[old thread] Trajectories : atmospheric predictions

[dlrk]

What's happening when it won't process atmospheric entry after transferring from a Mun with a planned maneuver node? It works fine after the SoI change.

[dlrk]

The latest release works in 1.2.2

[Kobymaru]

17 hours ago, dlrk said:

What's happening when it won't process atmospheric entry after transferring from a Mun with a planned maneuver node? It works fine after the SoI change.

Well it doesn't process atmospheric trajectories after SOI boundaries.

[brymck]

On 4/29/2017 at 4:13 AM, Kobymaru said:

It's not just about the decoupling force. The main issue in your case is that the craft has totally different aerodynamics after decoupling. If you ditch your heavy engine and only have a tiny light capsule, that is going to cause a lot more drag.

And there is really not a whole lot you can do to fix that. You could guesstimate how much it drifts after decoupling. The most precise way is to leave some RCS thrusters on the final capsule with which you can correct your trajectory.

Would it be possible to add to the mod a toggle button to show "final stage" trajectory based on the aerodynamics of the final stage to help guesstimate better? I think that plus a low force decoupler would probably make landing much easier. Usually I just quicksave and quickload a few times to try to get it close but...I don't know, maybe I'm the only one who's experiencing this.

[TheRagingIrishman]

1 hour ago, brymck said:

Would it be possible to add to the mod a toggle button to show "final stage" trajectory based on the aerodynamics of the final stage to help guesstimate better? I think that plus a low force decoupler would probably make landing much easier. Usually I just quicksave and quickload a few times to try to get it close but...I don't know, maybe I'm the only one who's experiencing this.

The problem is that it's very very hard (basically impossible) to find the aerodynamics of the final stage.

[Kobymaru]

1 hour ago, brymck said:

Would it be possible to add to the mod a toggle button to show "final stage" trajectory based on the aerodynamics of the final stage to help guesstimate better?

That would be super complicated, because one would have to keep a copy of the vessel object around and predict what it looks like after the staging events, and change all the aerodynamic properties accordingly.

Quote

I don't know, maybe I'm the only one who's experiencing this.

No, far from it. It's the most requested feature right along flight-window trajectory display.

But I'm no magician, don't know how to do these things and don't have nearly enough time. If you happened to be a programming wizard, however, I would love to accept pull requests for this feature!

 

ps.: Like I said, the way that people do this is is leave like 2 rcs thrusters on the final capsule with a few drops of RCS to correct the final trajectory.

We have bigger problems anyway, because right now the prediction is off by like 30 km.

Edited April 30, 2017 by Kobymaru

[linuxgurugamer]

@Kobymaru

I notice that when it's updating, it is posting a screen message around line 800:

Util.PostSingleScreenMessage("rho info", /*"preciseRho=" + preciseRho.ToString("0.0000") + " ; " +*/ "rho=" + approximateRho.ToString("0.0000") + " ; actual=" + actualRho.ToString("0.0000") + " ; ratio=" + (actualRho / approximateRho).ToString("0.00"));

I don't see any way to silence it.  If it's ok with you, I'm going to put in a flag which will determine whether or not screen messages get displayed, this way the Suicide Burn mod can avoid the messages while using it.

Let me know if you have any problems with this

LGG

Edited May 1, 2017 by linuxgurugamer

[Kobymaru]

1 hour ago, linuxgurugamer said:

I notice that when it's updating, it is posting a screen message around line 800:

This is only visible in the Debug build.

 

1 hour ago, linuxgurugamer said:

If it's ok with you, I'm going to put in a flag which will determine whether or not screen messages get displayed

No need, i already have a commit on my private branch that does that. If it really bothers you, I will merge that commit into the master repository.

[linuxgurugamer]

33 minutes ago, Kobymaru said:

 

No need, i already have a commit on my private branch that does that. If it really bothers you, I will merge that commit into the master repository.

Could you please, it's messing up my testing

Thanks

[Kobymaru]

2 hours ago, linuxgurugamer said:

Could you please, it's messing up my testing

Done. BTW, there's nothing that prohibits you fromkeeping local changes and private branches around, you know?  

[linuxgurugamer]

I know, rather busy right now

[dlrk]

Hey, @linuxgurugamer, would it be possible get some directional info in the trajectory window? Distance and direction instead of just distance? Also, could the trajectories' target be slaved  to the Mechjeb/ ScanSat target? Thanks for considering

BTW, the accuracy is fine with the FAR pre-release

[linuxgurugamer]

7 minutes ago, dlrk said:

Hey, @linuxgurugamer, would it be possible get some directional info in the trajectory window? Distance and direction instead of just distance? Also, could the trajectories' target be slaved  to the Mechjeb/ ScanSat target? Thanks for considering

BTW, the accuracy is fine with the FAR pre-release

Ask @Kobymaru, I'm not doing this mod

[Kobymaru]

24 minutes ago, dlrk said:

Distance and direction instead of just distance?

Doable, create an issue here: https://github.com/neuoy/KSPTrajectories/issues , try to describe what you want and what you need it for.

 

24 minutes ago, dlrk said:

Also, could the trajectories' target be slaved  to the Mechjeb/ ScanSat target?

Sorry, I'm not really interested in this feature. If you provide code, I'll merge it.

As a workaround, you could use the Waypoint functionality or enter the coordanites manually into the text field next to the set button.

[dlrk]

I'll write up an issue thanks?

What's the waypoint functionality?

[Guest]

On 12/23/2014 at 2:25 PM, Youen said:

In it's current version, the mod has some bugs that make it highly inaccurate with stock game aerodynamics. This is a known issue, and might be fixed in the future, somewhere between now and the end-of-time (we welcome contributions about that). But it should be fine with FAR.

 

FAR doesn't work in 1.2.x

[dlrk]

The semi-secret pre-release does

[Kobymaru]

13 hours ago, dlrk said:

What's the waypoint functionality?

You can create Waypoints in KSP and "activate navigation" to them. After you have done that, click "Active Waypoint" in the Trajectories targeting section.

Edited May 2, 2017 by Kobymaru

[juanml82]

I've tried this in 1.2.2 (stock aero) to calculate an aerobreak at Sarnus (from the OPM mod) and whenever I go to the map view, the FPS collapse to 3 fps while the console gets spammed by

WARNING: infinite loop? (Trajectories.MapOverlay.initMeshFromOrbit)

Is it an known issue with this mod? I'm also getting "reentry simulation timed out" in mechjeb's trajectories calculator, so maybe there is a conflict somewhere.

[Kobymaru]

6 hours ago, juanml82 said:

I've tried this in 1.2.2 (stock aero) to calculate an aerobreak at Sarnus (from the OPM mod) and whenever I go to the map view, the FPS collapse to 3 fps while the console gets spammed by

WARNING: infinite loop? (Trajectories.MapOverlay.initMeshFromOrbit)

Is it an known issue with this mod? I'm also getting "reentry simulation timed out" in mechjeb's trajectories calculator, so maybe there is a conflict somewhere.

First time I hear about this, but it sounds like a legit issue. 

I don't have OPM, so I don't know much about it. Is there anything special about Sarnus? Is it particularly big or something?

Does this happen all the time, or only on certain trajectories? Are there any other planets where this happens?

Could you please create an issue here: https://github.com/neuoy/KSPTrajectories/issues

Edited May 3, 2017 by Kobymaru

[Guest]

On 5/1/2017 at 7:21 PM, dlrk said:

The semi-secret pre-release does

Where is it?

[TheRagingIrishman]

58 minutes ago, dundun92 said:

Where is it?

ferram4 said that he only wanted people who know where to find the dev build to be using it. The logic behind this was that the type of people who can provide useful bug reports will, most of the time, know where stuff like this is and he really only wants dev builds to be used for bug hunting. If you haven't found it on your own then you are sol.

[linuxgurugamer]

@Kobymaru@Kobymaru

Have an odd question:  Right now, Trajectories calculates the impact position of the active vessel.  Do you think it's possible to do that for another vessel which isn't the current active vessel?

[Kobymaru]

4 hours ago, linuxgurugamer said:

Have an odd question:  Right now, Trajectories calculates the impact position of the active vessel.  Do you think it's possible to do that for another vessel which isn't the current active vessel?

Should be theoretically possible, but I don't know if the code changes would need to be too big to be realistically doable, I'd have to check.

Why do you need this anyway?

[linuxgurugamer]

2 hours ago, Kobymaru said:

Should be theoretically possible, but I don't know if the code changes would need to be too big to be realistically doable, I'd have to check.

Why do you need this anyway?

I was trying to have the Suicide Burn mod work on vessels which aren't the active vessel;  ie:  dropped probe, etc.