Blender Standalone Compositor
Hey! Don't know if it's possibile or if living in some running project... but it would be a game-changer to "extract" the Blender Compositor and create a new Standalone compositing software by maintaining the same interface (that is great) and all the nodes inside.
Useless or useful depends on the performances and would be focused on compositing by integrating the 3D engine of Blender and the VSE that has a lot of sense inside a compositor.
Maybe would be useful to create a "dynamic" link that can communicate with Blender in realtime when both the software are running.
Fusion and Nuke are great but have also downside (not to mention Natron..) therefore a new compositing sofware today is a must have!
Mostly if it can be supported and developed as Blender does today that would grab many users from other software, it could also exploit the potential of modern GPUs and would be open!
Just an idea..
9 Comments
This is not realistic. The compositor relies heavily on the render pipeline of blender. By separating a lot of code needs to be developed to support a compositor only pipeline. Blender power is that a lot is integrated and workflows can be done that is not easily done when it is stand alone. Don’t think this is desirable from developer and user perspective.
What is the game-changer you’re referring to?.
Not sure if it's not realistic..
"The compositor relies heavily on the render pipeline of blender. By separating a lot of code needs to be developed to support a compositor only pipeline. Blender power is that a lot is integrated and workflows can be done that is not easily done when it is stand alone. Don’t think this is desirable from developer and user perspective."
Sure but a standalone compositing tool will have much more performance speaking from the compositing side.. Who does use Blender as a compositor ONLY today for complex projects? Don't know exactly but i think not many.. We use Blender mainly for 3D and rendering that's why we have so many render passes available.
"What is the game-changer you’re referring to?" Today we have Nuke as main industry compositor, Fusion that is great (v.9) and AE but is better for other purposes.. All these software are great but have many downside like UI and performance and prices.. They have still half tools running on CPU and half running on GPU.. Optimizing the Blender compositor with this great UI (as it is) and boost the performance taking advantage of the modern GPU power... This is another game
but that was just in idea. The development team only knows if it's realistic or not and if it's a good choice
You may not be aware but you're replying to jeroenbakker, one of the original developers of the Compositor back in the days and a core developer of Blender nowadays working at the Blender HQ.
Jeroen knows what it takes to "simply extract" the Compositor and it's far from simple, besides adding a huge amount of overhead to maintaining different software, versions, libraries, etc.
People have been suggesting the same about the Video Sequencer, and even Sculpting. It sounds like a good idea until you think twice the amount of work it takes, how it harms Blender as a whole since now you have to deal with multiple software even "bridge" apps and so on and so forth.
Nice. I know is not a "click of a button" this is just a perception that the 3D realm can include all modules of Blender that deals with geometry creation and rendering but the 3D compositing realm should include and focus on compositing/video/tracking and so on. Sure is not easy now, i can imagine, but in the future..
The actual problem is that you have to launch a 3D software to do a Compositing work. This is like to open Davinci (video editor color and other..) to do a compositing project with Fusion and that takes down the performances to a half just to open the software. That is not efficient and the compositing capabilities are wasted.
As i wrote this is just a suggestion, Blender can remain as it is for life and that is good
"You may not be aware but you're replying to jeroenbakker, one of the original developers of the Compositor back in the days and a core developer of Blender nowadays working at the Blender HQ." kudos !
The current development team has enough of a hard time developing ONE software, and you want them to work on two distinct ones for no reason?
Why not? I've suggested to "extract" the Compositor it means the same development will no loger be inside Blender but outside. So thye could focus on a standalone application for compositing/videoediting and Blender could focus on 3d/sculpt/modeling/rendering and so on.. that will be the appropriare scope for a dedicated 3D application.
They could therefore focus development on 3D for Blender and focus on post-production on the standalone Compositor
If it ain't broken, don't fix it.
You think this split will allow developpers to "focus" more on their project. But they can already do that anytime they want. Any dev is free to develop what they want, from inside or outside the BF. And the fact that nobody so far tried to take out a part of Blender to "focus on it" is a proof on itself that splitting isn't a need. IOf a dev wants to work on something, they just do it, end of story. When Clement Foucault reworked Cycles' material view to make it into what became Eevee, he did it on Blender, not on a standalone renderer. When GeoNodes were made, they didn't make a different software.
The different editors are not different softwares developed by different teams to begin with, it's one software worked on by devs that have the freedom to work on any aspect they feel comfortable with and easily overlap whenever needed/wanted.
Dividing that would require to rewrite both Blender and "standalone" tools, split a dev team that is already small enough, split the user base, and force users to have several softwares to do the same thing.
Not diving too much in the kind of questionings this proposals leads to: why not split everything? Why even have a " 3D creation suite " if you can just split every department in a standalone software to "focus" development?
And then, what do you do with funds given by the community? Do you split them as well because user now might not want to fund standalone tools they don't use and care about? How do you split the already acquired funds? What do you do if one of the standalone tool just don't get enough funds to keep devs working on it?
The suggestion was just to separate the 3D and the compositing scope. That's it