Creating A Pontiac Trans Am
Hello everyone. In this tutorial I'm going to explain the process of creating a Trans Am. Before I start let me give you a brief review of this tutorial:
- Section 1 - Modeling the Car
- Section 2 - Car Paint, Lighting and Rendering
Section 1 - Modeling the car
First of all, let's talk a little about the blueprint or pattern we are about to use. When it comes to car modeling, some people always recommend using all three or four blueprints of main views. Well, in this case we are modeling a car whose blueprints are difficult to find. So we're going to try to get the idea from pictures instead of blueprints. I used some images as reference; generally the side view is always needed. You can try to figure out how the model will look in other views according to other images you're using as references. For my own way of modeling, I use only side images. Now this might seem difficult but when it comes to a model with no blueprints or a concept car with undone sketches it's great to have the skill to "imagine and design" the missing dimensions.
You have only got to set the images in the Max viewport. Just create a plane with the size of your image, and start modeling as described next.
The modeling method we're going to use is a simple and effective one. First focus on the part you want to model, and then start drawing on the side view with a simple Line tool. Fuse the end vertices, and easily try to make a wire patch. Take a look at the image below to get an idea what your spline should look like. Remember to create polygons in edit spline (by polygon I mean an area made with four segments and fused vertices) so that the Surface Modifier can turn the spline polygon into a 3D surface.
After creating the wire model apply the Surface modifier to the spline and set the Steps level to 0. Don't worry about the mesh; it's not supposed to appear nice and smooth. When everything is okay with your Surface applied mesh, right click on it and choose Convert to Editable Poly. At this point you're facing edit poly vertices. The key to success is the Cut and also the Quick Slice tool. These tools are active in any sub-object level. Use cuttings to make the model more and more similar to the main one (Fig.01 - 02).
There's something important about using the Cut and Quick Slice tools. Try to keep the mesh clean. What I mean is, don't use the Cut tool more than it is required. The more a surface is divided into polygons, the sharper result you'll get after smoothing, which is an advantage if you need some lines to appear sharper. But you are the one who controls this balance, so when you need a Hard Edge or a sharp line on your surface, use the Cut or Slice tools around that edge to make the smoother behave harder on that edge. Take a look at this image and notice the number of edges used to make the model appear non-smooth one around that part (Fig.03).
Alright, let's talk about the smoother a little bit. When you're editing the mesh in Editable Poly mode, there's a tab named Subdivision Surfac". Find it and check the Use NURBS option to smooth the model, with level one at least and level three at most. You can uncheck this option to continue editing, but do it every few changes you make to check out how your final result will look. Remember that the NURBS method performs a perfect smoothing on the model, but if you want the surface (especially a car body surface) to look nice, control the edges and polygons with the tools mentioned above.
I kept modeling a little more with the same method to get the result seen in Fig.04.
The method for modeling the rest of the object is the same. Start with a Line, continue your way to Edit Spline, add, move and shape new vertices and segments according to the image and make a spline cage ready for Surface Modifier to serve you with the third dimension. Set the Steps level to zero and convert the mesh to editable poly. Add, move, and shape new vertices and edges in editable poly mode and use NURBS to smooth the mesh. Finding and using a Front View Blueprint or image is very useful to guide us to how much we should move the vertices along the Z Axis. You can add this image later just to fix up the model and make it more accurate.
So we have this after a few steps (Fig.05 - 06).
Section 2 - Modeling the car
In this section we are going to talk about Car Paint Material, lighting, rendering and the Photoshop settings to use to get a good result. These steps lead us to get a good render without using a lot of time. I'm going to use the mental ray engine for the rendering passes, but the method could be used with any other engines too. I've chosen mental ray as it is a part of 3ds Max and so we can easily apply ready to use shaders and settings.
So, let's start with the car body material. One of the greatest materials included in mental ray is Car Paint Material, which could be found in the Materials list if you set the rendering engine to mental ray. So press F10 and in the Assign Renderer tab change the Production renderer to mental ray. Open up the material editor and choose an empty slot. Set the material to Car Paint Material. (Mental ray materials are shown with a yellow sphere).
There are a few tabs in Car Paint Material. Let's have a brief review:
- Diffuse Coloring: here you set the color of the car body, controlling weight and bias
- Flakes: here you set the density, color, scale and other options of the body paint through tiny metallic pieces called Flakes
- Specular Reflections: here you set how the body paint will respond against light and GI
- Reflectivity: here you set how reflective and shiny your body paint is
- Dirty Layer: here you can make the surface look dirty by applying a dirty layer
- Shaders: here you can apply any desired map to any parameter you need to.
We start with setting the Diffuse Coloring for the material. Change the Base Color to our desired one but to get better results let's use a Falloff map instead of a solid color. Also to make some tiny metallic flakes on the surface, set the flake color to a bright one (better use a Falloff map) and control the weight, size and density with the parameters available. This is how the material should look like so far (Fig.07).
When applying the Falloff map, change the Falloff type to Fresnel to get better results. Ok now let's set the reflections for the material. It's easy; first of all set the Reflection Color. We use a Falloff map for this part again. Remember, the brighter the color, the more reflective a surface we will get. But after setting the color you can change the reflection weight (both on facing and edges) to control how the surface behaves as a reflective object. There's also a very useful parameter named Glossy Reflections Spread, which I'll explain a little later, but just for your information this parameter helps us to make a blurry reflection on the surface, with the sample numbers under our control.
Another useful tab is the Specular Reflections. Here you can set the specular point's size, colors and weight. Keep the first color (Specular Color #1) near to your Diffuse Base Color, but to get different results you may also want to set a different color or weight. In Fig.08 you can check out the settings I've used for the reflections.
Now it's time to discuss the Pontiac Eagle logo on the hood. There are a few simple steps. Start with an empty material slot and set it to Composite. As you may know, there's a Base Material in the composite box, which we should set to the Car Paint we just made. The next step is to take the Mat1 and set it to a Standard material. This material which comes completely over the Base Material would be the Eagle Sign. So let's keep it simple; use a Blinn shader with any color you desire. Now if you render you'll see the hood completely covered with the upper layer, the Standard material. The only thing we should do now is to mask this material over the Car Paint with a texture like this (Fig.09).
Click on the Opacity Map Channel and set it to Bitmap. Select the black and white eagle texture and then click on the Show Map In Viewport option to set the coordinates. The final step is to apply a UVW Map modifier to the hood object, with simple planar mapping. We can also use the same method for the tires. Here you can check out the material settings and the result achieved (Fig.10 - 12).
After setting the materials, it's time to set the scene for rendering our model.
We start with creating the light sources. Add a few MR Spots around the objects; I created one and copied a few more as instances, from different angles. For the white boxes, just create some boxes (or any other standard primitive) with this material. Assign a standard material to the objects, with the Self Illumination parameter set to 100, and a Falloff map for the diffuse channel. Then set both colors of Falloff to pure white, scroll sown to the Output tab and set the RGB Level Parameter a number greater than 1.5. And we're done here (Fig.13)!
Set the Intensity of the lights to a low value (because of their quantity) and set the Hotspot Parameter to the minimum value with the Falloff Parameter at a higher amount. Leave the shadow type to the default type (Ray traced) and click Render. You should get a result like this (Fig.14).
As we are not using GI and Final Gather for this render, to get a better result we'll blend a few passes in Photoshop. Take Fig.14 as one of the passes, but for the smooth GI simulation and perfect shadow settings we will use another method.
Press F10 to reach the Render Panel. Click on the Processing tab and check the Enable option under Material Override section. Click the material and select mental ray material from the list. Open the material editor and drag the override material to an empty slot. Select the Surface map and assign an Ambient/Reflective Occlusion to its map channel. These settings will help you get a clear pass for the missing GI. There's a Samples parameter which could be set to 16 or 32 for test renders, 512 or 1024 for final renders. Check out the settings in Fig.15.
Let's render with the parameters set as above. We will get a result like this (Fig.16).
As we are about to blend the passes in Photoshop, we will also need an image for masking. For this, select the whole car and assign a solid black material with the background image set to white. Render the image, save it as the third image and keep in mind that all of three passes should have the same aspect ratio, size and sample quality (Fig.17).
Alright now we're done in Max. Open Photoshop and open the three images we just rendered in 3ds Max. The way we are about to blend these images is a very great technique but not a unique one. I mean you might blend them in your own way, if you just get the idea and basic knowledge about this stuff. Open the second pass, the image we got from material override (Fig.18).
Ok now we need to put the main render on this layer, but we only need the car not the environment. So let's use the third image to mask the image over the background layer. To do the mask effect, first drag and drop the main image on the background layer and then click on the Add Layer Mask parameter under the Layers panel. Then Alt + Click the second layer slot and just copy and paste the mask image to get the result needed. Then select the layer and choose Darken in the Blending Mode list. As I said you may get different results depending on different color types, shadow types and other factors so you may want to try other modes, but the process is the same. After applying the changes on the second layer we should have the result shown in Fig.19.
So far we have brought the missing smooth shadows back, but it's not yet the result we are looking for. So duplicate the layer with different Blending Modes and different Opacity values to get a better result. The new layers have various blending modes like Color Dodge, Lighter Color, Soft Light and Hard Light to get a clearer image. Take a look at this image to get the idea and take note of the blending modes and their opacity values (Fig.20).
So that was the studio render of our Pontiac. Remember that we chose this method to get a good render with the least time, but if you want to get the same result directly from mental ray with no post processing you can let the engine process and create the GI. To do so, using a Skylight would be good idea. Create a Skylight and activate Global Illumination and Final Gather in the mental ray settings. Test your renders with the lowest sampling quality and try out different lighting setups to get the result desired. Also using a HDRI map in the environment background will help your image to get more realistic reflections and quality.
And here's the finished image (Fig.21).