neuoy / ksptrajectories Goto Github PK

Not tested yet.

If I'm coming in for an aerobraking, having my orbit part turn white and atmospheric part red is great. However, when I'm not touching atmosphere the entire orbit is white.

I would prefer if this mod didn't come into play at all until part of my orbit touched atmosphere, and at that point showed the white/red.

Or even just drop the white altogether until you've hit atmo, and then instantly go red. That way, the white lines would only show after exiting atmosphere at least once.

Hi there, great mod.

Question: The angle of attack, is that relative to the surface velocity vector, or the orbit velocity vector. I'm guessing the first, since this mod seems targeted at planes.

Possible bug: I think -180 and 180 degrees give the same landing prediction? Currently they don't (Not off by much though)

For exemple a flag, or another vessel. KSP already has an interface for that, and other mods improve it (haystack for example), we just need to retrieve the current target position and use that to update the nav ball indicators.

Also would be nice to be able to set GPS coordinates (for example scansat can output these for anomalies, but currently no way to visualize it on the map view)

Ideally, this should be serialized in the game save files.

When pointing somewhere on the vessel trajectory, it would be helpful to see the target location at the same time. This is something stock KSP does with "nearest approach" markers, but sometimes it's not flexible enough (only one marker is displayed, and the player can't choose where on the trajectory)

Trajectories is causing the signal delay to break in RemoteTech.

Pods usually reenter backwards, but the current AoA is clamped to 30°.

I get incredible lags and stutter whenever I leave the ground (just after T/O with a plane/rocket).
There's a "Warning: Too many iterations." messsage at the bottom right, telling me prediction is stopped, but it still lags immensely, as in: One frame every 2-3 seconds.
Disabling trajectory prediction on the map screen options GUI doesn't change anything either.
The output_log doesn't give me anything related to Trajectories or any other mod, however, so dunno how to help reproduce this for you.

Among other things, this causes hyperbolic orbits to be rendered incorrectly, and makes it impossible to plan aerocaptures while still outside the target body's SOI.

A forum user reported inaccuracies when using this mod (but did not confirm if it was better without fairing or not)

For players that have a lot of mods, using the Blizzy toolbar makes more sense than the stock KSP toolbar

Also, add a setting to choose one, the other or both (like chatterer)

Currently, its size is constant, and it will get bigger than the whole kerbin system when scrolling out to view the solar system.
An other option would be to simply remove the marker when scrolling out too far.

Bug reported by marce on the forum and PM

"I created a new tiny vessel which only uses Stock+DR parts (maybe a few other modules, but I think KSP will just ignore them).
I also did a new album showing how much the prediction is wrong (I'd say 1/5th of the planet?). Hope that helps you!"

vessel file available on request if someone wants to test

KSP assumes velocity change is instantaneous at maneuver nodes. But we could compute the maximum vessel acceleration, and use that to predict a more accurate trajectories. This would mostly matter for low acceleration vessels, for example when using electric or nuclear propulsion.

Make the plugin work with NEAR aerodynamic model (auto-selection at game startup)

http://forum.kerbalspaceprogram.com/threads/84978-0-24-x-Stock-Drag-Fix

It should work, but sarbian reported it doesn't, so I have to double check.

Z-fighting between a impact marker and the surface was noticed in the hilly terrain of Minmus.
Z-fighting between the two markers was noticed upon setting the current marker as the target (on Minmus).

I'm on Linux, with nvidia, if that matters.

I wonder if it's requested but i wonder: can it also be used with maneuver nodes?
Because i want to plan maneuvers ahead and that will help even more with accurate landings.

Display a descent graph, indicating altitude over surface distance, and also ground altitude (would be handy to plan mountain fly-by)

Both stock brakes and other mods. Use the current state of brakes (even if the ship is still in space) to know if they will be deployed during entry or not.

Note: I'm currently working on this.

• put everything in the same window (for now)
• button to show/hide the window
• possibility to move the window + save position in file
• tooltips and more explicit presentation
• hide the window in tracking station

Even without atmosphere, impact prediction is not accurate. Looks like I'm not using the correct angular velocity or something.

By using e.g. a Verlet integrator, the simulation accuracy could be boosted significantly with little effort.

Currently, in map view, even if trajectories are not displayed, they are still computed

At this time, only drag is computed.

KSP x64 on linux. If "display trajectory" is turned on and the active vessel is on an escape trajectory, the game slows to a very slow crawl when in map view. Using most recent beta release of Trajectories.

To formalize (and second) a suggestion made in #18, it would be nice to be able to predict whether a given trajectory is survivable under DRE. Specific things I'd like to see:

• Peak temperature of the heat shield
  • whether any other parts will overheat would also be useful, especially for spaceplanes, but might be prohibitive for large vessels
• Amount of ablative used (or, at least, whether or not the shield will run out). Note that running out of ablative is not necessarily fatal, if it happens late enough, so any warning messages should be based on peak temperature predictions instead.
• Consistent use of custom DRE parameters (i.e., the contents of the REENTRY_EFFECTS configNode) and custom heat shields (like the RSS pack)

Using KAS, a Kerbal in EVA with a parachute attached makes the game lag in map view, and the trajectory seems incomplete

Or any other method to indicate how reliable the impact position is. For example by performing multiple predictions with slighly varying starting conditions to see if the trajectory is affected by the chaos theory.

Basically the title says it all: when I'm in the atmosphere, my trajectory is not shown.

This only occurs when I have not exited the atmosphere yet.

Profile the mod and search possible areas of optimization.

http://forum.kerbalspaceprogram.com/threads/93685-0-24-2-Trajectories-v0-3-0-%282014-10-01%29-atmospheric-predictions-with-FAR-NEAR?p=1453846&viewfull=1#post1453846

Example use cases:

• recovery mod that could simulate debris reentry trajectory, and estimate if the debris will be destroyed by heat or impact

This is noticeable mostly around the periapsis of highly eccentric orbits.
As a proposed solution, in addition to the 256 points that are spaced equally, time-wise, along the patch, add 360 points that are spaced equally, true-anomaly-wise. I.e. for each degree of true anomaly change, add one point.

reported by TMS on the forum : "the game begins to chug when the active craft is destroyed (by ploughing it into the ground). Lots of frames dropped until you leave the scene. "

Display what is necessary to follow the trajectory from IVA

At this time, ground is assumed to be horizontal near the crash location detected at sea level. A better solution would be to ray cast along the trajectory to find the first ground impact. In particular, it's important to detect mountains.

bug reported on the forum, and I've seen it from times to times in the log as well:
"My debug log was getting spammed with orange text messages like “Mesh.vertices is too small. The supplied vertex array has less vertices than are referenced by the triangles array.”"

Bug reported by Chimer4 on the forum : the trajectory is not displayed when he uses the space plane plus Explorer craft.

Find some way to specify which stage will enter atmosphere, and don't use the parts that will be discarded during predictions. Also need to check if FAR aerodynamics is more complicated when staging than just ignoring the discarded parts.

Add debugging indications in the GUI window:

• performance counter (time spent in computations)
• error counter (number of exceptions thrown during trajectory computations)

I don't know if it's kosher to post a bug report about a fork in the issues of the original project, but I didn't see a way to post it to the fat-lobyte fork. Also I believe this is the same issue mentioned on the forums and acknowledged today by baythan. I'm posting here to document the issue more methodically.

Trajectories prediction does not match FAR's latest model.

Defect reproduction method: Put a rocket in an orbit with a high apoapsis (~3Mm) and periapsis of ~65km. As the rocket passes through the periapsis the predicted next apoapsis will fall. Before I adapted my strategies this resulted in me setting up aerobrake orbits which were predicted to have two orbits after the first aerobrake, but actually ended up descending completely in the first aerobrake.

If this is expected behavior a warning would help, but I can also understand if this kind of surprise is by design and a warning would defeat that, in which case feel free to close this as "no issue" or "won't fix" or whatever. :)

I'm including the useful portion of my 'ckan list' for completeness' sake (why yes, I do have a lot of memory, why do you ask? :).

- AmbientLightAdjustment 1.3.1.1
- AutoAsparagus v0.8
- AviationLights 3.7
- CommunityResourcePack 0.3.2
- DistantObject v1.5.1
- DistantObject-default v1.5.1
- EditorExtensions 2.5.1
- EnhancedNavBall 1.3.4
- EnvironmentalVisualEnhancements-Config-Default 7-4
- EnvironmentalVisualEnhancements-HR 7-4
- FerramAerospaceResearch v0.14.6
- HullcamVDS 0.33
- InFlightWaypoints 1.2.7
- KAS 0.4.10
- KerbalEngineerRedux 1.0.14.1
- KerbalWeatherSystems 0.5.3-alpha
- LandingHeight 1.3
- MechJeb2 2.4.2
- ModuleManager 2.5.9
- PlanetShine 0.2.2.1
- PlanetShine-Config-Default 0.2.2.1
- PortableScienceContainer 1.2.1
- ProceduralFairings v3.11
- RasterPropMonitor-Core v0.18.3
- SafeChute 1.6
- SAVE 0.9.17-387
- SCANsat 8.1
- ScienceAlert 1.8.4
- SmartStage v2.5.0
- StageRecovery 1.5.3
- StationScience 1.4
- TCShipInfo 0.3
- TextureReplacer 2.2
- Toolbar 1.7.8
- Trajectories 1.1.2
- TWR1 1.17c
- USITools 0.3.1
- VNG-Parachute 1.2
- VNG-Parachute-EVALoading 1.0
- VNG-Plugin 0.7.2

left click to show/hide trajectories, right click to open GUI, and find a way to make it obvious for players so they don't miss the GUI completely

Also add an option to choose between stock or blizzy's toolbar (when it's installed)