About Bumpybox
Bumpybox is an animation company based in Cardiff, founded in 2011 by Sam Wright, Leon Dexter and Toke Jepsen. Bumpybox’s main project since 2016 has been My Petsaurus, an animated series broadcast on CBeebies.
Bumpybox is best known for animating a pet triceratops.
Currently, our main source of income and point of focus is My Petsaurus, a children’s television programme that combines live action and animation using CGI composition. We pitched the concept – the life of a young girl and her pet dinosaur – to the BBC, who now broadcast it on CBeebies. So far, we’ve made almost 50 episodes over five series and a handful of one-off specials.
We own the IP to My Petsaurus, but we struggle to monetize it further than the show itself.
While our series is quite small, it has a distributor and a broadcast deal with the biggest TV channel for young children. We would like to monetize our intellectual property through things like branding, licencing and merchandising, but we don’t have enough of a name to clinch such deals yet.
To build the following behind My Petsaurus, we spoke to branding and social media consultants about our situation. They helped us realise that we need to create a buzz on social media and gain the attention of any potential licensors. As so much of My Petsaurus is the BBC’s right to share, we can’t simply upload the episodes to YouTube to direct people to our IP. We need to create extra content to share on social media, to get parents talking about the series and to get children finding spin-offs and clips online on places like YouTube. If we had more people behind the brand, we’d be able to demonstrate to potential partners that it is a viable one.
As a small company, we want to find ways to innovate and speed up the animation process.
Animation takes so long. For example, the series of My Petsaurus that we’re aiming to finish in September 2021 started its pre-production stage back in January 2020. So much of our time is spent on animating the episodes; this makes it hard to find time to repurpose materials that are made as a byproduct of the show or to create other assets.
Clwstwr funding enabled us to do a feasibility study into how to improve our efficiency. We received seed funding from Clwstwr to research potential ways we could make our processes more efficient using our limited resources, with the knock-on effect of building our name to improve our chances of securing licensing deals. To hone the focus of the feasibility study, our technical director Toke identified the areas that were slowing us down. They were not creative areas, but they were taking up a lot of our time. We came up with three questions to answer, each one being about a specific area of inefficiency.
Our first question was: How useful could game engine technology be to shows like My Petsaurus?
We wanted to see if such technology could help us make more content from what we were already doing, or if it could make it easier for us to select content from our archive. We looked into the feasibility of using Unreal Engine and found that there are some things that it would work quite well for. There's a sequence editor, which could allow us to lay out an episode of animation more efficiently. There are also potentially useful tools for things like setting up camera work, continuity and a few fairly boring tasks. We like how it all happens instantly, and that’s something we could certainly bring over.
The other thing we considered was whether we could make animated shorts, a simple video game or some other kind of spin-off content with Unreal. All seem doable, but there are a few things we’d need to take into account. We've got some relatively advanced stuff on our animation rigs that doesn't quite carry over into Unreal. Also, we need to make sure that we can go from Maya, which is the software we animate with, to Unreal and back again.
Our second question was: How can we speed up the time it takes to compile edits for review?
A large section of our report focused on how long-winded our system for compiling edits is. Typically, we’ll get given the edit by the show’s editor, Gorilla. Then, we’ll ingest it into our managing software, where we take out all of the shots with the dinosaur in, animate these shots then export it as a completed video.
It’s such a laborious system that involves opening up editing software, bringing in every individual shot and sound then exporting it, which can take at least an hour for a two-minute video. If you then realise, watching back, that there’s a missing shot or something’s not quite right in terms of how it’s been compiled, you have to go back to square one.
We know that Netflix and Amazon have proprietary software that allows them to take two videos with the same codec and combine them together, without needing to use any other software. We couldn't find this sort of thing at an indie level, though. The closest software we could find is FFmpeg, which turns image sequences into videos.
Toke wanted to find a way to make FFmpeg combine multiple videos, but it kept creating errors and missing frames. So, he submitted a report to FFmpeg and, through the process of talking with them, actually found out a workaround that means that we can now do it. That felt like R&D in action! This means that we can now select certain scenes or episodes and put them together to make a new sequence or a longer video, which is really useful as it helps us make new video content quickly. In turn, this should help us build our audience and drive traffic to our IP.
Our third question was: How can we make compositing and rendering more efficient?
To answer this question, Toke developed a new step in our pipeline: pre-compositing. It allows us to view an animation before it's rendered in compositing software. I appreciate that probably doesn't sound very exciting, but it's quite a big deal for us. We spend many computation hours lighting and rendering the scene before it goes into compositing. If we are asked to change things at this stage, we have to go back to the animation stage to sort the content, redo the lighting and sound, render it again then send it to compositing for another review.
The fix was relatively simple. Essentially, we now have the ability to export an image sequence preview from animation with the formatting that the compositing software can read, which means we don’t have to wait hours to render things that might require changing. We’re currently using it on My Petsaurus; it’s saved us a lot of time and prevented little mistakes making their way into rendered versions, which has freed up time for us to do other things.
Overcoming some big hurdles, thanks to Clwstwr, has made us eager to continue doing R&D.
The Clwstwr funding has meant we could do important R&D alongside My Petsaurus. Somebody could take a little step back from My Petsaurus and do a bit of work on the Clwstwr project, then they could come back with something we could test out on the series. In some cases, our efficiency has improved to the point that we can take on more work. We did three My Petsaurus specials in time for Christmas thanks to our increased efficiency.
I’d like to do something similar again, maybe focusing on how we archive our work so that it could be easily found using tags. For example, if I needed a shot of Topsy the triceratops jumping, I could just search the archive and reuse a previously-made animation.
Improving efficiency makes a huge difference as it gives us pockets of time where we can free people up from production. It means that we have more capacity for someone to do some social media assets or make a video for YouTube or write a pitch to a toy company, for example. In time, these things will help us grow as a company. And, along with our growing social media presence, that can only be a good thing for Bumpybox’s future.