Jump to content

[WIP][1.8.x] SSTULabs - Low Part Count Solutions (Orbiters, Landers, Lifters) - Dev Thread [11-18-18]


Shadowmage

Recommended Posts

1 hour ago, Jimbodiah said:

I posted logs the other day, didnt get a response to those ;)   Petal thing is not usable in career mode.

What is 'not usable'?  Does it not show up in career mode, or other (crashes, does strange things, un-adjustable)?  (It has some built in tech-limit stuff, that hasn't really been tested much... so that could be part of it).

Link to comment
Share on other sites

It comes down to it not properly initializing the updated radius from the tech limits; for some reason it is giving a radius of 0; so it is either A: divide by zero error, or B: all radial vertices on the mesh get generated at the same point (which is why it has height, but no depth in either x or z axes, and hence, no colliders with any physical volume.. so no interaction).

Now that I'm aware of the problem I'll look into when I get a chance (likely later this week after I get the new mesh-generator code working; need to merge that code in, so will already be working on that module).

Link to comment
Share on other sites

Updated testing release is available:

https://github.com/shadowmage45/SSTULabs/releases/tag/0.3.26-pre1

Fixes log spam, actually adds new fuel tank variant, and adds a couple of new prototype testing parts (utility tab in the editor).  Have not yet fixed the petal adapter, it will require a bit more time to dig into it, after I finish the coding stuff I'm in the middle of.

Link to comment
Share on other sites

The fairings work on my end, even though I'm not to fond of fairing building and lean towards a procedural fairing. at least there's an option to make them 2 sided, instead of them busting loose all over the place. Sizing works as well, only got it to 5 m. You stated that the B tank was for Kerlox? Is that a RF thing? If that's so, then it looks different, of course and the textures work for it as well. I didn't see any obvious issues with this pre version. But as I stated before, it has to be real obvious for me to see it. I ran all the tests in career mode, so they work on that aspect. 

I have to remember to enable the LH2 patch every time you update, so unless you change that, I'm going to just copy the patch from the previous version. Unless it needs a new version upon every release. 

Edited by lynwoodm
added stuff.
Link to comment
Share on other sites

I just installed RF to test this out and the textures appear to be working fine for me. I don't know how RF works, but the only thing I see is that the volume of the tanks doesn't change from the original size to an adjusted size, it remains at 2000 for a 5 m x 21 m tall tank (size didn't matter, the volume remained the same). If you haven't done so, you'll probably have to re download the texture files again. 

Link to comment
Share on other sites

9 hours ago, 01010101lzy said:

 

Meeting a strange issue.

Resizing the tank(either height or diameter) when attached by side causes all meshes used to stack together, not sure if it is related to RF.

Hmm... that is occuring without symmetry enabled?  Not sure that I've seen that one before.... noted, will investigate when I have a few moments while not at work.  Any particular steps to reproduce?  1-select tank, 2-attach radially, 3-change size, 4-note duplicated models?

 

1 hour ago, lynwoodm said:

I just installed RF to test this out and the textures appear to be working fine for me. I don't know how RF works, but the only thing I see is that the volume of the tanks doesn't change from the original size to an adjusted size, it remains at 2000 for a 5 m x 21 m tall tank (size didn't matter, the volume remained the same). If you haven't done so, you'll probably have to re download the texture files again. 

Hmm... might be something wrong with the patch or tank setups then, as it -should- change volume as tank size changes... will investigate next time I have RF installed.

Edited by Shadowmage
Link to comment
Share on other sites

12 hours ago, lynwoodm said:

The fairings work on my end, even though I'm not to fond of fairing building and lean towards a procedural fairing. at least there's an option to make them 2 sided, instead of them busting loose all over the place. Sizing works as well, only got it to 5 m. You stated that the B tank was for Kerlox? Is that a RF thing? If that's so, then it looks different, of course and the textures work for it as well. I didn't see any obvious issues with this pre version. But as I stated before, it has to be real obvious for me to see it. I ran all the tests in career mode, so they work on that aspect. 

I have to remember to enable the LH2 patch every time you update, so unless you change that, I'm going to just copy the patch from the previous version. Unless it needs a new version upon every release. 

Fairings - Not sure what you are referring to; the stock fairings -are- procedurally built (procedural means, in this context, that the mesh is created at run-time and does not use a precompiled model; e.g. the model is built using a run-time process, thus procedurally).  And I don't have much choice in the matter; stock only offers one option for fairings, and so that is what I must use (when stock integrates P-fairing type code, I would gladly use that/offer that as an option; but I'm not recreating their entire codebase, nor interested in setting it up as a dependency).  However, I am investigating methods to un-confetti-ize the stock fairings, so they will be proper 2-part clamshells, and perhaps even offering a run-time option to change the number of radial segements.  The stock fairing module however is not written in a very moddable fashion, so it might take some time for me to get all of this stuff working (lots of hacking around stuff to make it work).

MFT-B - Its purely a visual thing; the location of the intertank-mesh.  An aesthetic variant.  All other stats, available fuel types, etc will be exactly the same as MFT-A.

LH2 patch - you'll need to keep an eye out for when it actually gets updated; which will likely be quite often while I'm working through the balance and setup of the parts.  Heck, I'm still not sure that I'll even be keeping the LH2 patch around -- it creates a few balance related problems that I really don't want to deal with (mostly boiloff related; I really don't want to implement boiloff code, but if I keep LH2 around, it is mandatory that it be implemented; which then mandates different types of insulated tanks, which requires a ton more coding support).  So... what I might end up doing is dropping the LH2 patch, and moving that balance stuff over to the RealFuels patch/setup.  No reason for me to reinvent the boiloff/insulation mechanics when a mod exists specifically for them (would just require finishing off RF support for all of the SSTU fuel tanks).

Anyhow, thanks for testing things and providing realistic feedback :)

Link to comment
Share on other sites

And in general dev news:

Got the new procedural mesh generator 'working' yesterday/last night.  It can now build meshes, and is a lot easier to work with than the previous (current) system.  The previous system was hard-coded to support only a single vertical panel segment-per-mesh; straight walls, with some very odd UV mapping.  The new system supports arbitrary arc/ring based mesh creation; specify a set of radius' and heights, and it will build the mesh to meet those specifications.  UVMapping is very simple, and will be config-adjustable.  Still need to add in pivot/rotation support (for stuff like the SC-GEN-IPA), and code to generate proper convex colliders for arbitrary meshes (yep, about as complicated as it sounds... so...might take a few weeks; thankfully, I won't need this functionality until 1.1/unity5).

This new mesh generator will replace the one currently used to generate fairings, the petal adapter, and procedural decouplers, as well as being used on the new yet-to-be-created interstage decoupler/fairing (the one with built-in ullage engines).  The old generator, while being very reliable, was just not adaptable, expandable, or easy to use.

The reasons for creating a new mesh generator were:
1.) Add ability for bi-conic/tri-conic/whatever type fairings in the future (after 1.1, needs a GUI); the petal adapter and interstage decoupler will use an early minimial implementation of this ability.  Yes, in the future you will be able to make more complex engine/etc fairings.
2.) Allow for less confusing UV mapping setup.  The old setup was 'ideal' insofar as maintaining a non-distorted 1:1 texture ratio at a constant pixel density, however it resulted in being unable to use anything but a flat-color texture.... which removed all usefulness from the non-distorted mapping.  This will allow for texture options for the fairings / texture switching; and allow for proper use of various patterned textures.  Additionally, the UV mapping will now be config-adjustable; to allow for even more variance in texture use/re-use and configuration.
3.) Ease of maintenance and coding.  The old system was very hard-coded for a specific use-case (the petal-adapter), and had been bashed/mangled/kicked around in order to make it work for the other parts (fairings, decoupler).  The new system is designed with all of these potential use cases in mind, as well as doing all that is possible to not limit or hard-code anything.  Also took the opportunity to abstract-out some of the more head-scratching bits of code, like triangle-winding and quad creation from an arbitrary list of vertices (previously I had to figure out what order to enter the vertices in;  now I just feed the generator a list of verts and it does the rest).
4.) Expandability -- the new system is designed in such a way that adding new mesh-generator types should be relatively easy.  Want a cube-mesh-generator?  No problem.  Pointed-cone?  Yep, could do that.  Could potentially even expand some of the generators to use config-defined setups.

 

Link to comment
Share on other sites

53 minutes ago, Shadowmage said:

Fairings - Not sure what you are referring to; the stock fairings -are- procedurally built (procedural means, in this context, that the mesh is created at run-time and does not use a precompiled model; e.g. the model is built using a run-time process, thus procedurally).  And I don't have much choice in the matter; stock only offers one option for fairings, and so that is what I must use (when stock integrates P-fairing type code, I would gladly use that/offer that as an option; but I'm not recreating their entire codebase, nor interested in setting it up as a dependency).  However, I am investigating methods to un-confetti-ize the stock fairings, so they will be proper 2-part clamshells, and perhaps even offering a run-time option to change the number of radial segements.  The stock fairing module however is not written in a very moddable fashion, so it might take some time for me to get all of this stuff working (lots of hacking around stuff to make it work).

MFT-B - Its purely a visual thing; the location of the intertank-mesh.  An aesthetic variant.  All other stats, available fuel types, etc will be exactly the same as MFT-A.

LH2 patch - you'll need to keep an eye out for when it actually gets updated; which will likely be quite often while I'm working through the balance and setup of the parts.  Heck, I'm still not sure that I'll even be keeping the LH2 patch around -- it creates a few balance related problems that I really don't want to deal with (mostly boiloff related; I really don't want to implement boiloff code, but if I keep LH2 around, it is mandatory that it be implemented; which then mandates different types of insulated tanks, which requires a ton more coding support).  So... what I might end up doing is dropping the LH2 patch, and moving that balance stuff over to the RealFuels patch/setup.  No reason for me to reinvent the boiloff/insulation mechanics when a mod exists specifically for them (would just require finishing off RF support for all of the SSTU fuel tanks).

Anyhow, thanks for testing things and providing realistic feedback :)

No problem. I do not have proc fairings installed on my windows version, trying to keep mods down to a minimum as to facilitate an accurate debug if there's a problem. When I refer to stock fairings, it refers to a fairing that is built rather than having a fairing that's applied to an attach point to facilitate shape versus building your own shape for stock fairings. As far as the LH2 patch, I've not noticed a boiloff. But, then again, I use the LH2 tanks at launch and not long term fuel system incorporation. But, as I said, I do not know how this RF works and if that's more responsible for boil off properties. Which it could be, but that's going to require more use for this and testing to see if does anything with boil off. Pretty sure someone else could answer this question about real fuels and boil off better than I could. 

EDIT: Removed RF from mod list and the volume of the tanks returned to normal, even with LH2 values. Boil off didn't occur with just the LH2 patch to incorporate LH2 to the tanks. If I didn't know any better, RF is a contributor to the boil off factor and the tanks not incorporating new volume with tank parameters. 

Edited by lynwoodm
added more information
Link to comment
Share on other sites

4 hours ago, lynwoodm said:

No problem. I do not have proc fairings installed on my windows version, trying to keep mods down to a minimum as to facilitate an accurate debug if there's a problem. When I refer to stock fairings, it refers to a fairing that is built rather than having a fairing that's applied to an attach point to facilitate shape versus building your own shape for stock fairings. As far as the LH2 patch, I've not noticed a boiloff. But, then again, I use the LH2 tanks at launch and not long term fuel system incorporation. But, as I said, I do not know how this RF works and if that's more responsible for boil off properties. Which it could be, but that's going to require more use for this and testing to see if does anything with boil off. Pretty sure someone else could answer this question about real fuels and boil off better than I could. 

EDIT: Removed RF from mod list and the volume of the tanks returned to normal, even with LH2 values. Boil off didn't occur with just the LH2 patch to incorporate LH2 to the tanks. If I didn't know any better, RF is a contributor to the boil off factor and the tanks not incorporating new volume with tank parameters. 

Indeed, I have not implemented boiloff yet; which is why you do not see it with my tanks.  I'm not even sure I want to go down that road, and may just mandate RealFuels for the LH2 patch.

RealFuels Tank Volumes -- as stated earlier, these -should- adjust as you adjust the size of the tank; if they are not adjusting, that is a bug that needs to be filed.

Link to comment
Share on other sites

14 minutes ago, Shadowmage said:

RealFuels Tank Volumes -- as stated earlier, these -should- adjust as you adjust the size of the tank; if they are not adjusting, that is a bug that needs to be filed.

Filed an issue on the bug tracker. Will upload and share any information needed to help in this matter. 

Link to comment
Share on other sites

On Monday, January 04, 2016 at 5:28 AM, JoseEduardo said:

have been playing too much Space Truck Simulator Elite Dangerous the past few days, and now that you spoke of a cargo C series (even though as a jester) it made me think, how cool would it be to have a Elite-ish freighter ship in the game? :P

Hehe, coming up next - the Kerbalon Type-9.  500t cargo capacity, with a 100kn main engine, and standard 1kn rcs.  Takes forever to accelerate, or decelerate, and generally drives like a cow :)  (Like a lead cow... pulling a ton of bricks)

E:D is actually what got me started on KSP.  Or rather, the lack of fulfilling content.  Someone had a post in the forums that went something like 'I'm playing KSP while waiting for updates;  1000x the content, but only like 1/10 the graphics'.  And I went 'Hmm... sounds worth trying out'.... d/l the free demo and checked it out for a minute, and I think I purchased the full version before I had even launched my first real rocket.

In ED, I had/have a half-fitted Python (rather, half-stripped), a maxed out Type-9, a fully A-class Cobra Mk3, and several tens of millions of credits in the bank... think I was like 5 hours away from moving into an Anaconda when I stopped playing.  I just couldn't do the trading-credit-treadmill any longer, especially when I had rockets to launch and moons to crash into :)

Link to comment
Share on other sites

The link gives a 404.

I don't use RF, it's outside the scope of the mods I use and do not see a point installing it just to use LH2. RF goes well beyond just substituting LH2 for LF.

Re the 2nd modular tank: wouldn't a texture set have been adequate, as the only difference is the placing of the pipes on the outside.

Edited by Jimbodiah
Link to comment
Share on other sites

13 hours ago, Shadowmage said:

Hmm... that is occuring without symmetry enabled?  Not sure that I've seen that one before.... noted, will investigate when I have a few moments while not at work.  Any particular steps to reproduce?  1-select tank, 2-attach radially, 3-change size, 4-note duplicated models?

It occurs when symmetry is on, and only the selected side meets this.

Reproducing steps:

(Maybe need Realfuels installed, not sure about it)

  1. Select a tank, change its size
  2. Attach it to a side decoupler(I've only tested with this), do not add anything in the bottom.
  3. Change its length, size, nose or mount for a few times
  4. Notice collided texture on one side

If it doesn't reproduce the issue, try detaching it and attach again. For me, it works.

Link to comment
Share on other sites

10 hours ago, 01010101lzy said:

It occurs when symmetry is on, and only the selected side meets this.

Reproducing steps:

(Maybe need Realfuels installed, not sure about it)

  1. Select a tank, change its size
  2. Attach it to a side decoupler(I've only tested with this), do not add anything in the bottom.
  3. Change its length, size, nose or mount for a few times
  4. Notice collided texture on one side

If it doesn't reproduce the issue, try detaching it and attach again. For me, it works.

Thanks for the details, will attempt to replicate this when I get off of work today.
 

3 hours ago, Jimbodiah said:

I reported this a few releases ago on github but did not see it since then.

Aye, and I could not reproduce it given the steps that were listed; notably, I was not trying it with symmetry enabled, as that was not listed as needed to duplicate the issue.  That is why all this information is important, and these little details matter. 

Now that I know that stack-symmetry is involved, that tells me what the general cause likely is, and should be able to get it cleaned up in short order.

 

Edit:  Have opened up an issue ticket for it for tracking: https://github.com/shadowmage45/SSTULabs/issues/142

Will work on getting this cleaned up for this weekends' update.  No guarantees, but I'll see what I can do.

Edited by Shadowmage
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...