Messages

31

Substance Designer - Discussions - Re: Morph between 2 images

 on: March 05, 2021, 08:36:06 pm 

What kind of images do you want to morph between?
Performing this type of operation without ugly in-between stages is really very complex. morphing between 2 images with very specific, meaningful details, like morphing between 2 people's faces (with good looking in-between stages) is not doable in substance designer. (this has only somewhat recently been done using advanced deep-learning algorithms)
But morphing between 2 visually similar noises for instance may be more approachable.

32

Substance Designer - Discussions - Re: How can I make tileable hexagon pattern without stretching

 on: March 03, 2021, 05:35:07 pm 

Hexagons are inherently difficult to tile on square textures because the ratio between the height and width of a hexagon is about 1:1.155.
So, the number of tiles in your tile generator should approximate this ratio as close as possible to minimize stretching.

6 by 5 tiles works reasonably well. (ratio= 1:1.2)
8 by 7 works really well (ratio~ 1:1.143)

Anything smaller will just visibly stretch.

33

Substance Designer - Discussions - Re: AO too sharp

 on: March 03, 2021, 02:59:01 am 

These are expected results honestly. The sharp edges appear because there are some parts of your heightmap that are steeper than you may realize.

Results of the AO node are really quite accurate. I have tested this before and they are almost identical to performing an AO bake on geometry that matches the same heightmap.
Perhaps the edges look odd on just the texture, but on your material it will look correct.

Here is a similar example where you can see the sharp edges in the aomap and results in the 3d view are as expected:

34

Substance Designer - Technical Support - Re: Do I need multiple tile samplers in graph to expose differnt rotation of pattern

 on: February 25, 2021, 07:49:42 pm 

I would first combine the 4 rotated shapes into 1 piece so you can then tile that more easily with the tile generator.
Something like this.

If you put that group of 4 into a tile generator and set the offset to 0.5 and play with the scale I think you should be able to tile it.

35

Substance Designer - Technical Support - Re: How can I apply LUT table for substance designer?

 on: February 21, 2021, 07:07:30 pm 

UE4 uses the ACES tonemapper, which has somewhat recently been added to SD in a tab under the 3d view.

36

Substance Painter - Technical Support - Re: Invibles and translucents faces

 on: February 09, 2021, 04:34:11 pm 

Looks like your model has some inverted normals.

You can enable backface culling in blender to see this here as well. (a checkbox in the 'shading' tab at the top of the 3d view)
In blender, select all faces of the model and hit shift+n to make normals consistent.
If this doesnt work, select individual faulty faces and use alt+n to flip their normals.

37

Substance Designer - Technical Support - Re: bad performance when using input value for switch nodes

 on: February 08, 2021, 07:52:36 pm 

Hi Luca,

That's indeed a bit of a frustrating limitation, and makes input values and value processors a bit less useful, considering that switches are a fairly core part of managing node networks and value processors should be as well I think. (or alternatively some different way to feed a sampled value to a switch)

That said, thank you for the clear explanation as always.
Automating my sample count may not be 100% necessary so I think I'll figure out some different approach

38

Substance Designer - Technical Support - bad performance when using input value for switch nodes

 on: February 08, 2021, 02:56:19 pm 

Im currently working on a filter which uses a number of iterations. More iterations increases rendering time as you would expect. So I have a switch set up to select a number of iterations. However, the selection is calculated by a value processor which sets the selection of a multi-switch.



I noticed that performance is always worst-case, as if all iterations are rendered no matter what selection is set by the input value.

I set up a simplified example to demonstrate the issue.

(direct gif link, in case it wont load https://i.imgur.com/mnbmRzJ.mp4)

As you can see, I have the exact same graph except the multi-switch is exposed through a value input on the 'bad' version, and gives always worst-case performance.

I attached the example graphs below.

39

Substance Designer - Discussions - Re: Edges not connecting when increasing the height scale

 on: February 08, 2021, 12:13:03 am 

This is expected behavior for sharp edges.
Try the rounded cube/cylinder or custom geo with rounded edges for better results.

40

Substance Designer - Technical Support - Re: Cannot enable post effects when rendering

 on: January 30, 2021, 10:42:25 pm 

It's under 'edit', next to 'file' in the top left of the screen.

41

Substance Designer - Discussions - Re: I need to randomly tile multiple image inputs with a 'copy' blend mode - how?

 on: January 25, 2021, 06:07:28 pm 

You can also use the 'shape splatter' node for you height and follow it up with a 'shape splatter blend color' to accurately color the individual shapes.

42

Substance Designer - Discussions - Re: Prevent Directional Warp Blurring? (Causes Issues with Edge Detect)

 on: January 16, 2021, 06:30:29 pm 

Thanks for the update. I'm glad it's done the job

43

Substance Designer - Discussions - Re: Get angle from Tile sampler?

 on: January 16, 2021, 06:29:41 pm 

No worries, I would have felt bad letting you struggle with this spiderweb of a node as your first contact with the fx-map :p
I hope it works out for you, let me know if you have any issues with it.

44

Substance Designer - Discussions - Re: Get angle from Tile sampler?

 on: January 16, 2021, 12:42:05 pm 

I decided to set it up for you.
I've added a rotation and rotation(vector) option to the color parameterization options.

45

Substance Designer - Technical Support - Re: Non uniform directional warp makes the texture have weird lines

 on: January 16, 2021, 02:28:01 am 

For once this isn't actually caused by bit depth, but by the node itself.
The way the trails are created is kind of by copy-pasting pixels a number of times in a row. Since your intensity is quite high, there aren't enough of these copies and gaps appear between them.

There isn't a perfect solution to this, other than modifying the node, but you can try setting the intensity lower and just put the same node a couple times in a row to get a cleaner result. This won't necessarily have the exact same result though. Some settings like a strong angle input multiplier and any amount of trail fade will cause some issues.