Mandrake0

There is Houdini engine with python api. That should solve the license issue's. Integration shouldn't be that hard. https://www.sidefx.com/docs/houdini/hapi/

Hi Denis, sorry i can't help you i am not a rigger. i think you have seen the goldfarb tutorials? https://www.sidefx.com/learn/collections/rigging-series/ you also can ask on the houdini discord sometimes goldfarb is also online and answers questions. what's also nice a new release of houdini should come out soon (i think this month). the character system should get a big update but i am not sure if this or next version. (mentioned in the end of the H18 presentation)

Sidefx won't remove MAC build that quick. The only problem with mac is the viewport code / shaders everything else is fine.

I would say like clarisse builder or better like katana.

Are there plans for some update? Don't forget the NLE functionality

Hydra is a defined API for render engines / viewport. If a render engine has a hydra implementation it can be used in houdini / katana and any tool that has hydra API implemented. It will just need USD no .abc but you can add a custom filehandler addon if required. I would say .abc will be obsolet with USD.

Siggraph comes near and some tiny information have been released. Did some one see that houdini will have a build for python 3.x? Have seen on vfx platform that python 3.x is set for 2020 but for 2019 preview builds could be relased. At the moment i play with some ML and python 3.x is manatory and to make a bridge to python 2.x is not so funny for learning.

It takes time and i'm sure something is in the works. For me the question is how they could make mantra gpu based, thats not a easy task.

Redshift and octane are under development for nvlink support. When houdini will add GPU rendering is a open question for some time. It could(speculation) be H18 or later.

Gpu rendering is much simpler to parrelize than fluid simulatens. Octane could use multiple gpus because the transferd all the scene data to each gpu. The gpu's rendered the temp. Rays data and merged the rays into a final image. So it's a simple task todo. Simulations need to be in sync for the 3d data and you must split it up on the gpu's because the data size gets quickly bigger. You also can's use a random seed and merge all the data like a gpu render octane does. So not easy todo(maybe in some years possible with vulkan) and even ocatne couldn't hand in the last decade custom attribute (more then two).

I can't anwser that... sorry.

I don't think it makes sense at all... You need to play with the solvers when you know them then you can ask your self do you need realflow. I would say in 99% you don't need realflow.

As the other said HDA is your friend. But if you want to work on the same file then there is one another way that i think thats possible. You need to build with python a custom multiuser setup. You need to use the .ascode() function to export the scenen to git or custom server and sync it back. It's not a small task but if you want to know more just ask.

The move from opengl to vulkan is a huge task and it takes years. SESI knows what's the best for houdini we only see some benchmarks and SESI has 30 years of knowlege in graphics development.

I know from last year a small talk about another tool where some functions should be in houdini. So i know LOP but not TOP even when i don't have a NDA i wont tell because i find a super suprise needs his moment to show. Btw. I guess TOP could be a game changer how CG works if it's that what i think it is!