To make cool shit? Nope. Think bigger.

Give up?

Whether you realize it or not, the sole purpose of all 3d art is to make something look believable.

That’s it. There are no exceptions.

If your image doesn’t relate to the real world in some way, the audience will feel disconnected and become disinterested. It doesn’t matter if your rendering an architectural fly-through of a house or an orc warrior fighting his way out of a volcano, you are still taking something that exists in your mind and producing it as a picture and hoping the audience believes it.

You’re doomed, unless…

Let’s say you spend hours intricately modeling each and every nail of a door frame, you piece together textures to create flawless materials, and you spend a solid week on the lighting setup.  Well guess what? If you hit render now your scene is still going to look fake. The reason for this is simple: You haven’t added camera imperfections.

It sounds like a joke, but it’s absolutely true. Photography is the single most important thing to understand when it comes to learning 3d, but for some reason most artists choose to ignore it.

Do I have your attention yet?

When you take a real photo with a real camera, do you realize how many flaws are being built into the photo?

Just to name a few:

• Chromatic Abberation
• Vignetting
• Soft Glare
• Light rays
• Reflecting glare
• Bloom
• Lens Flare
• Glare burnout
• Ghost glare
• Depth of Field
• Motion Blur
• Lens Distortion
• Lens dust, scratches, sweat, fingerprints
• Film developing artifacts
• Color grading

Now guess how many of these flaws occur when producing a computer generated image?

NONE. ZERO. ZIP. NADA.

When you hit F12 you will produce a perfect still. Every. Single. Time.

For example, take a look at this image by the talented artist, Marek Denko:

Aside from the perfect lighting, flawless modeling and impressive materials. What else has he incorporated?

Effects. And lots of them.

• Depth of field
• Chromatic Abberation
• Lens scratches, dust & dirt
• Color Grading
• Bloom
• Reflecting glare

These are all things are done outside of the 3d viewport and added in post production. That means that after he slaved over every piece of detail in the scene, he flipped to the compositor and continued working. That is what pushed this scene over the edge.

Have you ever wondered how Pixar achieved that authentic film look in Wall-E?

This was actually the result of many months of work. Before they began working on the meat of the production, they focused entirely on trying to replicate the look of live footage in their 3d software. They even went as far as to consult the Director of Photography from No Country for Old Men, on tips for creating real camera and lighting setups.

“We rented some equipment and used the live-action DP [Marty Rosenberg] who eventually shot some of the live-action elements. He helped us do some lens tests. Our depth of field, our cameras never look as we expect them to.

“Life is nothing but imperfection and the computer likes perfection, so we spent probably 90% of our time putting in all of the imperfections, whether it’s in the design of something or just the unconscious stuff. How the camera lens works in [a real] housing is never perfect, and we tried to put those imperfections [into the virtual camera] so that everything looks like you’re in familiar [live-action] territory.”

-Andrew Stanton

(source: AWN)

The result was a very slick, very believable environment for Wall-E to explore.

Why using the compositor is crucial to your success

The most common question people ask is, why would I want to incorporate camera flaws into my renders? After all camera flaws are exactly that, flaws. So wouldn’t an image that is clean from these flaws look better?

Nope. Let me explain why.

When you look at a white car sitting in the hot sun, your eyes expect to see a reflecting glare. When you look closely at your spoon during breakfast, your eyes expect the bowl of cereal in the background to be out of focus. And at night time when you look at a street lamp, your eyes expect to see rays of light.

I could go on, but I think you get the picture.

Our eyes have become so accustomed to seeing these imperfections that they have become apart of the object. And when they are missing, the viewer spots a fake.

Now don’t get me wrong, lighting, materials, textures and lighting are all important and I’m not pretending they aren’t, but until you learn how to replicate the look of a real camera in the compositor, you can kiss believability goodbye.

If you are interested in exploring this topic further, I am writing an eBook called The Wow Factor. It’s not available yet, but you can register your interest here: http://wowfactorbook.com

Keen to make my own, I experimented with different cloud shapes, trying to see what worked and what didn’t. Some of you may have already seen these attempts on my facebook fanpage.

But when I was stuck at Brisbane airport for 3 hours, I thought I might try my hand at…

Speed Modeling a Cloudscape

So armed with a laptop, I setup myself up at a vacant flight check-in desk and got to work. I honestly expected a flight attendant to kick me out, but amazingly no one bothered me. I knew I had less than an hour of battery life on my laptop, so I had to be fast.

Here’s the video:

Now that you’ve seen how easy it is, do you want to make your own?

Tutorial

Ready? Let's begin...

We will be using Blender 2.5 so if you don’t have it already, download it here.

Step 1

Create a new scene.

With the default cube selected, hit Ctrl+4 to add a subsurf modifier. Then go to the modifier stacker and hit ‘Apply‘. This will give us lots of vertices to play with.

Step 2

Press T to bring up the toolbar, and under the shading options hit Smooth.

Step 3

To make the cloud fluffy we will be using a combination of displacement modifiers.

But before we can add any displacement we need to add a new material. For now, we will leave all the settings as they are.

Step 4

Switch to the texture panel and click New to add the first texture.

Step 5

Leave the texture type as Cloud, but set the noise type to Hard and the Size to 1.00

Step 6

Next switch to the modifier stacker and add a Displacement modifier. Enter the name of the texture we just created and set the Strength to 0.40

Step 7

As you can see, the cloud texture we have just created is now effecting the geometry of the sphere.

To make it look smoother, add a Subsurf modifier underneath the displacement modifier. Change the Render count to 1.

Step 8

Add a second texture under the one we have already created.

Step 9

Set the noise size to 0.05 and the depth to 6.

Step 10

Add a second displacement modifier underneath the subsurf modifier we previously created. Set the texture name to the newly created texture and the strength to 0.30

Step 11

Add another subsurf modifier, making sure it’s underneath the last displacement modifier.

Step 12

Switch back to the texture panel and add a third and final texture. Leave the type to clouds  but change the size to 0 and depth to 6.

Step 13

Add another displacement modifier and set the texture field to the name of the texture we just created, and set the strength to 0.10

Step 14

Add a third and final subsurf modifier underneath that displacement modifier.

If you’ve done everything right, your object should now look like this:

Step 15

At the moment our cloud is spherical and looks very unrealistic.

Go into edit mode (TAB) and press O to activate proportional editing mode. We want to give the cloud a flat bottom, so select the vertice at the very bottom of your sphere and press G.

Step 16

Move the vertice upwards along the Z-axis until your sphere looks like a piece of dough that’s fallen flat. You can scale the amount of proportional editing by scrolling your mouse wheel.

Step 17

Use this method to shape the rest of the cloud. There’s no right or wrong shape as every cloud is different do whatever looks right to you. My cloud looks like a meringue

Step 18

We will now add the material.

Go to the materials panel, and set the material type to Volume. To give the cloud a thicker appearance, increase the density scale to 1.5. To exaggerate the effect of the bright sun, set the reflection to 1.3. Set the resolution to 100 and the transparency type to Raytrace. I’ve found these to give the best results. Finally set the Step Size to 0.02 so it doesn’t look too noisy.

Step 19

We don’t want any of the textures we created to effect the clouds material so turn all of them off. This will not effect the displacement modifier.

Step 20

With the modeling and materials out of the way, now we just need to light and render it.

Position the camera so it’s facing the side of the cloud. The easiest way to do this is to go into side view (Numpad 1) and hit Ctrl+Alt+0

Step 21

Next change the default lamp to a Sun Lamp and set the energy to 3. You may also wish to give it a slightly yellow color.

Step 22

Although the sun is the main light source, the sky also gives off a faint blue glow, and will make your cloud look a lot more realistic.

For this, add a second sun lamp and point it directly downwards. Set the colour to a light blue and the energy to 2.

Save your work. Render now and you should get a result similar to this:

Which is great except the sky is grey.

Go to the World panel. Check the Blend Sky box and set the horizon color to a very light blue and the Zenith Color to a light blue.

Render now and you’re finished!

As always I’m keen to see whatever creations you guys come up with, so if you make something, upload the render to pasteall.org/pic and post it in the comments!

Enter: The handheld video age. One of the newest methods for getting seen at the moment is, believe it or not, via the cell phone. The following is an excerpt from a discussion on getting a job at Crytek.

“Someone I met recently had his [demo] reel on his cell phone. It was really awesome to start up a conversation with someone and start talking about 3d and be able to see a reel mid conversation, really great idea I thought… you never know who might be sitting next to you at the bar having a drink.”

Touche. If you want to get noticed, you need get seen. Obviously iPods aren’t cell phones but I’m sure you get the idea. Speaking to someone face-to-face is far more personal than sending a link over the internet, and you are far more likely to get a call back.

In this tutorial, I’ll be covering the steps to render an animation to your iPod. To demonstrate, I’ll be rendering this simple fluid sim animation, but feel free to use whatever you like.

Render settings

Before rendering your animation it’s important to keep in mind that the maximum resolution an iPod is capable of playing is 640 x 480. The actual resolution of the screen is only 320 x 240, but it helps to upload the maximum incase you connect it to your TV.

So go ahead and set the dimensions of your animation to 640 x 480 and your output format to PNG.

This will export each frame as a separate PNG and place it in your output folder which by default is /tmp but you can change that by clicking the folder icon in the output panel.

When you are ready hit the big ANIM button (or press Ctrl + F12). When your animation has finished you should see each frame saved as a separate file in your specified folder.

Now to put all those loose frames into a single mov file…

Compiling the movie

To compile all our frames into a single video file we are going to make use of the Video Sequence editor. So with our blend file still open, change the 3d viewer window to Video Sequence Editor

In the new window, click Add>Images

Navigate to where you saved your frames and select every frame but holding down your right mouse button as you drag over them. Files will turn purple once they are selected. Once you’ve done that, hit ‘Select Images’.

You will now be taken back to the editor screen where you will see a purple box with a number underneath it. This is your timeline which contains all your frames. The number underneath is the frame that you will start the timeline at. Move the bar to frame 1.

In your render settings panel, change the output format from PNG to Quicktime. We also need to change the codec settings, so click Set Format.

In the window that opens next, you will need to set the compression type to MPEG-4, because it appears to be the only codec type that the iPod actually recognizes. H.264 does not work.

Go ahead and change the settings as follows:

Finally we need to click the Do Sequence button in your Scene panel (f10).

This will tell blender to follow the instructions in the sequence editor instead of rendering our scene.

Now all you need to do is hit the big ANIM button again. This will open the render window which will quickly compile all our frames into a neatly packed .mov file

Importing into iTunes

Open iTunes and click the Movies tab

Import the file by going to File>Add File to Library…

Navigate to where you saved your .mov file and click Open.

Connect your iPod and bobs your uncle!

So there you have it! A powerful promoting tool that tucks nicely in your pocket but can be whipped out at a moments notice. This will no doubt give you an advantage at the next Siggraph conference!

If you have an animation you have previously rendered, but would like to put on your iPod, I would highly recommend you use the open source program MediaCoder which will convert it to the correct format.

Do you know of any other unique ways to promote your work? Write it in the comments below!

For this tutorial we will be using Blender v2.48a, Indigo v1.1.18 and Blendigo v1.1.14 and Windows XP.

Let’s start with the basics:
The computer that starts a render is called the ‘Master’.
The computers that connect to the master are called ‘Slaves’.
A computer that is both managing the farm and chipping in it’s own CPU power is called a ‘Working Master’.

Step 1 – Installing Indigo

Download Indigo v1.1.18 and extract it to your C:\Program Files folder. It should look like this:

The exact location isn’t important, just make sure you alter the path in the next step if you decide to change it.

Step 2 – Creating the BAT file

BAT files are simply a set of written instructions for windows to follow. It produces the exact same result as typing into command prompt, but making a BAT file saves you having to retype it each time you use it.

You can quickly create a BAT file using notepad. Open it up and type the following:
cd C:\Program Files\indigo_v1.1.18
indigo.exe -n s

This is telling Windows to go to C:\Program Files\indigo_v1.1.18 and run indigo.exe in network (-n) slave (s) mode.

Save the file as ‘indigo_network_slave.bat’, change the type to ‘All Files’, and save it to your desktop:

Step 3 – Running the .BAT

Double-click the newly created icon on your desktop that looks like this:

This will open Indigo in ‘Slave mode’. The panel should say that it is ‘Listening on port XXXX’:

The slave is now ready! It will continue to wait until the master sends it some work.

Step 4 – Starting the Master

Go to your Master computer and open the scene you want rendered:

In the System Tab of Blendigo, turn ‘Working Master’ on and hit ‘EXPORT Scene’:

You are now rendering!

Step 5 – Check Slave and Master

Your slave machine should now be receiving instructions from the master machine and starting to render:

Your Master machine should display a message similar to this:

Receiving frame from 192.168.0.111:1105 :
num_samples: 1400000.000000
width: 1204
height: 904
Frame received. (lock wait time: 0.00 s, transfer time: 2.29 s)

You can kill the render on the Master machine whenever you like. When you export it again, the slaves will automatically reconnect.

That’s it! You are now part of the elite nerdy crew who can say “I value my rendertimes, more than my electrical bill”. Feels good doesn’t it?

Was this tutorial useful to you? If you got stuck at any step on the way, drop a comment below and I’ll be happy to help!