My work this month has focused primarily on QA of the C port of RADEv1 as well as updating the official FreeDV application to use it instead of the previous hybrid Python/C implementation of RADE. The recent C port of RADE, as mentioned here, has been a gamechanger for many reasons. One significant improvement users will see immediately is a reduction in the file download size. For example, the Linux AppImage for 2.2.1 is over 600 MB in size, while the most recent versions our CI system have been generating are only 69 MB (a nearly 90% improvement!)

Another improvement is in performance, both on startup and during regular use. With the Python implementation, for instance, it would take up to 30 seconds after pushing the Start button for my 2019 MacBook Pro to begin decoding OTA signals, while the most recent builds only take around a second or two. CPU usage is also dramatically reduced, particularly for more constrained systems; the FlexRadio waveform, for instance, only uses around 30% of a single core of a Raspberry Pi 4 now when not actively decoding a signal (whereas that would have been closer to 80%+ with the Python implementation).

In the latter half of the month, I’ve also been working on chipping away at our pending feature request backlog. In particular, these features have been implemented and are slated for release in 2.3.0:

• [Feature Request] Tune button for netting and ATU activation
• Add TX SNR / RX SNR fields to LogQSO and export RX SNR via UDP Type 5
• Easy access reporting ON/OFF switch needed in main GUI
• Tab for Frequency Delta
• [Feature Request] Add SNR plot to main window

For this month, I’ll continue supporting feature requests and any work required to improve the C port of RADEv1 / move RADEv2 forward.

More information can be found in the commit history below:

• freedv-gui
• radae
• radae_nopy

(Note that all commit logs above were generated with the following command line:)

git log --author="member@email" --after "Month 1, 2026" --before "Month 31, 2026 23:59:59" --all