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 - rraab

Pages: [1]
1
I am running into problems when exporting MDLs with exposed parameters that have specified group names in the "in Group" field.

I am have built an MDL material with a number of parameters exposed, adding all the detailed information for hard_range, display Name and In Group. And that is where I am running into problems:

When I export the MDL module, SD will always write out the "in_group"-annotation with two empty subgroups. This results in weird grouping when I reimport the exported material. The SD documentation is extremly scarce and does not give any hints on how this works (https://docs.substance3d.com/sddoc/exposing-a-parameter-mdl-145654033.html)

To my understanding, there should not be any empty groups created, unless the user types in the "In Group" field entries separated by a "/" sign. However, the emty subgroups are always generated. I know that I can easily fix this manually in the exported MDL code, but this proves to be tedious as I am making iterations over the MDL material and the materials I am building are way more complex than the one in this example here.


2
So reading the log file turned out to be helpful: Attaching two different diffuse_transmission_bsdfs is a bad idea since it generated two different transmissions  on both sides of the surface which will cause problems with the iray pathtracer. (As I learned this leads to different light interactions on the two sides which would cause the pathtracing algorithm to break)

Attaching one single transmissive bsdf is fine and will solve the issue.

3
When connecting a diffuse_transmission_bsdf to the frontside of an MDL material, the material turns black both on the front and on the backside.

Watch the bug being demonstrated here:
https://www.youtube.com/watch?v=965rGMDYceQ


4
Wouldn't it make sense to add a checkbox when submitting a substance that gives the person that submits it to resign from that clause? I guess some Substance creators of course want to be credited while others are fine with their work being reused. My 2 cents.

5
Substance DesignerSubstance Designer - Discussions - Re: Silk Material
 on: July 19, 2018, 11:06:17 am 
Hi there,

Fabrics depend heavily on shading them correctly. That said, I unfortunately cannot take the time to explain you all the details. There is more magic than just creating the right substance, it is also about creating the right shader for that.

As you know, Substance Designer takes advantage of using Iray with MDL (Material Definition Language). This topic can get very deep and maybe there is someone on the forum that is able to engage more. With MDL you can build a material that gives you the shading properties that you expect from this type of material.

there is a very well written chapter in the MDL Handbook that you can read about this:
http://mdlhandbook.com/mdl_handbook/index.html#example_fabric#example-fabric

I hope that is a good starting point for you.

All best,
Ruediger


6
Substance DesignerSubstance Designer - Showcase - Re: Surfaces
 on: July 12, 2018, 04:23:08 pm 
Neat, I like the third one. The first material uses too much cracks for my taste. You could somehow try to break it up a little bit more. For example, you could make the cracks go deeper and at other places they almost disappear. That hides the underlying cell pattern from which the noises are derived much more.

Cheers,
Rüdiger

7
Substance DesignerSubstance Designer - Showcase - Re: Mr: Flash
 on: July 11, 2018, 03:46:35 pm 
Nice Job!

8
I finally managed to do some iray renderings, after I notices that a material without a visual context does not work so well. Also I did some minor tweaks to the metal so that the hightmap is more balanced with the dents on the surface.


9
I want one of those in my kitchen to cook for me, do the dishes and clean the floor. Pretty cool demo, was this a project for fun or were you doing this for a customer?

Best,
Rüdiger

10
Substance DesignerSubstance Designer - Showcase - Pitted Steel Material
 on: July 06, 2018, 12:12:13 pm 
The best substance training is to work with photo reference material and I love browsing texture sites. So I found on textures.com this substance material (https://www.textures.com/download/substance0075/129446), that was just wiring up some textures apparently as these are available for separate download. I might be wrong though, maybe they just provide the textures there for those that don't use substance and the substance file itself is much more flexible. I did not spend money on inspecting the substance. I was just looking at the images and maps and I took my shot to created my material implementation from scratch with parameters for wearing out the metal, adding variable corrosion and color controls for the rust in the dents of the material as well as adding coloring as results of heat treatment.

The outcome can be seen here:





11
Since I spent the last days building grunge and noise maps I decided to put my experience to good use to create my first serious substance. So I decided to grab a reference image from textures.com (https://www.textures.com/download/floorscheckerboard0035/100036) and see how faithfully I would be able to replicate it. I have to say that I need to get used to some thing in substance designer, first the speed: Having done many effects and materials with Filter Forge I am not used to having imagery generated blazingly fast on the fly. This opens up new worlds of designing materials for me. But the speed comes with a certain prize. In order to get high quality normals, this is much trickier since there is no antialisaing in substance designer which I really miss. (Wink wink)
Anyhow, here is my first checkerboard floor material. I put a lot of emphasis on the specular and glossy maps. In fact, I worked for half a day to make wipe-grunge generators for floors. I am happy with the outcome. However, I haven't reached the photorealsim that I aimed for this time, but what the heck, it was my first serious try to auther a proper material in substance. You can't have it all.



12
That helps indeed!

So I just chose the one parameter where I can't use $depth... Which makes sense, that I cannot change the render pattern dynamically (though if it worked, it would open up new crazy worlds of FX-tree rendering )

I have to apologize for my laziness, not trying it out with another parameter and see if it works there. All I wanted to do is to be able to turn the rendering of a quadrant node off. So the switch is the solution for that.

Thanks for looking into this!

All the best,
Rüdiger

13
Pretty good! I like the peeling skin effect. Something that is not really material related: The renders could use a bit more light, so it is difficult to catch all the details.

  • When looking a bit closer, I found that the fine dark spots are being overlaid by the peeling bark effect. You could try to merge that somehow better. (See attachment)
  • It seems to me, that the normal are missing, or is it due to the lighting?

Good work though!


14
I should have attached it right away, dumb me  :o

So here comes the file with the graph showing the issue. When zou open the FX tree you will notive that I attached the same function to the "Pattern" parameter of the last four quadrant nodes. I was expecting to get different values for $depth, depending on the depth of the quadrant node. However, in each function the $depth evaluates to zero.

Best Regards,
Rüdiger

15
I am working on an FX map that activates/deactivates certain octaves depending on the size of the map that I am generating. To do that, I need to access the FX-Map system variables.

However, upon my first test it appears that no matter in which octave of my FX tree I am, the variable "$depth" always appears to return 0 instead of the level that I am editing. So either this is a bug, or there is something that I understood wrong about the $depth variable.

I attached an image of my FX tree and the function I created for the quadrant nodes.

This was reproduced with Substance designer 2018.1.1

Pages: [1]