-10 means “enable tuner AGC”, not -10dB.
I know, but AGC is not the best choice. May be later I’ll write topic about difference in gain profiles in cases of “standalone” and “with LNA” setup and so on.
It’s better on some setups. Due to a quirk of librtlsdr, enabling the tuner AGC actually sets the total gain higher than you can manually set (the AGC is not suited to ADS-B signals and so it sets max gain; but the fixed gain stages in AGC mode are set higher than librtlsdr will let you manually set). It certainly doesn’t actually give you real auto-gain control.
Well, I understand how RTL-SDR work and I know difference between “manual” gain control and AGC (I mean exactly how hardware work and how librtlsdr control first stage of input amplifier), but IMHO “It’s better on some setups” only when setup is placed far away from TV/FM towers, cellular/trank/radio transponders and so on. In “real world” it’s not so easy. For example, my RTL-SDR stick maximum gain is 49,6dB, but it get best result with gain 44.5dB with my antenna. And when I connect my home made simple circuitry LNA with 17 dB gain - the best result riched with gain 33.8 dB on RTL-SDR stick. 33.8+17=50.8dB - it’s more then 44,5dB without LNA. Strange? No. Input LNA of RTL-SDR stick have noise figure above 3dB and overload value about +25dB, but my homemade LNA - less then 1.6dB and more then +32dB so I catch more airplains with RTL-SDR + LNA setup.
AGC produces better results here on an unamplified antenna. I don’t have strong interferers nearby. So: it’s better on some setups. I agree that it’s not going to be better everywhere. You have to try it and see. It’s really just another gain setting, somewhere around 52-53dB from memory.
I have some patches floating around that give you full control over the whole gain range without having to invoke the AGC as an odd workaround, but I haven’t got around to cleaning them up yet.
As I explane above - in my case without LNA 44.5dB is better then 49,6dB and especially better then AGC (of course I try AGC mode). LNA makes sense at first - if it nose figure lower and in second (especially in a dirty environment) - if its overload capability higher then internal RTL-SDR input amplifier have. In other cases LNA is usable only for compensate signal lost in cable.
I use AGC because that’t the best configuration for my current equipment and location. That’s it, nothing special.
I have helicopter park nearby and interesting about how hight its fly (for example 
). In other cases A/C mode give me more aircrafts, for example right now I view IKAO 4245A8 flight and in airframes.org/ database it have [MODE-S] [ADS-B] [HFDL]:
Registration Manuf. Model Type c/n l/n i/t Selcal ICAO24 Reg / Opr built test reg delivery prev.reg until next reg status
VQ-BTG Boeing 737-8FZ (W) B738 41992 4908 L2J 4245A8 Dobrolet 2014 2014-05-13 active edit
Remarks: leased from BBAM
[MODE-S] [ADS-B] [HFDL]
but translate only mode A/C messages.
Why you think that --oversample is better way to have more positions reported? IMHO wider passband give less S/N ratio signal. Correct me if I make mistake.
Yes, I understand.
Empirically, it gives about 30% more messages. The 2MHz demodulator really suffers from (a) having one sample per bit and (b) being phase-locked with the transmitter clock. Going to 1.2 samples/bit makes quite a big difference. The IF passband is the same in both cases AFAIK. Try it!
Ok. I’ll try it out. Thanks for description.
Update main post about MLAT configuration.
Best regards,
Serge.
Serge,
I really would like to thank you for your “manual”
As for me, has to be considered as one of the “perfect setup guides for your RPI”
I just combined the mutability setup from this topic : http://discussions.flightaware.com/ads-b-flight-tracking-f21/what-is-easiest-way-to-install-dump1090-mutability-t35431.html and that made my day 
Greetings from EBBR/BRU
Leopold
Excelent Tutorial Serge…
I’m following it Step by Step…
Some questions.
1-What’s the difference between configuring FR24 MLAT on installation
Step 1.3 - Would you like to participate in MLAT calculations? (yes/no)
$: (I’m describe later why turn off mlat in this step)
Would you like to use autoconfig (yes/no)
$:
and doing it after…
- MLAT on fr24feed together with piaware MLAT.
8.1. Setup MLAT on fr24feed - open fr24feed configuration file:
Code:
sudo nano /etc/fr24feed.ini
and change mlat=“no” to
Code:
mlat=“yes”
2- FlightAware’s MLAT is configured automatically with the PiAware client (no modifications need to be done?)
FlightAware (PiAware) and FlightRadar 24 (FR24Feed) are two different companies.
Step 1.3 - Would you like to participate in MLAT calculations? (yes/no)
$: (I’m describe later why turn off mlat in this step)
Would you like to use autoconfig (yes/no)
$:
- MLAT on fr24feed together with piaware MLAT.
8.1. Setup MLAT on fr24feed - open fr24feed configuration file:
Code:
sudo nano /etc/fr24feed.ini
and change mlat=“no” to
Code:
mlat=“yes”
This is not correct. Changing the ini file without using autoconfig will feed incorrect data to FlightRadar at the worst and will have the FR servers drop your traffic at the best. Do not try to fool the FR24 system by messing with the ini files. The FR24 system is in beta and designed to be fed directly from a dongle and not from another instance of dump1090. There will eventually be a version (possibly) that can use another decoder but for now it is not designed to work as a secondary feed. If you want to feed FR24 use the non-beta (non-mlat) version. Please don’t feed in extra noise this way.
Edit: You are also bypassing the FlightRadar24 questions about station coordinate location and antenna height, which are needed to calculate the MLAT information, please edit your post and remove the MLAT information for other services, you are giving incorrect information.
Thanks dschaper for the clarification. I saw a magical inscription " ok ] FR24 MLAT AC seen:" and did not understand further.
I was temporary remove MLAT chapter from main post yet I will not understand in issue.
Best regards,
Serge
Thanks dschaper for the clarification. I saw a magical inscription " ok ] FR24 MLAT AC seen:" and did not understand further.
I was temporary remove MLAT chapter from main post yet I will not understand in issue.
Best regards,
Serge
Thank you, there already is discussion at other flight feeder sites that mention FlightAware feeders specifically as being misconfigured and feeding in MLAT data as ADS-B/Mode S data and causing issues.
I did like bringing up the gain issue, I’m now going back to my gain setup and checking to see if AGC is the best with my setup currently or if bringing it down a bit on the non-amplified antenna is better.
Minor updating in item 2.4 in the main post.
Hello…according to me If you just want to plug it in and send data to FA, then download their image and run with it. It is a good starting place. If you want to get deeper into the performance and intricacies of the RPi, mutability makes a better gateway. Think of it as the difference between an automatic and a stick. Have another SD card handy and use multiple images.