Hey Tim, That all seems like a decent workflow with the exception of retopo. I feel that would negate the benefits of using MOI if you are going to rebuild the mesh after export. I would just suggest adjusting the mesh on export to get it low poly enough for use in Unity.

-Shane

Hi, Tim. I use MoI precisely for game asset creation. I model my high-end prop in MoI, sometimes I export at a high level for additional sculpting in Zbrush, and I create my final, in-game asset based off of a low-res export of the object I made in MoI. I use Modo as my main pipe into Unity, but I have used Maya, Max and XSI(Softimage) for the same thing and they all work fine.

The biggest issue you are going to run into is whether you should retop (re-build) the high-end mesh or clean-up an export from MoI. It will depend on the object, but I usually start with a lower-res tessellation from MoI and proceed to clean the mesh up in Modo. However, sometimes on more curvy, ergonomic-oriented things, I will use traditional retop techniques (Modo added great functionality for this in 601). Cleaning up a MoI mesh can be time-consuming. I've found ways to speed things up by building purposefully lower-res versions of the same model in MoI. For instance, I'll take something like a pipe or a cylindrical shape that I used by sweeping a circle over a curve to make the high-end object and make an alternate version using a hexagon or an octagon (octagon is a little easier to work with). I might even make a simpler version of the curve I use to sweep (using corner points instead of curving points). Much easier to clean-up in your other DCC tool.

Some will argue that Sub-Ds are better and it might be for them. I find the time I would spend making Sub-Ds 'work' on most assets is time better spent making the asset cool and interesting and that's often the result of trying different things. In short, I get more bang for my buck, timewise, by running through a bunch of different iterations or prototypes instead of making a good high-end sub-d model only to want or need to change it later.

I'm not trying to plug my website for no good reason, but if you look at some of the weapons I've made, you'll see my workflow. I rely *heavily* on MoI. In fact, I really don't want to work any other way.

Here's a few links:

http://www.ronnieashlock.com/?p=999

http://www.ronnieashlock.com/?page_id=567

http://www.ronnieashlock.com/?page_id=345

I am working on a simple workflow tutorial showing the benefits of the MoI approach. Someday I'll get around to finishing it.

In short, MoI is not only a good alternative to Sub-Ds, I'd say for most hard-surface things it's superior in just about every way. The only negative thing I've found is that people on a production team unfamiliar with the tool often don't want to consider it for some unfathomable reason. But that's a topic for another conversation.

Hi Tim, yes NURBS modeling is usually a much more natural fit for hard surface modeling than sub-d modeling.

With NURBS modeling you're able to use cutting and boolean operations as a primary way of building things, and this makes it pretty straightforward for construction of a lot of man-made type objects that are constructed by having some piece of stock that then has holes drilled in it and pieces cut off to form the final shape.

Basically the strong area of NURBS modeling is when you are able to form a lot of your model by focusing on creating 2D blueprint like curves and then using those curves directly, some for constructing base objects (like extrusion/revolving/sweeping), and then other curves used directly as cutting objects to drill holes or slice off ends of things. Many types of mechanical objects are well defined by 2D profile curves and fit with this method of modeling.

Organic shapes like monsters, characters, and faces are very different - they are a bad fit for this technique because they are not primarily __defined__ by 2D profiles, a 2D profile of a face for example does not "define" the face, it only captures the face at one single angle and it's rapidly changing in shape as you move away from that silhouette. So when there is not any 2D nature with the objects involved sub-d modeling tends to be a much better fit for that.

There are some kinds of models that are in a bit of a gray area which are semi-profile-driven but also kind of organic, sometimes vehicles can end up in this sort of gray area where it might not quite fit in either bucket exactly.

But for a lot of props and hand-held objects and things like that, NURBS should work quite well for those kinds of things.

It tends to be quick to work in 2D and when you're able to construct large sections of your model by a small number of 2D curves that's when you can become very productive and time efficient with NURBS modeling. That's specifically where it really shines and has its greatest strength.

It is a very different approach than sub-d modeling though and there are some kinds of habits from sub-d modeling that are not productive to try and replicate in NURBS.

To get some more ideas on some of these things, I'd recommend watching the video tutorials that are here as a good overview of the techniques involved:
http://moi3d.com/2.0/docs/tutorials.htm

Those walk through step-by-step with creating a few simple models and you can see how big chunks of the model are formed by only drawing and editing a small number of 2D curves.


And also check out here for some links to discussions and general tips for people who are coming from a sub-d / poly modeling background:
http://moi3d.com/forum/index.php?webtag=MOI&msg=4865.2


Probably the biggest difference from sub-d modeling is that with NURBS modeling you want to be using booleans as a primary way of building the model, ideally with a lot of just 2D curves involved.

Often I see poly modeling people try to build a "scaffolding" of mainly 3D curves and then trying to fill that in with little patches by doing a lot of sweeps or Network surfaces - that technique can be used for a type of organic modeling using NURBS but if you're doing a model that requires that it is kind of a red flag that you may have been better off using sub-d for that model anyway.

Another general problem that I see from sub-d modelers is trying to build surfaces directly to some kind of complex boundary. When Booleans are one of the primary tools to work with you often want to construct some initially more simplified and extended shape and then use Booleans or trimming to slice off chunks of that simple shape and form some of the final edges by intersections or cutting rather than directly drawing them. If you have hand drawn every single 3D edge in your final model rather than having many of them generated as intersections or cuts then you're not really using the NURBS toolset to its full advantage.

Hope these tips help! Check out the links above for a bunch of other similar kinds of discussion.

I would think MoI should be a good fit for you from what you have described though, basically MoI's primary reason for existence is to give artists who do not have a big mechanical engineering background to be able to make use of a NURBS toolset for models where that is a good fit which is mostly mechanical type models.


Like others mentioned above, you also might be able to use direct output from MoI for several kinds of cases rather than always doing retopo - do you always need to provide Unity with quad sub-d surfaces?

- Michael

Wow, thanks everyone for the great replies to my questions. Your answers were very insightful and helpful to me.

@SurlyBird:

It is great to hear that others are using MoI successfully in their game asset creation pipeline. I checked out your web site and your game assets are top notch! Great work and I really liked the way you not only showed the end result but also the workflow that you used to achieve the final low poly game objects. I noticed that you are using Substance Designer for texturing and map generation…I just bought this application as well and think it will be the future of texturing (especially with apps that support it natively like Unity). I appreciated your explanation of the different techniques that you use to generate a low poly mesh from your MoI nurbs model. Just to summarize, it sounds like any of the following workflows would work (depending upon the complexity of the MoI object that you are starting with):

1) Export high-poly mesh from MoI to be used as basis to bake normal/ambient occlusion map for low poly object
2) Generate low poly mesh by one of the below methods:

- Simply export a lower poly mesh straight from MoI and use that “as is”
- Export a lower poly mesh from MoI and clean up the mesh in poly modeler as needed
- Retopo the high poly mesh in poly modeler to create a low poly mesh
- Build a “more simple” version of the nurbs model in MoI and export that simplified version to create a more game friendly object (this was a great idea that I didn’t think of)

@Michael Gibson

It’s great to see that the creator of the application is so active on the forums. Thanks for taking the time to answer my questions. The tips that you provided about techniques to avoid when coming from a subd modeling mindset were very helpful.

I understand that many people are extremely productive at hard surface modeling using subds, and while I can hold my own, it has never seemed as intuitive and easy as it should be. For me, subds force me into a mode of focusing so much on topology during the creative phase that I spend most of my brain power working out poly flow puzzles rather than just designing. I am going to buy MoI this evening and am excited to learn more about nurbs modeling.

To answer your question about what the Unity game engine expects…

Unity does not support the rendering of subdivision surfaces (most game engines do not) since they are processor intensive and not really appropriate for real-time game engines. Unity only accepts standard poly meshes (though it will honor smoothing angles). When Unity loads a mesh, the engine will triangulate all of the polys in the mesh (if all polys aren’t already triangles). Therefore, a typical workflow for someone creating a game asset for Unity using subd modeling would be: create high res subd model, create low res version of same model, triangulate the low res mesh in a poly modeler (it is better to triangulate yourself so your UVs and normal maps are optimal in Unity), UV map, bake normal and ambient occlusion maps onto low poly mesh (using the high res mesh to supply detail), texture, bake diffuse and specular maps, export low poly mesh (with uvs and all texture maps) to Unity.

Hi Tim,

> I understand that many people are extremely productive at hard surface modeling using subds, <...>

It can vary a lot depending on the particular qualities of the model - "hard surface" is a rather generic term...

If the objects being created have the overall form of something like one connected skin, then the sub-d modeling can progress in a similar fashion as it's done for organic models and it can go well. Things like space armor and vehicles can fit into this kind of stuff. These kinds of things can be in the sort of "gray area" that I was mentioning previously where it's not always 100% clear whether you'd want NURBS or sub-d for them and when that happens you may actually want to err more on the sub-d side especially if you're still getting up to speed with NURBS modeling strategies.

If the objects being created are more like engine parts that have a lot of holes in them, then suddenly the sub-d productivity level drops off a very steep cliff even for expert users. There is just fundamentally a tremendous amount of manual topology arranging work involved to try and accommodate precise holes when using sub-d, and it also gets easy to have a lot of imperfections like little bumps or ripples - those are often unwanted side effects when you're trying to force a sub-d topology into place that has to deal with a lot of constraints.

Check out this thread on the modo forum for a good example of this specific type of situation:
http://forums.luxology.com/topic.aspx?f=32&t=67986

There you'll see an example that's really clear, a whole lot of topology planning and just overall work needs to be done to do that in sub-d but meanwhile in MoI the same model is done more precisely in literally 20 seconds of work, see the youtube MoI screencap video that's linked to in the thread there.


So a lot of this depends on the particular qualities of the model at hand. When you get accustomed to using MoI you'll be able to recognize the kinds of things that will come together really quickly in it.


re:
> Therefore, a typical workflow for someone creating a game asset for Unity using subd modeling
> would be: create high res subd model, create low res version of same model,

So yeah it sounds like you've already got this figured out, but since you are not talking about retopo to produce quad topology for making sub-ds and are more focused on just making low poly, it is certainly worth a shot to just generate both a high poly and a low poly output from your same MoI model and see if that does the job for you at least for some particular cases.

NURBS models have a similarity to sub-d models in the sense that they define some "ideal smooth surface" and you can generate different approximations of that at export time, either high density or lower density output from the same base model.

- Michael

As I know, quad polygons only need where animation happens. So if I'm right for hard surfaces it's not necessary to have quads because hard surfaces do not bend like arms. And if I'm right it is possible to export from MOI High and low poly model for hard surfaces.

Hi Andrei, well it all depends on your own particular needs... it is pretty typical for hard surfaces to not animate by stretching or bending, they do tend to be animated by rigid body motion only and if that's the case then yes you're right.

But it could be possible for someone's particular project to need hard surfaces that bend and flex if they're going to try and make it move with some kind of personality or something like that... whether that applies to your specific project or not is something that only you can answer.

It's often better to base your requirements on what you are specifically trying to do. Ironclad abstract rules like "every polygon must be a quad" or the reverse don't really make sense, try to do what applies to your particular needs.

- Michael

hoi andrei!

for game props, engines and general game related modelling i would check out polycount.

http://www.polycount.com/forum/

tons of information in there.