@Staglaitor, if you dislike Houdini so much, why use it, why not use Maya or any other DCC Instead? 🤔

BTW, Houdini can be high level too, in a few clicks you can have a boat, on an infinite ocean, with flip fluids that are driven by the ocean and that are reacting to the boat, then you can dig-in and tweak the whole thing to the point that it will drive you crazy, that's the low level part!

But then again, you need to learn the tool before complaining, and a 3 weeks learning period is just an absurdly too short amount of time to even scratch the surface of what is arguably the most powerful DCC on the planet! 🤔

It's not true in the last instance.

It is true - Companies don't have infinite resources.

Developer's resources are spent on common (general) high-level tools

And they are also spent on 'low level' aspects of the program.
Again, since you made the following statement initially, of which I already responded in that context...

"The message needs to be conveyed to the developer that Houdini's tools need to be more high-level so that the artist remains an artist, rather than turning into a programmer."

That's a suggestion of directing more resource/time towards your preferences; Of which, IF it takes away too much time from lower level(compared to your suggestion/s)development; I'm not in favor of that direction.

Just for the record - the "Visual Programming" metaphor doesn't mean to imply that by using the software you're magically adorned with professional programming skills. Rather, it's suggesting that a Houdini node network can be run reliably over and over, with new input when needed, much like a regular computer program. The whole point of "Visual" is because it's possible to construct such a thing without writing code.

Staglaitor
This is false houdini-use

It's not. Houdini is a tool. There is no "false houdini use". Unless you somehow use Houdini to kill baby seagulls.

To avoid becoming a programmer you need high-level tools that I didn't find in houdini.

To avoid becoming a programmer you need high-level tools that I didn't find in houdini.

To avoid becoming a programmer you need high-level tools that I didn't find in houdini.

Copying & pasting your sentences 3 times doesn't make it more convincing.

Staglaitor
To us (artists). - No, that's not true. Houdini is not for us. Houdini is a programmer's world.

Yes and no.
You like the “plasticine” method - you easily sculpt your model using a certain set of actions, without thinking about the processes either in the plasticine or in your hands. This has the right to life. How to dig holes in the garden more conveniently with a shovel. And an excavator, which is a million times more powerful than you with a shovel, will not dig 1 hole faster than you. It's expensive and time-consuming. But if you need to dig the foundation of a building, you and your shovel are no good there. Because you need a drawing, measurements, mathematics, construction equipment. Formally, they dig in both cases, but there is a nuance... Houdini, or rather the procedural method itself - this is the same construction trust that allows you to solve a large number of construction problems if you know how to manage it. Just the ability to quickly wield a shovel cannot be compared with it. You will actually dig holes in your garden much faster with a shovel than you can activate a building trust.
The idea that everything can be expressed in numbers and allow the user to customize and insert these numbers into any parameters and customize any algorithmic interactions is a wide field of possibilities. In proceduralism, you easily scale a well-designed model to thousands of variations. While the manual method is absolutely dead here.
Any complex scene/model may consist of a huge number of submodels or copies. In proceduralism, you can easily play with any quantities, while with your hands - absolutely not.
Houdini's problem is an archaic, inconvenient interface with incredibly strange names, in which the user gets stuck like a fly in a cobweb and a huge number of bugs. But once you understand exactly how ordinary typical actions are done in Houdini, everything becomes much easier.
I myself constantly say here that all newcomers to Houdini start with modeling. But this is where Houdini can drive an unprepared user crazy with its bugs and oddities. You need to be very stubborn to learn how to figure out and work around bugs, and then all the advantages of Houdini will be revealed to you and you will fall in love with its capabilities. In a year, with a slight movement of your hand, you will make a project in which your entire garden will be dug up by a robot, and you will only correct something a little with your hands.

First of all I'd recommened people save their capitalized, bold and italicized letters for really important points.

Pacman972
You want to make organic modeling in Houdini ? It will certainly take 5 or more time than doing it in MAX. I've experienced/compared it.
Even if SideFX is providing Labs Tools in every version to ease artist's work, there is almost no tool dedicated to organic (destructive) modeling.
That's why you can find tiers plug in like MODELER developped by Alexey Vanzhula (https://alexeyvanzhula.gumroad.com). This plugin is the proof of the weakness of Houdini regarding to organic modeling and speed to achieve it.

Note that this is the opposite of what OP wants. OP wants a procedural asset, to a point where they called a tutorial "bad Houdini-use" because it's not procedural enough.

That's what people mean when they say SideFX can't please everyone.

This thread is ridiculous IMHO, you don't need any advanced coding skills to use Houdini properly, all you need is a good understanding of high-school level math, if you want a DCC (if you can call it that) that is a programmers' nightmare, just have a look at Blender, Blender users have literally created several thousand plugins in the hopes of making Blender better, and yet, it still sucks! 🤦‍♂️

tamte
the point of creating anything in Houdini is not that it should always be fully procedural and allow you to change any parameter without breaking anything downstream

This thought is not the truth in the last instance

"The fire in the soul" of many users is precisely built on the fact that they can make a living model - that is, a model with a tarnsfarsational reserve (potential)

tamte
and I'm sorry some individual "badly cheated" you into thinking Houdini is this magic one button solution for any imaginable problem, but taking it out on everyone in the forums just comes off as bitter

NO - "..but taking it out on everyone in the forums just comes off as bitter" - i feel absolutely good - just peaceful discussion

GCharb
But then again, you need to learn the tool before complaining, and a 3 weeks learning period is just an absurdly too short amount of time to even scratch the surface of what is arguably the most powerful DCC on the planet! 🤔

Yeah, I'd like to think so, too, but show an example of transformable model that has high visual aesthetics....
It feels like instead of developing visual aesthetics, you're spending your time on programming

BabaJ
...

Aa i said before - if you not a programmer - show us your high aesthetic works

HGaal
You need to be very stubborn to learn how to figure out and work around bugs, and then all the advantages of Houdini will be revealed to you and you will fall in love with its capabilities. In a year, with a slight movement of your hand, you will make a project in which your entire garden will be dug up by a robot, and you will only correct something a little with your hands.

The thing is, I don't want a whole garden of ordinary flowers - with small differences.
I want to make just one flower with huge transformative potential to achieve a visual uber aesthetic.

In this thread:

- "transformative"
- "aesthetics"
- blah blah

Staglaitor

GCharb
But then again, you need to learn the tool before complaining, and a 3 weeks learning period is just an absurdly too short amount of time to even scratch the surface of what is arguably the most powerful DCC on the planet! 🤔

Yeah, I'd like to think so, too, but show an example of transformable model that has high visual aesthetics....
It feels like instead of developing visual aesthetics, you're spending your time on programming

Use both hemisphere of your brain dude, it's not because you call yourself an artist that you can't solve a rule of three, or add two vectors. Just be curious and "open" to learn new stuffs and test your limit.

If you are looking for a tool to build procedural/parametric system (large transform reserve ???) with no mental effort then you might probably want to wait an other 2 decades or so.

this discussion seems pointless

- you are complaining that you are not able to create an ubermodel which you can configure using parameters into wide range of variants

- and for some reason you are thinking that Houdini is supposed to allow you to do that without any problem solving skills on your end, but that's not how it works and I don't know who told you it is

- I suggested to bring up examples of procedural systems (I mentioned a bunch) that have workflows and tools more aligned to your way of thinking and therefore allow you personally to create such parametric ubermodel easier than Houdini, since then it can be constructive discussion

But your lack of examples suggests that you haven't tried any of those and therefore it seems like you are just generally complaining that to create such parametric model you have to have some problem solving skills and experience, which yes, its unfortunately true, not just within Houdini (unless of course someone already did some work that fits your specific needs in form of a node, tool or addon, then you can just use it or build on top of it)

nobody here would say you have to be a programmer to do that, but if in your mind anyone that uses a bit of math to create relationships between the parameters and inner logic is automatically a programmer then probably for you majority of Houdini users will always be programmers

I feel like people here are way to shy to tell the truth.

I'll say it out aloud: if your end goal is to make highly procedural assets that fit your specific needs, you need to learn programming. That's it.

Complaining this is like complaining that water is wet. Programming is not the opposite of art. Programming isn't typing code into a text box. Programming is literally just telling computer what to do and automating repetitve tasks. If that's what you want then yes you want to learn programming in one form or another.

Don’t know if this is on topic, but if SideFX is smart they will incorporate some kind of AI asap, and programming or scripting will be replaced by writing easy to understand prompts, which would make Houdini much more accessible for mere mortals that can’t write code.

When you think about it, programming is just an easier to comprehend layer between human language and binary machine code. Writing prompts interpreted by AI would just replace that layer, which would make the whole process sooo much easier and more fun.

Besides that, looking at the new discoveries already made with AI and new possibilities opened up after such a short time, I think it will become a much more powerful tool than programming or scripting.

But that’s just me ;-)

toonafish
Don’t know if this is on topic, but if SideFX is smart they will incorporate some kind of AI asap, and programming or scripting will be replaced by writing easy to understand prompts, which would make Houdini much more accessible for mere mortals that can’t write code.

When you think about it, programming is just an easier to comprehend layer between human language and binary machine code. Writing prompts interpreted by AI would just replace that layer, which would make the whole process sooo much easier and more fun.

Besides that, looking at the new discoveries already made with AI and new possibilities opened up after such a short time, I think it will become a much more powerful tool than programming or scripting.

But that’s just me ;-)

You have to be realistic though and go with what's possible at this point
SideFX is smart and they will implement any technology they deem useful and reliable

That doesn't mean they will jump on the bandwagon of prompt controlled tools if that means they have a chance to produce tons of useless hallucinations and quickly become frustrating

So give it time, it's easy to speculate, but the industry is not there yet

If anything I'd guess NVIDIA Omniverse will be one of the first DCCs utilizing such tools as they have the most research and resources in that area specifically related to CG

Hey, of course if it’s useless don’t implement it, but to me, and I suspect a lot of other peeps, anything that works even at just a basic level would be useful at the moment. And Houdini’s procedural pipeline would be the best environment for it I guess.

I couldn’t write code if my life depended on it so personally I’d be happy if I could just write a prompt that would copy some geo to every Nth point and slowly scale it up, or create a basic rig for a character…for instance.

And I’m no expert in this field, but especially when you consider AI would need time to learn from all the code used in Houdini that is accessible before it would become really useful, it would take some time before it could replace programming completely, but it would really help a lot in the meantime.

Maybe someone should drop an RFE before they forget :-)

toonafish
Maybe someone should drop an RFE before they forget :-)

Trust me SideFX knows that lol. Everyone in tech is aware of LLM-based AI now. I mean literally everyone.

These days if you release an AI product and it becomes semi-popular, lawsuits will come after. It's inevitable in our current society, like gravity.

ALL of your favorite AI products are facing multiple lawsuits, guaranteed:

https://www.theverge.com/2023/12/4/23988403/getty-lawsuit-stability-ai-copyright-infringement [www.theverge.com]
https://githubcopilotlitigation.com/case-updates.html [githubcopilotlitigation.com]
https://www.bbc.com/news/technology-67826601 [www.bbc.com]

SideFX isn't a megacorporation with an army of lawyers, like Microsoft. It's not a 1-year-old startup that they can just shutdown if they fucked up, either. They have to wait until these cases settle down.

toonafish
Don’t know if this is on topic, but if SideFX is smart they will incorporate some kind of AI asap, and programming or scripting will be replaced by writing easy to understand prompts, which would make Houdini much more accessible for mere mortals that can’t write code.

That's pretty sad to hear. I wouldn't want SideFX to waste their time and go down this path of development using AI as an assistant inside Houdini.

Over the last 10 years there are already enough tutorials: articles, courses, blogs telling you that it's not that hard, you just need to spend some time to learn the basics, which shows that it's already quite accessible. After all, there are even tools that allow you to beautifully package snippets of code so that it can be reused with.
And I would not call it all programming as it is used now in the usual sense of the word. Real programming is much more complex than what is used in VEX, HScript and what is required of artists.

What is more interesting in my opinion is the possibility to apply AI invisibly to the user where Houdini may not be fast enough at the moment: simulation, rendering, data compression algorithms, improving DashBox's internal search engine, etc. As it was shown in the last H20 presentation that we can use ML for skinning and terrains.