Shader Nodes: Render Engine Switch node
Currently you can make make materials different between Cycles and EEVEE through the material output node like such.
However a big usability problem with it is that you basically need to duplicate the entire material tree when you wanna make the material behave differently between engines.
I believe this could be solved with a new node that preprocesses the shader based on what engine is chosen, like such:
Example usage:
This also effectively means that the engine dropdown can be removed from the material output node since you can replicate the exact behavior as shown here.
I believe this shouldn't be difficult to implement in a way similar to how Repeat Zones, Closures, Bundles and Menu Switch was implemented. projects.blender.org/blender/blender/pulls/141936 projects.blender.org/blender/blender/pulls/146896
Also an open question whether or not this should be a feature in Geometry Nodes.
3 Comments
It's been requested multiple times already, please prefer upvoting and commenting on existing proposals to make them more visible in the list instead of adding yet another proposal in the list that dilutes the topic's visibility.
Good to know that in the meantime, we can already have a render engine input node using drivers.
1 if engine == 0 else 0How does that make them more visible in a list?
When sorted by TOP, the first entry has 724 upvotes. The "6 year ago" thread has 11 upvotes. How far down the list are people regularly scrolling?
And "this was suggested 6 years ago" effectively means "and is now off the radar, in this website's suggestion box."
That discussion happened a few times .
Regrouping the upvotes and the activity one post instead of scattering it on several makes it go higher in both Hot and Top sorting.
Regrouping the information brought by different people, additional ideas and suggestions, and all kinds of discussions that happen on any topic, also helps fleshing out proposals, instead of having a bunch of them, many of which seem to be lazy oneliners or poorly thought out.
Hence when mods mark a post as duplicate, it becomes listed below the original so that both devs and users can trace back and forth duplicates to make sure no information is missed. But this takes moderation time which is limited, and even when done, the information is easier to fetch but still scattered.
In an ideal world, I'd see a platform that allows every similar proposal to be combined into one community-currated post that combines all the greatest ideas, mockups and designs people could figure out for each proposal. But for now we are here.
What do become easily off radar is yet another proposal that will get 1-5 votes and just go down in the list as it keeps receiving posts every day. But good proposals, quite the contrary, get traction over time.
The ones you see withhundreds of votes and many comments rarely got them in their early life, it's precisely because they were highly in demand or well done: they stood out, and time made the rest.
But many of those marked done were accomplished years after being posted. In fact, due to how long development can take beyond papercut-level proposals, I'd bet my sandwich the average time to delivery for proposals is counted in years. And to this day there are years-old posts about which development is planned or ongoing.
Just out of mine, this one took 6 years, and that one got partially done after three years but is still waiting for more three years after.