograde (and oscillation to die down if using SAS).
3. use RCS translation keys to see how my orbit is impacted.  This works best if your orbit already enters the SOI that you're headed to and you focus that body - in your case, Jool is sufficient.  As you maneuver you should see your track within the Jool SOI move around and possibly cross its moons' SOIs.

The farther you are from your destination, the more change that will be made by a small adjustment, so if I can get my initial track within a couple million meters (close enough to show the closest approach and target position at closest approach), then I will start fiddling with it even while I am still in Kerbin.

Danny

[Geschosskopf]

15 hours ago, Dizzle said:

So I have no idea if this will be addressed in the UI changes coming in 1.0 for real 1.1, but there has just GOT to be a better way to plan maneuver nodes. Having mostly kept my interplanetary missions to Eve and Duna, my experience with this has been present but limited until now. Right now I'm trying to get a probe out to Pol, and WOW it's difficult to see what I'm doing. I mean, what am I supposed to make of this?

.........

I haven't been around the forums for a while so maybe I'm beating a dead horse, but seriously, what do I do here?

The underlying problem is that the whole maneuver node system was originally designed just to get to Mun.  And it works pretty well for that, but it has major inadequacies when the node and the destination are separated by interplanetary distances.  Despite much complaining from the community, the system has never been revised to address this.  But this horse is far from dead.  With Squad updating the UI, now is the time to register your opinions on the failings of the existing system.  So sound off loud and proud.

To me, the biggest problems with the maneuver node system are as follows:

1. Node Collapse:  Any time you change camera focus (say from ship to planet) or click on anything other than the node (such as a Pe marker), the node handles collapse  And at even the distance from Minmus to Kerbin, it's often hard to get the node to open up again without major camera moves, especially if you have other ships already in orbit near the node.  It's also quite possible to accidentally delete the node when trying to reopen it.
2. Inaccessible Node Handles:  The node handles are 3D things so Z is a factor.  Often times at interplanetary distances, when you're fine-tuning your Pe at the destination, you can't reach the retrograge (or some other) node handle due to other handles being "stacked" on top of it.  This is PARTIALLY ammeliorated by the (relatively new) ability to use the mouse wheel in reverse on the opposite handle but even that is difficult when the node's back at Kerbin and you're eyeballing Jool or something from short range.
3. Non-Linear Mouse Wheel Inputs:  The amount of change in the burn you get from moving the mouse wheel is dependent on the speed at which the mouse wheel turns.  This means you have to be VERY, VERY careful to move the wheel only 1 detent per second or so, otherwise you WAY overshoot the small change you wanted and then you have to waste a lot of time hunting for where you were before.  The mouse wheel should give a fixed dV change per detent regardless of wheel rotation speed.

As a fix, I recommend making a pop-up window for each node that you can drag anywhere in the map view.  This way, the node never collapses until you close its pop-up.  The pop-up would give access to all handles for large-scale pulls/pushes and small-scale mousewheel tweaks, PLUS have text boxes for each component of the burn where you could directly type in new values instead of fiddling with the handles.  And you'd also be able to slide the whole node forwards and backwards along your path from the pop-up.  That way, you could have full control of a distant node while having the camera focused in close on the planet you're trying to go to.

Patched conics also have several problems:

1. The patched conics projections of your future path completely ignore the effects of passing through atmospheres so setting your Pe for aerobraking is totally a trial-and-error process using much F5/F9.
2. You are unable to get patched conics to display more than 1 encounter at a time (such as entering Jool's SOI, then Tylo's).  No, you have to wait until you're inside Jool's SOI before you can target Tylo and get the closest approach markers and such.
3. There are multiple draw modes for patched conics but you can only change between them by editing the settings.cfg file.
4. You can draw a variable number of patched conics but can't change the number without editing the settings.cfg file.

These just require a whole new system.

To deal with the situation as it stands requires mods.  MJ and @regex's Precise Node are big helps with the node issues (and should be starting points for the official fix).  I believe both of these can deal with the patched conics issues as well.

To know when to go to other planets, you can also use Kerbal Alarm Clock, which can figure (approximate) times to go and prevent you from warping past them.

[problemecium]

In general I agree with OP. The existing maneuver system is completely usable once you get used to it, but that doesn't mean it's intuitive or that getting used to it isn't needlessly hard. If nothing else, some of the PreciseNode functionality needs to be duplicated in the stock game, and maneuver nodes need to be editable in the Tracking Station.

[DBowman]

Being able to place notional maneuver nodes on any objects orbit would really help with interplanetary transfer planning. Drop a node on Kerbin orbit, prograde till the predicted orbit reaches Duna orbit, drag it around to get an  encounter. It would smooth out the learning curve between Mun shots and interplanetary shots.

[AbacusWizard]

I'd love to be able to plan out maneuver nodes for a hypothetical craft (at the tracking station, presumably, or maybe mission control) without having to have an actual craft out in space. It'd make so much more sense for planning transfer windows especially.

[Der Anfang]

On 12/15/2015, 11:42:03, Venusgate said:

Actually, KAC has transfer windows, not to mention Transfer Planner, but yeah, stock would be nice.

What is KAC? I'm interested.

[Dizzle]

3 minutes ago, Der Anfang said:

What is KAC? I'm interested.

Kerbal Alarm Clock. Its main purpose is to remind you when to check up on your flights, but you can set it to show you when the launch windows are. It's extremely useful, I could never go interplanetary without it.

[Der Anfang]

2 minutes ago, Dizzle said:

Kerbal Alarm Clock. Its main purpose is to remind you when to check up on your flights, but you can set it to show you when the launch windows are. It's extremely useful, I could never go interplanetary without it.

OH THAT. I have that already. It's amazing. I love that too. I just wish it could automatically set up nodes like mech jeb does. Know of any mods other than MJ that does that?