What holds back artists, decision-makers, tools from using Houdini?
Comfort? Complexity? Price? No Adoption runway? Stubbornness?

No wrong answers here.

If you are uncomfortable with answering publically, you can email me directly: shaned@sidefx.com.

Thank community!
Cheers

For me it was always about having the time to learn the necessary Houdini skills to be able to do meaningful work, and for that you need time and access to learning resources.
Over the past few years the amount of learning resources has grown massively, and many are excellent, and as a result of the pandemic the time side of equation was also made more possible. (probably always had the time, but events made you prioritise things differently)

I don’t think this is any different to any other software or skill, it's just that Houdini's learning curve is somewhat steeper than most. Once you get past that initial incline, it really pays off, and you can do stuff in Houdini that's either a massive struggle or just not possible in other software. And it's an absolute joy to use.

One of my biggest challenges on the journey was knowing what to learn when - a curriculum that best gets you to where you need to be as efficiently as possible. For me this was helped by and interview with someone who urged new users to try and forget DOPs and focus on SOPs, get really comfortable with manipulating points, really get this to become instinctive, then move on to the cool stuff with a solid understanding of whats happeing.

TL : DR - Curated learning paths.

shaned
What holds back artists, decision-makers, tools from using Houdini?
Comfort? Complexity? Price? No Adoption runway? Stubbornness?

No wrong answers here.

If you are uncomfortable with answering publically, you can email me directly: shaned@sidefx.com.

Thank community!
Cheers

I'm a long-time houdini user (13 years), and 3 years ago I started a small studio. Right now I still have the option of continuing indie, using indie licenses. I'm still figuring out how to grow and make the transition to workstation or studio licenses, since there is a huge difference in price. I won't say It isn't worth the price, since houdini is amazing, I love to work with it and it puts some studio-level tools in the hands of regular users, but as I small studio owner, working from brazil (with a huge exchange rate difference), making the transition seems almost impossible right now. I've talked to sales and there is no easy way, even transitioning directly to the maintenance cost (since I already have my own license?).

It feels like HIndie is too cheap and FX licenses are too expensive. Don't know how It would work, from the business perspective, to lower the studio licenses, but indie is pretty cheap compared to c4d and compared with a regular freelancer rate, and on the other hand, studio seems too pricy compared with a small studio's income, at least here in brazil

I'm a freelance motion designer weighing up whether to invest a large chunk of time to learn Houdini.

Pros
1. Pricing for an indie licence is super cheap. Even if I just end up using Houdini only for particle simulations its still cheaper than an XParticles subscription.


Cons
1. Houdini has a reputation for being very complex to learn. Am I setting myself up for a lot of frustration?
2. Houdini seems like a very specialised tool but what I need is more of a generalised 3D application (like Cinema 4D which I currently use). Sure I could use both but I'd be more inclined to commit to an application if I felt like I could achieve everything I needed to in that one application.
3. The UI seems a bit clunky and dated. All those bevels and gradients and the checkboxes are nasty.

guilhermecasagrandi
.Don't know how It would work, from the business perspective, to lower the studio licenses, but indie is pretty cheap compared to c4d and compared with a regular freelancer rate, and on the other hand, studio seems too pricy compared with a small studio's income, at least here in brazil

You can rent both Houdini Core and FX for as short as 90, 60 or even 30 days.

This should be affordable for any studio that is actually working on projects. Of course it would be great to have a perpetual license but these options should give any business the opportunity to use Houdini if you want to use it commercially.

For me it boils down to UI. I'm finding myself using a lot more Maya these days. Modeling, grooming, animation... all things Houdini can do extremely well, but all things that feel smoother in Maya. There's a lot of clicking and typing and wiring nodes in Houdini, plus so many non-essential features are front and center in every part of the interface. It's fine in small bites, but when working for hours all the little speed bumps add up and it becomes tiring.

For example, the new groom tools in Houdini are fantastic (with the exception of a few bugs), but Interactive Groom in Maya feels faster and more pleasant so it wins out for me. I would love to have things like live simulation brushes and repositioning hairs in Maya, but during the course of a project all those hundreds and thousands of small extra actions that are required in Houdini end up feeling like work.

I'd like to see a refreshed, "artist-friendly" UI in Houdini, including packaged-up tools that highlight relevant tasks for easy access. Designing a great UI isn't easy, but I think with some love Houdini could get there without too much trouble.

Hi. Although Maya does have Xgen etc, I gave up on that years ago when it stopped being developed. Its good until you hit a dead end that you cannot get yourself out of. When I look at things in Houdini now I would say, direct modelling. I know it's not really a Houdini thing. But so much of when you rig and animate relies on being able to model fixes to things. FACS facial setups often need quick modelling fixes even if you have made the shapes in Zbrush to begin with etc.

Maya's sculpting tools that were taken from Mudbox are super useful. The other thing Maya has that is great is the modelling toolbox. I know they bought that from someone else and added it in. But it is very easy to just use this part of Maya for modelling and kind of ignore other parts (for beginners).

The biggest easy win I would say for SideFX is just copy Maya's viewport navigation. Spinning on objects in Maya just works. With Houdini I have played with the prefs but I still get my camera flying off sometimes.

The other easy win is "industry standard" Maya keyboard shortcuts for move rotate and scale. Lots of other 3D programs have to option to use the Maya standard ways of navigation and Xforms.

Im saying this as I know Sidefx are looking at animation with KineFX. So at some point modelling will need to improve.


My 2C.

Best

Quick backstory for relevancy - I'm mostly a 3D motion designer that moved over from C4D about 5 years ago. I'm using Houdini probably in less common VFX ways than most, but I believe there is a huge potential to grab people from the C4D market if done right.

1. A lot of us motion designers are freelancers. The indie pricing is great, but it makes it so freaking hard to work with studios and because of that, studios and freelancers often stick to C4D despite wanting to move to Houdini. I understand the reasoning behind hiplc, but studios want full hip files. I'm now on the hiring side and when we have to bring in a freelancer and rent a node for a month, get them squared away with licensing, etc. etc. it can be a couple days of downtime for them. If they could work in Indie and hand over full hips, that'd save us having to rent seats and deal with this downtime.

2. The indie pricing for freelancers is great but it caps out way to early. Here in the states it can be relatively easy to break $100K gross and even easier to work with clients that make over $1M or whatever the limit is. IMO the cap should be $250K-$500k before they need to go up to a non-indie license and no cap with whomever they work with.

3. Hard to shake stigma of being a difficult program. There are so many reasons why motion designers should be using Houdini over C4D or other software. However, despite easier and easier tools being introduced every year, people seem to think Houdini is just for VFX and extremely hard to learn. There really needs to be some effort towards having motion design ambassadors (like Maxon does) for Houdini and showing the power, stability and workflow tools that make it a much better choice.

I think it comes down to simplifying things for art directors/designers like myself. I know c4d quite well, and I cannot get what I want there anymore cause of limited performance. On the other hand H easily beats any software I tried in that regard. Also the way it unifies everything from particles to liquids to cloth is simply amazing.

The reason I cannot switch from c4d is only one: its too complex.

While I can understand SOP level operations, nodes and vellum, I cannot understand DOPs, vex, vops etc.

If you keep developing node tools in SOP level exactly as vellum progresses so far, it will be amazing as it becomes really accessible for people like me and also easy to understand. I think simplification is super important. The majority of 3d artists, do not write code and prefer simple tools to help in quick setups/art direction.

C4d has excellent field system that is unified falloffs between deformers, forces, dynamics, anything is actually connected there. If every node had its own falloff (as it has its own attributes) eg. a bend node where you can only bend where the falloff is, this would be huge. Falloffs and decays and all those simple but essential things are difficult in Houdini. I love pop axis node cause there is a falloff that controls where the force is applied.
Do that in every node, add falloffs and probably whole c4d community will migrate to H Or a velocity field: a single node where u plug whatever object u want and creates a field on its normals or away etc etc. Xparticles in c4d makes those operations 2 clicks and in houdini this is rocket science.

Long story short, H is used by many motion design studios, but lacks the motion design toolset. Easy nodes, easy to apply falloffs like MOPs+, unified falloff system.

The last few months that I am fiddling with H I feel that every update is on the right direction as it keeps adding easy to use nodes and does not require to jump in and out of networks or write code.

To me I still find myself running back to C4D for ease-of-use regarding setting up lights and placing objects.
Also, Redshift material packs are a core part of my lookdev workflow. While I know there are ways to import material libraries from GSG (for example) into Houdini, it seems like a hacky and unintuitive workflow for now.

I know Solaris is trying to solve some of Houdini's limitations in those ways, but I haven't spent the time to learn it.
Other than that, I love Houdini and would prefer to just stay in it if possible.

I know it's probably a minor opinion, I'd like to see in Houdini is a real-time renderer. I don't think there is any real-time renderer that is well-integrated with Houdini. Now the only choice is to export everything to Blender Eevee/Unity/Unreal. To me Eevee is the only reason I'd consider using Blender, really (besides supporting open source).

Ideally, a real-time counterpart of Karma would be a godsend. Just like what Eevee is to Cycles. I'm not talking about Karma XPU, which is actually more like Cycles than Eevee.

Yeah, the viewport supports custom OpenGL shaders... but it's extremely hard to use. Like, literally the worst on the market. I've hand-written quite a lot shaders in Unity before it introduced shader graph. And it was a far more pleasing experience than Houdini's viewport.

Most people probably don't care about real-time renderer... a lot would say even Redshift looks fake to them. But again it's just my personal opinion.

On the Eevee topic. Quite a lot of 3D apps are moving away from OpenGL to Vulcan, so Im hoping all viewports will improve over the next few years. Best

motionpunk
I think it comes down to simplifying things for art directors/designers like myself. I know c4d quite well, and I cannot get what I want there anymore cause of limited performance. On the other hand H easily beats any software I tried in that regard. Also the way it unifies everything from particles to liquids to cloth is simply amazing.

The reason I cannot switch from c4d is only one: its too complex.

While I can understand SOP level operations, nodes and vellum, I cannot understand DOPs, vex, vops etc.

I don't think simplification would make our life easier. Being an art director|communication designer as well my clients do expect me to deliver sophistication, not simplification.

I'm asking the same from the instruments/applications I use — Houdini does deliver it, at a cost. What helped me at the beginning was to understand Houdini as a suite of applications, dedicated operators, so I can whittle down the steep learning curve into bite size studies. I can admittedly stomach a bigger chunk of SOPS than VEX or DOPS until today, given my deficits in programming languages and physics. Houdini is a no-nonsense tool. Just add technical skills and artistic abilities to get the work done.

Over the years SideFX has collected and published a remarkable amount of most useful learning materials. I no longer see Houdini as a dedicated FX tool but a fully-fledged CG application that does all I need, provided I can wrap my brain around it. The last versions have seen a lot of love for improving animation and rigging. I especially do embrace Solaris.

We will certainly see more improvements outside of Houdini's FX skills along the way. However, please leave simplification to other software vendors.

A modern, flat UI, like Nuke Indie is on top of my wish list.

Cerimon
Being an art director|communication designer as well my clients do expect me to deliver sophistication, not simplification.

I thought it's a no brainer when it comes to design to have simple tools in order to deliver sophisticated designs. An AD is not a TD. The best ADs around can do miracles with low level tools.

I run a patreon page with 500-600 students and no one wants to see an H tut unless I put it side by side with c4d and explain everything only with nodes. I was also afraid to touch H cause of watching tutorials online that include VEX/convoluted explanations of simple tasks. And I am quite technical person myself.

I disagree with some of the users regarding the UI, I like colorful icons and I'm not a big fan of the latest "flat UI" trend I see everywhere, in which everything is shades of gray or two colors.

IMO the biggest Houdini's issue is UX, which encompasses many aspects of the 3d pipeline. Buggy viewport interaction (persp. camera goes crazy, clipping, switching from persp to front/etc sometimes acts weird.
Incomplete/unrefined modeling toolkit and frustrating behavior in which, for example, some components are selected, call an operation (transformation, extrude, etc) and it acts as if the previous SOP's selection is active; having to press twice for a transformation (E,R,T) in order to update the current selection; many more.
Character rigging/animation has been a big weak point and still is with KineFX in its infancy, although I like KineFX's potential.
Mantra is shelved and Karma is no Arnold/Vray/etc killer, not yet at least.

Other than these, Houdini's great. Being technically difficult is not an issue since that's a product of reality being complex. If it were simpler, it would have been less powerful, if that needs to be said.
There are some things which are technically difficult only in Houdini, but this is relatively easy to solve with more high-level tools / refinement of existing ones.

a.The ugly interface compared to C4D.
b.Houdini is not a general 3d package,especially in product promo area.Using houdini to handle the cad model and build scene is a nightmare.
c.The lacking of modeling toolset.

Please, please don't make Houdini's node icons flat. At least not monochromatically flat like we've seen in other apps. Single-colored icons work in some apps because they don't have that many:





It might look ok-ish, but what if Blender has 500+ modifiers?



Would you prefer these 3 icons to be the same color?

Realtime fully interactive viewport rendering (EEVEE like... please...)

Also I strongly disagree with all the "if Houdini is simpler, it will be less powerful". Like... really? In what way? Do you have a different definition of "powerful"?

Let's see the progression from H17 to H19.

1. Vellum is far simpler to use than FEM. Does it make Houdini less powerful or more?

2. Realistic Shoulder VOP is simpler than making the clavicle rig by ourselves. Does it make KineFX less powerful?

3. RBD Material Fracture is simpler than stacking several Voronoid and Boolean fractures on top of each other. Does it make RBD fracture less powerful?

Having a higher-level feature doesn't make an app less powerful as long as you don't hide the lower-level features behind it.

I started with H17, and really like the direction SideFX have taken with Houdini since then. If they continue down that path of serving up accessible tools, while minimising the need for VEX for common tasks, Houdini will be in a great place (well, greater than it currently is).

In my opinion, the main issue is perception. Houdini has a massive image problem, and aside from people believing that VEX and VOPs are mandatory, the shorthand of SOPs, DOPs, COPs, LOPs, CHOPs, etc. doesn't help. It's an impenetrable barrier of jargon to new users, and almost immediately turns them away.

While I personally have no issue with it, I have seen first-hand many, many new users' eyes just glaze over when they encounter these terms, even people with plenty of 3D experience in other apps. I'm not really sure how to solve this, sorry.

Back to VEX/VOPs, the issue isn't with what they offer, because they are incredibly liberating, it's again - perception. There are many YouTube videos out there where instructors will jump to VEX or VOPs for the simplest things, when no VEX or VOPs are needed (Entagma is notorious for this, but they're certainly not alone). This, again, doesn't help with the perception, as people see simple tasks needing "complex" solutions, and assume that that's just the Houdini way.

Moving on, I don't really have issues with the UI, and think it's quite flexible and powerful, but I can definitely see it being spruced up a bit to not look quite as dated. This is extremely low priority for me, but again, for front-facing stuff, it's quite important.

What I think would really help is a series of tutorials aimed at transitioning. Showing how common tasks and effects done in C4D or Maya can be done as easily in Houdini, without needing VEX, and then perhaps, take it to the next level with some VEX/VOPs, if applicable.