Generative Design Visualization: From Fusion 360 to 3ds Max to the Client

STEVEN SCHAIN: Hi and welcome to CP500033, Generative Design Visualization from Fusion 360, to 3ds Max to the Client. My name is Steven Schain. I'm a designer and trainer. I've been teaching 3D animation since the 3D Studio DOS days gone by. And I've been an instructor since 1998, working with Autodesk since even before that.

And I trained under the 3D Professor. And I'm happy to say that this is my 10th year presenting at Autodesk University. I look forward to future Autodesk Universities, as well. So let's look at the learning objectives.

We'll take a look at the basic workflow for generative design in Fusion 360, discuss the connection between Fusion 360 and 3ds Max for an optimal design workflow. We'll talk about exporting from Fusion 360 and then importing a model into 3ds Max 2022. And we'll also discuss creating and rendering an animation of the final design in 3ds Max using the Arnold renderer.

So the project concept here is nothing unique. It's a simple bracket. The bracket design itself, though, will be unique. In that it's going to use four 1 by 2 cherry wood strips. And even though it will be a regular shape, standard shelf bracket, the client wants a unique design. So I decided that I would use generative design for this process to see if I can come up with something that's a little bit unique. And I also needed to use standard mounting hardware, so just easier to find a set of number 8 screws that I can use to put into the wall.

So let's look at generative design in Fusion 360. Right, so what is generative design? Well, it's a design exploration tool. It allows you to generate thousands of unique design ideas from one original concept. And it does this by sort of mimicking an evolution of the design. So it takes this evolutionary approach to how it creates the design itself, starting with a main structure, and then iterating from there.

In Fusion 360, it's cloud based. And it has an enormous number of applications, like automotive, aerospace, consumer goods, construction, artistic, and in this case, just an off-the-shelf design. So how does it work? Well, you define a set of parameters. Those parameters-- and these are the parameters, we're going to be using for this design. You have a set of design objectives. Are you trying to minimize mass or maximize the stiffness of the object?

And what are your limits? What's your safety factor for that design? Then, are there functional requirements? What are the constraints of that design? And materials, so in this case, I'm going to be using several different materials, plastic, and metal. So what kind of metal, or what kind of material is this going to be made from? And then performance requirements, so load forces. Does it need to hold 2 pounds or 200 pounds? What's the requirement for that?

And then lastly is the manufacturing method. Is this going to be machined? Is it going to be 3D printed? Here, we're going to set it up for unrestricted, which will just create an unrestricted design, freeform, no design-- no particular manufacturing style. We'll also do additive with x positive and z positive orientations, as well as a milling configuration for three axis milling.

So let's take a look at the generative design process. I have some prerecorded videos in this presentation with narration audio. So I will just be sitting here looking pretty for you guys. And enjoy this segment. This is going to be setting up a generative design in Fusion 360.

The basic parts of the shelf bracket have been created. I don't need to model the entire part. But I can unhide the original bracket that I designed. That was a blocky shelf bracket. I wanted to create something simple and quick to show the idea that I was going for.

When using generative design, you can start with bodies that are going to be used to preserve geometry and others to be used to make sure that geometry does not go into a certain part, or obstacle geometry. For example, this back piece here will be used as obstacle geometry and represent the wall. I want the mounting to be flat, where the bracket goes up against the wall. So I only need to create the parts that define the envelope of the design.

Opening the generative design workspace, I'll start by picking the preserve geometry parts first. I can do that by clicking on Preserve Geometry and picking the six bodies that will make up the essence of the shelf. That includes these four bodies here for the 1 by 2s and the two bodies for the mounting points. Then I'll select the obstacle geometry.

The obstacles are the areas that I don't want to intrude on. I'll select the bodies that make up the obstacles. Selecting the blocks here, and the bodies that represent the screw holes. There's one that's hidden. Opening the model component's bodies, I can hide body 13 and click on the hidden body. To see if all the objects have been selected for the right geometry type, they're color coded, with green being the preserved geometry and red being the obstacle geometry. And it appears that all the geometry is correct. So I'll move on.

The next thing I want to do is make a few minor changes to some of the geometry. I'll click Edit Model and enter the Edit Model space within the generative design workspace. The edits I make here will only affect the generative design model that I'm working on, not the original design geometry. Here, I've already made a few changes. So I'll just step through them.

The first two changes are offsetting the faces of the obstacle geometry for the screw holes. This is to ensure that I can get a screwdriver to the location, where I'll be needing to add the screws to both the wall and the 1 by 2 pockets. Moving the slider to the end, I decided to make the obstacle geometry for the wall mount access bigger. Then the last extrusion adds holes to the bottom of the 1 by 2 pockets. Without those holes, I can run into a problem when I'm trying to define a solution.

So I'll finish editing the model and move on to adding materials. I'll click on Study Materials. In the dialogue, I'll click over to Favorites for the library. And I created a PLA plastic material that I can use for this study. I'll drag that up into the study material. I'll leave it for all methods.

Then I'll change the library to the Fusion 360 material library. I'll open the plastic library subfolder and scroll down until I get to PET plastic. Then I'll drag that material into the study materials area. I want to have a metal in this study as well. So I'll open the metal folder. Scroll down until I get to aluminum 6061. Then click and drag that into the study material area. Once I have the study materials set up, I'll close the Study Materials dialogue.

Before I generate a study, I need to set design conditions. I'll hide the obstacle geometry. That will leave me with just the preserve geometry visible in the viewport. From the toolbar Design Conditions, I'll pick structural loads. In the structural loads dialogue, I'm going to pick the four faces that are on the bottom of the 1 by 2 preserve bodies. I could add independent forces to each 1 by 2 slot. But for this purpose, selecting them all works just fine.

Change the units to [INAUDIBLE] and set the magnitude to 10 pounds. There may never be more than a few pounds, but it's good to set a slightly larger value than what you think will be needed. Then I'll click OK and move on to adding constraints. From the toolbar, I'll click Structural Constraints.

I want to make sure the type is set to fixed. I'll rotate around the view a little bit. So I can see the backside of the green rings being used for mounting. Then, I'll pick the back face of the top and the back face of the bottom ring and click OK. This now sets up everything I need to begin generating a solution. In the Generate tab, I'll click Generate.

In the Generate dialog, it's important to make sure that you have enough cloud credits to generate the study. I'll select the study and click Generate. That opens the Explorer and begins processing the study. This is the part that can take some time. So take a break or close the browser and come back later. Once the studies are ready, the outcomes will show that they are converged.

You can choose one of the converged outcomes. Double click it and preview the results. From this point, you can create a design from the preview and export the model.

Once you've exported them, or once you have your design, you can use the Explorer in Fusion 360 to really get an idea of what you're looking at. Let's take a look at the Explorer built into Fusion 360 for your generative design study.

Before moving on to the next step, I wanted to take a moment and look at the Explore option in Fusion 360 to explore your designs. When working in the Explore option, or the Explorer, you have the ability to filter the outcomes. Those outcomes show the process status, the study that they were from, groups that are visually similar, and the manufacturing method and materials.

So I could isolate ones that are made from aluminum 6061. And I can see here that these are all the isolated solutions. If I wanted to see just the PET plastic or the PLA plastic, I can do the same. And clicking on them opens up a preview. This is the PET plastic. So the material itself is going to be the clear PET material. And if I wanted to check out, let's say, the aluminum, I can look at this one. This is pretty chunky.

And you can see where the holes for the screws are. That this is a bigger area for mounting. And it's made from one solid hunk of aluminum. Once you decide which one you want, then you can export it. This is how you explore your outcomes, see how they look. You can look at different properties. Let's reset that.

You can see a scatter plot of the mass versus the minimum safety factor. So here's one that's got a high safety factor and low mass. Or you could display the view as a table and also export the data for the outcomes to a CSV file. You can open up that in Excel. So there's a lot to do in the Explorer to see what your outcomes look like, how they compare, and then you can choose which outcome you would like to use.

So let's take a look at working with Fusion and getting a part to 3ds Max. So the Fusion to 3ds Max connection is done through a converted design. So the first thing to do is convert your design from the solution into a Fusion 360 editable design. Once that is converted, then you can export that from Fusion 360. Now, the file formats that are available, there's several that are available.

I like the Inventor file format. You can use Step, DWG, OBJ, there are several others that you can use. Personally, I like the Inventor file format, because it keeps the material if I want to use it. So let's take a look at converting a design and exporting from Fusion 360.

Now that I've chosen one of the outcomes and created a design from it, I can export the model to a format that can be imported into 3ds Max. For this, I'm saving to an Inventor IPT file. Exporting a model from Autodesk Fusion 360 is pretty straightforward. With the design from the chosen outcome open, I'll open the File menu and choose Export.

From the Export dialog, I can select where I want the file to go and what file type I want. From the File Type dropdown, I'll select Autodesk Inventor 2019 files, IPT file type. I'll leave the name as is, which is the name of the original file. Then click Export.

IPT files have to be converted in the cloud. So once you click the Export button, you'll see a job status dialog showing the progress of the conversion. Once the export is complete, you're ready to import the file into 3ds Max.

Now, let's take a look at importing the file into 3ds Max. You've exported the Inventor file from Fusion. And you want to import it into 3ds Max. And again, 3ds Max has a number of file formats that are available to use. Once you've imported it, you can position it, add materials, and do other things. Let's take a look at the import process into 3ds Max and how that works.

Once I've created my generative design, I can bring it into 3ds Max to be able to render it and show my client what it's going to look like. 3ds Max is an incredible tool. It's sort of a Swiss army knife of 3D animation packages. I can bring in a huge variety of 3D formats to animate or render them.

Before I import the file, I need to decide if I'm going to build the scene from scratch or use a pre-existing scene. In this case, I pre-defined everything in this scene, except for the brackets themselves. I've even created the wood 1 by 2s in the right location, added lighting, an animated camera, and an environment. This cuts down on time, because I have pre-defined the environments for multiple situations. This environment is the corner of a room that I can repurpose for other rendering jobs.

To bring the Inventor part file into 3ds Max, from the File menu, I'll choose Import, Import to open the Select File to Import dialog. Here, I've exported the design from Fusion 360 into the 3ds Max project folder I'm using for this project. And just a quick aside about project folders, use them. Unless the company you work for has a specific project management tool, using project folders in 3ds Max is a great way to keep yourself organized.

You can store your models, textures, IES lighting files, and other assets that are specific to the project you're working on. Back in the Import dialog, I'll open the Files of Type dropdown. And you can see just how many file types 3ds Max is able to import. If you keep all formats selected, 3ds Max will use the extension of the file selected to determine the type of file being imported.

Here, I'll select the generative design bracket and click Open. This opens the Autodesk Inventor File Import dialog. Depending on the needs you have, you can import the objects as either body objects or meshes. I prefer meshes. And I can increase or decrease the quality of the mesh by adjusting the mesh resolution slider left or right.

The real keys in this dialog are whether you're going to merge the object with the current scene or replace the current scene with it and the Inventor vertical direction. I'm going to merge this with the current scene and set the vertical direction to Z. You can rotate the files within 3ds Max. But having the initial orientation correct saves time. Once the file is imported, I can move it into place, make copies if I need to, and then apply materials.

And once I've done that, and I have the parts in the scene, I can then start working on the configuration for rendering. Since this is a pre-defined scene, I'll go through the material configurations, lighting configurations, and Arnold configuration, and then look at rendering the scene to a file. So let's take a look at that. And rendering the final scene.

In 3ds Max is a broad subject. However, in this case, I decided I wanted to do a simple camera move that showed off the look of the bracket and the shelf setup. After I imported the Inventor file, I positioned it and made a mirror copy that's placed on the other side of the 1 by 2s. When it comes to creating materials, there are two camps. One camp says create materials for each part as you build it. The other says build your objects then create your materials.

There is a third option. It's usually the way I go. And it's a hybrid approach, where I'll create a material for certain objects if it's easy at the time I build the object. And other materials that might take longer and would require special textures, I'll do later. I'll go ahead and open the material editor and review the materials that are used in the scene. When working with textures in 3ds Max and rendering with Arnold, my preference is to work with the physical material instead of the Arnold standard surface shader.

The only reason for this is that the physical material has fewer parameters. And it's quicker to work with. When setting up a material like the cherry wood for the 1 by 2s, I only worry about a few parameters. Those are the base color, reflections, and bump map. For this wood, the base color is coming from the advanced wood shader. So is the reflection roughness and the bump map.

I'm also a big fan of Autodesk's advance wood material. Clicking on it, you can see that it has an extraordinary amount of detail and contains several presets that really make life easier when creating wood textures. I also can make use of the three outputs for diffuse color, roughness, and bump, to plug directly into the material. Or in this case, I'm using an OSL shader to adjust the gamma a little bit and give the color a richer look.

Before moving on to how I set up the lights, I want to talk briefly about the environment shader that I'm using in the scene. I'll navigate over to the OSL HDRI environment material and select it. Aside from the photometric lights that are in the scene, I'm using this Studio Lighting HDR image to add additional light coming from a background image.

Opening the Environment and Effects dialog, I can use the environment shader in the environment map slot. Dragging from the output of the environment shader to the map slot in the common parameters, I'll instance the shader. So I can control it from the material editor.

The only change I made to the OSL environment shader is to bring the exposure down to negative 1. This reduces the amount of light contributed by this environment. Before I close the Environment and Effects dialog, I want to point out that this scene uses the physical camera exposure control. This allows me to set the exposure value differently for each camera in the scene. But here, I only have one camera.

I'll go ahead and close the Slate Material editor, the Environment and Effects dialog, and select one of the lights that I have placed in the ceiling. If I wanted to, I could use an Arnold light in this scene instead of a photometric light. The only time this really matters is if you want to switch renderers. Using the photometric lights tend to be more compatible with other rendering packages on the market.

This light's set up to use a photometric web for light distribution. Using a photometric web can accomplish two tasks with a light. The first is to make the distribution pattern more realistic by simulating real world light distribution patterns. And the second is to control the illumination value of the light. The IES or photometric data file contains that information.

The last element that was edited is the camera. When I select the camera look in the timeline, I notice that there are two keyframes. These keyframes indicate the start and end position of the camera animation itself. The animation lasts for 5 seconds. And when I scrub the time slider, I can see, in the three orthographic viewpoints, that the camera is moving, and the results of it in the camera view. It's not a complex animation. But it will get the point across to my client.

Remember, don't complicate things. It makes life easier if you just keep things simple. Lastly, I need to configure the Arnold renderer. Clicking on the render setup button opens the render setup dialog. Under the common parameters, I'm going to set the time output. So I render the active time segment. And to save time, I'm going to render every other frame. I can do this by setting the every nth frame option to a value of 2.

I'll set the output size width value to 800 and press Enter. That will automatically set the height to a value of 450. This is because I have it set to the HDTV video setting for the output size. And the aspect ratio is fixed at 1.77. Next, under render output, I'll click the Files option to set the file name and location.

I'm going to use a name template. From the Name Template dropdown, I'm going to pick the Scene Camera View option. And I will add an extra underscore at the end of the template. That will separate the frame number from the image file name. From the Save As Type dropdown, I'll select the PNG image file. Then I'll select a default render output path and just click Save.

In the PNG configuration dialog, I'll set the RGB value to 24-bit and turn off the alpha channel, then click OK. On the Arnold Render tab, I'm going to speed things up a little bit for the camera anti-aliasing and drop that value to 2, then bump the diffuse ray depth to 2 and the specular ray depth to 2. To speed the rendering up a little bit more, I'll turn on adaptive sampling for anti-aliasing. I find that the default settings work just fine for most of the renderings I need to do. So I'll leave them as is.

Before I click Render, I'll change the view to render the physical camera view and then lock the view. Now, I'm ready to press render and start rendering the animation. Depending on the computer, this could take a while.

Well, I want to thank you very much for attending my class. I hope you enjoyed it. Again, my name is Steven Schain, the 3D Professor. If you need to reach me, my email is in the slide. I look forward to talking with you. And enjoy the rest of your conference.