ReactOS Overhauls DirectX With Wine 10.0 Integration, Expands Game Support

The open-source Windows-compatible OS ReactOS has integrated a large-scale update to its DirectX stack. According to reporting by Michael Larabel of Phoronix, the merged implementation reflects the state of Wine 10.0. The change adds 498,000 new lines and removes 119,000 lines. Much of the previous custom handling has been removed, unifying the rendering path. This article is based on Phoronix (All Rights Reserved) and relies on fair quotation under Article 32 of the Japanese Copyright Act.

The update is in line with its policy of pursuing binary compatibility with Windows applications and drivers. ReactOS has reimplemented the internals of the NT family to preserve an execution environment for existing assets. Graphics has been a difficult area for compatibility, determining whether entertainment software runs. By incorporating the Wine-derived implementation, maintainability and extensibility have been improved.

Major Update to Wine 10.0 Base

The purpose of the merged pull request is to update the DirectX stack to Wine 10.0. The development was led by ReactOS developer Justin Miller. He explained that he has devoted at least more than 230 hours to the work so far. While parts remain unfinished, the work has reached a major integration milestone.

The update significantly cleans up previous custom handling. Rather than relying on individual workarounds, it makes greater use of the upstream implementation as is. Bugs in some tests were also fixed. The cleanup is seen as reducing the long-term maintenance burden.

The development history belongs to the long-term compatibility improvement trend traced in ReactOS Achieves Half-Life 2 Compatibility: 30 Years of Open-Source Windows-Compatible OS. A record of launches of entertainment software has accumulated, making the update of the graphics stack the next step. This integration is a milestone on that continuum. Reciprocation between base updates and application cases is driving development forward.

Implementation Scale of 498,000 Added Lines

Explained

The merged code amounts to 498,000 new lines added and 119,000 lines removed. It is a large pull request requiring hundreds of hours of work. Rather than a simple version bump, it centers on replacement involving removal of custom handling. The aim is to narrow differences from upstream and make future follow-up easier.

What was removed was ReactOS-specific DirectX-related workarounds. While effective for securing short-term operation, they had caused divergence from upstream. The replacement improves the clarity of the code base. Reducing duplicate implementations also helps make bug investigation more efficient.

Miller noted the scale of the work and the remaining unfinished issues in the pull request. The original text reads as follows.

This is an incredibly long time coming, that’s taken at least 230+ hours of my life so far.

He also wrote about the goal as follows.

The goal of this PR is to update our DirectX stack to WINE-10.0, while removing a lot of our hacks.

Test fixes are also included, raising the quality baseline. Detecting and preventing regressions from the large-scale replacement will be the future focus.

Translation Support via OpenGL and Vulkan

The new implementation enables a configuration that translates DirectX processing through OpenGL and Vulkan. It is a mechanism for relaying drawing commands that cannot run directly on ReactOS to the existing graphics infrastructure. Use of Vulkan is conditional, requiring either a custom fork or Mesa software rendering. Selection must take into account the constraints of each execution environment.

The notable point is that ReactOS’s DirectX stack has become easier to run on Windows itself. It is moving closer to a layer usable on Windows, rather than a closed implementation specific to ReactOS. The ability to test the same stack in both environments is a major advantage. It makes it easier to isolate bugs and share fixes.

This direction is also consistent with the history of code sharing between Wine and ReactOS. Wine has developed a compatibility layer for running Windows applications on Unix-like OSes. ReactOS has aimed to reimplement the entire OS, including the kernel and driver layers. Unifying the graphics stack is a rational choice that leverages the achievements of both.

Test Cases Toward Unlocking Hundreds of Games

According to Miller, combined with other related pull requests, hundreds of entertainment software titles will become usable. Even this update alone has confirmed launch cases on the NT5 line. The specific example cited was Dragon Age: Origins. It was reported that the software reached launch on the NT5 line.

The NT5 line corresponds to the Windows XP and Server 2003 generation. Demand remains strong for maintaining older application assets in current environments. Whether a title launches depends on coordination across multiple layers, including graphics, audio, and input. The DirectX stack update resolves one of those bottlenecks.

The developer said on X that support for DirectX 10, 11, and 12 games will soon become possible. Not all titles work at this point. Continued verification and fixes are required. Even so, the scope for expanded support has clearly widened.

The importance of a testing system overlaps with the discussion of OS testing infrastructure shown in GNOME OS Test Center, Inspired by Apple TestFlight. The larger the base update, the more systematic testing and distribution mechanisms determine outcomes. For ReactOS as well, recording launchability and sharing reproduction steps will be key. Interest in development infrastructure is also reflected in Agentic Infra Chosen by Four Companies: Infinigence Strategy.

What Improved Operation on Windows Itself Means

Improved operation on Windows itself is more than a technical side effect. It shows that the ReactOS implementation has moved beyond environment-specific code to become a more portable layer. If it can be tested directly on Windows, it becomes easier to separate ReactOS-specific factors from common factors. Upstreaming fixes and pulling downstream updates also become smoother.

Removing custom handling also relates to the sustainability of the development system. Individual workarounds are effective in the short term but increase long-term follow-up costs. Aligning with upstream Wine 10.0 simplifies the application of update diffs. It is an effective strategy for supporting a wide compatibility range with limited personnel.

The remaining issue is the constraint on the Vulkan path. The current reliance on a custom fork or Mesa software rendering limits performance and ease of adoption. Stronger coordination with device drivers and the graphics infrastructure is needed. Future progress in fork integration and performance verification will be closely watched.

Editorial Opinion

In the next 3 to 6 months, the Wine 10.0-derived rendering path will become an intensive testing target, and bug reports are expected to increase. In particular, the Vulkan path remains constrained by dependence on a custom fork or Mesa software rendering. Allocation of development resources and development of a verification system will determine the pace of progress.

In 1 to 3 years, expanded support for DirectX 10, 11, and 12 will become the focus. Dual operation with Windows will improve verification efficiency and improve maintainability of the entire compatibility layer. Its practical value as a means of extending the life of assets for older Windows versions may increase.

While expanded compatibility widens user choice, will it blur responsibility for rights clearance and operational guarantees. What is needed to resolve Vulkan dependence, and who should take responsibility. How should the scale of hundreds of titles being unlocked be verified and quality assured.

References

• “ReactOS Lands Significant DirectX Stack Update From Wine For Enabling More Games”, by Michael Larabel — Phoronix, 2026-09-24T21:43:49.000Z (ARR)
• Source URL: https://www.phoronix.com/news/ReactOS-Updated-Wine-DirectX