Skip to content

v2.0.0

Choose a tag to compare

View all tags
@github-actions github-actions released this 09 Jul 00:34
v2.0.0
af61a70

Added

  • Cameras (parity): the MonoGame and raylib Camera2D/Camera3D modules now offer the same set of operations. MonoGame Camera2D gains viewportBounds, screenToWorld/worldToScreen, smoothFollow/clampTarget, and the full render/withViewport/withClear/splitScreen* config-builder surface it lacked; raylib Camera3D gains lookAt/orthographic/orbit/screenPointToRay (wrapping Raylib.GetScreenToWorldRay). Closes the camera API-surface gap between the backends.
  • Docs: new "MonoGame type quirks" reference collects the raylib-vs-MonoGame type differences that first-time MonoGame users hit — System.Numerics vs Microsoft.Xna.Framework math (Core layout/spatial/light APIs take System.Numerics on both backends, so a bare Vector2 resolves to the wrong type on MonoGame), float vs int Rectangle, Color constructors, the IAssets namespace and asset-path conventions, and the live window size via ctx.WindowWidth/ctx.WindowHeight. The affected guide pages now cross-link to it.
  • Templates: the starters now steer AI assistants (and readers) to the API reference for exact signatures and to the guides only for general usage; the MonoGame starters additionally require reading the type-quirks reference before writing code.

Changed

  • MonoGame — Breaking: the camera modules are consolidated into a single Camera2D/Camera3D surface that mirrors the raylib layout. The standalone Camera2DConfig module is removed — its builders (render/withViewport/splitScreen*) now live in the Camera2D module, and withClearColor is renamed withClear. Camera2D.smoothFollow/clampTarget now return a new camera instead of mutating in place (the camera's fields are immutable).
  • Raylib — Breaking: the 2D camera readers (viewportBounds/screenToWorld/worldToScreen) and Camera3D.screenPointToRay now take the camera by read-only reference (inref), so call sites must pass &camera (the smoothFollow/clampTarget mutators already used byref). This skips copying the native Camera2D/Camera3D structs on per-frame reads. All raylib camera helpers are now inline.
  • Culling — Breaking: Culling.isGenericVisible is renamed isVisibleBox (it tests a bounding box against the frustum — the new name says what it does).

Removed

  • 3D — Breaking: removed ShadowAtlasConfig.ShowDebugOverlay and the raylib ShadowAtlas.RenderDebugOverlay overlay — a dev-time diagnostic that leaked into the preview builds. No config flag overlays the shadow atlas on screen anymore.
  • Cameras — Breaking: removed Camera3DConfig.PostProcessPasses and the Camera3D.withPostProcess/withoutPostProcess builders — the pipelines never read them (v2 post-processing is command-driven via Draw3D.postProcess), so they were no-ops. Also removed Camera2D.overlay/Camera3D.overlay — they only set a viewport and a black clear (no compositing); the equivalent is render |> withViewport |> withClear, and on-top layering is draw order.

Fixed

  • Docs: the Culling guide now covers both backends (raylib's Frustum vs MonoGame's native BoundingFrustum, and the raylib Camera2D.viewportBounds &camera form), instead of describing only the raylib types.

Full Changelog: v1.3.0...v2.0.0

Assets 6
Loading