Mike_A
a good example being the new 'Create HDA' dialogue

Ah yes, I noticed this last night. Such a great improvement!

I hate when the workflow changes for no reason. But most of the time it has a good reason. APEX/animation context are big workflow changes, but now I'm convinced they are necessary. People who prefer obj-rigging would be considered insane.

kodra
I hate when the workflow changes for no reason. But most of the time it has a good reason. APEX/animation context are big workflow changes, but now I'm convinced they are necessary. People who prefer obj-rigging would be considered insane.

Yes. Totally agree. The new anime stuff is awesome.

I think LukeP raises a valid point and should be listened to. We should also remember that there may be many more like him out there that could buy the program if some of his concerns were addressed.

Some issues:
- When a new release comes I often lack examples of the new features. I had some problems first testing the new cloud features but then found the documentation which was written well and had examples.
- I tried the new white water which I was really excited about. I could only find the wave system trigger some white water. However it would have been much cooler if you could actually render it and it would look good.
- Then I wanted to try out some of the new Karma features. I started but quickly wished there was an example file to try out, maybe there is one but I could not find it right away.

A few things that I believe would help
- Those Flip Setups like Lava is super helpful. Add more examples that setup basic things for you. It would be great if they would also be ready to render as well. This I think would be the biggest way to help new starters. It was very useful when they let you download explosion example files that looked great. So basically have more examples that look good, and maybe have descriptive tooltips for them too. We could have tons of examples, how to do bubbles, lava, volcanoes, explosions, smoke, water spray etc. They are not that hard to do when you know how to do them, but they can be super tricky to master the first time.
- I wish there were more example materials available. H20 seems to add some from AMD so this is promising.
- The errors could have more solutions in them. I went into Solaris and it begun spamming an error. It complained something about camera missing in the render pass. I had a camera setup so I did not understand how to solve it. If it would describe more how to fix it this would be helpful.
- More tooltips and comments would be helpful. Especially if the tooltips were a bit more in depth, not just a line of text.
- More tutorials. Right now I often feel that if there is a new feature it will not be possible to use unless there is a tutorial or a course on it. The example files would solve this. I think it would be a good thing if SideFX has an example file for every major new feature and at least a quick tutorial how to use it.

I think that SideFX are working on all of the above points, I just wish it was a bit more. A bit more examples and better tooltips.

Cheers,
Andreas

LukeP

Dougie0047
@LukeP Can you define "artist friendly"? What does that actually mean specifically to you?
Cheers

Great question. Realizing that my opinion might not be the most popular one and that there are other perspectives on this.
To me ‘artist friendly’ means:
- Intuitive UI
- Streamlined workflows or workflows that don’t change with every release :-)
- Mid-level features and building blocks that cater to the creative needs of artists, allowing for easy exploration of ideas without things getting overly technical or data-centric, at the same time offering ability to ‘deep dive’ for advanced users. In my opinion as a simple example - a user working with the clouds should not have to worry about converting to vdb from particles, then from polygons, activating vdbs, writing volume wrangles etc.
- Welcoming environment for new artists without sacrificing the depth and flexibility required for complex projects
- Documentation that holistically explains the software, explains the nodes and parameters, doesn’t reference outdated workflows (e.g. Rendering section still explains Mantra and many other legacy concepts)
- Nodes and especially parameters that are named in a humanoid-friendly way and also explained in that way
- Advanced parameters that are hidden under advanced options, in general, hide complexity. Make it accessible, but not at the forefront
- Examples that are up-to-date and simple enough to grasp core concepts
- Software that hides or clearly deprecates things that maybe aren’t that usable anymore

@LukeP - Thanks for the clarification. I like your reply, since often when these discussions take place not enough time is spent on actually defining what is being talked about.
So, in response I'll reply with two possible definitions of my own and some additional thoughts. By the way, sorry for the length of this post. I could go on and on about this but stopped it to not bore you guys too much.

The first definition I can think of would be a bit like this: "Artist friendly" means - direct access to data. This aligns fairly well with what you mention above. For me, in everyday terms that would translate to things like working directly in the viewport, "touching" what I want to modify etc. And this is where UI/UX look and feel comes into play in a significant way. I guess you could label that the 'ergonomics' of the program...
One note on mid-level tools - whilst I think I get what you mean: that you can effectively get further in exploring a given idea with less detail focused node building etc., it is not always so easy to achieve a good balance between speed, usability and function. I find that mid level tools quickly start to feel like "ueber tools" where there are simply too many buttons, tabs, folders to click on and wade through for what the thing is supposed to do. When you then add to this the fact that everybody and their uncle wants things done slightly differently I think it must be very difficult for developers to hit the mark that satisfies the most possible amount of people.

The second definition I've had in mind would go something like this: "Artist friendly" means - maximum access to creative power. I realise that is not a sentence a great wordsmith would construct... But, perhaps one can think of it as 'the biggest possible playground with the most powerful tools.' That way the 'friendliness' does not lie so much in the ergonimics but rather in that there is nothing in principle that stands in your way of creating whatever you want. Again, in everyday terms, for me this means things like being able to construct a given output from the ground up, always having the option of fine control and to choose whether I want to use a conventional or a novel approach, and also always having the ability to explore a concept as far down the rabbit hole as I want to go (fyi - I am not saying that I am an expert at this, just that this is my mindset). Being able to do this is incredibly motivation for a lot of people, even if it is also hard. The cost of this would be that as you delve into the nitty gritty of the details you will have to learn a bunch of things that are not exactly what you look for but are nonetheless connected and necessary.

Personally, I think that in the second category SideFX/Houdini excels with it's procedural recipe approach. I mean, what houdini has to offer for a single dcc is kind of insane if you ask me. And, you may disagree, but I see signs that the good people at SideFX are trying to address the first category. I think you've already mentioned the new APEX/Animation context in that light, yes? Maybe the main question here is to do with time and resources, which are scarce, rather than a lack of responsiveness?

Again personally, I think I prefer the second category over the first if I must choose. I would of course like to both have my cake and eat it at the same time, but that is not always an option. I also think that various programs or even various aspects of the same program attracts different crowds. A bit like with children: the kids that like to role-play and make a game of things might prefer Playmobil whereas those kids that likes to build stuff wants Lego. I like to build things.

When it comes to ergonomics - the 'right' way to do character animation will be something very different to the 'right' way to do, say, procedural building generators, simply because the tasks are so very different. To marry those things within one applications must surely, for developers, sometimes feel like trying to fit a square peg through a round hole, and I honestly don't feel that sidefx is doing a terrible job at this. The software feels quite coherent to me. For the most part. Not to say they are perfect though. I could add a long wishlist of things I'd like to seen improved (for example I agree with your points on UI), but it would be a wishlist and not a list of demands.

Another thing that strikes me is the question of resources and where to allocate them. For every mid/higher level tool you make, like a new hda, you now have to not only support the atomic elements the tool consists of but also the tool as a whole. With this you sort of add ever more 'moving parts' to the system and therefore more things that can go wrong. This needs added support which means more resources spent on it. And I'm sure that Sidefx's resources whilst solid also have their limits, right? Do you as a company want to focus on rnd or on workflows? I guess you do both, but where exactly should the balancing point be? Difficult.

I guess what some of this boils down to is that the most possibly welcoming environment for newcomers by necessity sacrifices some of the depth and complexity and vice versa. If you bury the complexity in higher level tools it becomes more difficult to find it when you need it. And you might not even realise that it is there. Having said that I don't think that this a reason not to improve. Because, you know, how amazing would it be if I really could both have my cake and eat it?

Cheers,
Dag

Dougie0047
I mean, what houdini has to offer for a single dcc is kind of insane if you ask me.

Truly, but this single sentence can also be a fly on the pie for a complex piece of software balancing the challenges of pleasing accomplished users and attracting new users.

If Houdini's documentation was far more expansive in terms of examples and samples I think silly artist's might dig this beast way more. For example, when I'm self-learning about how to apply wnoise to a plane using vex and when accessing the docs I see technical niceties for Houdini lords but not a single example of HOW to make the plane react to a wnoise vex function for the uber-creative numskull (me).

I mean, it's just a plane... I'd like few lines that would illustrate HOW vex can apply this particular function without chasing the Youtube dragons.

For me it is really helpful when the documentation is good. For example the new clouds has good documentation and some great examples. THAT is what we need more of. But some of the new parts have very little documentation and no help. I have found that most of the problematic things in Houdini can be helped with documentation, but sometimes there just is no documentation to get and then you are doomed.

Quelltroi

Dougie0047

If Houdini's documentation was far more expansive in terms of examples and samples I think silly artist's might dig this beast way more. For example, when I'm self-learning about how to apply wnoise to a plane using vex and when accessing the docs I see technical niceties for Houdini lords but not a single example of HOW to make the plane react to a wnoise vex function for the uber-creative numskull (me).

Whilst I agree with you that the docs could often be better, making something far more expansive will definitely be a matter of how much resources are available for Sidefx to do this.

Also, I think that we should offload some of the responsibility of this on us the users. I know that many users do not even bother to look at the documents. Certainly not reading things in detail. It seems to me that people cry out for video clips and tutorials on everything long before they make a serious try at reading the docs for themselves first. I'm also guilty of this btw. For vex related things I look at the docs a lot, but I've noticed in myself that when it comes to nodes and setups, I often try to find things on Youtube first. Feels a bit like too much of an effort I guess... So, guilty as charged.

And I'm sure that as soon as a short clip is released you can bet that people will say: but we want a 2 hour long Masterclass! Those things take weeks to prepare.

Having said that, I do agree that having more examples of HOW to do things would be great and would indeed ease the transisition into houdini for many people. I have a feeling that with H20, they are in fact uploading more and more stuff to their content library than they used to do. Or am I wrong? Anyway, perhaps that is a step in the right direction?

if a 2 hour masterclass takes weeks to prepare, you should consider living a bit further away from black hole

osong
if a 2 hour masterclass takes weeks to prepare, you should consider living a bit further away from black hole

Having done hundreds of tutorials in the last 20+ years, I can attest that a good masterclass takes a lot of work, and yes, it may take weeks, especially if it involves a lot of research and high quality project files!

Dougie0047
And I'm sure that as soon as a short clip is released you can bet that people will say: but we want a 2 hour long Masterclass! Those things take weeks to prepare.

I'm a 'fairly' new customer to Houdini so I will use my absurdly limited perspective as an example... your points are straightforward but not entirely accurate and you prove this by stating very correctly, "Having said that, I do agree that having more examples of HOW to do things would be great and would indeed ease the transisition into houdini for many people."

The Houdini world is blessed with decent teachers and most of them quite expensive. I know... and so does my wallet. Learning Houdini at higher levels is NOT cheap in spite of a small multitude of free lovely masters providing Youtube tutorials.

I judge none and appreciate all.

However, the mindset of artists seeking this platform will naturally utilize the excellent system of documentation built into the interface of Houdini and then these artists will quickly realize they do NOT need a 2 hour long masterclass but a quick perspective on how to apply worley noise to a plane, man, for example...

Documentation is a science but it is also how people LEARN to adjust to a complex product.

I think this is a valuable topic. Right off the top of my head I’d say for new artists using Houdini, it would be helpful to have:

1. More presets. More of an elaboration on shelf tools.
2. Material libraries, with Karma presets.
3. A continual evolution of the help docs, with more examples.

There are, however, many resources out there already that fill these gaps. To me, it’s a long, engaging journey to learn more and more.

T

Is there a pose library that works with the new animation context?

kodra
Is there a pose library that works with the new animation context?

Not yet, still TBD.

This weekend, I rendered the Houdini 20 splash at 4K to use as a wallpaper, this scene would make a pretty decent benchmark!