Improvising a 3d Animated Short in Blender

As 3d artists today, we have tools and services at our disposal that shorten the production pipeline tremendously. Online communities and 3d asset platforms like Sketchfab even make it possible to indirectly collaborate with skilled artists, and some renderfarms such as GarageFarm, offer generous free credits on top of already affordable rates. Combined with the relatively non-destructive nature of 3d, it feels like there’s almost nothing stopping us from creating the 3d animated stories that inspired us to learn Blender in the first place!

Except maybe coming up with the stories themselves.

Coming up with a concept for a 3d animation might seem pretty simple at first, but ten minutes of letting the imagination do its thing can very quickly turn into days of overthinking and eventually, creative paralysis. There’s also the fact that it’s always easier to conjure up ideas than it is to execute them.

Of course in an ideal scenario, developing a concept should be given as much time as it needs so that whatever reaches the production phase is worthy of all the effort to come. However, despite Blender being as amazing as it is, and all the wonderful addons and tools out there, going through the 3d animation pipeline can still be an arduous and sometimes thankless endeavour, and if you’re doing everything on your own for the most part, that might just mean you never get around to actually finishing any animation at all.

In making my last animation I found that perhaps an overlooked advantage in doing it within Blender (or 3d in general) is that we can take a more improvisational approach to craft stories without as much repercussion as other animation media.

If you worry whether your Blender animation will actually see the light of day, here are a few ideas that might help (they certainly helped me, at least):

1. Let what assets you already have dictate your premise and not the other way around.

Instead of coming up with an idea from scratch, go through your old project files and asset libraries, and see if you can repurpose assets to create a setting for your short.

For example, I happened to have a Japanese village themed kit that I bought from sketchfab for a previous project, a few assets I bought from Megascans, and a fox mesh from another older project. These elements helped me create a prompt: something to do with the fox as a figure in Japanese mythology.

2. Go through available mocap data on Mixamo or other libraries to help inform your story

Being able to audition different character movements made it easy for me to come up with a narrative based on certain movements that I thought looked cool, and it helped me plan ahead for any adjustments to the animations I’d be using.

Mixamo allows us to quickly rig and assign animations to our own characters, but we could also begin downloading the dummy characters as placers to create block-outs for shots.

3. Consider how to tell the story in the most effective but economical way

I knew that even with minimal modelling and texturing, this was going to be a time-consuming task, so I tried to think of what I could do to make the animation work without relying on elaborately set-dressed scenes or having to write dialogue. I found that using heavy volumetrics not only reduced the assets I needed on-set but also added to an otherworldly atmosphere.

Also, I used to make music, so I figured I might as well leverage that and have a custom track tie the sequences together.

Syncing animation to music is a quick way to tie everything together, and you don’t really need to know how to make music to do it. There are loads of well-made tracks available on the Youtube audio library.

At this point I knew I was going to make something set in a fantastical medieval Japan that featured a mystical fox and a chase sequence between two characters, culminating in a fight scene, all happening without dialogue and synced to music. Nothing deep or provocative, but short, action-packed and (hopefully) fun.

4. Create blockouts as quickly as possible

Once I had some loose ideas for shots, and I had some working characters, I decided to jump right into Blender and start laying out and my scene elements to know where to place my camera. This allowed me to quickly get everything in place and determine what parts of a shot would need more detail, and what assets would be obscured by shadow or depth of field blur.

5. Make some Open GL or workbench renders and iterate

Once I had some shots in place, I rendered them all out and strung them together in Da Vinci Resolve (you can use Blender’s video sequence editor as well).

Here are a few snaps from the open gl renders.

Doing this allowed me to get a big picture view of how the shots looked one after the other, and helped me anticipate what changes had to be made. This also helped put all the shots in context and gave me ideas for more shots that I would work into the comp to make clear what was happening throughout the animation, as well as know what could be implied instead of having to literally be shown in the sequence.

Obviously, this approach is not better than planning everything linearly, but it can be a great way to develop skits or short stories while also being able to practice all of the necessary technical 3d and film-making skills to see a project through to the end. It’s my hope that these tips at the very least embolden you to create or keep creating the animations you want, and that it seems more doable now. Ultimately, the most important thing is to not get stuck in your own head and just go for it! It’s a rewarding experience. Best of luck, and Happy Blending!