I've been reserving my use of Houdini for procedural assets, and doing all my basic modeling for static assets in Modo.
I love Houdini though, so I was considering consolidating my pipeline and trying to just do all modeling in Houdini.
I'm wondering if anyone else does this, and how well it works. I know they added a lot more direct modeling tools in 16, but I haven't used them very heavily.
I know that Houdini would be good enough to get by in, but the relative speed of the work is my real concern.
There aren't enough videos of this for me to get a good grasp of what people can do, and my own experience would obviously slow me down and throw off my evaluation of how fast and easily I can work with it.
I do primarily hard surface modeling. Organic stuff I generally handle in ZBrush.

here is a new video for you!
https://vimeo.com/286242378 [vimeo.com]

we'd really love more feedback from people about our modeling tools/workflow

Is there an expected release date for 17 when we'll get those tools?

I've finished the video, and the tool looks kind of… limited?
I'm glad for more modeling tools, and I can see it helping a bit with some tasks here and there, like more easily filling holes and cutting in edges. But the demonstration, plus the name “polydraw” gives the impression that it's focused on actually drawing vertices in an isometric view. Which is a very archaic and niche workflow. I was actually kinda shocked to see the presenter doing that at all. So having the tool focused on that just seems bizarre.
It sounds like the hotkeys for switching modes are kinda messy as well. I doubt I could use it for a long while without having the hotkeys listed on a second monitor at all times.

I watched the video and like where the tool is headed but it still has some work to be done. I came from Cinema4d and put in a request for a C4D-like poly pen tool, which really helps in hard surface modeling. I am hoping this will move in that direction.

Yeah, when I heard about this, I assumed it was basically the Houdini version of C4D's polypen but I guess it's really not.

Grimwolf
I've finished the video, and the tool looks kind of… limited?
I'm glad for more modeling tools, and I can see it helping a bit with some tasks here and there, like more easily filling holes and cutting in edges. But the demonstration, plus the name “polydraw” gives the impression that it's focused on actually drawing vertices in an isometric view. Which is a very archaic and niche workflow. I was actually kinda shocked to see the presenter doing that at all. So having the tool focused on that just seems bizarre.
It sounds like the hotkeys for switching modes are kinda messy as well. I doubt I could use it for a long while without having the hotkeys listed on a second monitor at all times.

In what way looks limited?
I don't know about “drawing vertices in an isometric view” being archaic - what has it been supplanted by? I mean, you did acknowledge the fact there's a need for this ability, be it a niche or not. It was not available before and now it is. I find this to be a very good thing, but if you think it's not sufficient, record your screen with a tool that works in a more modern way and post a .gif file (or link to a video). I would however wait until H17 is released in order to actually test it and not judge it by a single video.
Hotkeys in general are customizable and if you'll find that these aren't, you'll be able and encouraged to file a RFE about it.

Gee, I have to say … I just had a look at my Cinema4d version and I can tell you for sure that the modeling tools in there are in no way comparable to what Houdini is offering today.
OK, I admit, my C4d version is 1.something and I did only quickly skip through my C4d V2 manual, since I never got to install V2, I may not be up to date.

But wait.

Why should someone developing a tool that is OFFICIALLY being presented as a “work in progress” (look up the term if you are not familiar with it) *know* what exactly some arbitrary other tool is doing? When you are deep in your own project, it is quite uncommon to waste precious time by playing around with neighbor's toys.

Instead of being “underwhelmed” by something that is shown in its EARLY STAGES, why don't you just:
- give constructive feedback like “I would like to be able to do precisely THIS” (not “do it like in xyz”, because for that you can use “xyz”)
- show a demo of what you want to do if you want a copy, demonstrate how it fits into Houdini's “mind” of proceduralism
- explain in what way the tool you want to have would help a large majority of users to become more productive (as opposed to “a single user's wish”)

Yes, I know. This is not how many artists work. Yes, I would love to see “easier” workflows in Houdini all over the place (I struggle hard with nodal workflows, nodal workflows kill every joy of work for me) but I do not expect any developer to just copy something else she does not see any benefit in (for the tool at hand).
EXPLAIN the benefit to the developer. SHOW the productivity gain you would get to management (do a video comparison of a typical workflow in whatever-tool and Houdini). Maybe you get hints from other users at how to BETTER (faster) do it in Houdini?

Sorry for the text-wall. I am currently *underwhelmed* by the type of feedback I see around.

Marc

For me this is a tool that is welcomed because it was missing in Houdini.
What would be great is to enhance this tool for the third dimension - from the presentation it looks like you can only model in one plane.
Some functionality like in zbrush zmodeller tool, where you can easily block out some basic mesh - like making a 3dimensional sketch.
Here is a link to a video:
http://pixologic.com/zclassroom/lesson/poly-qmesh [pixologic.com]

Great would also be to have the possibility to curve selcted edges with interactive resampling of the curve. a little bit like in polygon pen in c4d.
Here the video:


I know this are all not procedural functionality - but would be great to sketch some basics form easily and fast without the need to change programm, export and import to Houdini.

Nico

goldfarb
here is a new video for you!
https://vimeo.com/286242378 [vimeo.com]

we'd really love more feedback from people about our modeling tools/workflow

I recently spent an entire day doing viewport modeling in Houdini, and it worked quite well. I found two main problems:

1) There seems to be no way to snap and fuse. Hitting X and snapping works great, but let's say I snap 10 different points. When I'm done with the snapping, I now need to select all those 10 points and use a fuse node. Yes, often times one can just use an overall fuse node on the entire mesh without selecting the 10 points, but even that is a hassle.
Also, the Fuse does not seem to be part of the “C” radial menus. These radial menus are super awesome by the way.

2) The bevel tool is limited, and it's sooooo sad because it would be the perfect companion to the boolean tool. On a dense mesh (e.g. you subdivide a geo and boolean a few holes into the smooth surface) the bevel tool becomes useless: it stops the entire bevel once the bevel collides with an edge anywhere. Coming from Softimage, I'm used to bevels working in these scenarios because the bevel tool would handle the colliding edges, merge them and apply the bevel anyway.

What's interesting, doing all this viewport modeling, I got into a very destructive mindset where initially I did everything non-procedurally. I hand-selected everything etc. However, this model had several similar but not identical features. Eventually I realized that parts of these similar features could be made procedurally. So I used a For loop, some procedural groups and extrusions inside of it and saved hours of time.

malbrecht
Gee, I have to say … I just had a look at my Cinema4d version and I can tell you for sure that the modeling tools in there are in no way comparable to what Houdini is offering today.
OK, I admit, my C4d version is 1.something and I did only quickly skip through my C4d V2 manual, since I never got to install V2, I may not be up to date.

But wait.

Why should someone developing a tool that is OFFICIALLY being presented as a “work in progress” (look up the term if you are not familiar with it) *know* what exactly some arbitrary other tool is doing? When you are deep in your own project, it is quite uncommon to waste precious time by playing around with neighbor's toys.

Instead of being “underwhelmed” by something that is shown in its EARLY STAGES, why don't you just:
- give constructive feedback like “I would like to be able to do precisely THIS” (not “do it like in xyz”, because for that you can use “xyz”)
- show a demo of what you want to do if you want a copy, demonstrate how it fits into Houdini's “mind” of proceduralism
- explain in what way the tool you want to have would help a large majority of users to become more productive (as opposed to “a single user's wish”)

Yes, I know. This is not how many artists work. Yes, I would love to see “easier” workflows in Houdini all over the place (I struggle hard with nodal workflows, nodal workflows kill every joy of work for me) but I do not expect any developer to just copy something else she does not see any benefit in (for the tool at hand).
EXPLAIN the benefit to the developer. SHOW the productivity gain you would get to management (do a video comparison of a typical workflow in whatever-tool and Houdini). Maybe you get hints from other users at how to BETTER (faster) do it in Houdini?

Sorry for the text-wall. I am currently *underwhelmed* by the type of feedback I see around.

Marc

What kind of a response is that? Hostility aside, it sounds like you're saying “Tell me how they could possibly have done better. But DON'T MENTION C4D!”.
And proceduralism? The exact purpose of this tool is to allow more effective non-procedural modeling.
Also, the main problem I have with this tool is that it does seem like “a single user's wish”. I've never had to manually place vertices, because I've always found box modeling and booleans to be faster. I don't think I've done that since I was 14 using GMax, and didn't know how to make complex shapes without manually placing points one-by-one.
The tool was presented as a modeling swiss-army knife that would make it much more convenient to model traditionally without constantly switching tools and generating a bunch of unnecessary nodes. But the main focus of the tool is being placed on something that is very rarely, if ever, used. And being done in a very bare-bones fashion compared to the competition, which does little to expand that niche value.
Like some others have mentioned, I think the best case scenario would have been to have it work more like this [www.youtube.com], giving it a wider range of application.
Barring that, I'd have liked to see the drawing tools replaced by more common tools relating to the manipulation of points, like drag-weld and slide.

SOmeone once said on this forum that Houdini is all about proceduralism, which of course is nonsense. Houdini is all about exactly what all the other 3d programs are about: creating compelling images (static or animations) and 3d models to be used in various industries.
It just happens that we, as a community and I'm willing to bet that SESI as a company too, just happens to like achieving those goals by using a procedural paradigm.

Here's my difficulty: I'm having a very hard time understanding either of the two extreme positions - everything's about proceduralism and some tools have to be non-procedural in order to excel in some viewport workflow.

Both of these positions that I see sometimes expressed on these forums rest on a logical fallacy or a lack of imagination, I'm not sure, which implies that going forward with one is going to affect negatively the other. I've seen absolutely no evidence thus far to be the case.
Here's a more balanced position: all tools have to be procedural, because we happen to like to have this option for when we need it, but tools also have to work well in viewport. This is IMO, the best way to think about non-procedural modeling in order to keep everyone's happy - those who prefer to stay in VP and don't care about nodes, don't have to deal with them, but those that do like to also keep the netview open for various tricks and workflows (see Fianna's video above) will also have a bunch of not unnecessary nodes to play with.

Did I say anything controversial?

Grimwolf
What kind of a response is that? Hostility aside, it sounds like you're saying “Tell me how they could possibly have done better. But DON'T MENTION C4D!”.

Yeah, welcome to Houdini-world, it's really weird around here with people having convinced themselves that archaic and convoluted ways of doing simple things is somehow awesome.

Sarcasm aside, have you looked at some of the 3rd party solutions for modeling in Houdini like this one?

https://gumroad.com/alexeyvanzhula [gumroad.com]

Hi there,


To me, hard surface modeling doesn't mean boxy surfaces, so at some point you need good tools to deal with edge loops/slicing for subdivision surface modeling.

Coming from Modo, I could probably request hundreds of features, but instead I will try to focus on the most obvious one.
Please let me know if some of this feature exist, as I might miss some functionality ?

1) Subdivision display mode.





We (I) badly need a subdivision display mode the doesn't just mix a cage display with the subdivision result.
Once in sub-d display mode (-, it's very difficult in Houdini to select points or edges, as they might be under the subdivision surface.

Modo version display the limit surfaces on the subdivision model, and you can easily select components and do your modeling in this mode.

2) Local LoopSlice.



They are few handy options in Modo :

- The slice is only done on the selected poly
- You have a visual feedback where the slice is created (%)
- There is symmetry mode than will really speed-up the edge looping work.

3) Edge Slide / duplicate



- This is really my favorite edge loop method
- You can specify a numeric distance for the sliding … when creating support edges for subdivision surface, you try to keep this distance
constant all over the model.

SDS modeling for hard surfacing is all about creating support loops to crease the model … and it's a very boring tasks.
That's probably the reason why solid modeling (Nurbs) get a lot of success lately, with product like Fusion 360, and MoI3D (by far my favorite tool for this tasks).
In Fusion you can do hybrid modeling, turning cages SDS model into Nurbs surface, and use all the solid tools to add details/trims and fillets.

It would perfectly fit the nodal procedural aspect of Houdini, but that's an other story : https://www.sidefx.com/forum/topic/55277/?page=1#post-248072 [www.sidefx.com]

Traditional DCC try to copy this solid modeling method too, with product like MeshFusion (Modo) and HardMesh (Maya) for example.
Because dealing with flat surfaces and booleans are really fun in Houdini, but the bevel tool is useless on curved surface.

In Houdini we have Soft-Boolean, part of the Direct Modeling from Alexey Vanzhula … and so far it's really my preferred product (partially because Houdini's node graph ).



I really hope some collaboration can be done with Alexey on the subject !

Pascal.

The name of the company says it clear SideFX. Could be called Side3d, makes it clear that houdini was created for the vacuum that was in quality Fx. Today is pretty good in terms of modeling, formerly it was worse and things were very good, I think it's the same as always. It is not the arrow, it is the Indian. Just get used to what you have and dry match. Greetings to all.

And why is Houdini's modeling better than before now? Pretty sure it has something to do with someone not willing to get used to what they had.

PaQ WaK
Hi there,


To me, hard surface modeling doesn't mean boxy surfaces, so at some point you need good tools to deal with edge loops/slicing for subdivision surface modeling.

That's spot on! Here's a video of one the most proficient Sub-D modeler on the planet.



Edge loop cuts/insertions, point/edge sliding, beveling&extruding, point snapping/collapsing, not a big number of tools. Just a few, but smart and reliable tools.

I rather would like to see the things mcnistor and PaQ WaK mentioned before seeing too many new half baked tools like polybevel 2.0 *cough*. Looking over some subd modelling sessions in modo/si could bring many ideas/needed features. One example would be inserting smooth/subd edge loops.

Personally I’d be happy if I were provided a “don’t keep history” context where I could do whatever I wanted to as a modeler without a new SOP being created for each operation.

PaQ WaK
We (I) badly need a subdivision display mode the doesn't just mix a cage display with the subdivision result.
Once in sub-d display mode (-, it's very difficult in Houdini to select points or edges, as they might be under the subdivision surface.

If helps, you're not only one who tried subd display in H 16.5, hehe. It reminds me to Maya Smooth Proxy (or something called like that) from 2002 or so…. Funny enough, now it can't do ‘display all subds’ mode anymore, probably not used that much by modelers, but usable for checkout by people who are creating animated deformations. Hopefully I got the habit to never ever uninstall the old version. Still 16.0.7 here…
Otherwise, with all due love and respect to H, don't know what to say, it seems to be multiple chicken - egg problem with anything ‘direct’ in H, including animation. From ‘chicken side’, Side FX should be waiting for real response first, from ‘egg side’, they seem to have a great talent to keep the app 5-10 years behind the rest of 3d world, at least when it comes to ‘direct’.
Yeah it's better than it was in times of H 13 or so, but starting point was a somehow catastrophic mix of ‘unique solutions’ like old Bend deformer, not able to turn plane into cylinder (and more, like ‘soft selection’ based on normals, move tool changing into tweak tool by own criteria, so on), and unbelievably weak engines like nurbs in H, asking to manually control the marching steps.

Perhaps this is a philosophy of app for developers, first to supply something quarter-done, then to proceed if there's enough of interest for full quality. But this is not how does it works anywhere else…

One day when they'll remove their IMO arrogant comments about tumbling style in preferences, when they'll seriously take apps like Blender, Max, c4d, Modo as an 1:1 reference for direct modeling (that is, anything, but seriously), there will be some hope. Otherwise, just use 3d viewport only as display and everything will be fine, of course with possible activities in that, node only way.