a. I read some other posts blaming rbfeeder
b. I have a bee in my bonnet about AN and data
c. It’s an age thing
d. All of the above
And after a dozen or so rewrites of /etc/default/dump1090-fa I’ve finally figured out it does go in the EXTRA OPTIONS and it is case-sensitive…lower case --no-modeac-auto so now it seems to have settled on @1750 messages per sec which equates with what I see in Graphs so thanks everybody for your help.
sudo systemctl status dump1090-fa can now see the flag --no-modeac=auto and a quick jq < /run/dump1090-fa/stats.json ‘.total.local’ returns;
{
“samples_processed”: 5040373760,
“samples_dropped”: 0,
“modeac”: 0,
“modes”: 34872558,
“bad”: 66795208,
“unknown_icao”: 14053876,
“accepted”: [
3414363,
144523
],
“signal”: -13,
“noise”: -22.5,
“peak_signal”: -0.5,
“strong_signals”: 32681
}
But now I’m left wondering if the gain of 42.1 is optimal and if the signal, noise, peak signal and strong signals shown above are the best that I can do!
Oh its all most definitely trial & error with me foxhunter!
I’m really pleased on progress I have made so far thanks to you guys in here as my uneducated, quick & dirty performance test has moved me back into the Top 10 RadarBox UK Rankings so it seems I’m on the right track but I was kinda hoping some knowledgable soul would cast their experienced eye over my graphs & signal data and point me in the right direction by saying something like;
You need to increase the gain a bit more
You need to decrease the gain a bit more
That’s exactly where the gain needs to be
For example the number of Messages > -3dBFS has dropped from 15% all the way down to 0.3% which would indicate I might have to increase it a bit as I see references to 5% being optimal?
As expected the ADS-B SIgnal Level has dropped since I adjusted the gain down from 60 to 42.1 but I don’t see a reduction in aircraft as far as I can tell.
It looks like ADS-B CPU Utilization has dropped from 22% to 16% with a corresponding decrease in Memory Utilization since I am not decoding Mode A/C any more but is not accompanied by a decrease in Core Temperature although this is not a concern as I am still safely within limits even when the outside temperature hit an all time record of 38.3c here!
Percentages from 0.5 to 5 are what you should be aiming for.
If you live close to an airport even 10% can be an optimal value, i’d still prefer 5% or less.
If most of your traffic isn’t close by but low / close by aircraft are important to you, go for 0.5 % or even lower to make sure the close by traffic is received without overload.
This is just a guideline, experimenting until you actual start losing close by planes due to overloaded receiver is encouraged.
If you’re using graphs1090, adjusting gain so your weakest signal is between -25 to -30 dBFS usually yields similar results, again it’s just a guideline. In case of focus on nearby aircraft you can also go lower with the weakest signal to something like -35 dBFS.
Source:
Please read the whole article.
There you will find also a short one liner how to check it at the command line without waiting for graphs1090 to be refreshed.
I’ve already read that and the 591 post ‘Thoughts on optimizing gain…’ thread as well as many, many more but as is usual for me the more I read the more confused I get!
Guess I’ll just have to experiment a little more. Problem is I have many things I’d like to try such as;
Optimise existing system
Compare FlightStick vs Pro Stick Plus with/without FA 1090 Dark Blue Filter
Convert across to Airspy Mini/R2
Kuhne 1090 LNA
etc.
All seems a bit daunting for me especially when I have a working system…
Appreciate your continued patience, help & support.
