Re: WS3.0 Philosophy and Status

Hi Alex,

On Thu, May 28, 2026 at 10:05 AM Alex Barrett wrote:

> Great, Stuart. Thank you for the clarity.
>
> The big question I'd like to ask, I suppose, is how can we best help with
> the effort? There's, I would imagine, a lot of people who'd really love to
> see WS3.0 come to fruition and enjoy the benefits. How can we mobilise the
> curious casual user to benefit the project?
>

I think there are a couple of areas that folks can really help:  Testing
the scenery quality, benchlining performance and using the scenery
generation tools.

However, while WS3.0 is available on 2024.1,  I would not recommend using
it for testing purposes as it is quite far behind "next" in terms of
development.  So that means either using "next" (e.g. build from source or
a nightly build), or waiting for the upcoming Tech Preview 1 release.  I
would recommend the latter for the "curious casual user", as I think
running "next" for most users is just going to lead to frustration right
now.

(Of course, if you're already running "next" then you're already seeing
WS3.0).

In terms of helping:

Scenery Quality: There would be a lot of value in folks checking out their
favourite or local area and providing feedback on the scenery quality.
 While WS3.0 uses regional textures, it has a single shader for all land,
as opposed to the multiple different shaders in WS2.0, so rendering is
quite different.

Performance benchmarking:  I'd like to get a better handle on performance
on different platforms.  This is also going to be impacted by the HDR/PBR
pipeline as well as Core Profile support, so it's difficult to make an
apples-to-apples comparison. However,  our hope/expectation is that
performance will be better than 2024.1, particularly on lower-end hardware
and systems with integrated graphics.  It would be really useful to
understand if that is the case or not.

Scenery Generating Tools:  Probably going a bit beyond the curious casual
user, but one big improvement compared with WS2.0 is the ease with which
new scenery can be generated.  Scenery developers can use standard tools
such as QGIS to generate landcover rasters, then a docker image to generate
scenery.  In the future we may have a database that will allow the scenery
to be updated as easily as updating objects in Terrasync. Wayne has already
been using these tools and uncovering lots of bugs and assumptions in
them.  Having more people try it out would be great.

-Stuart

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