Hey Kent and Team,
I saw you still use "the old" workflow of putting a b/w texture in the displacement slot and controll the size with a math node. Its still working fine.
However, now there is a "displacement node" to do this stuff with a bit more control.
Is there a reason you do this or are you just not aware?
maybe you are now aware and i could give some knowledge back or there is a reason and I learn something . No critique, because it works anyway. Great tutorial. Love it :-)
PS: One maybe connected follow up tutorial could be one about "procedural pile of tablets/Granulats/cylinders". I'm making a lot of animations in the realm of explosives and chemical catalysts, where most substances are pressed in the form of tablets/pills of different shapes and I didn't really found a way to make those procedurally. https://www.dropbox.com/s/2b40ks4y0yzvnxj/Pistol.mp4?dl=0 this is an proof of concept animation (WIP) for my next film about explosives, here an example of the way a pistol works (intended to have experts check if its right before i make it nice). I needed to make propellant cylinders as meshes, because procedural was not succesful. Just an idea :-D
bburnarakiss Honestly, I'd not heard of the new displacement node. Thanks for letting me know about it! I'll certainly check it out.
As for the pile of cylinders idea, I personally would try to do that with rigid body physics rather than a procedural material. That would produce a much better result imo. I recommend dipping your toe in the physics pool with the Fundamentals of Dynamic course, specifically with this lesson.
bburnarakiss The displacement node came up in this thread where it seems that it forces displacement to only be in the geometry normal direction whereas the displacement socket of the output (old method) allows for horizontal/diagonal displacement.
I think when they introduced vector displacement (https://www.youtube.com/watch?v=ZuSeof9IOxY this guy you may know ) they also put in the displacement node in. As i understand displacement move vertex along face normal, while vector displacement moves it along the normal given by the normal map.
Thx for the tip with the physics. That's what I actually do at the moment. But its time consuming compared to putting on a procedural material. Anyway after the latest fixes in 2.8 it became significantly faster. Not as fast as a procedural material, but its working and has some advantages for the procedural dissolving effect (old trick by Midge Sinnaeve)
love your courses and life streams (although I hardly have time to be live with you)
@theluthier I just came across the Vector Displacement Node which doesn't seem to change much over directly connecting the input. It just gives a couple more sliders and can change the coordinate system of the transformation.
@theluthier train30 I was wondering about this too. Specifically I was wondering how Kent was "getting away with" plugging a value output into the Displacement socket in the Material Output node, given that they recently changed this input of the Material Output node to be a vector.
I guess my own question is more along the lines of "how does that still work?". Taken from the release notes:
When using displacement in materials, a new Displacement node must now be used. This node will be automatically inserted into existing .blend files. The reason for this change is to support vector displacement, and to make it easier to tweak the displacement scale and midlevel. (4a5ee1a) (b129ea8)
https://wiki.blender.org/wiki/Reference/Release_Notes/2.80/Cycles