Lumiera
The new emerging NLE for GNU/Linux
Current Phase

Development

→ Git-flow
→ Release

Objectives

At the current stage of development, the focus is primarily on the overall integration of fundamental functionality at all levels of the code base. A sequence of Integration steps has been defined, each driven by some pivotal functionality, to be implemented as a »Vertical Slice«.

Our Goal is to integrate an ongoing render process, with a render pipeline in the engine and video output in the UI.

  • A network of Render Nodes can now be built and invoked ✔

  • 🗘 The provisional buffer memory manager needs to be improved next, to resolve a mismatch between the Buffer Protocol and the Output Sink Protocol, so that the target buffer of an output sink can be used seamlessly within the Render Node network.

  • Another point to be resolved next is to decide upon the design of the Vierwer Output Connection, which shall be used as an anchor point for the running playback process.

  • Next we need a solution how to integrate some Video Output solution into the Lumiera UI (→ see the recent research), so that it can be represented as an Output Slot.

  • After that, we could look into the integration of Job Planning this Output Slot with the Scheduler.

  • Once this has been accomplished, it should be possible to fill in the remaining parts into the Play Process implementation.

  • Furthermore we still need some provisional solution how to trigger all this new functionality from the UI.

Current Release

Lumiera is not usable yet.

Changelog

As of 11/2025 (0.pre.04)

This version is again motivated by a series of major upgrades and clean-up. Compilation now requires C++23 and Debian/Trixie is considered the reference platform. The GUI is connected to the core over an asynchronous messaging bus and the timeline display has been reshaped to accommodate flexibly nested structures of media content, instructed by the session through a series of diff messages. A scheduler has been implemented to drive the Render Engine, but the connection to the player and UI for output is yet unfinished.

As of 11/2015 (0.pre.03)

A lot of long standing maintenance work has been done. The codebase has been upgraded to C++14 and Debian/Jessie as reference platform, followed by clean-up of several workarounds now obsoleted. On the GUI side, the transition to GTK-3 was accomplished, which however necessitates a rework of our timeline widget, which is in progress but not finished yet. This work also spurred an effort to reorder the connection and communication between the core and the UI, which is expected to be asynchronous. Due to the limited developer resources, work on the Engine and Player part is stalled.

As of 10/2013 (0.pre.02)

The data models have been elaborated and some significant parts of the session are finished. Work has continued with time handling, a draft of the output connection framework, a draft of the player subsystem and interfaces to the engine and processing network. Unfortunately there was a considerable slowdown and decrease in team size, yet still the code base is growing towards 90k LOC. No tangible progress regarding the GUI and the backend.

As of 1/2011 (0.pre.01)

The project has created and documented a fairly consistent design, and some parts implemented — progressing from the technical foundations. The code base is approaching 65k LOC. Roughly half of this is test code. The Application can be installed and started to bring up a GTK GUI framework, but the GUI is rather preliminary and not connected to core functionality. The video processing pipeline exists only in the blueprints.

As of 2/2008

The project has been separated completely from Cinelerra, the parent project. The Community, which at that time was largely identical to the Cinelerra-CV community, choose the new project name Lumiera through a collaborative selection and vote. The basic project infrastructure was established, and work on the new codebase was underway. It is clear that no results other than a test suite can be shown in the foreseeable future.

As of 7/2007

A group of volunteers has formed to address some longstanding issues and limitations of Cinelerra, which we all used for video editing tasks requiring professional quality. The plan was to restructure some central parts of the render engine: introduce a Builder for render pipelines and use Placements to control arrangement of clips and a unified Data-Backend to coordinate usage of system resources. These innovations would serve as anchor, with the intention to propose the reworked codebase as a new version of Cinelerra, in the hope of turning Cinelerra development into a more participatory process going forward.