How do I make river flow along canyon?

[Shabazzy]

Ya could possibly use weight maps/gradient to fade it at polling area and make it a different geo then the pool area geo, and maybe uses a ripple displacement to look as it is slowing down and spreading out into the pool area. though it would be better if the pool area was alot larger. I would also mix texture animation ad displacement, as well as add some particle fx on the sides. Don't forget ya can also use fx in compositing like After Effect to enhance the image. All this in addition to what was posted above.

Yeah, I agree. I think this type of motion would require a mutli-layered approach.

[jwiede]

I have created an object of a canyon with a river flowing through it and would like the flow of the water to be in the direction the river is heading and terminating into a secluded pond.

A minor nit: The river looks much too wide relative to the scale of the (closed) pond. There's nowhere for all that water to go at the end. If it's an art direction thing, so be it (and ignore all of this), I just felt I should mention that their proportions look a bit "off".

One option might be to leave the pond "open" on the right and front sides, so it extends "offscreen" (implying it's much larger in scale).

Another option might be to shrink the width of the river. If you do, you might also want to add some slight (but visible) widening at beginning of U-curve bend -- flow would be constrained there.

though it would be better if the pool area was alot larger.

Ok, so not just my eyes noticing it. Good!

[erikals]

what's this ON Topic stuff... ?!

[gerry_g]

saying something like, "only 2D procedurals can be assigned to uv co ordinates, 3D procedurals lack that ability, so unless you bake out your procedurals as an image sequence your choice of procedurals is very limited, maybe the endomorph trick will get you round that problem, would have to try it to find out"

[Sensei]

saying something like, "only 2D procedurals can be assigned to uv co ordinates, 3D procedurals lack that ability, so unless you bake out your procedurals as an image sequence your choice of procedurals is very limited, maybe the endomorph trick will get you round that problem, would have to try it to find out"

But I just showed tutorial how to use 3D procedural textures with UV mapping...
https://forums.newtek.com/showthread...rial-with-subs

[KScott]

Should be able to use a gradient to control whats affected and whats not. texture wise, weight map, old school alpha map. Its LW, theirs like 10 ways. Getting back into after years myself, been a user since version 5.

Kevin

[MonroePoteet]

Particles, HVs and an Animation Path wind, maybe? A very quick-n-dirty LW2015 scene attached. I'll leave it up to the PBR experts to use LW2018 volumetrics instead of HVs, if desired. I *used* the Fast Fresnel shader for a quick water surface which has been removed post-LW2018.

MOV file: RiverFlow_PathAnimationWind_Particles_2_small.mov

mTp

[Greenlaw]

I think KScott and Greenlaw's suggestions are great, but the problem I see with these suggestions is with regards to the water pooling at the end of its journey.

For me, any movement and type of movement in the water that I bother with depends on scale and camera placement. From a high distance, as in the originally posted image, you really won't see much movement at all. From that distance, it would be more about how the light and sky reflections bounce as the camera moves. I might place some breaking water elements (usually just a masked fractal texture,) here and there for good measure.

If the camera is down closer to the surface, you won't see much of the transition where the flowing water enters the calmer pool. In this case, I'd mostly rely on displaced geometry, and when I entered calmer waters, I just reduce the wave animation.

In either case, I'm probably cheating stuff in textures with masks, and sweetening it all in compositing. Typically, I don't have a lot of time to get a task done, so it's usually about employing what serves the shot best in the shortest amount of time.

If you look through my demo reels over the years, you can see many examples of bodies of water from many camera angles and scale. There's usually a lot of cheating in these shots where I just rendered some basic elements in LightWave and then enhanced a final look in compositing (Ae, Fusion, or Nuke) with fake lighting effects, masks, particles, and natural comp elements. If you know the cheats, the work gets done very quickly and still look pretty decent.

None of this is meant to be a critique of trying to do it all 'correctly' within a single setup; I'm just describing how I approach a case like this.

[OFF]

https://m.youtube.com/watch?v=lTUCPq2pSYY

[erikals]

nice MonroePoteet




think i'd go the displacement route myself, but cool never the less, bit like paint perhaps?

[Greenlaw]

https://m.youtube.com/watch?v=lTUCPq2pSYY

Great demo! That fun to watch.

Edit: Just saw MonroePeet's, that's pretty cool too!

[shrox]

This is totally my plan, retire with the latest version of LW . . . lol
at least I know I am not alone.

I did that. Well, 2015V anyway...

[erikals]

i'm not retiring, and still running LightWave 11.6.

[jwiede]

i'm not retiring, and still running LightWave 11.6.

The whole bit about calling Bullet RBD "complete" without constraints, then selling Bullet constraints as a "big feature" in LW2015 paid version upgrade didn't sit well with many LW customers. Still, I'm glad overall that I upgraded to LW2015, though even at the end Bullet in LW2015 had some serious issues (f.e. dynacache limit, perf.).

[Greenlaw]

I'm using 2019-2020 more and more these days, but I still fall back to 2015 when I'm in a hurry (which is most of the time.) This is mostly out of habit and laziness, I guess.