Thanks for the reply.
Yes I know that. What I mean is to get the gain graph not the rest that are produced by dump/readsb in network mode. To have the gain I believe control of the dongle must be done either by readsb or dump and in that case stream1090 controls the dongle.
Hi nikall, thanks for the logs! I can see that stream1090 works at 36.4 dB in your setup. Which value do you see when you use dump1090-fa?
We need a comparison to understand whether that costs us distant aircraft or not. You have both a clone and an original V4, do you also have a splitter to run both at the same time from the same antenna?
Can you provide an I/Q capture when there’s a distant aircraft that dump1090-fa receives while stream1090 doesn’t, so I can see the noise and see if there’s anything we can do?
Which dump1090-fa version, with which settings are you running?
For the charts, I can probably create a JSON file for graphs1090. Are you running everything with docker or just normal binaries?
Thanks,
Enrico
Hi Enrico,
Thanks for looking into the logs and the info.
So my suspicion that your implementation exhibits similar results to readsb seems legit. The only difference is that readsb keeps gain way below 36.4.
Here is a 2 hours worth of gain values as being handled by dump1090-fa adaptive gain. That’s after the problem I mentioned with the high noise at 1078MHz. Before that, the same graph would have much wider range in values (between 24 and 45).
And here the setup for the adaptive gain/adaptive burst.
RECEIVER_GAIN=
ADAPTIVE_DYNAMIC_RANGE=yes
ADAPTIVE_DYNAMIC_RANGE_TARGET=20
ADAPTIVE_BURST=yes
EXTRA_OPTIONS="--adaptive-burst-alpha 1 --adaptive-burst-change-delay 1 --adaptive-burst-loud-runlength 2 --adaptive-burst-quiet-runlength 4 --adaptive-range-change-delay 1"
That way I can fully control how long it can stay at lower gain, at higher gain and how fast it can switch. That’s why I consider Dump1090-fa adaptive gain implementation superior to Readsb. It gives you full control and it really stands for its name “adaptive” meaning you can adapt it to your own needs.
I build from scratch, I don’t use pre-built debs.
dump1090-fa --version
-----------------------------------------------------------------------------
| dump1090 ModeS Receiver dump1090-fa 11.1~bpo12+1 |
| build options: ENABLE_RTLSDR |
-----------------------------------------------------------------------------
detected runtime CPU features:
selected DSP implementations:
magnitude_uc8 lookup_unroll_4_generic
magnitude_power_uc8 twopass_generic
magnitude_sc16 exact_float_generic
magnitude_sc16q11 exact_float_generic
mean_power_u16 u32_generic
count_above_u16 generic_generic
Unfortunately I don’t have a splitter. Can you suggest one to buy (preferably from China) ?
I also want to buy a 50Ohm dummy load to test if the noise is coming from the dongle side.
regards
Hey, thanks for the info. I haven’t studied it in detail yet, but in general you shouldn’t compare the gain level between different implementations, ad stream1090 is quite robust with clipped messages. In the meanwhile I can give you this link - I’ve measured it with a VWNA3 and the numbers are quite promising:
https://www.aliexpress.com/item/1005011591580045.html
Definitely an A/B test will tell us a lot here, please consider getting one and running your two receivers in parallel.
Hi,
thanks for the reply and the link. I will buy ti and let you know as soon as I receive it in order to do the comparison test.
I don’t compare the two implementations (although it would be nice to have the same customization level) I’m just saying that dump1090-fa is for the time being a better fit for my case (mountainous area, closest airport at ~35km, many overhead high altitude flights and more recently, high level of noise).
Alright, thanks.
Is there any RSP1B owner in this thread? If so, please check out: RSP1B owners: looking for testers for experimental SDRplay support · mgrone/stream1090 · Discussion #85 · GitHub
Is something for SDRplay RSP-duo or RSPdx user?
Hey, I don’t have either, but I don’t see why I can’t enable them.
The code is pushed now, can you give it a try?
I think yes Why not…
Στις Κυρ, 27 Σεπ 2026, 20:53 ο χρήστης enricolorenzoni59 <notifications@flightaware.discoursemail.com> έγραψε:
Some of you may have done some custom filter optimization for stream1090. One thing i always wanted to do, is optimize the filter by hand and see if i can do better than the automatic one.
In filter_utils there is now besides filter_opt.py also filter_edit.py. It is an interactive tool to design your own filters. I provided also some examples in filter_presets. Together with the plotting script this is actually quite nice.
Some more here:
