My Journey integrating Daz with Houdini
ggrancharov
Posts: 39
Prelogue
As most people here I started doing static art over the past several years (almost a decade actually) and sit on a quite heavy collection of Daz figures and accessories well in the mid 5 figure range in terms of expenses (and I sincirely hope the mods don't kill my thread having this in mind). The obvious step of anyone after doing a lot of static work is getting his feet wet in animation, and when you already have a massice library in Daz its pretty obviousl that you want to use it.
Like everyone else I had to make a decision 1-2 years ago if I go with iClone/CC which is the easiert way to animate Daz characters, or if I jump into Maya or C4D, because lets face it - if you do something for fun why not learn something (in this case a software) in the process that can land you some freelance work. With that being said, almost nobody is looking for Realillusion content creators, which you can very easily find a gig for Maya or C4D. I experimented a bit with both platforms, it's complicated compared to Daz, but not that much - so many freely availabe tutorials online and integrating Daz characters seems easy enough once you figure out how to do the export/import properly.
Yet C4D and Maya have their limitations. While rigging and animating is relatively easy, creating envioronments and especially special effects is a PIA except you download 100s of plugins and make your mac react slower to a simple transform than trying to get out of bad after a bad hangover. Another problem is the competition - if I do want to learn something new, do I really want to spend half a decade just to compete with someone who is doing that for over 10 years if I want to do some freelance work?
Then I stumpled upon Houdini. The "king" of special effects that all the major studios use (yes, I know, in a pipeline with C4d/Maya I get back to that later). Just type in Houdini VFX into youtube and you will be looking at the same rabbot hole I was looking over a year ago. The possibillieties are just breathtaking - and the learning curve - brutal to say at least. Houdini is the most complicated, the most difficult piece of software you can lay your hands on. At least thats what Reddit told me. They literally laughed at me when I asked them if investing 5 years into Houdini would be enough to be able to create some semi-professional animated videos. They laughed even more when I told them that I was planning on using Daz characters. "Impossible", "waist of time", "you're crazy" and some other nice replies.
The stubborn guy I am, I didn't give up. I started getting into the basics of Houdini following all the free tutorials I could get y hands on, then moving on to some paid ones. After around half a year I felt comfortable enough to combine bits and parts of what I've learned into my "own" set of animation and I started to really see why Houdini is so powerful.
One drawback of Hudini is their animation engine - however there has been quite some progress in the newer versions. Actually, people doing rigging and animating characters IN Houdini are almost non-existent and I see that as the "golden niche" so to speak if you are serious about investing time in a "fallback career based on your hobby" - and as an added bonus, if you were manage to do that with Daz characters you would be one of a handful of people that can pull it off. Because, why spend a week shaping a character in Zbrush when you can do the same conveniently in under two hours with sliders considering you have a vast variety of characters an morph packages? Well, the game was on.
The Now.
For me personally, I am working with a Daz -> MB -> Houdini pipeline, but over time I'd like to leave out MB completly and focus more on animating inside of Houdini. Using MB only makes sense if your animation doesn't interact with objects (collissions), but once you have a character riding a bike it seems like a better idea to do everything in Houdini.
Currenty I am in the first stage of creating realistic skin deformations when interacting with foreign objects and also experiment with grooming to make hair movement as close to realstic as possible. My next step is to fine tuning face expressions and hopefully create a script that can be used as "text to speech" and actally look realistic. That way we would be able to use Facecapp capture for facial expressions via blendshapes but text to speech for lip and jaw movement which again, will give you a much beter starting point for finetuning your scene.
So after this whole tirade, I am looking for people who already dipped their toes into Houdini using Daz chacacters/props so we can exchange experience and come up with solutions to problems as well as learn from each other. I did a uick search and found some existing threads on Houdini, so I am sure we can do a small team - I can' be the only one doing this.
Cheers,
George
PS: Lets keep this thread focused on Houdini, I don't want this to end up in a flame war between favorite softwares because the thread will be deleted. If you use CC/iClone or C4D/Maya thats perfectly fine with me and I am sure you make amazing results, but there are already many many threads on that topic.
Daz 3D is part of 
Connect
DAZ Productions, Inc.
7533 S Center View Ct #4664
West Jordan, UT 84084
Licensing Agreement | Terms of Service | Privacy Policy | EULA
© 2025 Daz Productions Inc. All Rights Reserved.
Comments
You aren't going to find many people on these forums who use software other than Daz, especially if it's paid and incredibly complicated software like Houdini. I think for the people able enough to use Houdini, they'd have already established a workflow around leveraging it's strengths and avoiding unnecessary headache e.g. rigging/animating. Developing those skills within Houdini is not conducive to an industry workflow -- and neither is coming within a mile of Daz if we're being honest (in reference to your fallback career comment).
Houdini is of course the default for everything simulations with its amazing vellum solver. Learning to rig characters in Maya and import alembic or vdb from Houdini is probably the best course of action. But as a neat hobby, I could see how rigging in Houdini might be a fun hole to dig.
Personally dont have time to learn Houdini, but i hope things work out for you. I have looked into Houdini several times because amazing simulation videos are produced all the time from conferences, and research papers, etc.
At one point, i wanted to simulate particle build-up like snow etc, and Houdini results were so much better than the Blender examples i was able to find.
Because Houdini is a kind of industry standard type of software, i think learning it would give you many options even if you move on from integrating it with Daz.
Very interested to hear about your process. I'd gotten Houdini doing cloth and jiggle-style physics over here: https://www.daz3d.com/forums/discussion/comment/4825921/#Comment_4825921
But I moved away from it because I could only use the low resolution FBX characters in the pipeline I had (I was going through iclone). But I recently discovered I could rig the OBJ characters in blender using Auto Rig Pro. It did a great job of rigging a level one subsurf OBJ straight out of Daz. Even adding a facial rig, although I had to apply some tweaks to the eyes.
Are you rerigging in Houdini? I found that quite difficult and slow myself, at least compared to Autorig. At some point I was going to take my Autorigged character and bring it into Houdini to try animating and physics simulations. I want to do animations in Houdini for the exact reasons you've mentioned, being able to animate with the physical environment right there.
I delved into trying to get daz peoples into houdini, didn't end up having luck with it. My roadblaock was setting up a good univeersal skin shader that I could use with the textures that come with daz peoples. After a while of that, iray came out and I gave up on the idea. I loved the mantra render engine, could do massive sized renders pretty fast lol.
I got Houdini 19 Indie for the GPU accelerated cloth sim. I haven't had enough time to learn much more than that, but the SDK is a dream... It is well documented and things that seem like they should be easy usually are, and the forums are full of knowledgeable, technical minded people who share info. I was able to port a hem pinning script from Python in Blender to C++ for Houdini in a few hours.
My current project is one to convert dForce Curly hair to a particle system in Blender, with the same styling, but my experience with Houdini was so positive that I may just try to do it in Houdini.
Houdini also handled an extremely large Alembic file, as Blender did, but Marvelous Designer choked on.
I've starred this thread... I hope there'll be more discussion because Daz to Houdini via Alembic works well.
Sounds cool. I don't do animation myself, but I'm looking forward to learning about your progress.
Wow, I am so relieved I am not the only one after looking at all the answers. Awesome!
Wow, I completely missed your thread - amazing work, especially the physics! Rerigging in Houdini manually is a nightmare, but thankfully a friend of mine is almost done doing a daz to houdini rig autorigger, I let him know about this thread because he can much better explain what his mod is doing and how awesome it is. He was working on that for over one year now and it seems to be almost finished. I has the same workflow you currently do with blender, it uses an OBJ export on high sub-HD level to keep the high resolution but importing it directly into Houdini.
To be honest all my attempts of converting daz hair into something usable in Houdini failed, without having to do a lot of manual work. It seemed easier to just use the Houdini particle or even cloth system. I am still working on this, but using the cloth system for the "under layer" and particles for hairs strains and top layer seems to work out nicely. I will post some examples soon, I am currently at home and have no access to them.
====
https://www.youtube.com/watch?v=CjfxO_136Bg&list=LLeLecLpgtm-BwZIeDrWRZJA
This is a good example of what I am trying to achive right now but a bit more 'dramatical' and I am almost done with the VFX part, now working on the character/hair on the motorcycle - in this demo it is too anime like, I want the Daz realism. I can help you guys out with the VFX part if you have any questions. We might even form a small group and share some hud files, i believe its much easier to learn from a hud file. I promise to get a bit more technical in my answers once I am back in the office and have access to my files, to be honest I didn't expct even one answer ;)
So long,
George
Whoa! That's amazing. I can't wait for it to be out I'd happily help test and pay to support a finished product ????
Thanks for the insight. I'm already loving this thread.
I had an idea to try using proximity to the cap mesh as a weight for pinning, and run a cloth simulation on a polyreduced version of the hair mesh. Then use the results of that to influence the hair. It seems like there should be an easy way to get at least subtle physics based movement on the hair in Houdini.
Very interesting. I'm also wondering whether to put time into Houdini. I've played with it a little. No tutorial I try to follow ever works out right and it's impossible for me to discover why (you need knowledge you don't have at first to "debug" networks). Some things are slowly clicking though... I, like you, can see it's immensely powerful. A high res Daz rig in there would be too good to be true however.
Played with this a little yesterday. Something that's great is that when you import an OBJ, you can create multiple groups from the imported material groups, then apply mats to those groups. After that, you can reimport as many different obj's as you want to that network and your groups will retain those mats, even if you delete an article of clothing or a hat or something. You don't get the vertex mismatch error importing a new OBJ (like in many other applications) and you do get to keep your face/torso/arms/legs/etc. mats assigned. You can even build this once and then put it in the shelf for easy reuse.
Maybe that was obvious to everyone but me, however I find it a real timesaver.
Moin,
I have an HDA (Houdini tool) in the works that imports arbitrary FBX rigs and converts FBX style joints-based rigs into current Houdini style bones-based rigs (read: You can animate right there in Houdini after conversion is done, which is pretty fast). The system also works with HD meshes exported from DAZ and converts the DAZ exported morphs into blendshapes (read: You can use both HD versions of DAZ characters and animate morphs).
Setting up materials depends on the render engine one wants to use and is, usually, a very subjective matter. I have a semi-automated skin material converter that works well enough for my non-artistic needs.
Unfortunately, so far public interest has been negligible and therefore does not justify a release version (which would require professional support, since I don't do "go-fix-it-yourself" open-source-style). My (limited) testing peers have confirmed that the system is doing what it's supposed to do, so maybe with the next Houdini release I can uppen the gates to something bigger :-)
Marc
Hi Marc,
I have been following your project in the Houdini forums. Amazing work like everything you do. I hope more people come out to support it, I for one would love to see it resurface. It's understandable considering the tremendous amount of work an HDA like this one takes to create. Keep up the great work!
David P
Hi, David,
I do not get notifications from this forum if someone replies to a comment - apologies for the late response.
As it seems, there is - aside from a very, very, VERY small number of people - next to zero interest in bridging DAZ products and Houdini in a smooth, back-and-fore way. As far as I can tell, most "bridges" offered by DAZ simply use the FBX workaround, which comes with too many issues for "productive" work (in terms of animateable morphs not correctly being transferred, HD information being lost and materials not getting converted/collected correctly). The solution I developed is reading DUF/DSF files and from that "re"creates geometry and materials natively inside Houdini. I considered that a cleaner, more future-safe solution, but, alas, I seem to be almost the only one.
Since neither DAZ nor Houdini is generating any income for me (I develop software in C++, CUDA, openGL etc. for the movie industry as one of my major routes of revenue), continuing work on the conversion system doesn't make any sense for me for the time being. I am not an artist, therefore, aside from solving the problems along the way, the project doesn't have any value for me now that I figured out most of the riddles :-)
Things may change if a sponsor pops up. However, mentioning the brand "DAZ" almost immediately shuts doors, as it seems.
Marc Albrecht