[1.3.1] Ferram Aerospace Research: v0.15.9.1 "Liepmann" 4/2/18

[Viatic_unicycle]

What they are saying Matuchkin, is that they are using an unstable beta that still has bugs but is available in Ferram's github. It's not intended for general release and is considered for "advanced users" since you can expect to come across a bug or something else wonky. The obvious benefits to doing these types of releases are the, hopefully, bountiful beta testers and subsequent bug reports that follow.

[ferram4]

@soulsource: Repro steps?  I don't think I'll need the log for this one, but I thought I covered all the ways that buttons could keep appearing.

@kcs123: If your findings are limited to non-specific, "it all looks good," just post it here, that'll be fine.  If you've got a specific issue, I'd prefer it if you put it on Github.  Makes tracking the issues easier and I should really get in the habit of using Github's issues more, so... Yeah.

@Matuchkin: Basically, what Viatic said.  It's a small barrier to entry to ensure that most of the people messing with the dev builds actually have an idea of how development works, either because they've been here for awhile or they've done their own development of stuff so that the feedback I get has a better signal:noise ratio.  Sorry, but this is to make things move faster; if you're able to figure things out yourself though, there's nothing stopping you from getting the dev build.

[Phineas Freak]

@ferram4 for the button duplication:

• Enter either the SPH or the VAB
• Launch a vessel
• Recover the vessel

At this point there will be two FAR toolbar buttons in the Space Center view. Additionally, you will get these buttons when you exit to the Main Menu.

[ferram4]

Try with the latest build.  I've messed around a bunch more with that and I think it might be fixed.

[Feni]

3 hours ago, ferram4 said:

 

@kcs123: If your findings are limited to non-specific, "it all looks good," just post it here, that'll be fine.  If you've got a specific issue, I'd prefer it if you put it on Github.  Makes tracking the issues easier and I should really get in the habit of using Github's issues more, so... Yeah.

Build looks good, played some scenario's. Aircraft behave as expected (recently built a plane with CoL behind CoM in stock but not in FAR, did not fly well.. ;-)). Kidding aside, I don't have multiple FAR buttons and no other issues with the way I'm playing. 

An addition I don't remember: I set up a plane with control deflection of lets say 20. When flying, open the window of the control surface and check the control deflection value. If you fly faster, the deflection value decreases, when flying slower the value increases again. I know this dynamic control deflection function was in FAR before, but it didn't visually alter the value if I remember correctly. I like it, it gives you more insight in the behavior of the craft.

[Phineas Freak]

37 minutes ago, ferram4 said:

Try with the latest build.  I've messed around a bunch more with that and I think it might be fixed.

OK, the multiple buttons are fixed. The only remaining "button" issue is that it shows up in the Main Menu but this is not critical for the correct operation of FAR.

Edit: found another way to make FAR duplicate buttons:

• Load a save from a fresh KSP instance
• Exit to the Main Menu
• Reload the previously loaded save.

At this point, two FAR toolbar buttons will appear. Upon exiting to the Main Menu again only one will appear though. if you load the save for the third time you will get two buttons again but no further duplications.

Edited December 20, 2016 by Phineas Freak

[Freds]

Can someone do a test? 
With FAR, launch a manned airplane, go as fast and high as you can and go EVA... (Keep the FAR display status window opened)
Now go to flight status and see if the falling Kerbonaut has 0 mach, 0 drag, 0 lift, and an ever increasing speed......  
I use 'EVA Parachutes & Ejection Seats' Mod, and last night i accidentally ejected my Kerbal, and this happened (I was deorbiting, it happened at aproximally mach 4 at 35000M ). 

I would test this myself, but i probabily wont have time to play anything tonight .

[Matuchkin]

8 hours ago, ferram4 said:

Basically, what Viatic said.  It's a small barrier to entry to ensure that most of the people messing with the dev builds actually have an idea of how development works

Okay then, sorry. I probably shouldn't touch it then, as I am not a developer.

[larkvi]

@ferram4 I tried the most recent version on a Mac, without any apparent problems. I do not know what the shader issue looks like, but I did not experience it. The doubled menu button from the previous version is gone. Everything appears to be working properly: my planes are transonically unstable, prone to flat spins, and aerodynamically disintegrate. (Have not tested water.)

One thing which I do not know if it is an issue or not (so I have not put it in GitHub): are some voxels supposed to be sticking out a bit from the plane? http://imgur.com/a/OxYrc Craft: https://www.dropbox.com/s/jnehzt4xkb1bydd/Dart.craft?dl=0

[tetryds]

14 minutes ago, larkvi said:

are some voxels supposed to be sticking out a bit from the plane?

Yes, that is normal, voxels can actually represent half of the space they are in, imagine a half block on minecraft, that kind of thing.

That airplane does not look very stable for supersonic flight, especially with a canard, I can provide other aircraft for you to test if you think there is something wrong.

[Clinton]

With Mechjeb I often look at both the instantaneous (m/s^2) and total atmospheric drag since launch (m/s) to judge the efficiency my launches. However with FAR, Mechjeb doesn't report these figures. Is there anyway I can see these figures with FAR (ideally both, but one would be good). 

[cantab]

The FAR Flight Data window shows current drag along with a bunch of other possibly-useful info. But I don't think it has a way to show total drag over a time period.

[qromodynmc]

Hello! Im currently having issue but not sure what causes to it. I couldnt decide if this is Far issue or Mfi, so i decided to share on both thread;

After launching a plane, i park it near landing strip, launch another plane, after launching second plane first one loses it's lift capabilities i suppose, because it just doesnt take off. Control surfaces looks to be working.;

I know Far build is dev build but, somebody gotta find the bugs i guess.

[kcs123]

Similar screenshot, but with FAR Flt Data window would be of more help. Those would show if there is NaNs on properties where should not be.
Even better, log file could tell with more accuracy what is going on. Without that, only if some other user repeat same bug and provide log files it will be a chance to have this bug fixed.

[ss8913]

@ferram4 - as requested, I installed the new dev build and tried it out - I don't have any problems with it.  at all.  everything works as documented for me.  Using strict area ruling.  everything flies as it should, if the numbers in SPH are reasonable, the craft flies that way.  the transonic graphs work, there's only one FAR button where it should be, everything looks spot-on from where I'm sitting.

[naught405]

Second Vessels spawned with VesselMover have has no lift bug.  Not sure which mod is the problem though, Just confirming what @qromodynmc was experiencing.

Repro: using VesselMover https://github.com/PapaJoesSoup/VesselMover/releases spawn a second vehicle, second vehicle has no lift.

screenshots with FAR Flt Data shown:

https://sli.mg/a/FoSKzj

Workaround Second repro involves 3 load screens.  Load first vessel normally, use vessel mover (or not) to place it off the runway/launch pad.  Return to Space Center, enter SPH/VAB, select vehicle and launch.  Switch back to the first plane ([,]), it has no lift as in the screenshot below:

 

This behavior breaks AI pilots/multiple aircraft.

*Re-read qromo's post and edited.

Edited December 23, 2016 by naught405
mis-read post I am responding too.

[larkvi]

On 12/20/2016 at 4:19 PM, tetryds said:

That airplane does not look very stable for supersonic flight, especially with a canard, I can provide other aircraft for you to test if you think there is something wrong.

If it was easy to fly, how would I know FAR is working?

(That is my story, at least. The truth is that, as much as I love FAR, I am terrible at both designing and flying airplanes. I did assemble that one in about five minutes.)

Edited December 23, 2016 by larkvi

[DoctorDavinci]

19 hours ago, naught405 said:

Second Vessels spawned with VesselMover have has no lift bug.  Not sure which mod is the problem though, Just confirming what @qromodynmc was experiencing.

Repro: using VesselMover https://github.com/PapaJoesSoup/VesselMover/releases spawn a second vehicle, second vehicle has no lift.

screenshots with FAR Flt Data shown:

https://sli.mg/a/FoSKzj

Workaround Second repro involves 3 load screens.  Load first vessel normally, use vessel mover (or not) to place it off the runway/launch pad.  Return to Space Center, enter SPH/VAB, select vehicle and launch.  Switch back to the first plane ([,]), it has no lift as in the screenshot below:

<image snip>

This behavior breaks AI pilots/multiple aircraft.

*Re-read qromo's post and edited.

Just a PSA ...

The Vessel Mover Spawn Vessel function is technically still experimental and it is known that there are gremlins that rear their ugly head when using it ... some attributes of a craft being spawned are lost in translation and sometimes things get a little garbled

Please open an issue on the Vessel Mover Github: https://github.com/PapaJoesSoup/VesselMover/releases

[ferram4]

The Vesselmover behavior is almost certainly an issue introduced on the Vesselmover end.  It appears that it is not spawning vessels in the same way that the stock game does, breaking everything.  Nothing I can do to change that.

[Freds]

@ferram4: I'm testing without vessel mover, and the problem happened. I'll post more details in a few minutes.

 

 

[Freds]

@ferram4:

This is what i have done:

New sandbox game, last version of KSP_UPDATE Branch of FAR installed (and lot of other mods).

Spawned an aeris3A by clicking on the runway and choosing aeris 3A. I drived the aeris 3 and parked it in the grass. FAR worked corretly. 

Back to spacecenter, i repeated the same procedure (click on runway, etc...) to spawn an Aries 4A. Drived the aries 4 to the grass, and FAR worked fine. 

The issue happens when i change from aeris 4A to aeris 3A (using , ) - FAR simply ignores Aries 3A. Lift, drag, all 0.

I was analysing the Logs, and i think this is relevant:When aeris 3A was unpacked for the first time, FAR updates the Voxels:
 

Spoiler

[LOG 20:14:47.525] Updating vessel voxel for Aeris 3A
[LOG 20:14:47.624] Unpacking Aeris 3A
[LOG 20:14:47.719] Std dev for smoothing: 3 voxel total vol: 221,726059173867 filled vol: 11,5487776052247

[LOG 20:14:48.355] ScaleModList: listSize 697 maxListSize 788
[LOG 20:14:49.169] Updating vessel voxel for Aeris 3A
[LOG 20:14:49.215] Std dev for smoothing: 3 voxel total vol: 221,726059173867 filled vol: 11,5487776052247

The same happens with aeris 4A

Spoiler

LOG 20:15:24.510] Game State Saved as persistent
[LOG 20:15:25.395] Updating vessel voxel for Aeris 4A
[LOG 20:15:25.458] Std dev for smoothing: 3 voxel total vol: 601,407793800689 filled vol: 27,4488191155767

[LOG 20:15:25.494] Unpacking Aeris 4A
[LOG 20:15:25.499] [Aeris 3A]: ground contact! - error: 0.000m
[LOG 20:15:25.499] Unpacking Aeris 3A
[LOG 20:15:25.939] [OD] --> ScaledSpaceDemand.LoadTextures loading NewMunSurfaceMapDiffuse and NewMunSurfaceMapNormals
[LOG 20:15:26.267] ScaleModList: listSize 697 maxListSize 788
[LOG 20:15:27.070] Updating vessel voxel for Aeris 4A
[LOG 20:15:27.105] Std dev for smoothing: 3 voxel total vol: 601,407793800689 filled vol: 27,4488191155767

At this time in log, FAR is working with aeris4A. But when i change ships, no more lines related to FAR appears in the log anymore, and aeris3A simply has 0 drag, 0 lift, no matter the speed it archives.

If i switch BACK to aeris 4A, FAR works.

If i return to the space center, and take control of aeris 3A, FAR works with aeris 3A. when i switch to Aeris 4A, FAR no longer works with aeris 4A.

I can post the logs if you want, or i can test with modless install (only with FAR and module manager).

Sorry for my poor english - its not my native language, so i'm still learning how to write in english.

 

 

[naught405]

@ferram4 I repeated the behavior with no other mods as described by @Freds with the same exact result. VesselMover is a little wonky I know; but I also showed that it happened during a normal load sequence.

[DoctorDavinci]

2 hours ago, Freds said:

<snip>

Please do testing in a clean install without other mods ... Having more than just FAR installed is a bad testing environment and only serves to cloud the issue of whether FAR is working correctly or not

Testing against other mods comes later on down the line as, for the time being, ferram is just trying to get it working on it's own the way he wants it to

Edited December 24, 2016 by DoctorDavinci

[tetryds]

I think that at this point testing it with other mods is pretty valid, but it always requires more work and nailing down which mod is behaving badly under which circumstances, then reporting the issues at the correct places.

[mikegarrison]

8 minutes ago, DoctorDavinci said:

Please do testing in a clean install without other mods ... Having more than just FAR installed is a bad testing environment and only serves to cloud the issue of whether FAR is working correctly or not

Testing against other mods comes later on down the line as, for the time being, ferram is just trying to get it working on it's own the way he wants it to

Yes but really I thought Freds was extremely helpful. He discovered that FAR correctly handles switching vehicles when they are loaded through the tracking station, but doesn't handle switching them on the fly using the keyboard refocus.

Now if this does work fine when FAR is the only mod installed, then OK, it's an interaction with some other mod. But it seems likely to me that this has nothing to do with any other mods, unless he has some that affect how different vehicles are refocused.