[1.4.0] Old School Fairings

[Virtualgenius]

Awesome thankyou

[Felbourn]

I tried it out a few ways. I checked the debug logs. I changed the mod. It still flips and I am not sure why this one rocket is doing it. Perhaps it's something about the model itself. I'm not sure how KSP handle aerodynamics in this case, where we have a hard hollow shell and things under it marked as "shielded from airstream." I suspect it's something entirely different from what we're trying to fix. I'm not sure how to proceed at this point, so maybe I will get inspired when I stop thinking about it directly.

[Felbourn]

I forgot to make a post about this when I did it, but 2-3 weeks ago I put a new version link in the Original Post that included the discontinued KW fairings as part of the installation. There are two reasons: (1) as I just said, KW has discontinued those hard shell fairings, (2) even though KW still has them in a Legacy folder, that folder does not contain all the parts. Some were missing, so my folder has them.

[Starwaster]

On 2/15/2016 at 11:45 AM, Felbourn said:

I tried it out a few ways. I checked the debug logs. I changed the mod. It still flips and I am not sure why this one rocket is doing it. Perhaps it's something about the model itself. I'm not sure how KSP handle aerodynamics in this case, where we have a hard hollow shell and things under it marked as "shielded from airstream." I suspect it's something entirely different from what we're trying to fix. I'm not sure how to proceed at this point, so maybe I will get inspired when I stop thinking about it directly.

Does each fairing piece have the cargo bay module? And have you checked the offsets to make sure it isn't occluding anything?

Or were you handling that in the fairing base with a single cargo bay module? (if so, again, check the offsets. The center of the bay should be halfway between the base and the top of the topmost fairing)

[Felbourn]

No, no cargo bays. The point of the mod was to enable hard shells again since cargo bays were not working on old school fairings at the time. See OP for details and caveats.

[politas]

Any chance of an actual release of this? I'd love to list it in CKAN, but CKAN ignores pre-releases.

[Starwaster]

On February 26, 2016 at 3:00 PM, Felbourn said:

No, no cargo bays. The point of the mod was to enable hard shells again since cargo bays were not working on old school fairings at the time. See OP for details and caveats.

Wait, those were the OLDD parts right? Whatever happened with doing the custom drag cubes for that?

I'd bet money that the model's normals are screwed up and the drag cubes are not being properly rendered. His deployable heatshield has the same issues and needs custom cubes. (Which I did for DRE then promptly overwrote my work with bad cubes I'd made previously and lost all my good cubes)

[Felbourn]

4 hours ago, Starwaster said:

Wait, those were the OLDD parts right? Whatever happened with doing the custom drag cubes for that?

Yes, this was in relation to this guys question:

Hi I was wondering if you could do the drag cube and config for the OLDD proton  I am trying to get this working again 

http://forum.kerbalspaceprogram.com/index.php?/topic/69061-8k82k-aka-proton-oldd-in-ussr-v-05-120514/&page=1

I didn't know normals could be set independently from just the orientation it appears in the VAB. I guess I could set some silly values for certain cube faces and then compare to the debug UI to see what var those silly values show up under. That would let me see how his fairings are rendered.

Unless you know a better way to visualize the normals in-game?

[Virtualgenius]

I PM'd Denny who created the OLDD models he is going to update them for 1.1 so please dont do any unnecessary work on my account.  I do hope 1.1 comes soon so many excellent mods are being held until its release, saying that so many mods will possibly break we are  stuck between a rock and hard place so its hurry and and wait

Edited February 28, 2016 by Virtualgenius

[Starwaster]

6 hours ago, Felbourn said:

Yes, this was in relation to this guys question:

I didn't know normals could be set independently from just the orientation it appears in the VAB. I guess I could set some silly values for certain cube faces and then compare to the debug UI to see what var those silly values show up under. That would let me see how his fairings are rendered.

Unless you know a better way to visualize the normals in-game?

AFAIR, It's the normals of the mesh itself that were screwed up in the case of the ADEPT heat shield. It confuses the drag cube renderer as to which way the surface is facing. 

There is a way to get the drag cube renders but I don't recall how.  @NathanKell would know.

[NathanKell]

Actually, Mike and I just looked into this a couple days ago. The issue is this: Unity is bad about non-uniform scaling. It applies the scaling matrix directly to the normals, which means a non-uniform scaling operation will have the inverse effect on normals.* So if the ADEPT's animation involves non-uniform scaling, that would explain why it's screwed.

 

*Let's work in 2D. Consider a 45 degree poly and its normal

\  /
\

Now consider it scaled by 30 in X and 1 in Y

___

Ah, but what happens to the normal? Same scaling applies, so the normal, too, gets flatter. Rather than, properly, pointing more upwards (normal to the surface).
 

[Starwaster]

2 hours ago, NathanKell said:

Actually, Mike and I just looked into this a couple days ago. The issue is this: Unity is bad about non-uniform scaling. It applies the scaling matrix directly to the normals, which means a non-uniform scaling operation will have the inverse effect on normals.* So if the ADEPT's animation involves non-uniform scaling, that would explain why it's screwed.

 

*Let's work in 2D. Consider a 45 degree poly and its normal

\  /
\

Now consider it scaled by 30 in X and 1 in Y

___

Ah, but what happens to the normal? Same scaling applies, so the normal, too, gets flatter. Rather than, properly, pointing more upwards (normal to the surface).
 

Can the drag cube renderer be modified to compensate? Or will that always be an issue if nonuniform scaling is used? 

[NathanKell]

That's an issue in Unity itself. We could go through every shader in the game and add a second transformation matrix that is calculated as the inverted transpose, and apply that to the normals...or we could release 1.1 and 1.2 and.... :]

[bassmannate]

I'm a little confused on how to install this. Do I unzip the Felbourn folder to my GameData or do I need to place the individual folders inside that into certain places?

[Delta_8930]

3 hours ago, bassmannate said:

I'm a little confused on how to install this. Do I unzip the Felbourn folder to my GameData or do I need to place the individual folders inside that into certain places?

Just unzip the Felbourn folder to GameData and you'll be fine. The folders in them contain ModuleManager configs that can be used when put anywhere in the GameData folder.

[InsaneDruid]

Will there be a 1.1 version of this mod?

[Felbourn]

Yea, it's probably just a quick recompile. I don't think anything is changing how it should work. I have not played yet because -- catch 22 -- I'm waiting for other mods to be ready for me to start my next series.

[the_pazter]

the attachment points on all of the fairing walls are missing

[Felbourn]

I'll check it out when I can. I use the same files for my own game and everything works, so maybe it's just one of the walls and I have not been using that one wall lately. I thought I regularly used all the bits though.

[Delta_8930]

Felbourn, I am having an issue with the T1500 fairing from AIES. Whenever I decouple that fairing, the four quarters collide with each other after separating. In other words, the meshes "pass through" each other at the top (pointy end). Sometimes, they knock one off to the side and it starts spinning like crazy. How should I adjust my cfg file to prevent this?

[Felbourn]

That sounds like the decoupler force is pointing the wrong way, or the parts are clipping by default and getting caught on each other. It might need the decoupler module to be changed, or the direction of the force changed if it uses an anchored decoupler, or the attachment nodes modes the tiniest bit to avoid clipping. Hard to say exactly without experimentation or at least seeing it happen, and seeing the CFG.

[Delta_8930]

I will send some pictures soon (I will be very busy during the next week and I probably wont have free time until Friday though).

[Delta_8930]

The decoupler force is in fact pointing the right way (separates like all other AIES fairings but they clip at the top (pointy end)). I'm not sure if it uses the anchored decoupler module, but it probably is due to the decoupler module and/or the attachment nodes.

[dunadirect]

I switched to KW for my new 1.1 career, and am 2 game years in when i finally realized... the fairings do nothing! so i searched around and found this and... of course it's from bob fitch (a big fan!).

If you haven't tried it with 1.1 it seems to work OK after a rebuild. thanks for doing this!

[Delta_8930]

My issue with the T1500 seems like an issue with where the decoupler module is placed. It might have to be moved either higher or lower than the current location. Can I change this by changing the attachment nodes in the cfg file?