#2713 C172/182: interior shadow renders on propeller disc

[Benedikt Hallinger]

Hi,
like discussed on the mailing list, NEXT has a rendering regression regarding interiour shadows rendered onto propeller discs when the sun is behind.
Affected is at least the C172p and the C182s/t.

Details:
- https://github.com/HHS81/c182s/issues/470 (screenshots/detailed problem description)
- https://sourceforge.net/p/flightgear/mailman/message/37404673/ (some discussion)

[James Turner]

We haev some model XML flag to say doesn't cast shadows, what I don't know is if we have a flag to avoid receiving shadows. I'll take a look now but I am not really familiar with this area of the code.

[James Turner]

A quick searching: model-interior-ALS defines a 'shadow-mapping-enabled' flag, which eg the C172 sets in a few places, eg: ./Aircraft/c172p/Models/Effects/interior/c172p-interior.eff: <shadow-mapping-enabled>true</shadow-mapping-enabled>

And the compositor shadows use the uniform 'shadows_enabled' which is controlled by modef-default.eff,

<shadows_enabled>
<use>/sim/rendering/shadows/enabled</use>
</shadows_enabled>

What you might need is to subclacc model-default.eff and make a model-unshadowd.eff for your prop disc? I don't know if this is something which makes sense to include in FGData effects.

[Benedikt Hallinger]

The problem also existe with the c172 so IT surely would be good to have this solved centrally…

Otherwise i‘m not into effects yet and only understand half of what you wrote g

Am 09.01.2023 um 16:23 schrieb James Turner jmturner@users.sourceforge.net:


A quick searching: model-interior-ALS defines a 'shadow-mapping-enabled' flag, which eg the C172 sets in a few places, eg: ./Aircraft/c172p/Models/Effects/interior/c172p-interior.eff: <shadow-mapping-enabled>true</shadow-mapping-enabled>

And the compositor shadows use the uniform 'shadows_enabled' which is controlled by modef-default.eff,

<shadows_enabled>
<use>/sim/rendering/shadows/enabled</use>
</shadows_enabled>

What you might need is to subclacc model-default.eff and make a model-unshadowd.eff for your prop disc? I don't know if this is something which makes sense to include in FGData effects.

[codetickets:#2713] C172/182: interior shadow renders on propeller disc

Status: New
Milestone: 2020.4
Created: Mon Feb 14, 2022 10:13 AM UTC by Benedikt Hallinger
Last Updated: Mon Jan 09, 2023 03:16 PM UTC
Owner: nobody

Hi,
like discussed on the mailing list, NEXT has a rendering regression regarding interiour shadows rendered onto propeller discs when the sun is behind.
Affected is at least the C172p and the C182s/t.

Details:
- https://github.com/HHS81/c182s/issues/470 (screenshots/detailed problem description)
- https://sourceforge.net/p/flightgear/mailman/message/37404673/ (some discussion)

Sent from sourceforge.net because you indicated interest in https://sourceforge.net/p/flightgear/codetickets/2713/

To unsubscribe from further messages, please visit https://sourceforge.net/auth/subscriptions/

[Benedikt Hallinger]

For the c182, the prop disc is effected as following:

<effect>
       <inherits-from>Effects/model-transparent</inherits-from>
       <object-name>FastProp</object-name>
       <object-name>SlowProp</object-name>
        <object-name>TaxiLight</object-name>
        <object-name>PlaneX</object-name>
        <object-name>Plane.002X</object-name>
        <object-name>Plane.003X</object-name>
        <object-name>LandingLight</object-name>
        <object-name>TaxiLight</object-name>
        <object-name>PlaneTailLight</object-name>
        <object-name>NavLightRed</object-name>
        <object-name>NavLightGreen</object-name>
   </effect>

It comes from the default FGData model-transparent.

How could i tinker with that (ie overload the shadow receiving)?

[James Turner]

Try adding (inside effect, after the object-names);

    <parameters>
        <shadows_enabled type="bool">false</shadows_enabled>
    </parameters>

[James Turner]

(I am guessing slightly, I am no expert in this stuff, just reading the code)

[Gijs]

Fixed per https://github.com/HHS81/c182s/commit/8364a309c215be4a482508331756e9bb11532d05 according to https://github.com/HHS81/c182s/issues/470

[Benedikt Hallinger]

Its visible again....
Reopened for investigation