The Making of Mixtape: How The Devs Synchronized Gameplay, Animation & Music
The Mixtape developers discuss Unreal Engine 5, MetaSounds, animation layering, cinematics, gameplay systems, and building an interactive experience driven entirely by licensed music.
Few games place music at the center of their design process. Rather than treating licensed tracks as background accompaniment, Mixtape builds every gameplay sequence around the emotional rhythm, pacing, and identity of each individual song. Every level, mechanic, camera move, animation, and visual transition exists to amplify the music, resulting in an experience that feels closer to an interactive album than a traditional narrative adventure. Critical reception says they absolutely nailed it.
Achieving that vision required an equally unconventional production pipeline. Built in Unreal Engine 5, the team developed custom animation layering systems, music sequencing tools, MetaSound workflows, and bespoke editor features that allowed gameplay, cinematics, lighting, and audio to evolve together while remaining perfectly synchronized with every track.
We spoke with the developers at Beethoven & Dinosaur about designing gameplay around licensed music, building flexible animation systems, creating custom Unreal Engine tools, balancing cinematic storytelling with player agency, and the production lessons learned while crafting one of the year's most distinctive narrative adventures.
Mixtape is built around music as a core storytelling device. What were the key design pillars when structuring a game around songs rather than traditional gameplay loops?
Johnny Galvatron, Writer/Director: The most challenging aspect of Mixtape was approaching each track as a unique experience that required the same kind of care and respect as adapting some classic piece of literature. Each song required something different, its own tone and mechanics.
We had been entrusted with these great pieces of art, some of the most beautiful popular music ever written, and there was no set of principles that could be applied across the board - that’s how you end up with so many different systems throughout the game, from playing softball to the Touch by Stan Bush or bombing a hill in a shopping cart to Rainbow or detonating fireworks to More Than This by Roxy Music.
The idea of reliving memories through music is central to the experience. How did that concept shape the game’s narrative structure and pacing? And did the team pull on specific personal memories tied to music for things like this?
Johnny Galvatron: Mixtape’s structure and its pacing derive from the Mixtape itself. We began with many of the songs that ended up in the game - we would arrange them in different orders to see what stories they told, how they flowed. The art of making a mixtape is an art of arrangement or collage - you can take these different pieces of art, from different eras, different parts of the world, from widely different experiences, lay them out side by side, and find the emotional through line that connects them. We absolutely leaned on personal experiences.
I’ve run from the cops on several occasions as a teen causing trouble, Woody got dragged around a field in a shopping cart pulled by dirt bikes, and various other exploits that cannot be named for legal reasons.
From a systems perspective, how do you design gameplay sequences that sync meaningfully with licensed tracks rather than just using them as background audio? In particular, the animation is incredible. I love the way you fluidly adapt the animations of clapping, headbanging, etc to the scene, even while the character is still moving and doing other things. The skating at the intro bit is a great example.
Roman Maksymyschyn, Technical Director: A pattern we found ourselves using often in Mixtape was to have a procedurally animated core layer that gives you the "main" pose, and then exposing override points that other systems can inject or layer animations into (usually via sequencer).
So in the case of the skating example you mentioned, the core would produce the base skating locomotion, handling things like aligning the board and feet to the ground via Control Rig, leaning based on the player's input, etc, while another sequence that gets scrubbed in time with the music lets us author hand claps at appropriate times with a high degree of position.
After all the layers have been processed together, there is a final post-process layer that will do things like move mouths while talking, and flapping hair/clothes in the wind.
Levels like headbanging work in a similar manner, where there is a master sequence that has the music track we want to sync to, which acts as a timeline for us to author almost everything in the level such as camera moves, animations, changing scenes, sound effects, particles etc all in time with the music with quick iteration/preview time without having to play the game up to that part in the song.
From a technical standpoint, what engine and tools are you using, and how do they support music-driven transitions and scene changes?
Roman Maksymyschyn: Mixtape is built in Unreal Engine 5, and since music/audio is such a core part of the game, we gave it a lot of attention, using Unreal’s provided audio systems as a base, but also adding more sophisticated tooling to make sure everything was as art-directable as possible.
One example of this is how we made the music "endless" during dynamic gameplay sections. We often needed the ability for a song to start, but to then seamlessly crossfade back into itself at a specific start/end point, so to save having to manually splice up files into an intro, then a loop, then sequence those together etc, we made a re-usable metasound that lets you provide loop start and end times, fade durations, and lets you preview how seamless the configured loop is by clicking play on the asset.
We also made a custom music track in sequencer that let us do some more advanced things with this, such as author where music starts in sequencer and be able to hear it in preview, etc, but also let it persist over level loads, or add effects like reverb / low pass etc and key the intensity of those effects in the timeline. We made lots of other custom systems and tooling to be able to author things like ducking levels, ambience, reverb, etc in a very granular/bespoke way, with a similar approach to aforementioned animation layering systems where there were reasonable defaults set up, but you could override most audio properties per level, per cinematic, or even per shot where needed via custom sequencer tracks.
Which animation software do you primarily use? Can you walk me through the process of creating a scene, from early blockouts (if you have visuals, that would be amazing) to a completed in-game moment?
Joe Toole, Principal Animator: We use Maya as our main animation DCC, though towards the end of the project some control rigs were created and animated using Unreal. There were four animators on the project, each with their own similar yet distinct processes. The process for cinematic animation was often some variation of: Receive layout/storyboards (done by Johnny or sometimes Lily), either film some reference or consult your mind palace, block in some rough action, take it to dailies.
From there we’d talk as a group and discuss what was working, if anything needed changing. Sometimes shots would be totally fine to go straight on to lighting; other times it was a full rework - and everywhere in between.
From there, once a shot is in Unreal, there was still often a bit of time spent making sure everything worked with the “steppy” tech, which relies on root motion and could often need tweaking and updating depending on the camera's movement.
Roman Maksymyschyn: We knew early on that we wanted to have a lot of variety for this game, so to facilitate this we made the decision early on to be very careful with what systems should be shared across different levels vs what should be bespoke, with the dream that we could make major modifications to each part of the game in isolation without accidentally breaking other parts of the game. While this did result in some level of duplicate boilerplate code to get new characters/gameplay up and running, ultimately it was worth it as it allowed us to go all out creatively and be free to fully change how things worked per level, rather than having to try to work within the rigid constraints of some shared base class that you might find in other more traditional games.
In the cases where we did want to share functionality between characters/levels, we would often build them as subsystems or components that you could opt into using as needed, or omit if not, which gave us the best of both worlds.
Lighting, color, and composition play a big role in conveying mood. What does your visual development process look like for these memory-driven scenes?
Harry Truman, Level / Lighting Artist: We had a lot of fun creating each of the memories and tried to draw inspiration from existing cinematic compositions in film and TV. The camera is constantly moving around and reframing the characters in our cutscenes, so each individual shot ended up having a completely unique lighting arrangement. Rim lights needed to stay in the background and fill lights in the foreground, and in some cases, lighting rigs would be hand-animated to follow or orbit the characters for the more dynamic shots. The creative freedom we had with the lighting in Mixtape was endless. It didn't need to be realistic. We just wanted to squeeze as much of the emotion the scene was trying to evoke out of our colour palette.
It should feel magical and vivid. In your fondest memories, the clouds are pink, the sun is golden, and the moonlight illuminates the whole night sky.
How do different disciplines—narrative, design, art, and audio—collaborate when building sequences that are so tightly integrated with a game like this?
Roman Maksymyschyn: I think an important part of how all of these disciplines come together is that we were all looking at the game as a whole rather than everyone going away and doing their bit, then trying to mix it all together at the end and see if it fits. So things like doing full playtests as a team really helped unify that vision and make sure we were all on the same page about where everything is at and where we want to head.
What were some of the biggest challenges in balancing player agency with a more curated, cinematic experience?
Harry Truman: There's obviously a fine line between having not enough interactivity and having too much. Do we want to challenge the player? What kind of friction will this create within the story we're trying to tell? Does introducing more freedom prevent us from having level design that’s intentionally synchronized with the beats and crescendos of the soundtrack? All these things play an important role in deciding what level of interactivity we feel is appropriate for the player.
We experimented with more expansive levels and failstates earlier on, and just wound up asking ourselves every time if it adds anything to the experience we want to create. Some parts of the game we simply just want the player to play. Make a slushie, TP a house, paint a d*ck on the door of your hideout. Just have fun with it. If you’re running through backyards and across rooftops to rescue your friend from the cops and Yesterday’s Hero by John Paul Young is playing, we want you to BE in that moment. There’s no need to restart if you fail. Some could argue that it would add to the stakes. But it’s an idealised, perfect scenario and you’re the main character. And in a perfect world, you're living in the moment, not thinking about the danger ahead; movement should feel effortless, and the song should end perfectly right as you get to your destination.
From a production standpoint, how do you test and iterate on something as subjective as emotional impact and pacing?
Harry Truman: Rather than building up a small segment of the game as a ‘vertical slice’, we opted for more of a 'horizontal slice' where we built the entire game at a very basic level until we could play from start to finish. This allowed us to cut and rearrange the entire game until the pacing of the story beats and the different ways the player interacted with the game were all spaced effectively. We played through everything front to back in single sittings countless times. Tested it ourselves and tested externally. It'd go from being a greyboxed level with minimal animation and unvoiced dialogue, to suddenly yanking on your heartstrings. Once everything came together towards the end, we just felt like everything was right where it should be, and it hit how we wanted it to.
Looking back, what were the biggest lessons learned while building a game so deeply rooted in music and nostalgia?
Johnny Galvatron: The biggest lesson was that each of these incredible songs deserved its own unique approach. That everything we added had to be in service of creating and conveying the heightened experience of the music. That it’s important to build narrative around the nostalgia of the heart and not nostalgia of things - a nostalgia that can be missed but never returned to, as we must all grow up and move on.
As the saying goes - you can return to the past, but there’s no one there.
Beethoven & Dinosaur, Developers of Mixtape
Interview conducted by David Jagneaux
Subscribe to 80 Level Newsletters
Latest news, hand-picked articles, and updates