Re: The endless influx of low-quality, low-realism, quantity-over-quality aircraft into FGAddon

The default filtering in the launcher already reduces the list of 
aircraft from ~750 to 62; and some of those should be variants.

There are some UI issues with how that filter works:

   1. The filter controls sit at the top of the aircraft browser and 
scroll out of view once you are into the list.
   2. Switching from a currently-installed aircraft into the browse view 
positions the list at that aircraft, which can leave the filter controls 
a long way off-screen above.
   3. Search ignores the filter entirely. It would help if results 
indicated something like "not showing N filtered aircraft" so the user 
knows the filter is still hiding results.

It would also help if the average star rating were shown alongside the 
favourite-star indicator - probably as "3.4 out of 5" with the stars 
drawn beside it. Implementation is left as an exercise for the reader.

Point (3) is probably the reason users still encounter low-rated 
aircraft that the filter would otherwise hide from them.

Alan asked whether we have any statistics on aircraft usage.

For the last 365 days (2025-05-22 to 2026-05-22), 1,858 distinct 
aircraft names appeared in that traffic and 165,666 flight-hours were 
logged.

Top 10 by flight-hours over that window:

    1. A320neo-PW       10,071 h    2,717 pilots
    2. F-16              8,394 h    5,034 pilots
    3. 777-300ER         6,930 h    1,197 pilots
    4. c172p             6,870 h    3,929 pilots
    5. 777-300           4,669 h    1,406 pilots
    6. c172-common       3,345 h    2,017 pilots
    7. A320-200-CFM      3,224 h      946 pilots
    8. F-15C             3,138 h    7,402 pilots
    9. A320neo-CFM       3,041 h    1,047 pilots
   10. A320-200-IAE      2,970 h      884 pilots

Top 10 by distinct pilots over the same window:

    1. F-15C             7,402 pilots    3,138 h
    2. F-16              5,034 pilots    8,394 h
    3. c172p             3,929 pilots    6,870 h
    4. A320neo-PW        2,717 pilots   10,071 h
    5. f-14b             2,599 pilots    2,493 h
    6. su-57             2,495 pilots       22 h
    7. c172-common       2,017 pilots    3,345 h
    8. 777-300           1,406 pilots    4,669 h
    9. 787-8             1,237 pilots    1,354 h
   10. 777-300ER         1,197 pilots    6,930 h

How activity is distributed across the catalogue:

   - Half the aircraft were flown by three different pilots or fewer
   - 356 aircraft (19%) were flown by exactly one pilot all year
   - 1,006 aircraft (54%) were flown by ten pilots or fewer
   - The bottom half of the list by hours accounts for 0.9% of all hours
   - The top 10 hold 31.8% of all hours; the top 50 hold 62.4%

The other useful angle is how much each pilot actually flies a given 
aircraft. Hours-per-pilot is a useful quality signal because a low value 
is an indication of an aircraft most pilots fly and forget, whatever the 
underlying reason.

Of the 1,201 aircraft that at least five distinct pilots tried over the 
year:

   - 246 (20%) averaged under 15 minutes of flying per pilot
   - 463 (39%) averaged under half an hour per pilot
   - 724 (60%) averaged under one hour per pilot

The mean across this set is 1.6 hours per pilot. The worst cases are 
striking:

   - su-57 sits at 0.01 h/pilot (an average of about 30 seconds per 
pilot before they switched off)
   - mirage2000 at 0.02
   - several helicopter passenger seats and AI-only models near zero

This is multiplayer traffic only, so solo offline flight is invisible; 
and the 1,858 figure includes hangars and dev builds as well as FGAddon, 
which is why it is well in excess of the ~750 in the catalogue.

The aircraft people consider low quality are generally not being flown 
all that much - either because almost nobody tries them at all (more 
than half the catalogue has ten pilots or fewer in a year), or because 
the people who do try them give up in a matter of minutes (40% of 
aircraft with five or more pilots average less than half an hour of 
flying per pilot).

The point worth taking from this is that what we currently have seems to 
be working. The catalogue and the players are largely self-regulating. 
Because FlightGear is free, people can - and do - try out whatever takes 
their fancy.

Anything we can do to improve the experience will be a good thing, 
whether this be UI improvements, fixing aircraft or updating ratings.

The point worth taking from this is that what we currently have seems to 
be working. The catalogue and the players are largely self-regulating. 
Because FlightGear is free, people can - and do - try out whatever takes 
their fancy.

Richard



_______________________________________________
Flightgear-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/flightgear-devel