• Which the release of FS2020 we see an explosition of activity on the forun and of course we are very happy to see this. But having all questions about FS2020 in one forum becomes a bit messy. So therefore we would like to ask you all to use the following guidelines when posting your questions:

    • Tag FS2020 specific questions with the MSFS2020 tag.
    • Questions about making 3D assets can be posted in the 3D asset design forum. Either post them in the subforum of the modelling tool you use or in the general forum if they are general.
    • Questions about aircraft design can be posted in the Aircraft design forum
    • Questions about airport design can be posted in the FS2020 airport design forum. Once airport development tools have been updated for FS2020 you can post tool speciifc questions in the subforums of those tools as well of course.
    • Questions about terrain design can be posted in the FS2020 terrain design forum.
    • Questions about SimConnect can be posted in the SimConnect forum.

    Any other question that is not specific to an aspect of development or tool can be posted in the General chat forum.

    By following these guidelines we make sure that the forums remain easy to read for everybody and also that the right people can find your post to answer it.

MSFS24 SimPropContainers

I am thus far unable to process the *.blend file via MCX to perform the analyses intended.


I shall still endeavor to post other info in reply to your query in advance of any resolution by Arno of the MCX import issue here:

https://www.fsdeveloper.com/forum/threads/mcx-fails-importing-blender-blend-file.460749/post-939211


FYI: Blender LTS 3.6.23 exports of a Kronos 2.0 *.glTF and a Collada *.DAE were also not able to be (fully) processed by MCX and the same Assimp library used in the "Universal Importer" Ruby Script plugin for Sketchup.

It is not clear at this point what may be required to process the Blender LTS 3.6.16 *.blend file you linked in your post above.

The analyses I intended to perform depend on inter-convertibility, and the latter Assimp library functions in MCX and Sketchup., as well as the ability of Blender to export a non-proprietary format Collada *.DAE that can be read by both MCX and Sketchup.

'No cigar' ...thus far. o_O


While we await a reply by Arno, please link to a MSFS SDK compiled package folder for a test build of the 3D models. :idea:

I believe that MCX 'should' already be able to import that MSFS "proprietary" content format successfully. :wizard:

GaryGB
 
Last edited:
I compiled / inspected the project in MSFS 2024, and have not seen any Red hangar flickering / shadow anomalies. ;)

I plan to run analyses of the objects to evaluate them for "Solid" and "Manifold" status next.

Nice 3D models. :)

GaryGB
 
Last edited:
Thanks :)
The only noticable problem with shadows is the transition between LODs, when the shadow suddenly disappears, but i think this is an offtopic.
 
Heermeyer said:
Thanks :)

The only noticeable problem with shadows is the transition between LODs, when the shadow suddenly disappears, but I think this is an off-topic.
Click to expand...

There certainly could be some details pertinent to how we make SimPropContainers which may best be addressed in separate threads ...eventually.

However, since we are early in the process of understanding SimPropContainers in this emerging implementation by Asobo via the SDK, it seems relevant.


My original concern was testing how well MSFS' compiler assembled "component" parts into a solid / manifold true 3D model without risk for display anomalies after compiled.


Do you have a screenshot of Shadow anomalies rendered before / after LOD transition in MSFS at run time you could attach / link to ? :scratchch

GaryGB
 
Last edited:
What is the status of this package ?

Following your link, I see a listing for what IIUC is another package for KPNN Princeton Municipal at flightsim.to.

Which package are we discussing here: KBST or KPNN ?

Are you inferring that both KBST and KPNN use SimPropContainers, but KPNN has the visible shadowing anomaly ? :scratchch


You refer to the screenshot in this post above:

https://www.fsdeveloper.com/forum/threads/simpropcontainers.460334/post-938194


AFAIK, this is at KPNN, and you are alluding to a visible shadowing anomaly in that image ?

If so, would you post another copy of it here with the visible shadowing anomaly circled or pointed at with an arrow ?

GaryGB
 
Please post a screenshot from MSFS of that KPNN Terminal SimPropContainer which demonstrates the shadow effect when AC unit switches between LODs.


FYI: When I download and de-archive the linked package, for the AC unit, only KPNN-terminal-ac_LOD2.gltf has clean geometry for the AC fan blades.

The other LOD glTFs are rather distorted (severely decimated ?), and no AC fan blades can be discerned. :scratchch

IIUC, one must have clean geometry for animated parts of a 3D model in all LODs that the animation is intended to be displayed.

[EDITED]

I traced distorted AC unit rendering to use of an older October 9, 2020 version of MCX on a laptop to view glTF's.

Having loaded the package from flightsim.to, AC unit 3D models all look normal in March 17, 2026 MCX version.

I adjusted ambient temperature in MSFS' weather dialog to a warm 75 degrees Fahrenheit, and the AC fan spun. :wizard:

Nicely done ! :cool:

[END_EDIT]

GaryGB
 
Last edited:
ac-shadow.png

ac-no-shadow.png


AC unit next LOD does not cast a shadow on the wall. Maybe this is the price for SimProps parts with different LODs. Or is it related to material without texture for an AC unit. In general, i dont know if this can be considered as "shadow anomaly".
 
Yep. The sim drops the shadows at some distance depending on the bounding sphere of the object. If you add a large invisible geometry (plane or cube) it will keep the shadow longer. I played with your xml a bit. I think you could have just used 2 LODs.
XML: 
<?xml version="1.0" encoding="utf-8"?>
<ModelInfo guid="{06a7f924-ff91-40af-9cca-3cf7eee8d84d}" version="1.1">
    <LODS>
        <LOD minSize="5.0" ModelFile="KPNN-terminal-ac_LOD0.gltf"/>
        <LOD minSize="0.0" ModelFile="KPNN-terminal-ac_LOD2.gltf"/>

    </LODS>
    <Animation name="acon" guid="7e0044b9-9bc5-4584-b14a-19257cc0b95e" type="Standard"/>
    <AnimGraph>
        <DefaultState name="prop-rotation"/>
        <BlendTreeState name="prop-rotation">
            <Animations>
                <Animation guid="7e0044b9-9bc5-4584-b14a-19257cc0b95e" loop="True" speed="1" threshold="18"/>
            </Animations>
            <Value>(A:AMBIENT TEMPERATURE, CELSIUS)</Value>
        </BlendTreeState>
    </AnimGraph>
</ModelInfo>

It worked OK. And maybe only animation for the LOD0. Really nice scenery package!
 
Last edited:
I am completely wrong about the cast shadows. The sim is computing visible geometry for shadows, and it is a hard limit per size of the visible part of an object, not controlled by the bounding sphere. We don't seem to have any control over the distance at which a shadow disappears. Invisible cubes or even large hidden 'solid' geometry have no effect. :mad: Big objects hold the shadow for greater distances.

If you change your Graphics Settings to Ultra, it will extend the display distance of shadows a small bit.
 
Last edited:
This interesting ...and baffling, at the same time. :scratchch

I was thinking of the mechanism that L-M was working on for P3D shadows with solids and transparency processing.


I suppose MSFS' "Cast Shadows" may be for directional artificial light sources rather than reflected global illumination.


This is an interesting thread regarding how MSFS lighting / shadows work:

https://forums.flightsimulator.com/...-cast-shadows-from-artificial-lights/756385/5


Heermeyer's image in post #2 in this same thread gives us clues about what the AC unit shadow may 'actually' be:

https://www.fsdeveloper.com/forum/threads/simpropcontainers.460334/post-938194

kpnn-terminal-simprop-png.98884


Although I am tempted to suspect a non-MSFS (HDRI ?) sky background may have been used in his image there, the image seems to show the Sun position as behind the building.

Also, the eaves extent makes it impossible for any directional sunlight to cast a shadow from the AC unit to ground.

Nor would sunlight reflected from snowy ground be bright enough to 'cast' a shadow from the AC unit onto the eaves.


IIUC, we are seeing Ambient Occlusion as a probable source for the shadow from the AC unit seen on the eaves.


Of course, in MSFS, all bets are off as to how Asobo arbitrarily generates Ambient Occlusion ("obscurance" ?). o_O

https://www.google.com/search?q=Ambient+Occlusion+"obscurance"&client=firefox-b-1-d&hs=7GYp&sca_esv=63d44ac5fb776b6b&ei=wWS8adnfEPHfp84Pn_C9sQg&biw=1200&bih=586&ved=2ahUKEwiZxJjD7qyTAxXx78kDHR94L4YQ4dUDegQIBRAP&oq=Ambient+Occlusion+"obscurance"&gs_lp=Egxnd3Mtd2l6LXNlcnAiHkFtYmllbnQgT2NjbHVzaW9uICJvYnNjdXJhbmNlIjIFECEYqwIyBRAhGKsCMgUQIRirAki0SlDGBVjtPnABeACQAQCYAW2gAdMDqgEDMy4yuAEMyAEA-AEC-AEBmAIGoALvA8ICDhAAGIAEGLADGLEDGIMBwgIUEAAYgAQYsAMYsQMYgwEYigUYjQbCAgsQABiABBiwAxixA8ICERAAGIAEGLADGLEDGIMBGIoFwgIFECEYoAHCAggQABiABBiiBMICBhAAGAgYHsICCxAAGIAEGIYDGIoFwgIFEAAY7wWYAwCIBgGQBgmSBwM0LjKgB7wQsgcDMy4yuAfoA8IHBTAuMi40yAcUgAgA&sclient=gws-wiz-serp


IIUC, there may be a way to better control AO on SimPropContainer objects in MSFS SDK Docs:

https://docs.flightsimulator.com/msfs2024/html/2_DevMode/Scenery_Editor/Objects/Scenery_Objects.htm

"Override AO Strength

When checked, this option will show a slider that permits you to override the automatic ambient occlusion settings. This can let you better match the lighting on the object with the rest of the scene."


Or perhaps use conditional display visibility for AC unit shadow as a SimObject inside a building SimPropContainer ?

Since the AC unit has a close proximity to the eaves surface, can a shadow can be baked onto the eaves Face texture ?



Years ago I was looking at MeshLab as a way to implement baked AO for FS2Kx objects,

MeshLab it is still being developed, and has new methods for AO / Obscurance.


Perhaps MeshLab might be an alternative to Blender's current workflow for lighting / shadowing / AO ?

If so, MeshLab might export a 3D exchange format file (glTF / glb) that can be processed via Blender and/or MCX ?


https://www.meshlab.net/

https://github.com/cnr-isti-vclab/meshlab

...also available for free on MS Store:

https://apps.microsoft.com/detail/9n4tl5dxc6c0?hl=en-US&gl=US


GaryGB
 
Last edited:
No, it is not HDRI :) Shading on the eaves is definitely pure AO baking by Substance Painter, ill make a dummy box especially for this. But i think that AO and 3d objects shadows are different things. AO depends on the COMP texture, and the shadow is a kind of raytracing.
 
Heermeyer said:
No, it is not HDRI. :)

Shading on the eaves is definitely pure AO baking by Substance Painter, ill make a dummy box especially for this. But i think that AO and 3d objects shadows are different things. AO depends on the COMP texture, and the shadow is a kind of raytracing.
Click to expand...

Could you post an updated screenshot showing how your adapted "AO Shadow" for the AC unit looks when done ? ;)

GaryGB
 
You must log in or register to reply here.
Back
Top