• Please see the most recent updates in the "Where did the .com name go?" thread. Posts number 16 and 17.

    Post 16 Update

    Post 17 Warning old domain name stealing credentials

  • Special Server Maintenance

    Just when you think you have everything running some what better...

    You get a message from the Hosting company. Short story they need to move us to a new server. I have been working with them to plan a good date for all concerned. Here is what we have come up with as a schedule.

    1. November 4th 17:00 -5UTC shut off the forums and start backup proceedures. During this time there will be no access to the forums or the Resource Section.
    2. November 5th Continue with Backups and download same until complete.
    3. November 6th 7:00 turn over server to hosting company to move system to new server
    4. November 7th - 8th restore backups and upgrade forum software.

    If all goes well we should be back online on November 8th.

    To keep you updated we will post updates on AVSIM forums.

    Thank you for your support.

DC-4 Skymaster

Thanks for your answer, Thomas but I have yet theses docs:
- AN 01- 40NM- 4A Parts Catalog for airplanes C-54 ...
- T.O. 1C-54D-1 Flight Manuel USAF series C-54 and Navy models R5D
Someone had indicated them to me at the beginning of the year and project, and I had downloaded them or been sent them to me.
 
And today even the seats are moved under their PBR texture:

1792746295415.png

On the other hand, I notice a slight problem regarding reflectivity.
a) In Blender and UcuPaint, the seats look perfect on the texture side: no metallic appearance and the roughness seems convincing.
b) On the other hand, after creating the COMP file to use them in MSFS, it’s not at all the same thing anymore; the COMP file seems to increases the metallic appearance of the leather too much...

Theses new seats under MSFS 2020 (Bloom is desactivated - DX12 beta used):
1792747564133.png

1792748381045.png
If it is the case, I will have to correct the COMP files after creation by modifying whatever the values in the roughness and metalness files.

PS: Unless it comes from my settings in the simulator, which is strange because I only use those from MSFS and not any shader modifiers external to the application.
 
Last edited:
@Hazza112,
Sorry if you signed up just to have a delivery date, but this addon is freeware and I never set myself a time limit to complete a project: the PBY was done in more than 10 years.
No doubt that the first version of DC-4 was made very quickly, but as for this one, the project started in January 26 and is not yet finished, that’s all I can say.

@Hollister,
As you said, just a little modification with Gimp (saturation) on the albedo texture for the seats and the result seems much better (too tan looking => a brown more darker).
The flight commands are also improved: less noise and less roughtness on the black layer in Ucupaint, a "bake on all the textures" to build all the various textures neeeded and now I have a "black paint with less wear".

1792776135716.png
 
Last edited:
Hello,

Today, I’m going to show you a Blender modeller’s trick to fix certain Blender behaviors that aren’t appreciated by exporting to MSFS.

In Blender when you create a material, if you look via the "SHADING tab" (in the top menu of Blender view) at the structure of the nodes that make up this material, you have a box whose name is gLTF Settings.

1793096499293.png

In fact, when you duplicate an object with a material assigned to that object, Blender normally duplicates this node and renames it "gLTF Settings.001" and so on. This behavior is completely normal for Blender, but when exporting to MSFS, it causes problems and some COMP textures are not processed. In the end, the texture file COMP_PNG.DDS is very often missing.

There is no easy way to find the objects that have this problem except going into the node tree after copying an object and via the drop-down list in this node changing each gLTF Settings.00x to gLTF Settings.
If we don’t pay attention, we can end up with a multitude of nodes of this type and consequently not have a COMP.PNG.DDS file during a compilation for all these objects.

Nevertheless, I have developed a trick to solve this (I insist: the best way to do it is to rename the node with each copy) when you are pressed by modeling when you don’t pay attention to it. This trick requires using the scripting tab and manually entering a few lines of python:

Code:
import bpy

target_name = "glTF Settings.001"
matches = []
for obj in bpy.data.objects:
    for slot in obj.material_slots:
        mat = slot.material
        if not mat or not mat.use_nodes or not mat.node_tree:
            continue
        for node in mat.node_tree.nodes:
            if node.type == "GROUP" and node.node_tree:
                if node.node_tree.name == target_name:
                    matches.append((obj, mat))
                    break  # Avoid reporting the same object twice for this material

# Print results in the System Console
for obj, mat in matches:
    print(f"Object: {obj.name} | Material: {mat.name}")

print(f"\nFound {len(matches)} object/material match(es).")

Then you have to run this script via the RUN button you can then in the console window of Blender read the objects that can be the source of these problems.
1793097041281.png
Then you need to click on each of these listed objects and then via the Shading tab to check if they are compliant or correct.

By modifying the gLTF Settings.001 target name in the script for each item encountered (.001, .002, .003, ..., .0010, .0011, etc.), we can clean up your Bllender file of these "slags" and obtain that each object in your Blender file is associated to a simply node group "gLTF Settings" without any suffix. Then you have an export that will allow us to compile all the texture files correctly.

You can find several days on Internet: this trick isn't documented but I can certified that it really function :cool:
 
Back
Top