Get ShadowMaps with RaymarchGBuffer

Hi guys! I would like to use SuperPhysical (Deferred) with proper ShadowMaps, however my objects are not Geometries but RaymarchGBuffers from SF3D Noises. On the other side, the ShadowMaps node needs a geometry input. Any ideas? Thanks!
GBuffer ShadowMaps

Unfortunately, this isn’t possible in vvvv beta, unless you can develop the whole shadow pipeline for SDFs in HLSL.

But FUSE in vvvv has solved this problem. Definitely, time to have a look at it. You can also share GPU textures, including the depth buffer, with Spout, So you can develop a system using both vvvvs.

maybe the cubemarcher from fieldtrip works for you.

come to the dark side, we got it all working here with gamma and fuse!

2 Likes

Thank you! Yes, this one worked for me! Sometimes the geometry is quite “jittery” though, probably because of the changing number of vertices, particularly when using SDF noise. But in general, it works, and I do get my shadows…
I already tried to make this project in gamma, but got stuck in some beginner’s problems (converting values?). And because have to deliver this project quite soon, had to get back to beta for a while. But with one foot I am already on the dark side ;)

2 Likes

May I ask what you exactly mean with this?

Well, I don’t know exactly what the point was from where I couldn’t go on anymore, but I got confused with different kinds of Vectors, Floats, GpuValue, GpuVoid. The problem often was feeding the output from one node to another node (because it accepts another type of values? - and how to convert between them). My impression was that in beta it is more convenient because there is less distinction between “values”, but please don’t kill me for these words. These things are probably clear to the most of you but since I have a completely different background it is sometimes very hard to find why something is not workling. But I keep trying :-D

A basic introduction is here:

1 Like

In your case you should definitely also watch parts 2 and 3 of the same series.

1 Like

Yes, I will do! Many thanks for these great tutorials!