Hi FMOD team,
I was wondering if it would be possible to get a sneak preview of some upcoming features in FMOD Studio 2.04. I heard it was showcased at GDC 2026, but unfortunately I wasn’t there to meet you.
Thanks in advance — fingers crossed for flexible output routing. 
Maybe it is still too early to share details publicly, but if anyone attended GDC and saw the presentation, I’d really appreciate any general feedback or impressions.
Here’s a few highlights that we showed at GDC:
- The new automation precision editor window will let you see the exact shape of your automation curves at larger scales than ever before, so that you can move automation points exactly where you want them.
- Automation curve presets will allow you to save frequently-used automation curves for later re-use.
- Shared effects and shared effect chains will be available in the mixer, allowing you to add them to any bus in your project.
- The loop markers of assets will be displayed on assets’ preview graphics, both in the assets browser and in the deck. FMOD Studio will let you edit and even remove the loop markers from an asset without having to bounce out a new copy of the asset from your DAW.
- The sync points of assets will also displayed on the asset’s preview graphic. Like asset loop markers, you’ll be able to edit your assets’ sync points without having to bouce a new copy of the asset out of your DAW.
- You’ll be able to assign release tails to single instruments, including single instruments in the playlists of other instruments. Whenever such an instrument is untriggered, it’ll automatically jump ahead to play the part of the waveform defined as the release tail, allowing a graceful and timely release.
- Each sandbox scene will have its own timeline. By automating the positions and event states of events and listeners on this timeline, you’ll be able to schedule event instances and listeners to move, start, and stop however you want. Then, with the press of a button, you’ll be able to set the scene playing. This will let you imitate the complex layered soundscapes of your game without using live update, making them a useful tool for mixing and auditioning events in context.
There’s more, but we don’t want to give away all our secrets just yet. You’ll just have to wait and see.
Thank you vzlomvl and joseph !
These all sound like very welcome additions. May I ask about the sandbox scene timeline: will it be possible to trigger or control a sandbox scene from the game/runtime? That could be very powerful. Local parameters in the mixer seem also very interesting…
Thanks again — looking forward to discovering the remaining secrets when 2.04 is ready 
The sandbox’s timelines are designed to be useful when your game isn’t available to trigger events and set parameters; if you have your game on hand, you won’t need it.
If you want your game to be able to trigger events and set parameters in a complex sequence, you don’t need the sandbox for that: You can simply create snapshots, events, and command instruments that do whatever you need.
Yes, of course, and these new features are very welcome ! but sandbox triggering could still be useful in some specific cases — for example, for custom event movement around the player’s head, or for building an audio-only cinematic sequence in 3d or simply to check audio inside early prototypes.
The problem would be how the sandbox scene listener relates to the actual game listener (maybe considering relative position can work).
I hope the idea makes sense.
There are already specialized tools for creating those kinds of behavior in both FMOD Studio and the Studio API. For example, to simulate an event instance moving around the player’s head, you can automate the direction and extent properties of a spatializer effect. If you let us know what kind of in-game behaviors you want to achieve, we’ll gladly suggest tools you can use for achieving them.
We have no plans to add support for including sandbox scenes in your built banks and triggering them from your game’s code. There are already other tools that are specialized in playing content in a game; the sandbox is specialized in its current role as an auditioning tool, so giving it a secondary role as a tool for implementing content would necessarily involve compromising its ability to fulfill one or both roles for the sake of the other.
Thank you Joseph for this reply. I understand the position. FMOD has been inspiring me for a long time, and I often tend to explore native features in non-standard ways.
Thanks again for the clarification.