Originally posted by Mike
View Post
Announcement
Collapse
No announcement yet.
Selection of Radar sites for web display?
Collapse
X
-
Originally posted by rik130 View PostIt was pretty simple : Mike said that out of 150 feeders in a highly populated area, 140 to 145 of them are ignored, so it was a perfectly valid question what happens to the Piper at 1500ft down in a valley, for example, if those 5-10 feeders are all located on the other side of the hill and can't "see" it because of the terrain.
Comment
-
OK, but why do "drop" the perfectly good data from a site at 5 miles out, and "pick" a site that is 40 miles away, and not even in the direction of the flight?
Especially when it seems that those few sites are "favored" consistently. What makes them more valuable for you? I gave examples in my first post.
Because I want to see what I can do better.
Obviously, of course this is about ADS-B only.
Originally posted by Oblivian View PostYou have to understand how databases work to get a handle on what actually goes on at the server to get the stuff you see publicly displayable. And fast
....
In that sense. Working out who sent the most recent data and making sure its valid down to 1 person. Is very, very low priority.
Is it something to do with the receiver mode selection? Is some data format "better"? I'm using ModeS Beast TCP for example.Last edited by SoNic67; 2018-09-21, 23:47.
Comment
-
We are trying to keep it as simple as possible. Instead om making a complex model that try to calculate distances, angles och directions to pick a receiver, it just uses random data. It doesn't really make any difference if we get the position from a receiver 1,50 or 150 km away. It's still exactly the same value. And no we haven't added any complexity to give some feeders higher priority. All is random and handled auomatically by the system.
Comment
-
Well, it doesn't look random to me.
At any given time I look at the FR24 web page and apply the filter for my radar site, I almost never see a plane. And, even when I hear a plane taking of from airport, and I see that plane, in short time it gets switched to another radar site.
I see some sites with multiple planes all the time. As we spoke, F-7NC21 has 10 planes in my radius, and several more further away. True, it has better range than me, but you said that didn't matter. F-KIAD1 had a bunch of planes next to IAD, but it also has three planes literally in 15 miles of me. F-KCHO1 has 8 plane in my range.
I can see those planes in my local Piaware page.
That's why I have asked if something else makes them to be favored.Last edited by SoNic67; 2018-09-23, 13:47.
Comment
-
Originally posted by SoNic67 View PostWell, it doesn't look random to me.
At any given time I look at the FR24 web page and apply the filter for my radar site, I almost never see a plane. And, even when I hear a plane taking of from airport, and I see that plane, in short time it gets switched to another radar site.
I see some sites with multiple planes all the time. As we spoke, F-7NC21 has 10 planes in my radius, and several more further away. True, it has better range than me, but you said that didn't matter. F-KIAD1 had a bunch of planes next to IAD, but it also has three planes literally in 15 miles of me. F-KCHO1 has 8 plane in my range.
I can see those planes in my local Piaware page.
That's why I have asked if something else makes them to be favored.
Filtering on my radar I show zero aircraft as well, even when they're overhead. I think FR24 might give the higher ranked feeders the credit for tracking radar. I'll only be able to confirm this as I rank up but that's my observation right now.
Anyway, this is fun and very interesting. I'm glad I got in to it!
Comment
-
Originally posted by MrBill View PostHi guys, I'm a newbie to the FR24 feeding and was wondering how the radar selection works as well. So my rank is obviously low on the scoreboard. Being in southern New England, I'm tracking on average with position, around 50 aircraft at any given time.
Filtering on my radar I show zero aircraft as well, even when they're overhead. I think FR24 might give the higher ranked feeders the credit for tracking radar. I'll only be able to confirm this as I rank up but that's my observation right now.
Anyway, this is fun and very interesting. I'm glad I got in to it!
Comment
-
Originally posted by SoNic67 View PostWell, it doesn't look random to me.
At any given time I look at the FR24 web page and apply the filter for my radar site, I almost never see a plane. And, even when I hear a plane taking of from airport, and I see that plane, in short time it gets switched to another radar site.
I see some sites with multiple planes all the time. As we spoke, F-7NC21 has 10 planes in my radius, and several more further away. True, it has better range than me, but you said that didn't matter. F-KIAD1 had a bunch of planes next to IAD, but it also has three planes literally in 15 miles of me. F-KCHO1 has 8 plane in my range.
I can see those planes in my local Piaware page.
That's why I have asked if something else makes them to be favored.
If you have better coverage, of course the chance of you showing up is increasing. User1 is tracking 100 aircraft and User2 is tracking 10 aircraft. User1 of course has 10 times higher chance of seeing his code, as he is tracking 10 times more, if feedring from the same area.
But if User2 has any unique coverage, of course his code will show up more often in these areas where he has unique coverage.
Comment
-
I get that this is how it should be. But something else is happening IMO.
Because, in my original comparation/example I can see those two radars:
Rank Radar Score Uptime Max Range Average Range
5,058 T-KPHF1 9,935 716 316 108
1,529 F-KFKN1 10,590 719 337 179
That's why I am trying to see what other factor is there. Like I said, I am "feeding" using ModeS Beast TCP from a Pi having an 8 bit dongle - not sure if matters or not. Looks like either the rank or the average range makes much more difference than that "random".
I am not saying that this is necessarily in the code, maybe is mathematical distribution function. Maybe ModeS Beast over TCP doesn't have the same payload as other type of data (like direct USB connection).
What's the difference between Direct, Beast, SBS1/1e, SBS3, AVR in that respect?
In my setup AVR TCP doesn't work for some reason, shows 0 aircraft tracked.Last edited by SoNic67; 2018-09-24, 09:27.
Comment
-
An interesting (to me) part of this question is one which sort of came up earlier in this thread: An aircraft will move in and out of the coverage of each receiver. So if a receiver has been asked not to send data for an aircraft, but it moves out of the coverage of the receivers that were asked for data on it, do you ask that first receiver to restart sending the data? If so, how do you know it can still see it? Or do you simply rely on the aircraft getting picked up by receivers in it's onward path?
Am I right to think that this explains why FR 24 range stats are lower than my local (dump1090) ones: because my receiver has been asked to stop sending data on my long range tracks? Here in the S.E. of U.K. we're not short of coverage! Please don't think this is a whinge about stats being "right" - I'm just curious about how they're generated. :-)Last edited by JFreddieF; 2018-09-24, 18:09.JFreddieF
T-EGWU97
Comment
-
Not sure why this topic is getting so much attention. Feeding data gets you a free Business account. Feeding more or less data makes no difference to that. The graphs on the feed stats page help you optimise your feed. Surely something sorting out FR24's excess load on and incorrect usage of the volunteer NTP pool network is more deserving of available resource than going over and over this old chestnut. Hopefully FR will remove the Radar from the webpage at some point then there will be nothing to complain about
Time for more caffeine
Comment
-
Originally posted by elljay View PostNot sure why this topic is getting so much attention.
Time to move on. This is a totally "dead" and irrelevant topic.
Comment
-
Understand why people ask the question - I asked the same question when I first set up my Pi. You've spent time configuring it and you want to see if your data is good. And you get a buzz when you see your radar name "in lights"
If so many people are asking the same question then it sounds like the data they need isn't available. Flightaware's data is very useful - stats on all sorts of things.
Hopefully this person will have the resources to put the data people want in place.
Comment
-
Sorry. My reply was a tongue-in-cheek response. Out of 18,000 plus feeders, only a handful ask about the radar label.
Maybe FR24 should have a "sticky" post?
"Do not ask about the radar label. It's totally meaningless. Check your feeder stats instead."
Comment
-
Or better: "Stop feeding data to us, we don't care".
I explained clearly why I care in my first post. Probably you didn't care enough to read that, but you cared to label this "meaningless". Nobody forces you to read this or comment on it.
But, well, if that makes you fell better about your self... what can I say?Last edited by SoNic67; 2018-09-26, 16:44.
Comment
Comment