GNOME 49 fixes laggy cursor with vrr enabled

A recent change to Mutter lets the cursor move at the display´s maximum refresh rate when variable refresh rate is enabled, improving responsiveness when the desktop frame rate drops. GNOME 49 is due September 13, 2025 and will default to Wayland with X11 as a fallback.

GNOME 49 is scheduled for release on September 13, 2025, and developers are finalizing last-minute fixes during the beta cycle. A recent merge request to Mutter, GNOME´s Wayland display server and X11 compositor and window manager, changes cursor handling when variable refresh rate is enabled. The update aims to keep pointer motion running at the display´s maximum refresh rate even if the desktop frame rate falls, which should make mouse input feel more responsive when the desktop environment is under load.

The merge request states that if the fullscreen surface frame rate drops below 30 fps, the cursor should be allowed to move at maximum refresh rate instead of being limited to 30 hz or less. The change was inspired by a description of how KWin, the Wayland compositor for KDE, handles the same situation, highlighting cross-project learning in open source desktop development. The new behavior brings GNOME closer to the level of support that KDE Plasma has historically provided for features such as variable refresh rate and hdr.

In addition to the cursor update, the article notes a broader platform change planned for GNOME 49: the release will disable the X11 session by default and use Wayland as the primary session, keeping X11 available only as a fall-back for compatibility applications. Other technical details, implementation timing, and testing results are not stated in the article. The merge request itself is referenced as the source of the change, indicating the update is already present in the project´s review workflow ahead of the stable GNOME 49 release.

65

Impact Score

Introducing Mistral 3: open artificial intelligence models

Mistral 3 is a family of open, multimodal and multilingual Artificial Intelligence models that includes three Ministral edge models and a sparse Mistral Large 3 trained with 41B active and 675B total parameters, released under the Apache 2.0 license.

NVIDIA and Mistral Artificial Intelligence partner to accelerate new family of open models

NVIDIA and Mistral Artificial Intelligence announced a partnership to optimize the Mistral 3 family of open-source multilingual, multimodal models across NVIDIA supercomputing and edge platforms. The collaboration highlights Mistral Large 3, a mixture-of-experts model designed to improve efficiency and accuracy for enterprise artificial intelligence deployments starting Tuesday, Dec. 2.

Contact Us

Got questions? Use the form to contact us.

Contact Form

Clicking next sends a verification code to your email. After verifying, you can enter your message.