Hi!


First of all, I want to show my appreciation for the amazing tools the developers at SideFX has created. And I hope this feedback is seen as friendly input from my humble perspective.


So. I'm having some mixed feelings about Solaris, and would like to start the discussion so that I might be proven wrong - or maybe SideFX could find the feedback helpful?

I'm feeling that Houdini is close to being a wonderful and magical software that our company one day could switch over to, but there are some tools and solutions that's just making it way too unintuitive at this stage. Among the quirky tools for modeling, there's the “scene management” aspect of it.

For example; in 3Dsmax, Unity, Unreal, and Blender (which I have quite some experience in) the “top level” of the scene management consists of a hierarchy list, folders, layers, etc. And I think that really works well - compared to a node-based solution. The ability to collect lights, models, dummies/helpers and hide/freeze, etc. entire folders/layers, are tools I find much more intuitive than managing your assets with nodes. There's never been a case where I thought “oh this was convenient with nodes” at the top level.

And then there's the scene manipulation and dressing tools. The ability to drag and drop an asset while holding a button, and make a copy or an instance, as well as being able to select and manipulate with the same tool is something I really miss in Houdini.

So, here are my thoughts about Solaris. I feel that some of the issues I've dressed above are somewhat trying to be solved in LOPS, but the direction it's heading is making it way more complicated than it could be. The tools in LOPS are wonderful, being able to have an IPR in the viewport, set dressing tools that can quickly move many assets in a pile, aligning lights and highlights, light mixer, etc.

But wouldn't it be so much more convenient to have these tools in the “good old scene view”? It feels like I'm wasting so much time having to import my assets to LOPS, connect lights and materials, it's like I have to do double work. I can't think of any scenario where the LOPS workflow could ever benefit me or my studio's pipeline - compared to how much more time it takes to manage.

Curious to hear other's thoughts about this as well.

I think it depends on the scale, amount of the shots and time you have to finished them.
Working in object/rop context on multiple shots (the old way) with takes/rops and stylesheets gets complicated very quickly.

USD is not easy to understand, but it is very powerful.

Imagine it the other way around. Object context can go away to be replaced by stage (LOP) context.

(As object context doesn't have interactive history of edits, collaborative layers, and other features that LOP context has - there are only takes and Mantra's stylesheets so I see object context as subset of features that LOP context can do.)

I would submit RFEs for workflows that you feel need improving (too many clicks for material assignment for example). It’s important those bread and butter operations are just a key press for sure. The multiple contexts make some of the seemingly simple things a little more work to solve in Houdini compared to other tools but I’m sure it will get there.

antc
I would submit RFEs for workflows that you feel need improving (too many clicks for material assignment for example). It’s important those bread and butter operations are just a key press for sure. The multiple contexts make some of the seemingly simple things a little more work to solve in Houdini compared to other tools but I’m sure it will get there.

Thanks for your input! Where could I submit that?

pezetko
I think it depends on the scale, amount of the shots and time you have to finished them.
Working in object/rop context on multiple shots (the old way) with takes/rops and stylesheets gets complicated very quickly.

USD is not easy to understand, but it is very powerful.

Imagine it the other way around. Object context can go away to be replaced by stage (LOP) context.

(As object context doesn't have interactive history of edits, collaborative layers, and other features that LOP context has - there are only takes and Mantra's stylesheets so I see object context as subset of features that LOP context can do.)

Thanks for your input!

Yeah I get that LOPS is great for what it's supposed to do, it's just that I'd love to see a re-design of the more every day “production tools and tasks” before LOPS, in order to be able to use Houdini more easily in production

johanbogeberg
Thanks for your input! Where could I submit that?

https://www.sidefx.com/forum/topic/70943/ [www.sidefx.com]

Yes please! We are always looking for feedback and suggestions, from the mundane to the grand!

We are certainly aware of many shortfalls in /obj, and in /stage, and in trying to combine or unify or simplify these two (currently very separate) worlds. We have lots of plans and ideas of our own, but we are always looking for fresh perspectives and new ideas, and help in filling in our blind spots.

mtucker
We have lots of plans and ideas of our own, but we are always looking for fresh perspectives and new ideas, and help in filling in our blind spots.

This is great to hear. I have a couple thoughts that I'll put into an RFE. SideFX is the best!

johanbogeberg
But wouldn't it be so much more convenient to have these tools in the “good old scene view”?

Hi Johan,

I've heard similar arguments in recent months as we introduce Solaris to traditionally direct-manipulating, ‘state-ful’, destructive workflows and it's a valid concern as many productive disciplines within a VFX pipeline are built on that model. (Not certain if you're approaching from a VFX perspective?)

The power of procedural workflows is the development of a recipe for reapplying that recipe with different inputs, at the cost of dealing with the cognitive and physical overhead of designing and implementing that recipe. It's this feature that will let you create systems that can work against multiple environments at the same time, with dynamic results and at almost-unbounded scales. That recipe can be tweaked and maintained and improved with time, leading to much rewarding automation and possibly increase in quality.

You might want to look at the Stage Manager LOP as an example of a node which allows you to treat the stage as a (rather plastic) scene. This is an example of a node where SideFX has recognized the need for stage which steps out of the narrow procedural functionality we generally see in other LOPs.

What we've seen in the past is SideFX implementing workflows that are first and foremost procedural in nature. However we've also seen SideFX recognize the “artisanal” nature of direct modeling and thus greatly increasing tooling and frameworks to try to break bread with procedural and direct workflows. Some of the direct-manipulation workflows take a little time to appear in the toolset, so I'd say continue to give SideFX constructive feedback on specific workflows and see how they respond over time.

Hope this helps,
Jason