Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Messages - romullus

Pages: [1] 2 3 ... 478
General CG Discussion / MOVED: Oculus Quest 2
« on: Yesterday at 13:56:44 »

In the context of the Corona Layered Material, why does it make sense for Layer 1 to inherit displacement information from the base layer, but not bump information?

Actually it doesn't make sense at all, but that's a technical limitation and we all have to live with that until it gets fixed.

It would be great if there were a way to control the influence of bump information from an underlying layer (e.g. base material) on a subsequent layer (e.g. Layer 1).

But you already can control that with masks, isn't? What else control you're expecting there? Maybe couple examples how it should work?

Real world scale is not an obstacle, because it's just a scale, it doesn't change anything. You just need to multiply tiles count by your fabric texture RW size and you will get exactly the same result. If your fabric has 50 by 50 cm RW size and you want 10 unique patches, you just need to set tiles to 500 by 500 cm RW size.

[Max] I need help! / Re: jagged edges from displacement
« on: 2021-09-21, 16:11:36 »
It all depends on how big the object will be in the rendering and how big will be the rendered image itself, in other words, how many pixels will have to cover your texture. 2K displacement might be enough if you're rendering to HD resolution and/or the object in picture is pretty small. I'd suggest to try increase displacement texture resolution to 4K and see if that helps with the issue. Also it's worth to note that usually it's not a good idea to have very high contrast in displacement textures (sharp transition between whitest and darkest pixels), you will get much nicer displacement if you will slightly blur the texture.

[Max] I need help! / Re: jagged edges from displacement
« on: 2021-09-21, 14:26:50 »
It looks like your displacement texture doesn't have enough resolution for smooth displacement.

That's physics, a light source with non zero surface area can't cast perfectly sharp shadows. Maybe it would be helpful to know what outcome you're expecting?

You don't clutter the thread at all. If there's a need to improve funcionality of Corona tools, then users discussions is one of the way to make developers notice that.

[Max] I need help! / Re: Mesh/UVW or Material problem
« on: 2021-09-20, 15:21:30 »
Maybe i don't know where to look at, but i don't see any issues in your screenshots. Maybe it's worth to make them more obvious, paint a big red arrow perhaps?

But why would you want straight square random gamma/hue patches on ground material? Doesn't sound very realistic to me :] Wouldn't it be better to apply same principle like above, just replace tiles map with noise, or some grunge texture? I'm struggle to think about scenario where random texture transformation would also require HSL randomisation per tile, but even if there are such, they probably are highly specific and wouldn't justify cluttering already crowded UI of UVWRandomization texmap.

Okay - figured it out. But this will not work with the UVW-Randomnizer on top.

Personally I would love a solution where one could patch and randomnize sveral different bitmaps on one surface.


Why would you need UVW randomizer in this example? You can't randomize transformation of fabric texture and get sensible result anyway.

How is this set up?

There's nothing to it, really. Bitmap is tiled 10 times, so to match this i set tiles count to 10 by 10 (note that tiling in coordinates rollout needs to be left at default in order to get proper randomisation), then it's up to you how you want it to mix it with bitmap, in this case i used Corona mix in softlight blend mode, but you can use native mix, composite, or even RGB multiply, mask and any combination of them.

Yeah, thought about this as well, but you need the bitmap to be bigger than what you'll show because this one will tile as well. Might work depending on the case, but it's probably really easy to setup.

The bitmap is tiling the same number of times as the tiles map in this case, but it's not crucial, since the bitmap is seamless so you don't have to match their tiling. One can be tiled 5 times and the other 7 times and you won't notice anything strange.

Maybe you should apply KISS principle here? You can randomize everything that way, gamma, hue, saturation, etc. Imagination is the limit :]

Pages: [1] 2 3 ... 478