Re: Updated volumetric clouds on next

Hi Stuart,

I had a peek at the voxel bitmaps, cu.voxels.png has some colour that hits the edge of the bitmap before it graduates to black (or clear) would that hard edge be causing the stepping effect?

Even though I've read the wiki I still need to get my head round how those voxel bitmaps translate to 3d so I could be barking up the wrong tree.

Chris



-------- Original message --------
From: Stuart Buchanan <[email protected]>
Date: 23/06/2026 09:35 (GMT+00:00)
To: FlightGear developers discussions <[email protected]>
Subject: Re: [Flightgear-devel] Updated volumetric clouds on next

Hi Chris,

On Tue, Jun 23, 2026 at 8:53 AM Chris Boylan wrote:
Agreed, very nice work and thanks for posting your progress.

When I came to upload the photos I was pleasantly surprised at how much better they were than the screenshots that I posted in January.  Being heads down and improving things incrementally meant I really hadn't noticed the progress in the last couple of months.

One thing that's perhaps not very obvious from the screenshots I shared is that the clouds shadow each other.  So if you have a low layer of stratus with a layer above you get lovely shadows of the upper clouds against the lower.  I'm hoping to extend this to shadowing the rest of the scene as well.  Because this is all within the HDR pipeline  I can just apply the shadowing to the right stage and it will automatically apply to the entire scene.  Even better, the light balance exposure etc will be adjusted automatically.

An observation: In the images at this development stage, I see see duplicate cloud shapes and some banding artifacts,
and I'm assuming you're aware while you work on the oversll effect rather than scaling across larger areas with some sort of random seed.

Yes, these are known issues.

The individual cloud shapes are defined as 3D textures of density and wispiness/billowiness (see https://wiki.flightgear.org/3D_clouds#Voxel-based_clouds_(post_2024) ).  Right now there aren't many cloud definitions, and there are some particularly "blocky" ones that need to be changed.

There's also a bit or mirroring of the cloud field itself under "basic" weather.  The cloud field is approximately 100km across and then is mirrored.  When Advanced Weather is supported that problem should go away.

The banding is an ongoing issue I've still to resolve.  It's proving quite stubborn to diagnose!

-Stuart

_______________________________________________
Flightgear-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/flightgear-devel