Jump to content


  • Posts

  • Joined

  • Last visited

Personal Information

  • Name

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

Jaanus's Achievements


Newbie (1/14)



  1. I'm guessing this thought might have already come up, but just to mention if no one has: Integration with the rapidly developing ML image generation solutions. This could be an opportunity to also upgrade the COP2 context, now with enhanced ML features. I don't fully understand what is shown in this video linked above, but it appears stability.ai are working on some node-based machine learning image generation solutions. Would be great to see what could be done in Houdini with some similar tools.
  2. Hello, This is more of a question of curiosity at the moment. Although compositing is usually done after rendering multiple passes, how would one go about if one wanted to composite more procedurally at render time, using shaders? Sort of bringing the compositing phase into the SOP/MAT context. I do not mean grabbing a pre-existing image from disc or creating something procedurally and then combining them in a shader with layering or the Composite VOP, I mean making the same calculations, but on the pixels that would be rendered if the object weren't there, something like transmission and ray-bending calculations on glass like surfaces. Basically I'm talking about getting "what would be rendered if the object weren't there" information into the shading context as a possible layer to be made calculations on, blended with, etc. Would there be a specific VOP, something I haven't encountered yet? This could provide some interesting possibilities, I think, for example the option of controlling blending modes at SOP context with attributes more intuitively without having to promote some attributes to shaders, render them as separate passes, and then try to do procedural operations on them in COPs or some compositing application. A more specific example would be a lot of grids, pointed towards the camera, moving around in world or object space, and having materials applied to them that dictate, based on geometry attributes, weather whatever is used as their surface color or texture, would be "overlayed" over everything in the 3D space behind it, added to it, multiplied, used in a difference or subtract calculation, or something else. I hope I communicated clearly enough. Anyone tackled this?
  3. Hello, In Cinema 4D there is a Xpresso node called "pmatterwaves" that lets one spawn particles based on light rays that hit the surface of an object. What would be the most straightforward way to achieve a similar effect in Houdini? I understand baking out light maps is an option, and also coding some custom ray trace calculations in VOPs or Wranglers, but if one didn't want to do neither, would there be some straightforward ways to get this information into SOPs, based on light values and positioning in the scene, a setup with just a few SOP VOP nodes. I've had a few ideas, but many of the lighting model related shader VOP nodes seem to be absent in SOP-context VOPs. I'd assume there have been enough people to want light-surface interaction effects, that there would be some relatively simple way. This is more of a theoretical question. I'm not stuck on any current project, and my aversion to using light maps is mostly because of wanting to find possible alternatives.
  4. Thank You for the replies! @matEvil: Thanks for the suggestion. Have seen it a couple of times, but it was some time ago, might watch again. @kleer001: The notes of it not being used in actual production often, definitely useful. If anyone knows about what was used on the Dr. Strange movie, where in an hallucinatory scene Dr. Strange's hands were turning into fractals (new hands were grown out of his fingers), that would also be great. I thought it should have had something to do with L-Systems at first, but of course it could also be a combination of other methods or some in-house tools, L-System definitely is not the last word on fractal systems. Unfortunately though, for me learning doesn't necessarily go with what is useful, what is interesting and intriguing seems more often to be a path to learning something, and I do like to learn about different ways of doing things and what benefits each brings, or even perhaps, how different tools give slightly new ideas. Am not interested in L-Systems as a single method for doing something, but an option among others. In fact, am quite fond of using SpeedTree myself and am currently going through Hossam Aldin Alaliwi's informative course on getting the most control over the models in Houdini ( https://www.sidefx.com/tutorials/tree-rigging-for-feature-film-tutorial_houdinifx/ ), yet even watching that course I come to feel I should know L-Systems and how they interact with wire-solvers, or how they form shapes, better, and if perhaps some L-Systems+Wire Solvers solution might not give better or quicker results in some cases. Of course I can build and actually am building tools to make the handling of SpeedTree trees in Houdini faster, and the benefits of the L-Systems built-in attributes and possibly imaginative and powerful ways of interaction with wire-solvers and polywire-setups (that are perhaps built in, but I don't yet know about them) might not end up amounting to much productivity-wise, but I usually do benefit from going through different ways of thinking about a problem, even slightly as if a technology historian, stitching together a timeline of what tools some people for some time somewhere were using to solve a problem, before perhaps a better (or an arguably better) tool was devised. In productivity-sense, I might perhaps end up using L-Systems for something else other than trees, some tentacle-growths, or geometric patterns, though, although I do not know yet. Rohan Dalvi has a course on this tool in pattern creation, and it does seem like a niche this node might be good in even productivity-wise. The CmiVFX course though does seem like perhaps the most in-depth course on the tool and it's possible uses ever made. Thank you for the advice on particles and a SOP solver as well, have ended up using them in combination with l-systems on a few occasions, and will definitely not shy away from using these tools if seemingly more appropriate at a given time, by my own evaluation though, not of a production house's who chose a way at some time and couldn't perhaps remember anyway why a method was chosen over another a long way back. @Librarian: If making me only interested in useful stuff were this easy, someone would have done it much earlier, myself probably included. So, no worries, my interest in L-Systems was never based on it necessarily being the best tool for doing something other people think people should be doing with a tool. My own aims for using this tool need not perfectly align with what some productivity-oriented institutions see as the most important aim, in fact the aims better not, or I was doing something less than what I by my own criteria should be doing. Anyway, as they say, often best new ideas come from forgotten old ones. I am not necessarily interested in L-Systems because they seem forgotten, but that might very well be a way I find them useful - to see possibilities other people persuaded out of using them might have become incapable of seeing. A big thanks for the links also! I will be diving into them, as well as other posts dealing with this node here on odforce, if I don't get too side-tracked by some other less-than-mainstream use for the program - the COPs, making sound simulations in CHOPs or hopefully not the using of Houdini as a tool for making vector graphics compatible with Illustrator, as the Entagma guys gave me an idea for using this program as. Anyway, a big thank You!
  5. Hello, I understand that these tutorials were made for an ancient version of Houdini, but I suspect the L-System node hasn't changed too much in the meanwhile. I became a paid user of cmiVFX, but it seems the site became inoperative before I really had the chance to get too far into that course. All the courses on the site became unviewable, and any attempts to contact anyone responsible for keeping it working, failed for quite a few weeks, and so did the ones that Paypal made on my behalf, so I got a refund. The wish to see these tutorials still remains, though. Does anyone here have an idea about how to contact someone to get access to these videos, or the DVDs. As I understand, they have been released on DVDs. Perhaps someone could sell these to me if collecting dust somewhere not being used? Would really appreciate the advice. Any other suggestions for tutorials dealing with L-Systems, would be greatly appreciated as well. I have gone through the Pluralsight course, the Tyler Bay course, Ari Danesh First Steps video, have the digital version of the Algorithmic Beauty of Plants and am playing with the example files with L-systems (and wire solver) from the SideFX docs, and scanning this very site on this subject, but still, if there is a valuable compiled resource for learning the different methods of using this node (like the cmiVFX course seems to be), I'd be interested.
  6. Inside of the actual Houdini program though, I'd say the Sculpt SOP, Creep SOP and Font SOP could do with a slight update (like the Bend SOP and Sweep SOP seem to be getting in this release), and I would love the idea of them expanding on the sound based CHOP tools (as could be seen in this video: ). A robust and intuitive tool set for experimenting with 3D audio setups or presets for running 3D audio simulations would perhaps appeal to a very small niche in the user base, but I do think this niche has great potential to grow, once the developers put some light on this side of using the program. Yes, many 3D programs have the option of animating something based on sound, but there is perhaps some potential in running actual sound simulations, giving physically based material properties to 3D objects, etc, the procedural music generation capabilities of Houdini.
  7. Looking at the amount of ground Houdini developers can cover between releases, I'm going with the idea that they should develop a full blown desktop operating system for desktop computers next. I'm only partly joking though. When learning Houdini, it is often advised to think of it similar to an operating system. That comparison makes me hurt everytime I have to do anything in Windows instead of Houdini. My thought is that a team of SideFX developers, or a group of people more knowledgeable of how Houdini is built should start creating an actual operating system for people who value the amount of control over what processes are currently running on their machine and the team would use the same customer feedback based approach for developing innovative and intuitive never-before-done state-of-the-art tools. Just distilling down the development of new releases of an operating system to the action of the expanding of a tool set that would be useful and innovative, in accordance with the user needs, that is centered on the interests of people who value having a robust access to different processes running on their machine, would be a huge deal. Just the dividing of the "application" abstraction down and concentrating on the managment of processes could be a game changer in that field, and that is something that SideFX seems to have ample experience with. One could argue one of the main powers of Houdini is the ability to divide surface level objects(programs) into individual processes. Seriously, I have been told by people who do not know Houdini, that the answer for such an operating system would be Linux, but I don't think it is quite correct. To me the analogy runs more in the vein of Windows equals Maya, Mac equals Cinema 4D, Linux equals Blender. I haven't had much to do with either Linux or Mac, perhaps there would be an analogy for Houdini, but there doesn't seem to (unless it exists in a not-very-well-known Linux version). Not such that is being designed with the philosophies, aims and innovative/experimental ideas of SideFX. The energy of SideFX would probably follow, because doing ground-breaking things, I imagine, is exhilarating. I think this is a true pain and a great loss to the whole of humanity that there seem to be no such forces working at that market, at least to the eyes of an average user. In my imagination, it would be unimaginably powerful if an operating system ran separated into different contexts as Houdini does - There would be the surface level, or object level, quite similar to Windows based desktops of today, where folders and links and programs exist, but I can't stop imagining different contexts for managing what objects on the surface would actually do. I imagine it a node-based programming context, like contexts in Houdini, and you have different types of information running through the stream of each context, you can manage them by attributes, and have different types of tools, with the tool set constantly evolving and expanding. Getting to understand a complex environment similar to Houdini could of course become a never-ending journey, but just to imagine what one could do with such a set of toolboxes, the boos it could be giving to users' imagination and ability to automate the building of innovative new tools for themselves and others. The idea of exactly into what contexts such an operating system would be best divided into, would probably develop in the making, but I envision a context (1) for file and folder manipulations, a context for managing network connections(2), and then perhaps a combination of contexts for image, video, text manipulations. Each context would be a tool set, just as in Houdini, where one could automate all sorts of processes running on their machine. In the context of network connections(2), one type of data that would be running through data streams would be the list of incoming/outgoing connections, and for example a user could generate node-based firewall programs, that would define what would happen to different connections based on some attributes or criteria, people could perhaps also write automated programs to upload some things from their system or create scheduled programs to download some things from certain servers when some criteria is met. In the file/folder managing context(1) I would imagine a user having access to a wide range of options to edit the information and contents of files and folders on the device. An example would be that perhaps he/she would have the tools to create for example small programs that would change the colour of a folder icon based on the medium age of files in the folder, the age of the files relative to a specific time, or based on the the number of files in the folder, the file type, etc. Imagine files and folders being created en-masse like particles, with different attributes attached that could be used to determine aspects about the creation or just the look of them. Seriously, at least the colour and size of folder icons should be connectable to their age, or different math operations based on different parameters of the included files. This kind of operating system would perhaps introduce new kinds of security problems, but if these would be difficult to overcome, is difficult to envision, but I'm really annoyed by the lack of innovation in the operating system world and the unwillingness to try things differently from the standard parameters by any of the big players (Android for mobile devices included), at least from the point of view that SideFX developers have for their products - not many people seem to be innovating with the aim of giving users robust sets of tools and control over what their computer is doing, as the innovation seems to be instead mainly in the direction of giving companies, governments and other paying entities control over the data that flows through someone's computer, and understanding and accessing of all that data is being made more and more difficult for the users, the process of trying to get some control over what happens in your computer is made progressively incomprehensible, unintuitive, etc. Seriously, what is taking SideFX this long to get into the operating systems business?
  8. Thank You! I will go through what You have suggested. I have seen the linked panel aswell, didn't remember there being a mention of using Houdini for audio in this one, but perhaps I'm just misremembering and merging two different videos in my head. The video I thought it was in, doesn't seem to be the one, so perhaps this is indeed it. Anyway thank You!
  9. Thank You! This was a very helpful response indeed. Although the size calculation doesn't work for me for some reason yet, although triple-checked the spelling (will probably give a go at finding a fix myself later), Your solution to my main question indeed was straightworward and a lot more elegant than any route I thought I would have to go through. In addition You gave me some ideas on how to tackle some other problems in the future. Thank You!
  10. Hello, I decided to have a go with this question. Anyway the basis is that in a video about learning Houdini, which for some reason I am not able to locate at the moment, it was almost casually noted that one of the talkers has used Houdini for generating music. In an imaginative fellow I sometimes tend to be, interested in weird new ways to use tools and, as perhaps Moritz from Entagma would say, abuse them, this small remark sparked huge ideas in my head about all the possibilities. I imagined an undiscovered area of Houdini, where You can do with notes, tones and audio samples similar things to what You can do with pixels, colors and points in a 3D space, in most other areas of Houdini. I imagined being able to place audio sources in 3D space, place microphones as You do cameras and run audio simulations where one sound (or a simulated band member) sends out audio waves and these react to the environment and preset material properties (audio waves reflect different from surfaces with different properties). One could make different sounding accoustic rooms, made of simulated wood, tin, or other materials, You can create sounds by making wind pass through tight areas in 3D space. Obviously I was sure that it was quite inplausible that an area this complex could have been hidden in Houdini without getting much attention. Yes, indeed, I could find almost no other mention of using Houdini to generate music, atleast outside of Houdini docs, it is sometimes taught to be used to make things happen based on audiowaves (3d objects move, things change colour), but no tutorials in sight of building great node-based procedural audio generation trees with the output designed to be audio-files, rather than images. In Houdini docs however I have found tools that seem to be precisely meant for some such use. The Beat CHOP: http://www.sidefx.com/docs/houdini/nodes/chop/beat.html The Record CHOP: http://www.sidefx.com/docs/houdini/nodes/chop/record.html The Spatial Audio CHOP: http://www.sidefx.com/docs/houdini/nodes/chop/spatial.html The Acoustic CHOP: http://www.sidefx.com/docs/houdini/nodes/chop/acoustic.html The Voice Split CHOP: http://www.sidefx.com/docs/houdini/nodes/chop/voicesplit.html The Voice Sync CHOP: http://www.sidefx.com/docs/houdini/nodes/chop/voicesync.html So, does anyone know of any tutorials on how to use Houdini in this way, or perhaps could hint at some programs, similar to Houdini (procedural, node-based), but aimed more specifically at dealing with audio? Admittedly I know close to nothing about generating music on computers, the traditional way. Perhaps there are a lot of people making sound effects and music this way and it is a common practice, just haven't found a right way into that area of knowledge. I have found a post on Fractal Forums about some programs similar to what I'm hoping to find, however. Music generated this way seems to be called generative music. http://www.fractalforums.com/fractal-music/fractal-and-generative-music-software-list/ So anyone share my enthusiasm for finding out ways to do something like that or have some knowledge to share on mentioned practices, other programs?
  11. Hi, I have quite a specific question. What would be a good and straightforward way to make a grid's divisions and aspect ratio depend on an image I am to use in an Attribute from Map SOP to color it? My aim at the moment is a HDA that let's the user choose an image from the hard drive, adds it's colors to the grid and calculates the divisions of that grid and the aspect ratio of that grid based on the image. Currently it seems to me I have to jump through using either Python, COPs context or VEX shader context, although it seems like something that perhaps is somehow still doable in only SOPs and VEX/VOPs. I don't have nothing against doing it in COPs, shader VOPs or Python, I just haven't thought of a straightforward way to get information from one of these areas to SOPs. Under shader VOPs there is a Resolution node that should give these numbers perhaps in the most straightforward way, and there are the COP specific VEX functions ixres and iyres and inputs in the COP VOP node, but at the moment the way VEX and VOPs work in COPs seems different enough from SOPs that I haven't found out a good way to get these numbers into SOPs. Same problem with using Python node in SOPs. Any ideas on how best to tackle this?
  12. I haven't read through all of the of the newer comments, but if it hasn't been mentioned, I think this new three-inputs-three-outputs node structure would be great for the rigging workflow. It would mean having to move some of the things thus far done on the object level to geometry level, but I personally imagine a single node structure with separate streams for say the geometry, bones, and capturing geometry could offer benefits compared to the workflow now.
  13. Thank You, Paolo for the answer and super thanks to Ben for offering the solver! I will definitely give it a go. I do sort of think I understand the nature of Houdini by now, but also think the creators' new way of offering presets and easier one-button/one-node solutions to complex ideas/workflows is positive not only in a way it makes the program easier to learn and more exciting to new users, but also by giving a speed-boost to new ideas and solutions developing inside the community. I think I have learned Houdini to a degree where, although I am far from being an expert at anything yet, I do get a somewhat superior feel and begin to feel I would benefit if the program remained somewhat inaccessible to outsiders, sort of a secret society, where to get in to, one would have to really sweat. Because I have the will to do it and I've done it to a degree already. Perhaps I'm just weird in feeling that, but I think it's somewhat normal and understandable and good for the community if it remained that way to a degree. However, scientifically speaking, whenever processes are made shorter and more intuitive, both workflows and the generation of new ideas becomes faster. The creators didn't have to give Vellum presets for hair, balloons, cloth, for the hard-core users to appreciate this new tool, but I think that by doing that they also benefited those hard-core users, same as with all the other such moves in the program (material-based fracturing system, specialised attribute noise node), by helping them do some things in a faster, more streamlined/automated way, allowed for development of new ideas and options to reach new heights. By a growth-solver, growth nodes, growth presets-tab I mean exactly that - a shorter more streamlined way of doing things that are already perfectly doable in the program. People with stronger will will still go under the hood and create systems that novice users will have to sweat for months trying to wrap their head around, but perhaps on a new level and by having gotten some fresh new ideas on the way. I don't doubt that when people start using these tools making things that mimic what Anastasia Opara, Ben Watts, Matt from CGWiki, and a few others are creating today, Anastasia, Ben and Matt will already be propelled to new heights by ideas these tools might have helped spark. Alright, as an end note I will have to admit that I am still on Houdini Apprentice (most of the free money going for tutorials) and because of the limitation of use between versions, haven't yet gotten into building my own tool set with HDAs, and am probably doing much more of the "building from scratch every time" than I would if I were actually using the program professionally. I definitely benefit from the repetition in my learning, but perhaps this request would have lost some of it's meaning to me personally once I actually get the hang of what these tools offer, however by the point of benefiting the community I would probably still stand by, same as for, say, a specialised lightning/electricity tool set.
  14. Ok, have read a lot of great tips on these forums, this is my first actual post. I don't know if it's a good tactic to start by wishing something new and more, but seeing people generating future possibilities always makes me want to contribute. I would second many ideas already mentioned ( reworked file browser, a faster - perhaps gpu-based renderer), but to add something new, my biggest wish would probably be a dedicated growth-solver and/or growth effects toolbar. I do think L-System and SOP-solver contain concepts and potential that hasn't seen it's fair share of attention from SideFX. The Cinema 4D particle plugin X-Particles has started to give the growth-effects side of particle effects some attention, but probably SideFX could make something truly unique and powerful to gain again some new ground in this territory and give Insydium a run for their money. I suppose a "growth solver" or a "growth effects" tab or a set of presets as such could take multiple intriguing forms, but my own over-ambitious imagination envisions a tool-set that would somehow allow the creation of both natural growths (trees and other plants, infections) and abstract shapes. The SOP-Solver is obviously really powerful and one could still do awesome things with L-Systems if he/she really took the time to learn it in depth, but to me these tools lack some of the excitement they should have. Ease of use looks like something SideFX has started to place value on and not having to make new users necessarily reinvent the wheel everytime. A dedicated button "grow inside" or "grow out of" would go a long way on that road, I'd think, especially if one could add nodes that define whether the growth attaches to a surface, avoids a surface or a force, changes it's behaviour based on the availability of light or space, or perhaps even if some simulation activity is happening inside it's vicinity, - imagine a growth slowing down or changing it's direction, based on whether something is moving, be it a character or some destruction simulation, all through some dedicated and well thought-out workflows. That is an area I'd really love to see some developments made in and it's one in which Houdini has quite a good base for reaching new ground in, compared to other packages. It atleast seems that way to me. As a second idea, I'd probably love to see something similar to Blender's Grease Pencil integrated.
  • Create New...