This is general wishlist and propably would take longer than one major release but Houdini must start attracting animators and artists instead of developers and technical minds only.

With more artists this software will have great synergy in evolution and i see hope for that!

• Better UI for keyframe animators - similar to Cascadeur one!
  
  
• Camera sequencer, better camera rigs and settings, light sequencer.
  
  
• Option for Importing NURBS from modeling software called Plasticity, with possibility of animation it (to every element like shape or point), and then render without losing NURBS quality.
  
  
• Option to work with NeRFs, allowing you to hand paint them (every feauture fe. Opacity or rotation of points) project textures, making NeRF materials - usefull for relighting. apply geometry modifiers, and animate it in easy way, hell.. mayby even rig it using bones and APEX? . Basically, everything that was possible with meshes but now using NeRFs. NeRFs are the new mesh converting NeRF to particles so VFX could be done with NerRF and exported to game engine.
  
  
• NanoVDB !
  
  
• Housecleaning of Houdini, new UI (artist friendly look and functionality), enhancments for dvelopers for quicker bug fixing Python errors, new UI for animators!
  
  
• Better looking gizmos with enhaned functionality (its not 1999 anymore)
  
  
• Spectral rendering in Karma XPU, also introducing Neural rendering.
  

And more Tool level wishes:

Stroke recorder in TIME – a tool where you can record changes to geometry, color, or any other parameter, using Wacom tablets with pressure or touch sensitivity.

You could then edit those strokes in a UI similar to editing software like Adobe After Effects or DaVinci Resolve. You could group paint strokes together, stretch them over time, fuse parameters and perform other non-linear animation operations.

I see Houdini as a platform that fuses everything together using .USD (kind of Nvidia Omniverse vision but fck big corpos xD, where quick fixes and tweaks should be easy to make by hand. This non-destructive stroke recorder could help with that. It would also be a cool tool to experiment with motion design. Something similar to OTOY's Sculptron.

Maybe you could paint with entire takes? Like applying changes of values using hand with Wacom tablet to all nodes ("take" is this dropdown menu on top right) ? I'm not sure


.

In LOP we can drag&drop everything from the table to the parameters, but we can't drag&drop anything from spreadsheet to sop. This is incredibly strange, considering that you are in sop much more often than in lop.
Why not finally make it convenient for users in sop?

Fix/re-write the weight painting system. Showcasing and working on ML solutions should come after the basics have been covered. As it stands, the basics are shit right now. I mean "worse than 20+ yrs solutions" shit. I hope H21 will address this area, otherwise I'm not sure why SideFX is even working on making Houdini a powerhouse for CFX. Unless I've misunderstood and Houdini is still meant to be positioned as a finish-up tool in Maya pipelines when it comes to character animation.

citizen
Fix/re-write the weight painting system. Showcasing and working on ML solutions should come after the basics have been covered. As it stands, the basics are shit right now. I mean "worse than 20+ yrs solutions" shit. I hope H21 will address this area, otherwise I'm not sure why SideFX is even working on making Houdini a powerhouse for CFX. Unless I've misunderstood and Houdini is still meant to be positioned as a finish-up tool in Maya pipelines when it comes to character animation.

How do you really feel lol?
What's wrong with the current weight painting system? Could you elaborate? Might be helpful.

Physics-based ML is the basics for current industry CFX. Has been for some time now.

2014 Subspace Clothing Simulation Using Adaptive Bases
https://graphics.pixar.com/library/SubspaceClothing/paper.pdf [graphics.pixar.com]

2016 Dynamic skin deformation simulation using musculoskeletal
model and soft tissue dynamics
https://studios.disneyresearch.com/wp-content/uploads/2019/04/Dynamic-Skin-Deformation-SimulationUsing-Musculoskeletal-Model-and-Soft-Tissue-Dynamics-Paper.pdf [studios.disneyresearch.com]

2016 Real-time Physics-based Motion Capture with Sparse Sensors
https://studios.disneyresearch.com/wp-content/uploads/2019/03/Real-time-Physics-based-Motion-Capture-with-Sparse-Sensors-1.pdf [studios.disneyresearch.com]

2018 An Empirical Rig for Jaw Animation
https://studios.disneyresearch.com/wp-content/uploads/2019/05/An-Empirical-Rig-for-Jaw-Animation-Paper.pdf [studios.disneyresearch.com]

2019 Sharp Kelvinlets: Elastic Deformations with Cusps and Localized Falloffs
https://graphics.pixar.com/library/SharpKelvinlets/paper.pdf [graphics.pixar.com]

2019 Tangent Space Optimization of Controls for Character Animation
https://studios.disneyresearch.com/wp-content/uploads/2019/06/Tangent-Space-Optimization-of-Controls-for-Character-Animation.pdf [studios.disneyresearch.com]

2020 Interactive Sculpting of Digital Faces Using an Anatomical Modeling Paradigm
https://studios.disneyresearch.com/wp-content/uploads/2020/06/Interactive-Sculpting-of-Digital-Faces-Using-an-Anatomical-Modeling-Paradigm-1.pdf [studios.disneyresearch.com]

2020 FaceBaker: Baking Character Facial Rigs with Machine Learning
https://graphics.pixar.com/library/FaceBaker/paper.pdf [graphics.pixar.com]

2024 Stochastic Computation of Barycentric Coordinates
https://graphics.pixar.com/library/StochasticCoordinates/paper.pdf [graphics.pixar.com]

2024 KineFX with APEX Rigging and ML Skinning
https://www.sidefx.com/contentlibrary/kinefx-with-apex-rigging-and-ml-skinning/ [www.sidefx.com]



CURRENT MOOD: https://soundcloud.com/thagodfahim/love-n-hate [soundcloud.com]

1.

My biggest wish is a tiltable (right now you can only shift it) camera backplate with live preview in the viewport.

For architectural projects this would make things so much more streamlined and creative. Especially when you're trying to put models into already distorted backgrounds, this could speed up the whole process by far.


2.

More (and robust) support of CAD formats for import.

Right now I have to rely on several 3rd party tools for conversion. This is just cumbersome and error prone.


3.

An option to switch the animation editor into a speed/acceleration mode.

Right now the editing is all value based. So, when editing value changes over time you're stuck with using the bezier handles or adding more keyframes which can get quite cumbersome, as well.

In my opinion this is a huge minus. As much as I dislike after effects, it's one of its bigger strengths for motion graphics, imho.

Its more general idea than wish for software feauture.

Create community feedback HUB where you could vote for NODES that need more love.
With option to add detailed sugestions with screenshots or other visualizations.

Fe. I will upvote NODE -> bend and add detailed sugestion: add logaritmic slider.

That way community will decide wich nodes needs more attantion. Ah.. it should be idea and improvement driven platform not bug report.

That way Houdini could become quasi open source. So it will be easier to prompt LLM for writing new imrovements (iam joking)

I hate Autodesk but they had good idea with their community feedback Hub where you could vote for certain ideas so developers knew better what should be updated.

Houdini should be reimagined!

I remember when i was testing Houdini back in day and first thing i was doing was connecting bend node to check how fast is Houdini compared to MAX and then there was this weird unintuitive bend envelope coordinates xD so i was like no fcking way iam gonna use 3D that that way. I know that other artists did the same check heheh

If community feedback hab existed back then i would upvote this and explained what should be rearanged and updated.


.

I do like this idea and I am working on something in relation to documentation. What I have found is that Forum 'style' knowledge repos are easily manipulated. This is an artifact of the feedback that is allowed with buttons and then also the Forum culture that exists in general. Social media is designed to be divisive and the rating facilitates that. I think there are ways to develop community without the pitfalls of toxic culture.

I do not have a new alternative yet, but I liken it to the current educational environment at school and see a better pathway to knowledge through the documentation that is already there. In order to avoid cynicism and sarcasm when seeking educational dialogue. These are confusing when just interpreting understanding through a 'Forum' style content repo.

I think that something that fact checks or provides resources in response and then how those resources are interacted with can be measured. Something that chatbots cannot spam.

Let's get back to the sinful earth from the clouds of meaningless fantasies.

So, what we really need in the next version from real things is hotkeys for flags for different nodes or node sets like QuickView. As an example - you have a huge tree of nodes, you edit a branch and want to see the final merge which is very far away. And now you use QuickView to move there, press the flag, look, move back again (lucky if you didn't forget to put QuickView in the old place), put the flag again. Long and inconvenient. Compare - you just pressed the hotkey and saw the result of the flag/flags display on that distant node/node set. Pressed again - the flags returned to the current state. You did not move around the tree and were not distracted. And so you can collect somewhere the final set of nodes applied to the result - color, bevels, subdivs, etc. And at any moment, without being distracted from the current tree, see how it looks.
And yes, SideFX, be sure to take into implementation the script from Vanzhulla for installing a hotkey display of the flag and activity on the node closest to the cursor. This incredibly increases the usability of nodes.

Again, sarcasm and cynicism. I really hope that in the future there are better alternatives than toxic forums to engage and form constructive community conversations that proactively develop solutions which can only be community driven. Companies rely on communities for these solutions to develop. The vibes here are just terrible all the time... But then again, this is my meaningless fantasy in a "wishlist" post on a Professional Forum.

Wishes can lead to self actualization but Who cares?

citizen
Fix/re-write the weight painting system. Showcasing and working on ML solutions should come after the basics have been covered. As it stands, the basics are shit right now. I mean "worse than 20+ yrs solutions" shit. I hope H21 will address this area, otherwise I'm not sure why SideFX is even working on making Houdini a powerhouse for CFX. Unless I've misunderstood and Houdini is still meant to be positioned as a finish-up tool in Maya pipelines when it comes to character animation.

Like the Capturepaint spreadsheet crashing Houdini to the desktop for instance ? https://www.sidefx.com/forum/topic/98003/ [www.sidefx.com]

In H21, it would be good to have the ONNX nodes for M-series use the Apple Neural Engine with the ONNX CoreML Device instead of the CPU or GPU. Frees up space on the CPU/GPU and is purpose built for inference efficiency.

Possibly even lean into hardware accelerators instead of device agnoistic OpenCL GPU frameworks. These device accelerators are being used in the Linear Solver SOP and could be brought to COPs and Volumes. Accelerate Lib on Mac does not crash COPs like OpenCL can.

https://www.sidefx.com/docs/houdini/nodes/sop/linearsolver.html [www.sidefx.com]

Seems the Linear Solver SOP is much more full featured than OpenCL with all the proven performant solvers in it.

This SOP node performs common numerical linear algebra operations, including linear system solve, least squares solve, and matrix decompositions, using real matrix and vector stored as geometry attributes, and outputs the result as a geometry attribute. This node supports both dense and sparse matrix inputs encoded in several ways. A dense matrix is a matrix where entries are fully populated with non-zero entries, whereas a sparse matrix is a matrix with a small proportion of non-zero entries. Both dense and sparse linear systems can be solved using direct or iterative solvers. Direct solvers first decompose the input matrix and solve the system using the decomposition, while iterative solvers multiply the input matrix to vectors multiple times until convergence to find the solution. Refer to Numerical Linear Algebra and Apple Accelerate Framework: Sparse Solvers for the summary of numerical linear algebra.

This node is implemented as a wrapper of linear solver and matrix decomposition classes available in Eigen and Spectra.

oldteapot7
Its more general idea than wish for software feauture.

Create community feedback HUB where you could vote for NODES that need more love.
With option to add detailed sugestions with screenshots or other visualizations.

Fe. I will upvote NODE -> bend and add detailed sugestion: add logaritmic slider.

That way community will decide wich nodes needs more attantion. Ah.. it should be idea and improvement driven platform not bug report.

That way Houdini could become quasi open source. So it will be easier to prompt LLM for writing new imrovements (iam joking)

I hate Autodesk but they had good idea with their community feedback Hub where you could vote for certain ideas so developers knew better what should be updated.

Houdini should be reimagined!

I remember when i was testing Houdini back in day and first thing i was doing was connecting bend node to check how fast is Houdini compared to MAX and then there was this weird unintuitive bend envelope coordinates xD so i was like no fcking way iam gonna use 3D that that way. I know that other artists did the same check heheh

If community feedback hab existed back then i would upvote this and explained what should be rearanged and updated.


.

There is nothing stopping you right now from submitting an RFE or even a new tool suggestion. There never has been, it's been an open dialogue since the beginning. Even an apprentice user (non-paying customer) can make suggestions for improvements to nodes, or new feature functionality.

Maybe you don't know about it?

L

lewis_T
Even an apprentice user (non-paying customer) can make suggestions for improvements to nodes, or new feature functionality.

I must admit that now, for example, I have absolutely no idea of ​​the logic behind the developers' decision-making.
On the one hand, they bring some new functions that no one asked them for, and on the other hand, bugs and inconveniences that users have been screaming about for years are completely ignored. But sometimes they are fixed. No predictability.
Perhaps, if there was a bug poll, the developers would better understand the users' priorities?
I wrote hundreds of bug reports and RFEs with the same feeling that I write to Santa Claus and with the same hope for success. Dear Santa, I want to get a properly working polybevel and properly working groups in polyextrud in the new year. In response - silence...

lewis_T
There is nothing stopping you right now from submitting an RFE or even a new tool suggestion. There never has been, it's been an open dialogue since the beginning. Even an apprentice user (non-paying customer) can make suggestions for improvements to nodes, or new feature functionality.

Maybe you don't know about it?

The "wish" articulated was for a public dialogue with community that the developers can get info from. Not a back channel convo that is not visible.

Maybe you do not know how to positively contribute feedback in community? Professional communities build new things and at the pace faster than a company-governed development cycle can.

Narrowing the innovation pipeline through ignorance should not be prioritized in H21. More than ever the community should be empowered to directly contribute so that Houdini keeps pace. There is no future that relevance is sustained without conviviality.

CURRENT MOOD:
"For What It's Worth"