Closed gvanem closed 1 year ago
I don't think it's an option for ifile .... not sure.
In general ... can you test the exact same command on linux? Then i know if there is an actual issue in readsb and you know if it's due to the port.
I don't think it's an option for ifile .... not sure.
The readsdb --help
page says it is:
ifile-specific options, use with --ifile:
--ifile=<path> Read samples from given file ('-' for stdin)
--iformat=<type> Set sample format (UC8, SC16, SC16Q11)
--throttle Process samples at the original capture speed
In general ... can you test the exact same command on linux?
I have only Windows Subsystem for Linux which is incapable of building readsb
. Even a sudo apt install zlib1g
fails?!
Run a VM.
I don't have one. Besides I'm not friend with Linux and my port did nothing to effect the cmdline parsing. IMHO there are far too many options for non-interactive mode.
And too little features for interactive-mode.
For example, I'd expected a --lat=60.3045800 --lon=5.3046400 --interactive ...
to show the distance to a plane. Would that be possible?
OK. I managed to build and run in WSL.
./readsb --device-type ifile --ifile samples-bin.iq --throttle
./readsb --throttle --device-type ifile --ifile samples-bin.iq
I.e. It shows the same in the 1st case:
<3> priorityTasksRun didn't run for 1681887115.5 seconds!
<3> removeStale didn't run for 1681887114.5 seconds!
<3> priorityTasksRun didn't run for 5.0 seconds!
....
And this SDR type '(null)' not recognized:
in the 2nd case. Not a porting issue AFAICS.
Yeah i just tested it myself and i needs to go after device type. I've made some improved error messages for it as i don't intend to change the entire system.
In response to more features for interactive and the port in general: Of course it's possible. Do i have interest to implement it? No.
readsb is meant to go with a nice webinterface (tar1090). You could just use the --write-json option and display it in any way you like.
The webinterface looks something like this: https://globe.adsb.lol/ Most people run it on a raspberry pi with the webinterface. A low power computer suits the 24h nature of how people run this stuff. It can also go in the attic close to the antenna.
I'm happy you have ported it to Windows, but i don't think i have much interest in maintaining a port, you'd have to put that into a fork on your github and maintain it yourself. I suppose it depends on how much code would need to be in ifdef sections.
You can just ignore this ...
<3> priorityTasksRun didn't run for 1681887115.5 seconds!
<3> removeStale didn't run for 1681887114.5 seconds!
<3> priorityTasksRun didn't run for 5.0 seconds!
I suppose ... maybe not i'll check how to fix that.
The webinterface looks something like this: https://globe.adsb.lol/
This is awesome. But no planes over my town Bergen (Norway).
you'd have to put that into a fork on your github and maintain it yourself.
That's what I'm doing; readsb works extremely good with the 2.4MS/s decoder. I'm hoping to port that over to my own dump1090 port. But hope to avoid all the thread-usage in readsb.
readsb is meant to do a lot of stuff. Thread usage isn't a bad thing, isn't there a pthread implementation that works on windows?
I've fixed the errors and hiccups with ifile ... at least most of it. It still pauses when it's done before it exits but that's an issue for later.
If you've already ported readsb ... not sure why you'd use dump1090 to be honest. readsb is just so much better ....
You can feed the aggrgators to add data. Possibly check the sdr-enthusiasts discord, it's a nice place: https://discord.gg/sTf9uYF
https://globe.adsb.lol/ https://globe.adsb.fi/ https://globe.adsbexchange.com/ (i worked on this until the owner sold it and now it still has good data but is no better than fr24 / fa .... might even start blocking soon no one knows)
I think I'll stick to https://planefinder.net/. I only wish they'd support distance-rings.
I've ported readsb to Windows and now trying to use it with the
--ifile <IQ-file>
and--throttle
options is rather confusing.Apart from
epoll()
andO_BINARY
issue with binary data, it was quite easy to port to Windows. First I generated IQ-data usingrtl_sdr -f 1090M -s 2.4M -n 1E8 samples-bin.iq
, and runningreadsb.exe --device-type ifile --ifile samples-bin.iq
shows:But adding the
--throttle
as inreadsb.exe --throttle --device-type ifile --ifile samples-bin.iq
, gives:But it runs fine without throttling.
Experimenting with adding it last
readsb.exe --device-type ifile --ifile samples-bin.iq --throttle
, takes 45 sec to complete. Makes sense (since areadsb.exe --device-type ifile --ifile samples-bin.iq
, takes 10 sec to complete).Why can't I have the
--throttle
option first?BTW. using
--interactive --throttle ...
,readsb.exe
crashes insideinteractiveCleanup()
since my PD-curses was not started.