my wish list is they put more tools to speed up (game asset) modeling. And probably a tighter coupling with their newly adopted stepbrother Unreal

These days you need to speed up content creation process for games.

tsiwt
Obviously SideFX needs to revamp its U.I/U.X I have to jump between so many windows just to render. This is absurd. Material assignment and the material interface is such a drag visually. Parameter windows needs better organization not everything on the same hierarchy. Just making some text bolder than other to differentiate the operations can go a long way. What if you hover over a node and there is another icon that if you click it gives you an example of how to use that node in a small window.

These are things at the top of my head, what if improving houdini was my job? hummm.

I really like Houdini UI.
It has got one (maybe the best)of the best node system connection/organization, and the interface is fully customizable.

I think you just need to spend a little bit more of time with it to learn it better : )

Houdini's UI is mostly fine. It's extremely functional and quite versatile. They can tweak, but no overhaul is needed imo.

The only thing I'd really want to see is the ability to easily dock panels that span across the full width of the UI. Since the current paradigm is to split existing panels first and then assign the type, it makes it impossible to create a new panel that spans across without rebuilding an entirely new set with that layout in mind.

This is what I'm referring to:

As we are discussing UI here maybe it would be better to create separate topic and pitch ideas there? Here it's hard to follow and the topic is not meant to discuss UI!

SIgor420
As we are discussing UI here maybe it would be better to create separate topic and pitch ideas there? Here it's hard to follow and the topic is not meant to discuss UI!

Don't know. We've quite exhausted whatever could be wished for in an H19 release (and way beyond). It's Sidefx' turn now. I guess we'll see a sneak peak in the next one or two weeks anyways...

eikonoklastes
Houdini's UI is mostly fine. It's extremely functional and quite versatile. They can tweak, but no overhaul is needed imo.

The only thing I'd really want to see is the ability to easily dock panels that span across the full width of the UI. Since the current paradigm is to split existing panels first and then assign the type, it makes it impossible to create a new panel that spans across without rebuilding an entirely new set with that layout in mind.

This is what I'm referring to:

this took me under 10 seconds to do.
and I can save this as a custom desktop aswell. A bit of effort will take one a long way

sekow
this took me under 10 seconds to do...A bit of effort will take one a long way

Hot damn! That's a pretty remarkable bit of condescension on a professional forum. You seem really comfortable behind that veil of online anonymity. But let's move on from that.

The other day, on the Houdini Discord, someone asked if there was a way to copy a node and have a hotkey to immediately paste it as an Object Merge node with the correct reference.

The current method is to copy the node, drop down an Object Merge, and then paste the path as text in the Object parameter - something that takes about 2-3 seconds to execute. However, a SideFX staff member acknowledged that the proposal would make a good RFE and encouraged us to submit it as such.

And I have to say, I'm really glad that SideFX employ people with that sort of mindset — where they are keen to remove as many barriers as possible, regardless of how tiny that barrier may be. I also really hope they don't employ people with mindsets like yours, that seem straight out of the Adobe or Autodesk playbook, where having the user employ workarounds to overcome the app's shortcomings is an acceptable state of affairs for them.

Actually, this links us nicely back on topic — hopefully we see a lot of little enhancements in H19 that only took 10 seconds in H18, so that a bit of effort will take one an even longer way.

eikonoklastes

sekow
The current method is to copy the node, drop down an Object Merge, and then paste the path as text in the Object parameter - something that takes about 2-3 seconds to execute. However, a SideFX staff member acknowledged that the proposal would make a good RFE and encouraged us to submit it as such.

The correct way for doing this is writing a few lines of Python

Tongue-in-cheek factor aside, investing some time in learning how to quickly make these adhoc automation tools can make a world of difference for anyone's workflow.

copy: shelftool/hotkeyed script to copy selected node path string to regular clipboard

paste: shelftool/hotkeyed script that checks is clipboard content is a path and resolves to a node, check if in SOP context, create a merge node, set merge path to clipboard contents. (and optionally fancy stuff like auto rename the merge node to reflect the source node's name, toggle rel/abs using shift-clicks etc)

Once you're up to speed on these things it will take about 10 minutes to make.. first tool will involve crying in a corner for a few hours but after that it will go faster and faster.


edit: not to say useful general tools shouldn't be added, more a matter of not being depended on SideFX for these little things. They are busy enough with core stuff I hope

Jonathan de Blok
Once you're up to speed on these things it will take about 10 minutes to make.. first tool will involve crying in a corner for a few hours but after that it will go faster and faster.

I think we can all agree that this applies to anything Houdini, in general.

I also think that something is getting lost in translation here. I'm pretty sure most Houdini users are aware that there are multiple ways to get things done, and a lot of them can be automated to an absurd level. That's why we all love Houdini in the first place.

I'm certainly not pushing against the "you can hack that" ability of Houdini. That's a huge part of what makes it great.

With their last few releases, the Labs initiative, and the recently vastly expanded learning resources, SideFX has shown a very keen intent on making their software more artist-friendly and more accessible. This does not entail taking away low-level control from those that desire it, but it enables higher-level control for those who prefer that route. I can list a string of newly introduced SOPs that do just that, but it'd be a long list.

I am very comfortable with VEX myself, I started off my computing journey as a programmer before shifting to CG. I just prefer using available tools and direct UI, and I'm glad that SideFX keeps bring more and more functionality into the app, that requires less tinkering with its guts to get the same result.

Ultimately, I want Houdini to be an easier app to work with, shed its "you need a PhD to work it" perception, attract more users, grow in popularity, and eventually be The One App That Does it All. Is that too much to ask? Maybe with H20...

eikonoklastes
Ultimately, I want Houdini to be an easier app to work with, shed its "you need a PhD to work it" perception, attract more users, grow in popularity, and eventually be The One App That Does it All. Is that too much to ask? Maybe with H20...

+100 for that one!!!

complicated things are much easier in Houdini than elsewhere. working with attributes (as an example) directly in other DCCs is gonna change your perception fast

my main "travolta lost.gif" was about working with USD, but that's not designed with me in mind anyways

osong
complicated things are much easier in Houdini than elsewhere. working with attributes (as an example) directly in other DCCs is gonna change your perception fast

EXACTLY! All one has to do is look at what the Blender Foundation is doing over there - they added a Geometry Spreadsheet, they are changing Blender to be 'all nodes'.

Other DCC's as well are gonna need to adopt something similar as USD spreads its wings and MaterialX hatches.

Houdini is way ahead of the curve.

Jonathan de Blok

eikonoklastes

sekow
The current method is to copy the node, drop down an Object Merge, and then paste the path as text in the Object parameter - something that takes about 2-3 seconds to execute. However, a SideFX staff member acknowledged that the proposal would make a good RFE and encouraged us to submit it as such.

The correct way for doing this is writing a few lines of Python

I think the point is that some things are so common, like that copy merge thing, that everybody does their own version and we have like a dozen half baked scripts floating around the community.

There's a bunch of those ui things, like the default vop parms always being in the way, the double bind import bind export nodes I always put down when in a vop, wanting see the end of long parm expressions rather than the beginning, having line numbers in wrangles, or resetting that damn scene view several times a day, etc.

Those are the kind of UI things that could be improved. They are minor cosmetic things but they're neglected and it would help.

Still waiting for Houdini 30 with seamless integration between UI and my room. AI suggestions for nodes and vex snippets to place. Mind as alternative input method. Quick export to usdnft format.

I think the Houdini UI is mostly fine. It's very customizable and they've done an amazing job keeping an insane amount of data easy to navigate. The only thing I'd say is that it would be nice if it looked a little more modern. Somebody shared a nice flat theme here, but more could be done in that department. Compared to most other popular creative apps Houdini is pretty dated-looking.

I'm not even concerned about core Houdini being made more accessible now that Labs is releasing so many powerful, artist-friendly tools. Leave the core alone for the pros who need it and keep adding Labs stuff for the rest of us. Well... leave the core alone except for making Karma the best renderer. That's my only request!

eikonoklastes
Ultimately, I want Houdini to be an easier app to work with, shed its "you need a PhD to work it" perception, attract more users, grow in popularity, and eventually be The One App That Does it All.

Careful what you wish for. Looking at Autodesk's offerings I get the impression that the popularity of an application is inversely related to its feature set evolution.

There used to be a tribe in Africa that always set up there camp at least a day's walk from water, even in the rainy season. That way they were always prepared for drought. Making things easier for artist is setting up camp on the riverbank.. all nice and comfy until the water is gone on no one knows anymore how to get by without a constant supply!

(Maybe to the strongest argument to start the above with 'There used to be' but you get the point!)

I'm sorry but no. The idea that making something easier and faster somehow fosters complacency or laziness in creative artists is simply wrong. The fact that we are doing what we are doing, and that the state of CG is the way it is today is absolutely 100% due to these tools becoming easier and faster than they were 20-30 years ago.

The CG artistic community is thriving and getting better and better by the day exactly because the art form is more approachable. Anytime that a tool comes out that makes something easier and faster, the net result is artists pushing creativity even more -- not less.

Quite honestly, that imaginary tribe in Africa sound like a bunch of idiots, unwilling to take advantage of a resource when it's available and use their extra time to build a reservoir for when there is a drought. Thankfully most of us are not that dumb!

Ok, maybe they were idiots All I'm saying is that there is a subtle cost by making things more accessible.

I gained a much deeper understanding of Houdini because I actually had to write a lot of tools myself, discovering a lot of stuff while going through API docs or even hacking stuff beyond the api. And looking back the most rewarding and enjoyable jobs I did are those involving both epic technical and artistic challenges, growing both skill sets in the process.

It a balancing act between challenging the users vs scaring them away.. Just like in videogames, no feel of achievement makes a project quite boring.

Jonathan de Blok
Ok, maybe they were idiots All I'm saying is that there is a subtle cost by making things more accessible.

I gained a much deeper understanding of Houdini because I actually had to write a lot of tools myself, discovering a lot of stuff while going through API docs or even hacking stuff beyond the api. And looking back the most rewarding and enjoyable jobs I did are those involving both epic technical and artistic challenges, growing both skill sets in the process.

It a balancing act between challenging the users vs scaring them away.. Just like in videogames, no feel of achievement makes a project quite boring.

Buddy, just because you loved it, enjoyed it, and so on, doesn't mean everyone else will. I don't want to be a programmer, I don't want to suffer for days searching for solutions and then to realize I missed one letter in the code and that's why things didn't work. I want to be an artist that can do cool stuff without ever touching the code. If I wanted to a programmer, I would go to programming school. If Houdini wants to be an artist-friendly software, then programming should be left for geeks and technical artists, and should not force anyone into it. As much as I respect programming as a calling, I simply don't want to do it, it's still a pain for me and I don't enjoy it. Luckily for me, I still find Houdini very enjoyable to work with because we can do a lot even now without any kind of Vex or Vops. I am sure things are going to get only better for those seeking simplicity over tehnicalitiies.