The main idea of the project is to quickly and efficiently create a compatible environment for the existing software and provide the best user experience of a personal computer.
- CPU
- UEFI boot
- x86-64-only mode (dropping x86-32)
- Multi-core CPUs and scheduler
- GPU
- Software Vulkan
- Hardware Vulkan
- OpenGL
- D3D
- Engine
- New one in Hexa
- Stable memory manager
- Initial .exe support
- GUI-capable .exe support
- Browsers
- Wi-Fi & Bluetooth
- Networking protocols
- Third-party drivers support
- Moving to a modern filesystem
- Unix subsystem (only software)
- Android subsystem
- LiveCD/USB
- GUI for OS installer
- Easy system updates
- Visuals
- New default theming and visuals
- HiDPI
- Theming engine
- New flexible theming engine and toolkit
- Overhaul of theming engine (GPU-accelerated, shadows, blurs, effects)
- Shell and UI
- Initial implementations of shell, tray, explorer, start button and others
- Greentea widgets and GUI elements
- Stabilization of shell, tray, explorer, start button and others
- Shell overhaul to add modern features and look
- Customizable themes and widgets with external theme-packs
- Stable and finished user space software
This project started as a controversy to undefined future (and past) of existing operating systems. Our team decided to define precise list of the most useful features to the wide audiences. Some features aren't really useful in any real manner right now, and their priority may be lowered.
Update from 2023: seems like M1 CPUs finally make ARM support valuable? Update from 2025: M2+ CPUs seems like nearly impossible to fully support due to hardware changes
Features, like LPT printing, has so small applicability (LPT ports in 2025+ anyone?), so can't be considered in any manner real target for Greentea OS team and use case for our users. Also, multiply that by a enormous number of bugs, hacks and workarounds, which we should fix now, to at least make non-academic project! And then improve implementations, the real things. Otherwise it is a waste of time.
Features are highly dependent of the API version. Also, the ecosystem defines it's own distribution rules. For example: while it is possible to run Vulkan API over virtually any (even 20 years old) operating system, no hardware or middleware (LunarG) distributors actually did it. Some features also, like native Wi-Fi or BLE support, were non-existent on old versions. So we need to declare timings and dependencies per feature.