I'm getting reports that reporting high frequencies to pskreporter is broken. This is in fact due to the protocol specifying 4-byte integers. Now I'm told that WSJT-X actually packs more bytes. @n1dq I think the documentation https://t.co/ZyuJ3Afht7 should be updated accordingly

Installed #OpenWebRX to try while working on drivers. Must say I'm pretty impressed with it. Code quality is good, and documentation is too (!). My SoapyAirspyHF+ driver worked right out of the box. No problems at all. @airspy

Hey, I made it to Friedrichshafen! Looks like you'll find me at the HAM in the next few days. When I'm not roaming, I'll be probably be hanging out at the ARDC booth in A1.

Small Delights ( https://t.co/tZbDF9evNF)

Thanks to @Tomek_SP5LOT : #rpitx running on pizero sends sstv pictures / M17 / aprs from stratosphere on a balloon. https://t.co/F8oZbjS4MR

M17 voice transmissions of the SP5LOT-11 strato balloon received at the SR4DON OpenWebRX site, more than 200km away from its current position. https://t.co/ZbZfIsYrOe SSTV images are coming in beatifully, too!

let the merge begin.

Turns out that if your decoding performance is slowly degrading over time, it's a good idea to check the NTP setup. 🤦‍♂️ Switched the RPis on GPS PPS duty, didn't update their IP allocations. Issues took weeks to show up.

Today, for the first time, I witnessed embedded GPS coordinates in a DMR voice transmission. Talk about rare... still, the format appears to be documented, so it should be possible to parse it. Like, you know, for next time.

I have a pretty bad concurrency headache right now.

OpenWebRX Updated to V1.1.0 https://t.co/uSwM1JiR8O

Alright, here you go: OpenWebRX 1.1.0 has been released. All details: https://t.co/JmgXXjbNYk - Enjoy! 😀

Introducing python bindings to your library seems to be the best way to double the number of installation steps. Why isn't there be a way to build python bindings with cmake without the need to shoehorn setuptools into it?

So... I switched to using dynamic_cast in the function, and now I get a segfault from dynamic_cast. The same method must have passed earlier. Suspecting memory corruption. Can't valgrind because valgrind breaks mmap() and the application can't run.

It maybe too early to take measurements, but the first readings do look very good. OpenWebRX with an rtlsdr @ 2.4MS/s on a Pi3B, single client decoding NFM. Old: 35-36% CPU, new: 25-27% CPU.

Listening to narrow FM using the new pipeline components for the first time. Working beautifully, the result is on par with the current pipeline. Next up: Measuring the performance benefit. Oh, and of course, fixing all the stability problems...

So... it looks like I managed to get a segfault from the function declaration (note line 37). I'm failing to understand this one... 😵‍💫 Anybody who does?

Back in business 😀 When reading from the network, it's important to know if you're dealing with bytes or samples... the difference is * sizeof(T)

"mostly working"... no idea what those lines are, need to figure it out. but hey, it's a good feeling to see a waterfall again after a day of debugging segfaults

@G4ELI Hi As antenna manufacturer we are interested in your 8 elephant antenna. Q's: * Are the elephants aligned along the boom lengthwise or sideways? * How about the trunk, is it rolled up or straight? * Do other animals work as well? We don't have many elephants in Germany. 😀