For .NET developers and users, Ubuntu is finally adding the latest .NET 10 support for Ubuntu 24.04 LTS, Ubuntu 25.04/25.10, and next Ubuntu 26.04 LTS.

.NET 10 is the Long Term Support (LTS) version that was released a month ago. It features 3 years support until November 10, 2028.

The new version features JIT enhancement with various inlining improvements, improved code generation for struct arguments, and loop inversion.

It added Advanced Vector Extensions (AVX) 10.2 support for x64-based processors. And, it updated NativeAOT’s type preinitializer with support for all variants of the conv.* and neg opcodes, further optimizing runtime performance. For more features about .NET 10, see it in MS website.

Microsoft provides official .NET packages for Linux through portable tarballs, which are available to download at this page. User may select download it, decompress, and run the executable directly from the extracted folder.

Why installing .NET package from Ubuntu Repository

As you may know, Ubuntu announced the collaboration with Microsoft more than 3 years ago, to provide enterprise-grade .NET support through native .deb packages in the universe repositories.

Canonical has packaged .NET 6/7 for Ubuntu 22.04, .NET 8 for 22.04/24.04 & higher, and .NET 9 for Ubuntu 25.04/25.10.

For .NET 10, it’s already made available for Ubuntu 25.10 and Ubuntu 26.04, while the support for Ubuntu 24.04 and 25.04 will be available soon.

The .NET packages in system repositories are well tested by Ubuntu developers, and keep receiving security updates. They provide better integration with your system, as they are built with Distro specific patches and PATH and environment variables setup out-of-the-box.

And, user may ask for community support (from e.g., AskUbuntu and Ubuntu Discourse) and report bugs for the packages in launchpad.

To install .NET 10 in Ubuntu from 24.04 to 26.04, open terminal (Ctrl+Alt+T) and run command:

sudo apt update && sudo apt install dotnet10

Besides installing the full SDK and run-times, you may select install only the SDK or runtime that you need:

sudo apt install dotnet-sdk-10.0

sudo apt install dotnet-runtime-10.0

sudo apt install aspnetcore-runtime-10.0

NOTE: For Ubuntu 24.04 and 25.04, .NET 10 is being rolled out gradually, so far for one-fifth of users. If you can’t get it, just wait. It will be available in next days if everything goes well.

What about Ubuntu 22.04

For Ubuntu 22.04, the maintainer team has made .NET 10 into this PPA repository.

Both the PPA and official Ubuntu packages are maintained by same team members. However, the PPA package is NOT officially supported by Ubuntu.

To add the PPA, open terminal and run commands:

sudo add-apt-repository ppa:dotnet/backports

sudo apt update

After that, you may run the apt install command above to install .NET 10 SDK and runtimes.

The 25.12.0 version of Kdenlive video editor is finally available to download!

The new version of this KDE’s software added a welcome dialog at app start, allowing to open or create new project, or create new from a profile.

For new users, the welcome screen also provides options to select prefer color scheme, and configure default profile resolution, frame rate, layout orientation, and video/audio tracks number etc.

The new version also introduced new widget docking system, allowing to easily re-arrange the video editor components by dragging on their tab-bars.

Simply click down on the tab-bar of the target component then drag, it will show you the layout thumbnails. Just drag the cursor icon onto target layout position, then you’re done moving the component. Or, you may double-click (on tab-bar) to cancel the dragging, or even leave it to be outside of the video editor window.

Besides that, the release also updated the clip monitor and project monitor (video preview section) with enhanced audio display, 1080p monitor scaling, and hamburger menu options to create mark from zone.

The add marker dialog now has time span support, letting to easily set the end time and duration. Media browser now has preview support, making easy to find out the file that you want. And, Safe Zone button is added in monitor toolbar, vertical safe zone layout and zoom reset.

The new version also updated audio waveform support by adding reset/cycle actions, and moving zoom control from A1 track header to status bar.

It brings back snapping when resizing for non-rotated frames, supports snapping when moving a rotated frame, and snapping for all edges when moving instead of only top-left.

And it added power management option to disable sleep while rendering and playing, switched av1 encoder to the highly optimized software dav1d encoder, and added SVT-AV1 to codecs which use crf option.

Other changes include:

• Two new frei0r filters.
• Add xsd for effects.
• Add vertical editing profile.
• Quality and speed control to svtav1 preset.
• Duration support in MarkerListModel import/export.
• Add XML files for LADSPA TAP effects.
• Allow appending video clips directly to a project in command line.

There are as well numerous other improvements and tons of bug-fixes in this release. For details, see the change-log in the KDE announcement page.

Get Kdenlive 25.12.0

The official packages for Linux, Windows, and macOS are made available in the KDE website via the link below:

For Linux, it’s an AppImage for AMD/Intel platform, which can be run directly to launch the video editor, after adding executable permission.

There’s also an official Flatpak package for most Linux on both amd64 and arm64 processors, though not updated at the moment of writing.

For Ubuntu user who prefer native .deb package, the new release has been made into Ubuntu 26.04 repository, and I’ve uploaded the package into this unofficial PPA for Ubuntu 25.10 and Ubuntu 25.04.

To add the PPA and install Kdenlive deb package, run commands below one by one:

sudo add-apt-repository ppa:ubuntuhandbook1/kdenlive
sudo apt update
sudo apt install kdenlive

OpenShot, the free open-source and easy-to-use video editor, released new 3.4 version yesterday.

It’s been almost a year since the last! The new release improved the overall performance with 32% speed up, and introduced many new effects.

According to the official announcement, the new version has 14.9% to 63% speed up for the performance of frame mapper, mask, crop, ffmpeg reading/writing, and other effects.

It also features 23% faster export by integrating the caching thread into the Export dialog, faster waveforms for long audio plus time-curved clips, and faster FFmpeg decoding.

Besides the significant performance improvements, the release also introduced interactive cropping support in preview.

Simply, right-click on your video clip and choose a crop method, then you may drag resizing and moving the rectangular selection in preview to crop the video clip from current position in timeline. While, you may also input the corresponding values in left to crop using properties.

The release also added new experimental timeline which can be enabled in Preferences dialog. The new timeline added buttons to do move up/down, lock, delete actions in track header. It as well features smoother scrolling/zooming on large timelines, new Keyframe panel and in-timeline keyframe editing.

OpenShot 3.4 also added following new filter effects:

• Sharpen – boost edge contrast to make video details look crisper.
• Color Map (LUT) – adjust colors using 3D LUT lookup tables.
• Spherical Projection – Flatten and reproject 360 or fisheye inputs into a rectilinear view with yaw, pitch, and FOV. Support Equirect and multiple fisheye lens models.
• Lens Flare – simulate sunlight hitting a lens with flares and spectral colors.
• Outline – add outline around any image or text.

Other changes include:

• Timing toggle button to re-time by dragging clip edges, similar to trimming.
• Dragging keyframes on the timeline with live preview while dragging.
• Add new Repeat / Loop / Ping-Pong options into Time context menu.
• Copy/Paste files into Project Files from system clipboard.
• New Benchmark tool in libopenshot.
• Improved EDL and Final Cut Pro XML import/export.
• Update AppImage to work better on newer Linux Distros.

How to install OpenShot 3.4

The video editor provides official packages for Linux, Windows, ChromeOS and macOS. Along with the source code, they are available to download via the link below:

For Linux, it’s an AppImage for AMD and Intel platforms. It can be run directly to launch the video editor after adding executable permission.

For Ubuntu, there’s also an official PPA, though it’s not updated for the new release at the moment of writing.

It’s also available to install as Flatpak package, which runs in sandbox environment and support both amd64 and arm64 CPU processors.

The second RC development release for the GIMP image editor 3.2 is out!

The new release introduced less features compare to the last RC1, but focused on improvements to performance and existing features and vulnerability fixes.

First, for user with a stylus pen, the MyPaint Brushes tool now support barrel rotation when relevant to the specific brush selected. The feature however is not tested because the developers don’t have a stylus to try it out.

Unfortunately I can’t test as we don’t have any stylus with such barrel rotation feature (and unfortunately this hardware feature is discontinued and getting a stylus with this is extra-hard now), but that should work. And at the very least, it does match with MyPaint’s implementation.

This RC also improved the Paint Select support. It’s an experimental feature introduced since version 3.0. User may try the feature by starting GIMP with --show-playground option, then go to preferences -> playground to enable it. Though, the gegl library needs to be compiled with workshop enabled.

In the release, it added progression feedback and improved local region computation to this experimental feature. And, it now processes more events, reset the scribbles at every stroke.

GIMP 3.2 RC2 also removed the restriction that prevented saving and exporting when no drawables are selected, raised the clipboard brush and pattern max size limit to 8192 for 64-bit processors, while it stays 1024 on 32-bit architecture, and it greatly improved the speed for initial font loading.

It now has a much more robust path importing from SVG by latest C-based librsvg code, and uses opaque path IDs for the ‘id’ attribute for SVG exporting, but not the GUI-visible path name anymore, to match the SVG specs.

To prevent infinite loops with cycling links, it also added a few safety checks when creating new link layers and when loading XCF files with link layers.

Other changes include:

• New startup splash.
• Set default PS/EPS export unit to millimiter instead of inch.
• Import legacy PSD Outer Glow layer style.
• Add Bash completion for the command line interface.
• Vulnerability fixes: ZDI-CAN-28311, ZDI-CAN-28273, ZDI-CAN-28158.
• Other fixes and build updates.

Get GIMP 3.2 RC2

The new RC2 is not announced at the moment of writing, though you may keep an eye on its news page for announcement.

For the source tarball, as well as the official packages for Linux, Windows, and macOS, they are available to download via the link below:

For Linux, select download the AppImage for either x86_64 (AMD/Intel) or aarch64 (for e.g., RasPi or SnapDragon X processors), then run to launch the image editor after adding executable permission.

Though, Ubuntu since 22.04 needs to install libfuse2 package for being able to run AppImage:

sudo apt install libfuse2

For those who prefer the Flatpak package, it’s also available to install via:

flatpak install --user https://flathub.org/beta-repo/appstream/org.gimp.GIMP.flatpakref

And, users who already installed the package can update it via command:

flatpak update org.gimp.GIMP//beta

Scribus, the free open-source desktop publishing software, released new 1.6.5 version few hours ago.

It’s been 8 months since the last point release, and the new 1.6.5 is also a maintenance release that contains only bug-fixes and improvements.

First, the new version added negative values support for the rotation in the properties palette. For example, you may input “-30” to automatically get “330” when hitting enter or leaving the properties.

For scripters, it added the document for scribus.ImageExport: transparentBckgnd, which corresponds to the ‘No Background’ checkmark in the ‘Save as Image’ dialogue, and, the function getBoundingBox() which was missing from HTML docs. And, it fixed that unknown/misspelled named parameter in API call crashed Scribus instead of giving an error message.

It also updated the print dialog, which fixed the inconsistency between output filename extension and printing language, and crop marks not present in output if it’s the only option selected in the “Marks”.

For fonts, the Additional Font Path Setting was cleared if any other setting is modified and saved while one document is open in Scribus. And, font rendering with non-integer font sizes (e.g., 9.90pt) went different after PDF export. Now, both the issues are fixed in the new 1.6.5 release.

The users who prefer dark color scheme, the new release fixed the poor menu appearance, though 3rd level menus are still in-visible in my case in Ubuntu with kvantum-dark theme.

Other fixes in the release include:

• NTLM hash leak.
• Crash when loading a .sla file created by v1.4.6.
• Crashes when exporting PDF with PDF-Push-Button or other PDF-thingy via Python.
• Color picker (eyedropper) always returns “black”.
• Layer disappears after editing inline object.
• Documents become partially invisible when zooming or scrolling.
• Not able to set “line spacing mode” for sub-style.

For more, see the Changelog file in source tarball (see the link below).

Install Scribus 1.6.5

The source tarball as well as installers for Linux, Windows, and macOS are available to download in sourceforge via the link below:

For Linux on AMD/Intel platform, it’s an AppImage that can be run directly to launch the software, after adding executable permission.

For Ubuntu, Linux Mint, Zorin OS users who prefer the classic .deb package format, I’ve uploaded the new release package into this unofficial PPA for Ubuntu 22.04, Ubuntu 24.04, Ubuntu 25.04 and 25.10.

Simply open terminal (Ctrl+Alt+T) and run the 3 commands below one by one to add PPA, refresh cache, and install the package:

sudo add-apt-repository ppa:ubuntuhandbook1/scribus
sudo apt update
sudo apt install scribus

Also, there’s a community maintained Flatpak package for most Linux, though it’s still at v1.6.4 at the moment of writing.

Ubuntu Team announced the second development release of Ubuntu 26.04, Resolute Snapshot 2, yesterday afternoon (UTC time).

The date was moved up by a week, because of end-of-year (EOY) shutdown and the developers need more time to fix issues before holidays.

Whilst the snapshot was planned for next week, we wanted to do it a week earlier so as to give ourselves some time to fix anything that might break ahead of the EOY shutdown from w/c December 22nd.

Ubuntu 26.04 is a Long Term Support (LTS) release with a total of 15 years support: 5-years standard support, 5-years Expanded Security Maintenance (ESM), and 5 years of security coverage (for paid customers).

The snapshot so far is still powered by Kernel 6.17 and features Gnome Desktop 49. There’s NO visual difference compare to the last snapshot release. Though, you may keep an eye on this on-going release note page for the updates.

According to the road map, Ubuntu 26.04 will have 2 new default applications: Showtime video player, and Resources system monitor and task manager.

It will continue improving the NVIDIA on Wayland experience, introduce Security Protocol and Data Model (SPDM) based fingerprint authentication, add more control for TPM-backed full disk encryption (e.g., ability to add/remove PIN or passphrase after installation, re-encrypt a disk via Security Center), unified software management, and Ubuntu Pro on WSL.

And, it will sadly exclude Ubuntu MATE and Ubuntu Unity as the official flavors in next LTS, because the two flavors lack of maintainers and miss the deadline of LTS re-qualification.

Get Ubuntu 26.04 Snapshot 2

NOTE: This is a development release that’s NOT ready for production use!

The .iso images for Desktop and Server, as well as Netboot tarball, WSL, and pre-install server images are available to download via the link below:

For desktop and education users, the snapshot also includes the .iso images for all the official flavors, which are available to download via the link below:

For Ubuntu MATE and Unity, the iso images are automated builds without maintainers behind them.

• EdUbuntu 26.04.
• KUbuntu 26.04.
• LUbuntu 26.04.
• Ubuntu Budgie 26.04.
• Ubuntu MATE 26.04.
• Ubuntu Unity 26.04.
• Ubuntu Cinnamon 26.04.
• UbuntuKylin 26.04.
• Ubuntu Studio 26.04.
• XUbuntu 26.04.

If you’ve already run your machine with Ubuntu 26.04 daily build or snapshot 1, then simply install all the available updates to get to the new snapshot.

For current Ubuntu 25.10, I’ve tried to upgrade to the new 26.04 Development release by first installing all updates:

sudo apt update && sudo apt full-upgrade

Restart if required, then run:

do-release-upgrade -d

For future releases, see the table below or the official release schedule.

January 29, 2026	Snapshot 3
February 26, 2026	Snapshot 4
March 26, 2026	Beta (mandatory)
April 16, 2026	Final Freeze, Release Candidate
April 23, 2026	Final Release

After more than two months of Beta testing, System76 finally announced the release of Pop!_OS 24.04 LTS.

As the version number indicates, Pop!_OS 24.04 is based on Ubuntu 24.04, but features Linux Kernel 6.17 and its new COSMIC desktop environment.

Pop!_OS is developed by System76, a American Linux computer manufacturer. It features COSMIC desktop, which was previously a customized GNOME desktop, but now a standalone desktop environment built from scratch.

The new desktop environment is written in the Rust programming language for high performance and memory safety. It replaced Nautilus, Gnome Terminal, Gnome Text Editor, Totem with its own COSMIC specific files, terminal, text editor, and video player apps.

Pop!_Shop is replaced by COSMIC Store, which supports both Flatpak and Deb out-of-the-box. And, the GDM display manager is replaced by new COSMIC Greeter with a redesigned login appearance.

The new desktop improved window tiling support. It added a toggle in system tray indicator, allowing to tile all current workspace windows by single-click. The icon changes automatically according to the window tiling status. And, it supports drag’n’drop exchange window position in tiling mode.

Besides using indicator menu, it also support tiling windows via keyboard shortcuts, and tiling in per display basis.

The workspaces can now displayed in either vertical or horizontal orientation. Multiple displays can have separated or spanned workspaces. And, it supports pinning workspace and drag’n’drop re-arranging workspace or even moving to another display.

Pop!_OS 24.04 also improved multiple monitors support. Besides per display window tiling and multi-display workspaces mentioned above, it also improved the experience for displays with mixed HiDPI and standard resolutions.

Displays are automatically scaled based on pixel density and display scaling can be fine-tuned in Settings. And, it will remember your display settings and restore automatically next time you plug it in.

Moreover, it added hybrid graphics support. Apps request the discrete GPU will automatically run on the correct GPU, while user may right-click on app icon to choose which GPU to use at launch.

Other changes in Pop!_OS 24.04 include:

• New stack windows feature to combine multiple app windows into single in tabs, just like web browser.
• NVIDIA 580 driver out-of-the-box.
• ARM computers support.
• Support reinstalling the OS anytime while keeping files, settings, and Flatpak user applications

For more, see the official release note.

Download or Upgrade to Pop!_OS 24.04

Pop!_OS 24.04 is available to download in its website via the link below:

It recommends 4 GB RAM, 16 GB disk space, and 64-bit ARM or X86_64 processors. While, NVIDIA edition asks for GeForce 16 series graphics or newer.

And for Pop!_OS 22.04 LTS, user may run the single command below in terminal to upgrade to Pop!_OS 24.04 LTS:

pop-upgrade release upgrade -f

Though it’s always recommended to back up your files first.

For users of Ubuntu MATE and Ubuntu Unity, the 2 Linux Distributions will NOT be the official flavors for the next Ubuntu 26.04 LTS.

Ubuntu announced the LTS qualifications for Resolute Raccoon in last week:

We haven’t seen LTS requalification requests from Ubuntu MATE nor Ubuntu Unity, so these will release subject to release team approval, and they will be non-LTS.

As you know, the Ubuntu flavors are the community maintained distributions with different desktop environments or specific intended use case. So far, Ubuntu has 10 official flavors, they are listed in this Ubuntu web-page.

Ubuntu always requires the official flavors to go through a qualification process for every releases, especially for LTS. For next Ubuntu 26.04 LTS, the following flavors have been approved with three year LTS qualification.

• Ubuntu Budgie
• Ubuntu Kylin
• Xubuntu
• Lubuntu
• Edubuntu
• Ubuntu Studio
• Ubuntu Cinnamon

What About KUbuntu 26.04

KUbuntu is not mentioned in the announcement. And, I didn’t see the qualification request for KUbuntu 26.04 in the Technical Board.

The KUbuntu developer, Scarlett Gately Moore, was in a car accident in last July. Though back to work now, she needs a little help.

Rik Mills, another KUbuntu developer who’s also Ubuntu core developer, is actively building packages for Ubuntu 26.04 and backporting Plasma desktop and KDE frameworks for current Ubuntu release through KUbuntu PPAs.

IMO, KDE Plasma is so popular and maintainers are still actively working, there’s no reason to exclude KUbuntu 26.04 as one of the next official LTS flavors.

Why Ubuntu MATE and Ubuntu Unity Excluded

Ubuntu MATE is Ubuntu with MATE desktop environment, a continuation of the classic GNOME 2. It was firstly approved as official flavor when it was in version 15.04.

MATE in current Ubuntu releases is still at version 1.26, while the latest MATE 1.28 has been released for more than a year. Debian upstream has started packaging MATE 1.28 for Forky and Ubuntu 26.04, though Ubuntu MATE developer missed the LTS qualification.

According to the future of Ubuntu MATE discussion, it seems that the project lead, Martin Wimpress, is now having other interests and obligations (Nøughty Linux), so he didn’t apply the re-qualification requests.

For Ubuntu Unity, the official flavor since 22.10, the project lead Rudra Saraswat, is too busy because of university exams and tests.

Other members lack of knowledge to maintain the project, and Maik Adamietz asked for help by this thread in discrouse few months ago, but without luck. And, they didn’t even know the requalification process so missed the deadline for next 26.04 LTS.

For current Ubuntu MATE and Ubuntu Unity users

Without Ubuntu MATE or Ubuntu Unity, Ubuntu 26.04 can install the 2 desktop environments from the system repositories, though they will be less stable due to lack of desktop specific maintenance.

According to this discourse page, 24.04 users may stay with it until 2034, by enabling Ubuntu Pro for security maintenance, though Canonical does not offer phone or ticket support for Community Flavors.

Microsoft Visual Studio Code announced the November release, version 1.107, yesterday for Windows, Linux, and macOS users.

The new release of this free open-source code editor integrated the agent sessions into the Chat view. When working in a workspace, it only shows sessions related to the current workspace, while all sessions across workspaces are shown when you are in an empty window.

The agent sessions can be displayed either in “compact view” that lists 3 most recent sessions along with “Show All Sessions” button, or in “side-by-side view” when Chat View is wide enough. While, there’s an orientation setting to set to always stacked (compact view), always side-by-side, or switch automatically according to Chat View width.

The local agent now continues running in the background when you close the local chat session. And, it’s able to see the status of the running agent in the sessions list and switch back to the session at any time.

When creating a new background agent, user can now choose to run in either the current workspace or a dedicated Git worktree. And, when running a background agent in a worktree, the changes is isolated in a separate folder, thus you can run multiple background agents simultaneously without conflicts.

The background agents now support multiple context attachment types. You can attach selections, problems, symbols, search results, git commits, and more to any prompt.

Other Agent HQ changes include:

• Introduce new “Continue in” option to continue a local chat with a background or cloud agent seamlessly.
• Define custom agents at the organization level for your GitHub account (experimental).
• Bring your own custom agents into Background Agents (experimental).
• Run agents as subagents (experimental).
• Reuse your existing claude skills (experimental).

VS Code 1.107 also introduced Language Models editor which provides a centralized place to view and manage all available language models for chat in VS Code.

It can be opened either from the model picker in chat or via the Command Palette with Chat: Manage Language Models. And, it can manage model visibility and add more models.

Moreover, the textSearch tool now supports searching in ignored files/folders specified by files.exclude or search.exclude settings or .gitignore files. And, Azure model provider now uses Entra ID authentication as default.

Other AI related changes include:

• Introduce collapsible chat sections for non-reasoning chat output.
• Ask for confirmation when chat attempts to edit sensitive files.
• New auto approve option to allow all future commands for the session.
• Rich terminal output in chat.
• Bind keyboard shortcuts to each custom agent individually.
• And more.

Besides AI changes, VS Code 1.107 also added support the latest revision of the MCP specification, and provide GitHub remote MCP Server as a built-in MCP server (Preview).

It as well added Intel Macs and Debian-based Linux Distros support for Microsoft Authentication, enabled Terminal Suggest for stable users, introduced new model for next edit suggestions.

Other changes include:

• 3-finger swipe on trackpad to navigate between editors in macOS.
• Preview next edit suggestions outside the viewport.
• Attach variables, scopes, and expressions to chat context.
• Classic Microsoft authentication no longer available.
• Rename suggestions for TypeScript.
• Ability to disable automatic hover popups in the editor.
• Add Stashes node in the Source Control Repositories view (Experimental), allows to see the complete list of stashes, view, apply, and pop each stash.

Get Visual Studio Code 1.105

For more about the new release, as well as download links for Windows, macOS, and Linux, go to VS Code website via the link below:

For Ubuntu users, besides download & install the deb package from the link above, there’s also official Snap package available to install in App Center (or Ubuntu Software for 22.04-).

NOTE: The source code for VS Code is open-source, but the packages above are proprietary freeware.

For open-source package, there’s a community maintained flatpak package is also available for choice. See this guide for how to install them in Ubuntu.

For Ubuntu, Fedora Workstation, and other Linux with recent GNOME Desktop, there’s now a new extension to enable the top-bar in multiple monitors.

As you know, GNOME top-bar by default only shows in the primary display for multi-monitors working in the “Join Mode”.

There was a multi-monitors-add-on extension which can add multiple monitors overview and panel. It’s however discontinued and support ends at Gnome 3.38.

Some forked that extension making it work in GNOME from v42 to 26, and I’ve written about how to install it in Ubuntu 22.04 & 24.04.

Now with the new Multi Monitor Bar extension, all Linux Distributions with recent GNOME v45 ~ 49 (e.g., Ubuntu 24.04, Ubuntu 25.04/25.10, Fedora 42/43, Debian 13, and Arch) can easily enable the top-bar on multiple monitors.

As you see via the screenshot above, besides the top-bar, it also shows following items in the non-primary displays:

• Activities button (dot and pill icon).
• date and time menu.
• system tray icons and indicators.
• the overview screen.

However, there are still some downsides. There’s neither overview search-box nor app grid to launch apps from the non-primary display. And, following things do NOT work properly either:

• Scroll on dot-and-pill icon does not switch workspace.
• Some indicator icons do not show in non-primary display.
• The built-in screenshot UI options only available in primary display, and “Screen” selection only take the primary screen, though area selection works in all screens.
• System tray icons display incorrectly when you full-screen something in primary screen. See this issue. In which case, you need to restart the extension to reset.

Install the extension to Enable Gnome Top-bar in Multi-monitors

For Ubuntu 24.04 and higher, simply launch App Center, search & install Extension Manager (filter by Debian package).

Next, launch Extension Manager and switch to “Browse” tab, finally search & install the “Multi Monitor Bar” extension.

After installed the extension, Gnome top-bar should display automatically on your external monitors. And, you may switch to “Installed” tab in Extension Manager to configure that extension with following options:

• Turn on/off panel, activities-button, date and time button in additional monitors.
• Enable/disable hot-corners function.
• Add more indicators to additional monitors.

For other Linux, simply launch web browser and visit the extension on EGO:

Then, install the browser extension (if it asks) with the link in that page and refresh. Finally, use the ON/OFF toggle to install the extension.

Tips: besides installing the browser extension, Debian/Ubuntu need to also install the agent package by running the command below in terminal:

sudo apt install chrome-gnome-shell

Finally refresh the page to see the toggle option.

After installed the extension, install “Gnome Extensions” in GNOME Software or your system package manager and use it to manage the extension preferences.

That’s all. Enjoy!

JetBrains announced the 2025.3 release of its IntelliJ IDEA JAVA IDE yesterday.

Like PyCharm 2025.3, the new version does no longer provide the Community Edition. Instead, it introduced an unified version that bring the Community and Ultimate editions together in a single product. While, Community users can continue using the IDE for free.

Also, it updated the UI with new default Islands theme for both light and dark mode.

Besides JetBrains Junie AI agent, the release also added Claude Agent to the same chat interface. While, BYOK (Bring Your Own Key) feature will be coming soon, allowing to connect to OpenAI, Anthropic, or any OpenAI API-compatible local models using your own API key.

To improve your workflow, the new version introduced command completion as new way to access context-aware actions directly from code completion.

You can now type . to see code completion, postfix completion, and action suggestions, type .. to filter completion suggestions to actions only, and select the action to see the exact context and preview what the action will do if applied.

This feature was initially introduced in last 2025.2 as Beta feature, now enabled by default in this 2025.3 version. While, a new “Enable command completion for read-only files” is available as new beta option.

IntelliJ IDEA has Java 25 support from the day when it was released. Now, with the new version, all bundled tools and libraries are compatible with Java 25 runtimes. And, JDK 25 builds from all major vendors are available for download directly from IntelliJ IDEA.

Moreover, the Develocity IntelliJ plugin brings profiling into the build process, allowing to view detailed build characteristics directly in IntelliJ IDEA, identify performance bottlenecks, and observe how build changes affect speed and resource utilization.

So far, the plugin supports Gradle builds, while Maven support is coming soon.

Other changes in IntelliJ IDEA 2025.3 include:

• Ability to view remaining AI Credits, renewal date, and top-up balance inside the IDE.
• Replace the term query console with query file.
• And Ultimate only features:
  • First-class support for the latest improvements in Spring Framework 7 and Spring Boot 4.
  • First-class Spring Data JDBC support.
  • Spring Debugger updates with remote applications debugging support, and auto-connect to user database on application starts.
  • Full Vitest 4 support, including its test discovery logic.
  • Kubernetes experience improvements.

Get IntelliJ 2025.3

The official release note and download link for 2025.3 are available via the link below:

For Linux, it’s a portable tarball includes executable file and most runtime libraries. Just download and decompress it, then run the executable file from extracted folder to launch the IDE.

Ubuntu users may choose install the official Snap package from App Center (or Ubuntu Software for 22.04 and earlier).

For choice, there are also community maintained Flatpak packages that runs in sandbox environment. And, I’ve written a step by step guide show how to install it in Ubuntu for beginners.

After 4 months of development, JetBrains finally announced the release of PyCharm 2025.3 today.

The new version of this Python IDE does NO longer have Community Edition, but introduced an unified version that brings everything together in a single product. While, Community users can continue using the IDE for free.

The IDE updated its UI with new default Islands theme for both light and dark mode.

It introduced new welcome screen for new users, with ability to create new script or notebook, clone a repository or connect to a remote development environment, and import files into your workspace.

For existing users who would like to try out this welcome screen, just enable “Show the Welcome screen in non-modal mode by going to Settings -> Advanced Settings -> Welcome Screen.

Besides the built-in Junie by JetBrains, the new version now supports Claude AI Agent from a single chat. And, it’s going to introduce Bring Your Own Key (BYOK) feature, that supports OpenAI API-compatible local models in next point releases.

For users with uv installed, Pycharm now automatically suggests uv as the default environment manager in the New Project wizard. And, for projects managed by uv, uv run will be used as default command for run configurations.

And, uv will use user selected Python version either by selecting the corresponding system version or downloading and managing the right one for you.

If you’ve previously selected a different environment manager (e.g., venv or Conda), it will remember that preference and continue using it by default.

The release also expanded its language server protocol (LSP) integration with new tools from the Astral, Microsoft, and Meta ecosystems, including ty, Pyright, and Pyrefly.

Other changes include:

• Support for PEP 765 (Python 3.14+) that disallow using return, break, or continue to exit a finally block.
• Support the asdf runtime version manager.
• New dedicated Python Process Output tool window that displays every command run by the IDE along with its full standard input, output, and error streams.
• Package and version completion for any filename that includes “requirements”.

There are also professional only features, including:

• Fully Jupyter notebooks supporte in remote development
• Latest updates from the DataGrip 2025.3 engine.
• Latest improvements from WebStorm 2025.3.
• Automatically analyze pandas DataFrames to detect the most common data quality issues.

Get Pycharm 2025.3

The official release note for version 2025.3, as well as the download link are available in the link below:

For Ubuntu, user may choose install the official Snap package from App Center (or Ubuntu Software for 22.04 and earlier).

For beginners, I’ve written a step by step guide shows how to install the IDE in Ubuntu via different ways.

After a month of beta development, Firefox 146.0 is available to download for Linux, Windows, and macOS users.

The new release of this web browser finally added native support for fractional scaling on Linux with Wayland session (e.g., Ubuntu 22.04+ and Fedora Workstation).

Mozilla Firefox merged Wayland fractional-scale-v1 support more than 2 years ago. The feature was however disabled by default due to bugs.

Now with Firefox 146.0, Wayland fractional scaling is natively supported and enabled by default, thus text, icons, and UI elements will render more effective on HiDPI displays with 125%, 150% etc scaling.

For user in the EU and some other countries, there is a New Tab Weather opt-in workflow available. And, user can choose whether to enable location detection or manually search for a location.

When type in search bar, Firefox 146.0 can now show you the direct results in the drop-down suggestions box. Though, this is a progressive roll-out feature that’s only available for small percentages of users.

For Windows, the new version introduced a scheduled backup feature, that automatically backup the passwords, bookmarks, and browsing data daily, and save them on local device.

The backup can be encrypted with a password and used to restore a fresh install of Firefox on any operating system to the status where you left off. Though, this is also a progressive roll-out feature so far only works in Windows, and will be coming to other operating systems soon.

Also for Windows, the Direct2D hardware accelerated 2D graphics rendering support has been removed! For those who still require this feature, it’s recommended to switch to Firefox ESR 140.0 or higher.

And for macOS, Firefox now has a dedicated GPU process by default. It includes WebGPU, WebGL, and Firefox’s own WebRender. With it, fatal errors in graphics code will no longer crash the browser, instead transparently restart the GPU process.

Other changes in Firefox 146.0 include:

• Firefox Labs is now available to all users, for trying out experimental features.
• Improve ‘Colors’ dialog in Settings to make it easier to understand.
• Show English-language suggestions for holidays and other important dates for users in France, Germany, and Italy using the English-language versions of Firefox.
• Support the contrast-color() CSS function.
• Add support the @scope rule.
• Introduce text-decoration-inset property.
• Support legacy -webkit-fill-available keyword for CSS width and height properties.
• Support ML-KEM for WebRTC.
• Support compressed elliptic curve points in WebCrypto.
• Update Skia graphics library with improved rendering performance and compatibility.
• Hide unused CSS custom properties.
• Provide full keyboard and assistive technology support, when the timepicker is enabled for <input type=\"time\"> and <input type=\"datetime-local\">.
• Various security fixes.

Get Firefox 146.0

This post is written according to the Github release-note, while the official announcement and download link will be available soon via the link below:

If you can’t wait, then go to this page to get Firefox 146.0.

Flowblade, the free open-source Linux video editing software, released new 2.24 version few days ago.

The new release of this multi-track non-linear video editor focused mostly on UX improvements, bug fixes and GTK 4 port, while there are also some new features.

First, the new version improved support for Compound Clips, the way to group multiple clips into a single container clip on the timeline, by adding 2 new ways to create it.

Previously, user may either select multiple clips (via Ctrl or Shift + click) and use right-click menu to create Compound Clips, or go to menu Project -> Create Container Clip -> From Current Sequence to use the entire current timeline/sequence.

In the new version, user may also select a timeline range (press I in timeline for start point, and O for the end) or use box selection method to create the container clips.

It also added new Auto Sync Trim Single Childs feature, so that child clips now follow the their parent clips to be trimmed automatically when trimming the parent clips, though there’s an option available in “Syncing” menu to toggle on/off the function.

The release also improved the Rotomask editor with new Box Select Move edit mode, allowing to move a selection of edit points as a group. And, the line color is now editable through a new color picker tool.

Moreover, new “Add Title” option is added to both “Project” menu and Media panel context menu, allowing to launch the titler dialog. And, the titler dialog added options to set the title name, and close the dialog automatically when clicking save.

Other changes include new Alt+J keyboard shortcut to mark selection range, new “Warning on Disk Cache Size” limit (4 GB and 8 GB), warn if estimated size exceeds 60% of the available disk space, as well as following changes:

• Allow moving audio master meter position to bottom row.
• Add ‘To Marks In’ and ‘To Marks Out’ buttons to player button panel.
• Fix Audio Sync Split undo bug.
• Add progress bar dialog for “Duplicate sequence…” action.
• Make keyframe info bigger and move it to the left side in keyframe editors.
• Fix ripple trim undo.
• Clear box move data on undo/redo.
• And various other minor improvements and fixes.

As mentioned, the release also includes some GTK4 port, such as new keyboard shortcut framework with per widget shortcuts, and more. See the RELEASE NOTE for details.

How to Install Flowblade 2.24 in Ubuntu & other Linux

Flowblade provides official installer for Linux through Flatpak package, which runs in sandbox environment.

Linux Mint and Fedora workstation (with 3rd party repository enabled) can search & install it from either Software Manager or GNOME Software.

While, Debian/Ubuntu and other Linux may do the following steps one by one to install the package.

• First, Debian/Ubuntu users can open terminal (Ctrl+Alt+T) and run command to install the Flatpak daemon:

  sudo apt install flatpak

  For other Linux, follow the official setup guide to enable Flatpak support.

  

• Then, run the single command below to install the Flowblade flatpak package:

  flatpak install https://dl.flathub.org/repo/appstream/io.github.jliljebl.Flowblade.flatpakref

  

After installed the software package, either start it from start menu (log out and back in if app icon not visible). Or, run the command below to launch it from terminal without re-login.

flatpak run io.github.jliljebl.Flowblade

And, you may replace run in last command with update to check & install updates.

Uninstall Flowblade

To uninstall the Flowblade Flatpak package, open terminal (Ctrl+Alt+T) and run command:

flatpak uninstall --delete-data io.github.jliljebl.Flowblade

Optionally run flatpak uninstall --unused to remove useless run-time libraries.

Audacity, the free open-source audio editor and recording application, released new 3.7.6 version few days ago. PPA updated for all current standard Ubuntu releases.

The new release of this cross-platform (Windows, Linux, and macOS) computer software added the FFmpeg 8.0 multimedia library support.

With the most recent FFmpeg library, it supports loading more audio codecs, such as Sanyo LD‑ADPCM, G.728 (.g728), and ADPCM IMA Xbox. And, custom FFmpeg export options now supports exporting to OGG/Opus audio formats using the libopus codec.

For cloud users, besides uploading your audio to audio.com, it now also supports loading your audio projects or audio files from cloud (meaning audio.com), though you need to log in with a free account.

According to the release note, Audacity 3.7.6 also added first simple implementation of Spectrogram Wavelet analysis. It’s a new visualization mode allows to analyze audio using wavelet transforms with more adaptive time‑frequency display than the traditional Fourier spectrogram.

The new feature should add a “Wavelet (1/6 Octave Hann)” algorithm mode into the Spectorgram Settings dialog. However, I cannot see it in my case in Ubuntu 24.04.

I’m neither good at audio editing nor coding. It could be that I can’t meet the new “if condition” added in this commit (see code below) to make it visible, or the feature is just disabled.

#ifdef WAVELET
      XO("Wavelet (1/6 Octave Hann)") ,
#endif

The new version also backported the 4.0 feature that user can middle-click on waveform and drag moving left/right just like the default behavior in GIMP, Kdenlive, Blender, etc applications. And, the scrollbar and viewport will update in sync for natural movement.

Other changes are mostly bug-fixes. They include:

• Fix “Share Audio” (upload to audio.com) function.
• Fix incorrect import 16-bit flac files.
• Fix audio list could not be accessed with keyboard.
• Handle utf-8 file names correctly on windows.
• Keep next/prev buttons focused on page change.
• Tool libraries and translation updates.

How to Install Audacity 3.7.6 in Ubuntu & other Linux

Option 1: Audacity provides official Linux package through AppImage, which is available to download at the link below:

It’s a non-install Linux package format, that has most runtime libraries in bundle. Simply add executable permission from file properties dialog, then run it to launch the audio editor.

The AppImage is marked for 20.04/22.04, though it works in most Linux. And, Ubuntu 22.04 and higher need to install libfuse2 library first in terminal (Ctrl+Alt+T).

sudo apt install libfuse2

NOTE: Both Flatpak and PPA packages below have the networking support disabled, meaning no audio.com upload and import support.

Option 2: For choice, you may install Audacity 3.7.6 in most Linux through the community maintained Flatpak package. It runs in sandbox environment, and supports both amd64 and arm64 processors.

Linux Mint may simply search for and install it from Software Manager, though you need to first enable “unverified Flatpaks” in the preferences dialog.

While Debian/Ubuntu may run the 2 commands below one by one to install the package.

sudo apt install flatpak

flatpak install https://dl.flathub.org/repo/appstream/org.audacityteam.Audacity.flatpakref

Option 3: For those who don’t like running Audacity in sandbox, I’ve uploaded the new release package into this unofficial PPA for Ubuntu 22.04, Ubuntu 24.04, 25.04 and 25.10 on amd64 and arm64/armhf platforms.

To add the PPA and install the native .deb package, run commands below one by one:

sudo add-apt-repository ppa:ubuntuhandbook1/audacity
sudo apt update
sudo apt install audacity audacity-data

Uninstall

For AppImage, simply delete the file from file manager. While, you may run the command below to uninstall the Flatpak package:

flatpak uninstall --delete-data org.audacityteam.Audacity

Also run flatpak uninstall --unused to remove useless runtimes.

To uninstall the PPA as well as the .deb package, run commands:

sudo add-apt-repository --remove ppa:ubuntuhandbook1/audacity
sudo apt update
sudo apt remove audacity audacity-data

Use apt or apt-get to install/upgrade PPA packages but the download speed is slow? Here’s a workaround to speed it up in Ubuntu or Linux Mint.

By choosing a download mirror that nears to me using “Software & Updates” usually can fix slow downloading issue when installing packages from official Ubuntu repositories.

But for app packages from Ubuntu PPAs, in most time it downloads at a speed of few hundreds KB/S.

It’s OK for small apps, but for PPA packages with large file sizes (e.g. NVIDIA driver from Graphics Driver Team PPA, or Plasma Desktop from KUbuntu PPA), it can take half an hour and even more time to install due to slow download speed.

In the case, apt-fast can help!

It’s a free open-source shell script wrapper for apt/apt-get, by using aria2 multi-connection download utility. It works by splitting large file into multiple small pieces, then downloading them from multiple sources in parallel.

Though, the launchpad PPA does not have mirrors, it can still speed up the large file downloading probably due to multi-connections. In my case, it reduced the nvidia-590 packages downloading time from about 30 minutes to less than 3 minutes.

Step 1: Install apt-fast

apt-fast has an official PPA that contains the most recent packages for all current standard supported Ubuntu releases (e.g., Ubuntu 22.04, Ubuntu 24.04, Ubuntu 25.04/25.10), as well as the old Ubuntu 20.04, 18.04, 16.04, and even 14.04 LTS releases.

To add the PPA, press Ctrl+Alt+T to open terminal and run command:

sudo add-apt-repository ppa:apt-fast/stable

Type your user password when it asks (no visual feedback) and hit Enter to continue.

All current Ubuntu releases will automatically refresh cache while adding PPA. But for Linux Mint and old Ubuntu, you may run the command below to update cache manually:

sudo apt update

Finally, install the wrapper by running command:

sudo apt install apt-fast

After installed the package, it will automatically run the post-install script to configure the basic things, like which package manager (apt, apt-get, or aptitude) to use for installing packages, maximum allowed number of connections, and, whether to ask for confirmation before downloading.

Step 2: Configure apt-fast (Optional)

Besides the basic settings you set above, you can configure more about apt-fast, including:

• add/change downloads mirrors, if you want to use it for getting packages from Ubuntu repositories.
• change the minimum size of each piece, and piece selection algorithm.
• adjust the ‘aria2c’ download command.
• change the temporary download folder, and apt cache directory.
• enable/disable APT authentication support.
• change the colors, and more.

Simply, run the command below to open the configuration file via nano command line text editor:

sudo nano /etc/apt-fast.conf

Then edit according to the description text if need, finally Ctrl+S to save and Ctrl+X to exit.

Step 3: How to use apt-fast

You can use apt-fast just by replacing apt or apt-get in all install, upgrade, etc commands with apt-fast.

For example, install app packages:

sudo apt-fast install package1 package2 package3

Install all available updates, but never remove existing packages:

sudo apt-fast upgrade

Install all available updates, allow removing existing packages if required:

sudo apt-fast full-upgrade

Clean download files in cache directory and temporary download folder:

sudo apt-fast clean

For more, see manual by running man apt-fast command.

Uninstall apt-fast (Optional)

If something goes wrong, or you don’t need it anymore, then open terminal (Ctrl+Alt+T) and run command to remove it:

sudo apt remove apt-fast aria2 libaria2-0

And, remove the PPA either by using “Software & Updates -> Updates”(or Software Sources for Linux Mint) utility or by running command:

sudo add-apt-repository --remove ppa:apt-fast/stable

Also, you may need to run sudo apt update to refresh cache afterward in Linux Mint or old Ubuntu 14.04.

NVIDIA 590, the next new feature branch driver for Linux, is available for beta testing.

This beta driver is NVIDIA 590.44.01, which however does not introduce any new features, but raised the minimum system library requirements and fixed some bugs.

The new driver now requires wayland >= 1.20 for the modern Wayland session, xserver 1.17 or higher for classic Xorg session, and glibc library >= 2.27. For Ubuntu, 22.04 and higher match the Wayland requirement, while 18.04 and 20.04 with Xorg are still supported.

Besides the minimum requirements update, the release also removed support for GeForce 700, GeForce 800M, GeForce 900/900M, GeForce 10 series, as well as GeForce MX100, MX200, MX300 series notebooks.

And, it improved the performance of recreating Vulkan swap chains. The swap chain is essentially a queue of images waiting to be presented to the screen. By improving the performance of swap chains recreation, it can help prevent stuttering when resizing Vulkan application windows.

Others are mostly bug-fixes. They include:

• Fix that “PowerMizer” preferred mode drop-down menu in the nvidia-settings control panel didn’t not function correctly on Wayland.
• Fix that the Dots Per Inch (DPI) to be incorrectly reported for some monitors, such as Samsung Odyssey Neo G9.
• Fix Vulkan applications not working in VMs using Venus VirtIO virtual GPU.
• Fix system freezes on PREEMPT_RT real-time kernels

For more about NVIDIA 590.44.01 as well as the official .run installers for x86_64 processors, see this nvidia webpage. Or, go to this page for aarch64 processor, FreeBSD, and Solaris.

How to Install NVIDIA 590.44.01 in Ubuntu

If nothing goes wrong, Ubuntu will add the official NVIDIA 590 driver package for next Ubuntu 26.04, and perhaps backport it for current Ubuntu 25.10 and 24.04/22.04 LTS releases. Though, the process can take few months. Keep an eye on this page for the official NVIDIA 590 package by Ubuntu.

For choice, you may add the long standing Graphics Driver Team PPA, which usually adds the new driver package earlier that the official package mentioned above.

For those who can’t wait, the Ubuntu developer Jacob Martin has made NVIDIA 590.44.01 into this personal PPA for Ubuntu 22.04, Ubuntu 24.04, and Ubuntu 25.10 for testing purpose!

NOTE: the PPA package is for testing only purpose. Don’t try it on production machine!!! It breaks my wi-fi in my case when trying to restore NVIDIA 580.

To add the PPA, open terminal (Ctrl+Alt+T) and run command:

sudo add-apt-repository ppa:jacobmartin/nv-graphics-2

Then, try launching “Additional Drivers” utility to install the driver. Or run the command below instead in terminal:

sudo ubuntu-drivers install nvidia:590

When done, reboot computer.

If the driver does not work properly for you, then try running the command below to uninstall:

sudo apt --purge remove '*nvidia*590*'

OnlyOffice desktop editors, the free open-source office suite for Linux, Windows, and macOS, released new 9.2 version yesterday.

The new release added AI agent support, customizable keyboard shortcut, macro recording for repetitive actions, and Windows on ARM support.

In the bottom left of the app window, an “AI agent” option is added, allowing to connect to either Anthropic, OpenAI, TogetherAI, or OpenRouter AI models through API key, or Ollama local hosted AI model.

After that, you may chat with AI to ask questions, find and open files on your computer, create new documents, presentations, or PDFs, generate text or HTML from files without open them, fill forms, and do more actions.

Besides that, the Document, Spreadsheet, Presentation, and PDF editors have AI toolbar options, allowing to do AI-powered translations, summarization, grammar correction and spell-checking. Though, they all need an AI model connected as mentioned above.

For those who often perform repetitive actions with OnlyOffice, the new version added View -> Record macro menu option, allowing to automate your actions by recording them as macros.

The new version also added support changing the keyboard shortcuts. Simply go to File -> Advanced Settings -> Keyboard Shortcuts, then click on “Customize” button. In pop-up dialog, find out your desired shortcut and either double-click or click ‘edit’ icon to change it.

Though, many shortcuts are locked as unchangeable, and a “Restore All to Defaults” option is available to reset them all.

Moreover, the new version added official support for Windows on ARM64 computers, such as Surface Laptop 7, Galaxy Book4 Edge, etc laptops with Snapdragon X series processors. And, ARM64 support for Linux is coming in future releases.

Other changes in OnlyOffice Desktop Editor 9.2 include:

• Custom color support for the Redact feature in PDF editor.
• Ability to add descriptive text labels to checkboxes and radio buttons.
• Assign specific roles to new fields when inserting them into form.
• For more, see the official release note.

Get OnlyOffice Desktop Editor 9.2

For Ubuntu, the office suite can be installed easily through the App Center (or Ubuntu Software for 22.04). It’s the official Snap package that runs in sandbox environment. Though, at the moment of writing, it’s still at version 9.1.

For choice you may download the app package for Windows, Linux, and macOS from its website. Where, Ubuntu user may choose download:

• native DEB package, then click open with App Center or Ubuntu Software to install.
• non-install AppImage, that can launch the office after adding executable permission.
• or Flatpak package which runs in sandbox environment.

TLP, the popular battery power saving tool for Linux laptop, released new 1.9.0 version yesterday!

The new release of this free open-source software introduced its own power profile daemon called tlp-pd, which in certain respects added back the GNOME/KDE/Cinnamon’s built-in power mode settings.

As you may know, TLP conflicts with power-profiles-daemon, the background service for common Linux providing ability to switch between Performance, Balanced, and Power Saver power profiles. Thus, installing TLP will remove the built-in power mode options in GNOME, Cinnamon, and KDE desktop environments.

Now with TLP 1.9.0, tlp-pd is introduced to replace power-profiles-daemon by implementing the same D-Bus API that GNOME, KDE, Cinnamon use. So, the power mode options in your desktop seem just like they are never gone.

Besides that, it added new tlpctl command. Without sudo permission, it allows you to switch power profile from command line, query current power profile, or run custom command using a specific power profile.

Meaning that you may start your games or certain apps in performance mode, while leaving all others running in balanced or power saver.

For example, the command below will start the SuperTuxKart game in performance mode, no matter what power profile you set system wide.

tlpctl launch -p performance /usr/games/supertuxkart

Moreover, it updated the /etc/tlp.conf configuration file. Now, all parameters end with _AC will work when Performance profile is active, parameters end with _BAT will work for Balanced profile, and parameters end with _SAV will work for Power Saver profile. While, all other parameters work for all profiles.

And, it added new TLP_AUTO_SWITCH option in configuration file, that controls the automatic switching of the power profile. It can be set to:

• 0, never switch power profile, and use the value of TLP_DEFAULT_MODE if set.
• 1, auto switch to performance when AC connected, or balanced when running on battery.
• 2, retain the manually selected profile (e.g., via your desktop’s built-in option) even when the power source changes.

Other changes in TLP 1.9.0 include:

• Add Battery Care (battery charge start/stop threshold) support for Tuxedo/Clevo laptops.
• Add TLP_DISABLE_DEFAULTS to deactivate all intrinsic defaults of TLP, so only parameters activated in config file applied.
• Fix tlp discharge to a target percentage for Chromebooks, Framework laptops.
• Fix Thinkpad X201, X220 discharge.

For more, see the changelog in Github.

How to Install TLP 1.9.0

TLP has an official PPA that supports Ubuntu 22.04, Ubuntu 24.04, Ubuntu 25.04/25.10, and their based Distros, e.g., Linux Mint 21/22, Zorin OS 17/18.

To add the PPA and install TLP 1.9.0, press Ctrl+Alt+T on keyboard to open terminal and run the commands below one by one:

sudo add-apt-repository ppa:linrunner/tlp

sudo apt update

sudo apt install tlp tlp-pd tlp-rdw

After installed the software, you may just forget it, as the default settings are already optimized for battery life. Though, you may follow the official documentation to install it on other Linux Distributions, or configure it to fit your need.

Uninstall:

To uninstall tlp in Ubuntu or Linux Mint, open terminal and run command:

sudo apt remove tlp tlp-pd tlp-rdw

Also run the command below to remove the PPA:

sudo add-apt-repository --remove ppa:linrunner/tlp

And, install back the system built-in power profiles daemon package:

sudo apt install power-profiles-daemon

Linux Kernel 6.18 was released yesterday! According to the version history, it will be probably the next Long Term Support (LTS) kernel release.

Linus Torvalds announced this kernel release on this lkml.org page:

So I’ll have to admit that I’d have been happier with slightly less bugfixing noise in this last week of the release, but while there’s a few more fixes than I would hope for, there was nothing that made me feel like this needs more time to cook. So 6.18 is tagged and pushed out.

…

What’s New in Linux Kernel 6.18

On CPU side, the new kernel updated Intel Trust Domain Extensions (TDX) to work with Kexec for being able to load and boot into a new kernel from another currently running kernel. It updated FRED with patch for the late breaking incompatible change. And, it added a workaround for Xeon 6 to avoid performance issue.

For AMD, it updated Error Detection And Correction “EDAC” driver with next Zen 6 “Venice” server processors support. And, added patch for the old Bulldozer processors with X86_NATIVE_CPU option support for maximizing performance.

The Kernel also added support for Sheaves, a new opt-in, per-CPU and array-based caching layer, updated IOMMU driver with ACPI and MIPS vendor extensions support for RISC-V, and introduced Loongson Security Engine support for handling offloaded RNG, TPM2, and various crypto acceleration on LoongArch processors.

It as well added patch to reliably handle 255+ vCores on AMD EPYC servers, enabled Wildcat Lake graphics support, introduced new Rocket accelerator driver for the NPU found on newer Rockchip SoCs, and added Google’s PSP encryption for TCP connections.

The Bcachefs code is removed from Kernel 6.18, users should now use DKMS kernel modules instead. And, exFAT file-system driver has been optimized leads to 16.5x speedup for loading time.

The new Kernel also added many new hardware support. They include:

• Haptic touchpad support.
• New sensor driver for GPD handheld gaming devices.
• Awinic AW86927 haptic driver for Fairphone 5 smartphone.
• Hynitron CST816x touchscreen control.
• New EC driver for Lenovo ThinkPad T14s Gen6 With Snapdragon X Elite SoC.
• AI button and Fn keys support for on Xiaomi Redmibook laptops.
• Red and green status LEDs support on QNAP NAS devices.
• Sensor monitoring support for Dell OptiPlex 7040, ROG STRIX X670E-E, X870-I/X870E-E GAMING WIFI, etc motherboards.
• Apple M2 Pro, M2 Max, and M2 Ultra.
• Qualcomm Snapdragon X1 laptops of the Dell Thena, HP OmniBook X14, Dell Inspiron 14 Plus, Dell Latitude 7455, HP Omnibook X14, and Lenovo ThinkBook T16.

For more about Linux Kernel 6.18, see either kernelnewbies.org or phoronix.com pages.

How to Install Kernel 6.18 in Ubuntu 24.04+

The Mainline Kernel PPA has built the new kernel as .deb packages, which are available to download through the link below:

Depends on your CPU architecture type (run uname -a or dpkg --print-architecture command to tell), select download either amd64 for Intel/AMD, armhf/arm64 for 32-bit/64-bit ARM, or ppc64el for IBM POWER, s390x for IBMz or LinuxONE.

NOTE: the Mainline Kernel packages are NOT signed! You need to disable secure-boot to boot the kernel. And it may be not work properly with NVIDIA proprietary driver.

For Intel/AMD platform, you may open terminal or connect to the server console, then run the commands below one by one to download & install the Kernel:

cd /tmp

wget -c https://kernel.ubuntu.com/mainline/v6.18/amd64/linux-headers-6.18.0-061800-generic_6.18.0-061800.202511302339_amd64.deb

wget -c https://kernel.ubuntu.com/mainline/v6.18/amd64/linux-headers-6.18.0-061800_6.18.0-061800.202511302339_all.deb

wget -c https://kernel.ubuntu.com/mainline/v6.18/amd64/linux-image-unsigned-6.18.0-061800-generic_6.18.0-061800.202511302339_amd64.deb

wget -c https://kernel.ubuntu.com/mainline/v6.18/amd64/linux-modules-6.18.0-061800-generic_6.18.0-061800.202511302339_amd64.deb

sudo apt install ./linux*-6.18.0*.deb

After installed the packages, restart your computer or server and run uname -a to verify.

Uninstall Kernel 6.18

For any reason, you may remove this Kernel by firstly rebooting into the old Kernel (usually under Grub -> Advanced options of Ubuntu)., then run command:

sudo apt remove --autoremove linux-headers-6.18.0* linux-modules-6.18.0-*-generic