Making Of 'A Mazda CX-9'

Bernhard Rieder and Michael Seidl:
We both are very happy about our good teamwork and progress in the 3D industry, and it was a long journey to get our skills and experience and to know that we fit perfectly together. There are several reasons for this, like our passion for photo-real rendering, and the fun we have every time working for productions. We also know that we are reliable and ambitious and we take our job very seriously. After several years we created our own company and started our way in the 3D industry. One of our strong points is our passion for learning and understanding different render systems, and developing material and lighting setups for photo-real rendering. But the most important thing is, that we love to do 3D "stuff", and this keeps you "alive" if something goes wrong in a project. You can see the car in action at: http://www.mazdausa.com, simply by clicking on the Mazda CX-9.

Getting started

Welcome to the project overview about the Mazda CX-9. A good start is always to collect as many reference images as possible. It is very important to have many pictures of the car from different views, to make sure that the car is modelled as accurately as possible (Fig.01).

Fig. 01

Fig. 01

Setting up the blueprints

The first step is setting up the blueprints, therefore we have to make sure that all blueprints align. You can check this easily in Photoshop, just drop the side and the front view in one sheet and follow the major car lines. Don't panic if your blueprints are not 100 percent correct. Do the same comparison with the other views (top-side, back-side, and so on) (Fig.02). After that we align the different pictures to planes in 3DS Max. Make sure that their measurements are the same as their dimensions in Photoshop (Fig.03).

Fig. 02

Fig. 02

Fig. 03

Fig. 03

Modelling

There are many different ways to model a car. A good way is to start modelling in front view. Make a plane and align the major lines with the hood of the car. After everything is correctly aligned, add a symmetry modifier on top of it. The next step is to chamfer the edges where a sharp seam is necessary. Last but not least, add a Turbo Smooth modifier to it (Fig.04-05). With this technique you can do the whole modelling process. Always start with a rough shape of the car, and then add more and more detail (Fig.06).

Fig. 04

Fig. 04

Fig. 05

Fig. 05

Fig. 06

Fig. 06

Materials

When I start doing the materials, I always think about how they would act in real live. This makes the creation process much easier. For example, with the rim material (Fig.07), I used simply Vraymtl with a grey diffuse colour and a Reflection glossiness of 0,78 (Fig.08).

Fig. 07

Fig. 07

Fig. 08

Fig. 08

Tyre material

The tyre material has a fall off in the diffuse slot going from black to a dark grey value, to give the tyre material a little bit more depth (Fig.09). I always get asked about the headlight material, which was very easily done, simply take a look at the screenshot shown in Fig.10. In the bump map slot I put a gradient ramp, to give the reflection more variation (Fig.11).

Fig. 09

Fig. 09

Fig. 10

Fig. 10

Fig. 11

Fig. 11

Environment

I always start by applying UVW to all parts in my scene, and with the help of the free Texporter plug-in I can export the wire to Photoshop. After that I browse through the 3DTotal Texture CDs and pick the textures that I would like to use and then save them in my local project folder. The great advantage of those textures are that they are perfectly seamless, and so the texture process is so much easier. To give the wall a more "random" feeling I used the stamp tool to add some more structure to it. In Fig.12 you can see the side wall. Now to add some dirt to the bottom of the image, so I use a few of the dirt images from 3DTotal's V5 Dirt & Graffiti" CD. First of all I added a new texture (3DTotal Texture Aged & Stressed) (Fig.13). In the next step I added a layer mask to our previous created layer and put the dirt map in the alpha slot (Fig.14-15). So far, so good. In Fig.16 you can see the added dirt mask. With the alpha slot still selected you can then use your pen tool set to black to paint transparent areas. If you want to see more dirt in some areas set the pen tool to white, and draw some dirt on the wall. The same process works if you want to add some graffiti to the wall. Just load up your image and apply your alpha map and adjust your blending from both surfaces with the pen tool in the alpha slot (Fig.17). I used this same technique for all other parts in the scene.

Fig. 12

Fig. 12

Fig. 13

Fig. 13

Fig. 14a

Fig. 14a

Fig. 14b

Fig. 14b

Fig. 15

Fig. 15

Fig. 16

Fig. 16

Exterior Lighting Setup

In this project I used HDRI for lighting. When you load your HDRI map, a window opens. I always move the White Points in that way, so that most of the main light source (in this case the sun) is purple. I paste this value into the RGB level output of the bitmap channel where the HDRI map is loaded (Fig 18-19). As you can see, I put an HDRI in the skylight and in the reflection environment slot. As you might also have noticed, when I used HDRI I always use the bitmap map, not the Vrayhdri map. This is because I have much more control over the HDRI. For example, with the RGB offset I can easily chance the contrast of the HDRI (Fig.20). The next step is where I rotate the HDRI map so that the main light is coming from the front. The preview what you see in the material editor is that which is opposite your model (Fig.21).

Fig. 17

Fig. 17

Fig. 18

Fig. 18

Fig. 19

Fig. 19

Fig. 20

Fig. 20

Studio Lighting Setup

For the studio lighting setup I always use "Lichtwannen" (Licht = light, Wannen = bowl) (Fig.22). This gives me a nice control over the reflection and the setup is pretty simple. As you can see, I only used a tube and a plan and applied a Vraymtl material to it. In the diffuse map I put an output map with an intensity of 1,5. In the output map I put a gradient ramp going from black to white and black again (Fig.23-24). Make sure that you turn off "Cast Shadow" in the object properties for both "Lichtwannen" (Fig.25). For the studio light I only used a skylight with the intensity set to 1. This gives me those soft, studio car shadows (Fig.26).

Fig. 21

Fig. 21

Fig. 22

Fig. 22

Fig. 23

Fig. 23

Fig. 24

Fig. 24

Fig. 25

Fig. 25

Rendering Setup

If you are using GI for your light solution, make sure you turn off "Default Lights" (Fig.27). The reason why I used a secondary rays bias value greater than 0 here is that, I had some overlapping meshes and even some objects with a height of "0", and VRay tends to make ugly artifacts when this happens. So I used 0,01 to get rid of those render errors. For the image sampler I used Catmul Rom, because this filter gives me a sharp render result (Fig.28). Indirect illumination settings can be seen in Fig.29, the Irradiance map settings can be seen in Fig.30, and finally my settings for the light cache are in Fig.31.

Fig. 26

Fig. 26

Fig. 27

Fig. 27

Fig. 28

Fig. 28

Fig. 29

Fig. 29

Fig. 30

Fig. 30

Colour Correction

A very important part of creating CG images is the post work. With this image, I need to correct the colour range of the image in Photoshop. OK, so our images have come straight out of 3DS Max. As you may have noticed, the image has a very dominating blue tint, and in the next few steps we need to correct that (Fig.32). First of all we add a threshold modifier to the image (Fig.33). After that we move the slider to the left side. Now you can see that only the darkest parts of the image are remaining. Hold down the Shift key (the mouse icon changes into a cross) and select the darkest area (Fig.34). Now move the slider in the opposite direction and do the same. You should now see two crosses, marked with "1" and "2". Hit the cancel button. OK, now we have set the range for the white and the black points. Now it's time to select the midtones (grey). Add a layer on top of it and fill it with 50 percent grey and set the layer blending to "different". After that add a threshold modifier again to the new layer. Now move the Slider to the left. The remaining colours you should see are grey. Hold down the Shift key and select them again. Finally press the cancel button again. You should now have 3 crosses, marked from "1" to "3" and you can now delete the new layer (Fig.35). Now it is time to add a Curve Modifier to the image. Under the button "Option" you see 3 icons for the eyedropper tools. The first is for the black points, second for grey, and the third for the white points. Select them in that order. To select them you have to hold down the Shift key or press Caps Lock on your keyboard (your cursor will change into a cross) (Fig.36). Now hit OK, and you will see the colour corrected image. (Fig.37)

Fig. 31

Fig. 31

Fig. 32

Fig. 32

Fig. 33

Fig. 33

Fig. 34

Fig. 34

Fig. 35

Fig. 35

Fig. 36

Fig. 36

I hope you have enjoyed reading through this brief project overview. If you have any questions, please don't hesitate to contact me. You can see more images on my website and can also download the discussed shaders.

Fetching comments...

Post a comment