Author Topic: [SOLVED] Exposing the Random Seed of Individual Nodes  (Read 3222 times)

I created a stylized texture that features rocks but I don't want to randomize the underlying base texture along with the rocks because I want to maintain tiling integrity throughout multiple variants of the same material, but I am unable to expose the individual random seed parameters for the shape splatter nodes used for generating the rocks. Is anyone able to explain why access is limited to the random seed and any possible work arounds?

At the end of the day it is not that big of a deal because I can simply play with the random seed and export the individual maps for each of the variants I decide on, but it would be nice to be able to just make a few variant graphs with just a few clicks and not have to do everything manually.
Last Edit: June 07, 2019, 09:42:36 am

Hello,

You can actually assign exposed parameters to drive random seeds in your graphs. This can be done by following these steps:
  • Create the input parameters you need at the graph level. Please note that only native Integer values will work reliably;
  • Create an Empty Function for the Random Seed at the node level;
  • Read the value of the parameter you created with a 'Get Integer' node.
Additionally, please keep in mind that operations such as multiplications and divisions should be avoided in Random Seed functions. Therefore, please stick to additions and substractions for this type of function.
Also, by 'native Integer values' I mean that converting a Float value to an Integer with a 'Cast' node should not be done, as the value will remain processed as a Float which will not provide reliable outputs. As such, we recommend working with Integer from the outset.

Best regards,
Luca Giarrizzo
QA Analyst
Substance Designer Team

You can actually assign exposed parameters to drive random seeds in your graphs.

Luca, thank you for your rapid reply!

I took some time to experiment with and learn about this way of doings things; at first, I did not think this was going to be an accurate solution but that was because I did not fully understand how to use custom inputs. It was not until recently that I came to understand that a custom input variable needs input value node to drive it. After discovering the input value node within the atomic nodes it all became clear!

Thank you for pointing me in the right direction!

Hello,

We are happy to help, the value nodes added with the new v7 engine are indeed powerful and exciting new tools!
For your information, you can find learning resources to hone your skills and further your understanding at Substance Academy.

Have fun using the Substance tools!

Best regards,
Luca Giarrizzo
QA Analyst
Substance Designer Team

Hi guys,

I'm lost here:

Quote
Read the value of the parameter you created with a 'Get Integer' node.

How? The names of the Input Value nodes from the main graph don't show up in the dropdown list of the Get Integer node.

By the way I really don't understand why the random seed value cannot be exposed as simply as all others?

Thanks for any help!

Cheers,
Miklos

Make sure your new variable is an integer1.

I made a little gif of the proces if that's helpful to anyone:
https://i.imgur.com/FCGv47E.mp4

Thanks Eggfruit!

My mistake was that I didn't make the integer as a new parameter, just as a new input value node in the main graph.