Three years after the Unity runtime-fee shock, Game Developer pairs new survey data with interviews from Godotcon 2026 to answer whether developers actually stayed on Godot or drifted back to familiar tools.

The numbers say they stayed:

  • 12% of surveyed developers now name Godot as their engine of choice — an 8 percentage-point year-over-year increase (Omdia, via the Game Developer Collective)
  • Godot was the engine of choice for 47% of GMTK game jam entrants, ending Unity’s nine-year run as the jam’s most-used engine
  • Both figures have climbed continuously since the 2023 runtime-fee announcement

Nobody interviewed switched for a feature Unity lacks. They switched for the cost of one think-edit-test cycle:

  • Godot starts up and begins editing immediately, where a fresh Unity project takes visibly longer to open
  • The debugger never suspends the engine — only the user’s script pauses — so you can jump in and out of a running game. Miguel de Icaza of Xogot: “The feedback loop is incredible.”
  • A composition-focused object model lets programmers test things atomically without harnesses or scaffolding
  • The hardware floor is low enough that a student team can treat it as free (“It can run on a potato”)
  • Open source acts as documentation backup: when the docs run out, devs read the source and describe it as legible

Legitimacy transferred through a single studio. MegaCrit moving to Godot and shipping Slay the Spire 2 quickly is what convinced newer developers the engine was not a toy.

The gaps are specific and small: GDScript remains controversial enough that some devs prefer C#, and one attendee’s request to the Godot Foundation was multi-pass 2D shaders, since today you either hack around it with SubViewports or bake every pass into one shader. Unity 7’s roadmap, meanwhile, promises to compete on exactly this axis — faster project startup and iteration.

The useful reframe: engine choice is a feedback-loop decision before it is a rendering-pipeline decision. A 47% share among jam entrants is likely a leading indicator, which makes 12% commercial adoption look like a lagging confirmation rather than a plateau.