View Issue Details

IDProjectCategoryView StatusLast Update
0000093Falcon BMS Known BugsAvionicspublic2021-03-24 22:57
Reporterairtex2019 Assigned To 
PrioritynormalSeverityminorReproducibilityalways
Status newResolutionopen 
PlatformPCOSWindowsOS VersionWindows 10
Product VersionBMS 4.35 
Summary0000093: AFThrottleUp/Down callbacks too coarse .. seem identical to AFCoarseThrottleUp/Down
DescriptionBoth AFThrottleUp/Down vs AFCoarseThrottleUp/Down callbacks seem to do exactly the same thing -- they both seem to produce very coarse steps, up/down. For example the smallest increment I can make while taxiing is from ~1000 lbs/hr (idle) is to about 3,200 lbs/hr. That's about a 5-point step in engine rpm .. from say 70% to 75% rpm.

I gather these throttle callbacks are not commonly used, but it would be nice to have some fine-grain throttle control .. for those of us with crappy old 8-bit sliders on their joystick that don't respond well to small inputs.
Steps To ReproduceTaxi...

Hit AFThrottleUp callback: [ctrl+alt+UpArrow]

Observe: fuel-flow jumps to around 3,000 lbs/hr. Expected: fuel-flow rises about 500 lbs/hr.

Return throttle to idle.

Hit AFCoarseThrottleUp callback: [shift+alt+UpArrow]

Observe: fuel-flow jumps to around 3,000 lbs/hr, as expected.
TagsNo tags attached.
Theatre of OperationsKTO

Relationships

related to 0000104 new Problems using mousewheel as Throttle 

Activities

anonymous

anonymous

2021-02-24 04:07

viewer   ~0000278

Also there is an AFIdle callback in BMS-Blank and BMS-Full key files, that appears to not work.

Not sure about AFABOn and AFABFull. They seem to sometimes work, sometimes not.. eg. maybe only when throttle axis is mapped to keyboard or mousewheel?
airtex2019

airtex2019

2021-03-03 02:23

reporter   ~0000318

Edit: AFIdle does sometimes work, just not reliably .. similarly to AFABOn/Full
airtex2019

airtex2019

2021-03-24 22:57

reporter   ~0000428

I've been experimenting with different approaches, eg. mapping my 4th/5th mousebuttons to keypresses to manipulate AFThrottleDown/Up.

I can send these keypresses programmatically (using Win32 SendInput API) to sometimes achieve smaller inputs. But it's not controllable or deterministic .. if the keypresses are shorter than 1 frametime, they sometimes get missed. And the shortest reliable keypress duration (~20ms) is enough to move the throttle +/- 1,000 lbs/hr. That's better than I can do by hand, but still too coarse to be useful.

Taking a look at the import table of FalconBMS.exe I see it is calling GetAsyncKeyState, as I suspected -- which means short keypress durations of ~5ms get missed a great deal of the time.

I think this has been happening to me, with other keys.. I'll make a note to press keys firmly for at least a full frame. lol

But in meantime .. whatever multiplier you're applying to the keypress duration, to move the throttle, needs to be greatly reduced.

Issue History

Date Modified Username Field Change
2021-02-12 00:19 airtex2019 New Issue
2021-02-24 04:07 anonymous Note Added: 0000278
2021-02-24 23:04 Blu3wolf Relationship added related to 0000104
2021-03-03 02:23 airtex2019 Note Added: 0000318
2021-03-24 22:57 airtex2019 Note Added: 0000428