github preston-peterson/aerodrome v3.4.197

28 minutes ago

A speed glitch your receiver repeats can no longer be the fastest record or Today's fastest. Since 3.4.157, a speed counts only if the aircraft's reading just before or just after it is within 40 kt. But a receiver keeps reporting an aircraft's last speed until a new one arrives, so one bad reading is often saved twice in a row, and the copy agreed with itself: 539 kt, then 1,362 kt twice, then 540 kt. Now identical readings in a row count as one reading, and the readings on either side of them have to agree. A real speed held for several readings still counts, because the readings before and after it agree.

A wrong record already stored corrects itself. On the first start after this update, a fastest record set by a repeated glitch is replaced by the best reading that passes, and the date shown is when that reading happened. You won't get a notification. Today's fastest and its detail panel use the same rule, so a repeated glitch drops out of both.

Operational notes

This is an ordinary update with a normal restart and no database change. The detail panel behind Today's fastest takes about twice as long to open, around 0.2 s on a busy install.

Behind the scenes

_reading_agrees and the live _prior_speed_agrees in collector.py treat a run of the same speed as one reading. The detail panel numbers each aircraft's runs with window functions and checks the readings on either side of each run. Measured on a real install with about 210 days of data, the rule removed four repeated glitches from the top of the all-time list (1,362, 1,071, 662 and 654 kt) and no real reading, and the record went back to the 668 kt jet-stream reading that the 3.4.157 check had found. The self-heal took 1.2 s. Replaying every day, the card and its panel agreed on the top speed each time, and the card changed on three days, each time to drop a repeated glitch.

Don't miss a new aerodrome release

NewReleases is sending notifications on new releases.