Direct Modeling in Houdini - Journal

   7049   40   8
User Avatar
Member
330 posts
Joined: Sept. 2018
Online
Hi folks, me again. This time with trying out DM modeling with vanilla Houdini stuff. At this point, I am not really sure why people are saying that DM modeling sucks, or it's not possible or I should not do it in Houdini. From what I have learned in the past two weeks is that things are quite the opposite. I have found a lot of nice solutions for different modeling tasks. I have to spend a lot of time trying out things and I am quite happy with what I can do with Houdini DM. It needs a bit more functionality to really make it awesome, like for example, Proportional Beveling. Now, I can see how people might have an impression that DM sucks in H. I have watched a gazillion of tutorials and none did show DM modeling in a way I managed to model this. So it's not about that H sucks in this regard, it's more about others simply not demonstrating what's possible and how.

This is still a work in progress. I plan to push this as far as possible.
Edited by SIgor420 - Sept. 21, 2021 06:45:37

Attachments:
wooden_carts1.png (1.4 MB)
wooden_carts2.png (398.5 KB)

QA Specialist | 3D Artist | Core4D Community Manager | https://twitter.com/ISeveric [twitter.com] | https://www.youtube.com/watch?v=x78cuYgegHc [www.youtube.com]
User Avatar
Member
633 posts
Joined: Sept. 2013
Offline
Looks really impressive. A screen recording or a video tutorial about your technique might convince even more people!
User Avatar
Member
330 posts
Joined: Sept. 2018
Online
Konstantin Magnus
Looks really impressive. A screen recording or a video tutorial about your technique might convince even more people!
Thanks Konstantin, I was thinking about it, but I am not sure would people actually be interested.
Edited by SIgor420 - Aug. 10, 2021 11:42:57
QA Specialist | 3D Artist | Core4D Community Manager | https://twitter.com/ISeveric [twitter.com] | https://www.youtube.com/watch?v=x78cuYgegHc [www.youtube.com]
User Avatar
Member
299 posts
Joined: May 2016
Offline
SIgor420
Hi folks, me again. This time with trying out DM modeling with vanilla Houdini stuff. At this point, I am not really sure why people are saying that DM modeling sucks, or it's not possible or I should not do it in Houdini. From what I have learned in the past two weeks is that things are quite the opposite. I have found a lot of nice solutions for different modeling tasks. I have to spend a lot of time trying out things and I am quite happy with what I can do with Houdini DM. It needs a bit more functionality to really make it awesome, like for example, Proportional Beveling. Now, I can see how people might have an impression that DM sucks in H. I have watched a gazillion of tutorials and none did show DM modeling in a way I managed to model this. So it's not about that H sucks in this regard, it's more about others simply not demonstrating what's possible and how.

This is still a work in progress. I plan to push this as far as possible.

really nice and well modeled. I know I will probably get slammed for this, but modeling in Houdini is not as simple as a "make it pretty" button, you really have to work at getting it all quad modeled and is not show stopping like FX seems to be. yeah yeah I know FX isn't as simple as a press of a button but you do have a basis for a start, with modeling and especially quad modeling you do not.

I would be interested in a short video of technique
Edited by bobc4d - Aug. 12, 2021 08:47:41
User Avatar
Member
330 posts
Joined: Sept. 2018
Online
bobc4d
really nice and well modeled. I know I will probably get slammed for this, but modeling in Houdini is not as simple as a "make it pretty" button, you really have to work at getting it all quad modeled and is not show stopping like FX seems to be. yeah yeah I know FX isn't as simple as a press of a button but you do have a basis for a start, with modeling and especially quad modeling you do not.

I would be interested in a short video of technique

Well, I cannot agree, I would say it's quite opposite in some cases and very convenient to have Houdini's strengths on your side. So far the most important thing is to figure out how VP works with shortcuts and how Networks behaving in relation to geometry. Once that barrier is overcome, modeling becomes really a good experience. One good example is merging meshes and there are actually 3 ways how you can do that. Two come from Networks directly while the 3rd one can be achieved within the viewport, you just need to know how and that's something I figured on my own, didn't see that in all the tutorials I was watching. Then its also important to understand how nodes can interact with geometry directly from VP and which one to you use in which mode (Point, Edge, Primitive). In short, once this is figured, modeling is really enjoyable and in some cases much more convenient then in other DCC apps.

I will try to record something meaningful, but I am just not sure where to upload it.
Edited by SIgor420 - Aug. 12, 2021 09:18:20
QA Specialist | 3D Artist | Core4D Community Manager | https://twitter.com/ISeveric [twitter.com] | https://www.youtube.com/watch?v=x78cuYgegHc [www.youtube.com]
User Avatar
Member
633 posts
Joined: Sept. 2013
Offline
SIgor420
I will try to record something meaningful, but I am just not sure where to upload it.

Looking forward to this. I have used to be really fast in direct modeling with Cinema 4D, too, but never transitioned these skills to Houdini. I assume a summarizing video of your findings would be really interesting.
User Avatar
Member
330 posts
Joined: Sept. 2018
Online
Konstantin Magnus
SIgor420
I will try to record something meaningful, but I am just not sure where to upload it.

Looking forward to this. I have used to be really fast in direct modeling with Cinema 4D, too, but never transitioned these skills to Houdini. I assume a summarizing video of your findings would be really interesting.

I am coming from Cinema 4D too and I found that H has even more to offer than C4D ever did in the modeling department. One good example is placing and orienting objects around. One thing I wish they implement is proportional beveling, that would be great and I would not have to use knife tools that much and things could stay procedural even with DM. Another good example would be patter selection.
Edited by SIgor420 - Aug. 12, 2021 12:03:09
QA Specialist | 3D Artist | Core4D Community Manager | https://twitter.com/ISeveric [twitter.com] | https://www.youtube.com/watch?v=x78cuYgegHc [www.youtube.com]
User Avatar
Member
633 posts
Joined: Sept. 2013
Offline
SIgor420
I am coming from Cinema 4D too and I found that H has even more to offer than C4D ever did in the modeling department.

If this is true, then this is even more impressive! In C4D the better I got the fewer tools I needed. Mostly just extrude, inner extrude, bevel, loop, ring cuts and the usual selection modes, which were easy and fast to combine.

However Houdini had a bad rep for its DM capabilities back then. So, if your modeling process is as efficient as the topology shown in the screenshots, then I think you would do the community a great favour by giving some insights.
User Avatar
Member
442 posts
Joined: Feb. 2017
Offline
SIgor420
Another good example would be patter selection.
There is pattern selection in Houdini.
Video on pattern selection [www.youtube.com]

Cheers
CYTE
Edited by CYTE - Aug. 13, 2021 02:43:01
User Avatar
Member
330 posts
Joined: Sept. 2018
Online
CYTE
SIgor420
Another good example would be patter selection.
There is pattern selection in Houdini.
Video on pattern selection [www.youtube.com]

Cheers
CYTE

I know, C4D is the one that is missing that. Sorry if I have confused you.
QA Specialist | 3D Artist | Core4D Community Manager | https://twitter.com/ISeveric [twitter.com] | https://www.youtube.com/watch?v=x78cuYgegHc [www.youtube.com]
User Avatar
Member
1880 posts
Joined: Sept. 2015
Offline
I am not really sure why people are saying that DM modeling sucks

It depends on what the user is aiming for.

For myself I like to create models that are dimension/measurement related forms to each other, that are also organic for the purposes of 3d printing.

I've used SolidWorks and Rhino and I know those programs could be considered more parametric but that doesn't mean knowing the tools you can't take a DM approach with that software; which I have done.

I found, especially with SolidWorks so many things are much easier to set up and do, both from viewing things in terms of an aesthetic viewpoint and establishing dimenstional relationships, e.g. the ease to set up very quickly, save by name for resuse - contruction planes which are synonemous with view planes. Along with drawing tools to create things like tangents and perpendiculars to curves, etc. from those viewing/construction planes. It makes it so easy for a very refined and 'absolute' dimensioned design - rather than doing some perspective viewing for a 'good enough' approximation. From my experience in the end, the difference between the two products designed/modelled that each software allows for is small but noticable. It takes much more work in Houdini to get your own tools refined enough so that the price point makes it worth the while ( 300 Indie vs 6k SW). And that is where I am, still developing my own tools and refining the approach to take with Houdini. It's a good thing I like fidgeting around making tools. Also once the tools are refined and you have your workflow established - You don't have to do it again, and can enjoy modelling at a much lower price point for the software.

One thing I do miss is SolidWorks awesome capacity to fillet, bevel and chamfer smoothly across very complex geometry that also changes in size; even with Rhino I found it's capacity can be missing in that department in comparison.
Edited by BabaJ - Aug. 13, 2021 14:43:12
User Avatar
Member
330 posts
Joined: Sept. 2018
Online
BabaJ
I am not really sure why people are saying that DM modeling sucks

It depends on what the user is aiming for.

For myself I like to create models that are dimension/measurement related forms to each other, that are also organic for the purposes of 3d printing.

I've used SolidWorks and Rhino and I know those programs could be considered more parametric but that doesn't mean knowing the tools you can't take a DM approach with that software; which I have done.

I found, especially with SolidWorks so many things are much easier to set up and do, both from viewing things in terms of an aesthetic viewpoint and establishing dimenstional relationships, e.g. the ease to set up very quickly, save by name for resuse - contruction planes which are synonemous with view planes. Along with drawing tools to create things like tangents and perpendiculars to curves, etc. from those viewing/construction planes. It makes it so easy for a very refined and 'absolute' dimensioned design - rather than doing some perspective viewing for a 'good enough' approximation. From my experience in the end, the difference between the two products designed/modelled that each software allows for is small but noticable. It takes much more work in Houdini to get your own tools refined enough so that the price point makes it worth the while ( 300 Indie vs 6k SW). And that is where I am, still developing my own tools and refining the approach to take with Houdini. It's a good thing I like fidgeting around making tools. Also once the tools are refined and you have your workflow established - You don't have to do it again, and can enjoy modelling at a much lower price point for the software.

One thing I do miss is SolidWorks awesome capacity to fillet, bevel and chamfer smoothly across very complex geometry that also changes in size; even with Rhino I found it's capacity can be missing in that department in comparison.

Unfortunately, I never had a chance to play with SW. On the other hand, that's the software exclusively made for modeling in more precision and CAD, so I would expect it's simply a better experience, in Houdini on the other hand you have more aspects to consider than just one. Houdini definitely shines in many aspects but it's not perfect as all other DCCs, in the end, you use what suits you the most and what you find fun to use. I choose Houdini after almost a decade with C4D for few reasons. Inide license and price of it, fact that SideFX is a still private company and on top of that, it's truly a DCC that can do so much more with the vanilla stuff than any other DCC. For all other DCCs you need a lot of plugins to be able to do what you can do with Houdini, so for me, it was really an easy decision, and on top of that even the fact that it was more technical at the beginning I knew Ill benefit later on and simply know more in technical terms of 3D which is a win-win combination due to my daily job and personal future development as QA and artis. In the end, it's all good fun especially when you are at the point where you finally get to understand the nature of the beast such as Houdini. I've been hearing over and over how Houdini is super technical, but I found that it doesn't have to be and it's just a totally different approach to everything which comes with a big advantage and some disadvantages.

Ans an update:

Attachments:
ss1.png (500.0 KB)
ss2.png (379.0 KB)

QA Specialist | 3D Artist | Core4D Community Manager | https://twitter.com/ISeveric [twitter.com] | https://www.youtube.com/watch?v=x78cuYgegHc [www.youtube.com]
User Avatar
Member
371 posts
Joined: Nov. 2010
Offline
SIgor420
Hi folks, me again. This time with trying out DM modeling with vanilla Houdini stuff. At this point, I am not really sure why people are saying that DM modeling sucks, or it's not possible or I should not do it in Houdini. From what I have learned in the past two weeks is that things are quite the opposite. I have found a lot of nice solutions for different modeling tasks. I have to spend a lot of time trying out things and I am quite happy with what I can do with Houdini DM. It needs a bit more functionality to really make it awesome, like for example, Proportional Beveling. Now, I can see how people might have an impression that DM sucks in H. I have watched a gazillion of tutorials and none did show DM modeling in a way I managed to model this. So it's not about that H sucks in this regard, it's more about others simply not demonstrating what's possible and how.

This is still a work in progress. I plan to push this as far as possible.
It's not that it sucks (with the additions in the past two years that is) or that it is not possible. It is just that modelling is still faster and more direct in other tools. Like XSI....
But as I said a lot has been done and I find myself using it more often for the entire workflow.
The biggest issue for me is still the lack of multi object editing. Having a model consistent of many parts, all in one tree with only tedious methods to directly select and transform a part in the viewport is slowing the flow down a lot. Being able to just grab all upper points of windows, doors and walls and adjust the y position - while having them being separate geo - is something you don't want to miss...

In the end what makes you the money is never what you can do but how fast you can do it and how many people you need.
Edited by OneBigTree - Aug. 17, 2021 10:05:10
User Avatar
Member
330 posts
Joined: Sept. 2018
Online
OneBigTree
SIgor420
Hi folks, me again. This time with trying out DM modeling with vanilla Houdini stuff. At this point, I am not really sure why people are saying that DM modeling sucks, or it's not possible or I should not do it in Houdini. From what I have learned in the past two weeks is that things are quite the opposite. I have found a lot of nice solutions for different modeling tasks. I have to spend a lot of time trying out things and I am quite happy with what I can do with Houdini DM. It needs a bit more functionality to really make it awesome, like for example, Proportional Beveling. Now, I can see how people might have an impression that DM sucks in H. I have watched a gazillion of tutorials and none did show DM modeling in a way I managed to model this. So it's not about that H sucks in this regard, it's more about others simply not demonstrating what's possible and how.

This is still a work in progress. I plan to push this as far as possible.
It's not that it sucks (with the additions in the past two years that is) or that it is not possible. It is just that modelling is still faster and more direct in other tools. Like XSI....
But as I said a lot has been done and I find myself using it more often for the entire workflow.
The biggest issue for me is still the lack of multi object editing. Having a model consistent of many parts, all in one tree with only tedious methods to directly select and transform a part in the viewport is slowing the flow down a lot. Being able to just grab all upper points of windows, doors and walls and adjust the y position - while having them being separate geo - is something you don't want to miss...

In the end what makes you the money is never what you can do but how fast you can do it and how many people you need.

I see what you mean, you are right, but with Houdini, you need to take a different approach. It's hard to compare H with all other DCCs when it works on a totally different level or workflow. There are plenty of things you cannot do in other DCCs while in H is really easy. Not to mention some things are impossible to do in other DCCs, so...its hard to argue about these things. So the best thing to do when coming to Houdini from other DCCs is to leave all things behind and not try to compare something is impossible to compare. Would be interesting to see how other DCCs would approach to nodal workflows. Both Viewport and Nodal Network have a lot of features to play with. I guess Houdini's advantage is more than 20y of experience which will be hard to beat.
Edited by SIgor420 - Aug. 17, 2021 11:04:08
QA Specialist | 3D Artist | Core4D Community Manager | https://twitter.com/ISeveric [twitter.com] | https://www.youtube.com/watch?v=x78cuYgegHc [www.youtube.com]
User Avatar
Member
330 posts
Joined: Sept. 2018
Online
Here are two ways you can quickly change your object size or replace them with one which was edited.

Attachments:
Houdini_Ways.mp4 (2.6 MB)

QA Specialist | 3D Artist | Core4D Community Manager | https://twitter.com/ISeveric [twitter.com] | https://www.youtube.com/watch?v=x78cuYgegHc [www.youtube.com]
User Avatar
Member
146 posts
Joined: Jan. 2018
Offline
SIgor420
Konstantin Magnus
Looks really impressive. A screen recording or a video tutorial about your technique might convince even more people!
Thanks Konstantin, I was thinking about it, but I am not sure would people actually be interested.
Direct Modeling is one area we definitly don't have enough tutorials and videos, so yes, I vote for that video tutorial
User Avatar
Member
85 posts
Joined: Oct. 2015
Offline
Whoa this is amazing!
User Avatar
Member
330 posts
Joined: Sept. 2018
Online
Berk_Erdag
Whoa this is amazing!
Thank you Berk, that's very kind of you, much appriciated.
QA Specialist | 3D Artist | Core4D Community Manager | https://twitter.com/ISeveric [twitter.com] | https://www.youtube.com/watch?v=x78cuYgegHc [www.youtube.com]
User Avatar
Member
1 posts
Joined: July 2021
Offline
I recently started using Houdini for more modeling work and have found its core features more than competent. The promise of being able to edit your object's history was what made me embark on this experiment and I found it to be absolutely amazing when it worked. Mixing in "opportunistic" proceduralism became surprisingly time efficient. I found myself using deform + raycast/detangle when working with curved surfaces constantly. This saved a massive amount of time that I would otherwise spend trying to align objects.

From my experience, the problems with modeling in Houdini have less to do with missing features, and more to do with a lack of quality of life. Some existing plugins in the ecosystem go a long way to filling these holes, but you will still find that any direct modeling with PolyDraw/Split/Fill/Knife/etc will quickly fill up your node graph and make it impossible to modify upstream nodes.


Then there are the things that are just different about Houdini and take some getting used to. As already mentioned, editing more than one object at a time isn't possible without merging them into a single geometry node. Using generated groups can help preserve proceduralism but can slow down modeling considerably because of switching between the viewport and node/properties windows.

Realistically as I get more experienced with modeling in Houdini I am finding it to be a time saver in more usecases than I originally expected. That being said, I still find myself falling back on Machin3 for cases where the topology is too complex for nodes to be very helpful.
User Avatar
Member
330 posts
Joined: Sept. 2018
Online
DasBurdock
I recently started using Houdini for more modeling work and have found its core features more than competent. The promise of being able to edit your object's history was what made me embark on this experiment and I found it to be absolutely amazing when it worked. Mixing in "opportunistic" proceduralism became surprisingly time efficient. I found myself using deform + raycast/detangle when working with curved surfaces constantly. This saved a massive amount of time that I would otherwise spend trying to align objects.

From my experience, the problems with modeling in Houdini have less to do with missing features, and more to do with a lack of quality of life. Some existing plugins in the ecosystem go a long way to filling these holes, but you will still find that any direct modeling with PolyDraw/Split/Fill/Knife/etc will quickly fill up your node graph and make it impossible to modify upstream nodes.


Then there are the things that are just different about Houdini and take some getting used to. As already mentioned, editing more than one object at a time isn't possible without merging them into a single geometry node. Using generated groups can help preserve proceduralism but can slow down modeling considerably because of switching between the viewport and node/properties windows.

Realistically as I get more experienced with modeling in Houdini I am finding it to be a time saver in more usecases than I originally expected. That being said, I still find myself falling back on Machin3 for cases where the topology is too complex for nodes to be very helpful.

THey are definitely competent and to be super solid I would need few more added features for some tools and I would be super happy/ One of those that I really miss is proportional beveling.

Now, you can check here and see a bit of how I model stuff in Hoduini. Have in mind this is my first tutorial ever so please don't expect perfection, heeh, but it might give you an idea how to speed up your modeling a bit more.

https://www.core4d.com/ipb/videos/view-253-bake-normal-map-from-geometry/ [www.core4d.com]

Around minute 5 I will show how I approach to modeling.
Edited by SIgor420 - Sept. 16, 2021 12:17:45
QA Specialist | 3D Artist | Core4D Community Manager | https://twitter.com/ISeveric [twitter.com] | https://www.youtube.com/watch?v=x78cuYgegHc [www.youtube.com]
  • Quick Links