Loophole Posted March 24, 2022 Posted March 24, 2022 I haven't really been entirely happy with the existing trim modes, and had been thinking of ways that might improve the experience. As a result I have come up with an idea for a different algorithm for the "trim" mode for the Apache (and potentially other DCS helicopters, and even other aircraft that have similar trim mechanisms). If you have already tried something like this internally and found it didn't work, that's fine - this is all hypothesis - but if it hasn't been tried, perhaps this approach might work better? The Existing Trim Options Of the current three options: "Instant trim", "Central Position" and "Without springs", I am going to ignore the "without springs" option, as I have no experience with it and expect it works just fine. I have found the other two options to both have certain disadvantages, and I hope this new approach might alleviate both and potentially replace both options. Disadvantages of the current trim options "Instant Trim": Unless you are very quick to re-center your controls, you will get a "bobble", as the system reads your off-center stick position as command input. "Central Position": Eliminates the bobble, but in my experience can eventually kill you, as the algorithm locks out all of your inputs until it sees a "perfect" (0,0) from the controls. If you don't use a significant dead zone then eventually you will hit a circumstance - typically, when trying to arrest a slide into a building while trying to dodge an IR SAM - when your frantic inputs just keep missing that magic (0,0) point and you die. My proposal here - if it actually works as I am hoping it might - would combine the best characteristics of both the existing algorithms - no "bobble", but also never locked out of control. Heuristic-Centering Trim When considering the control inputs there are three factors to consider: Your physical joystick position, The current trimmed control position, and The current joystick (0,0) reference position. The existing algorithms both calculate the Effective Control input at any time as: (current trimmed control position) + (physical joystick position) - (joystick reference (0,0) position) In both existing algorithms, the (joystick reference (0,0) position) is always at physical joystick (0,0). "Central Position" mode ignores all input after a "trim" command until observing (physical joystick position) == physical (0,0) Heuristic-Centering Trim would use the same logic as above, but with two additions: When a "trim" command is issued, the (joystick reference (0,0) position) is set to the current (physical joystick position) The Effective Control input calculation has one added step: If the (physical joystick position) is between the current (joystick reference (0,0) position) and the actual physical joystick (0,0), then set (joystick reference (0,0) position) to (physical joystick position) Calculate Effective Control input the same as before: (current trimmed control position) + (physical joystick position) - (joystick reference (0,0) position) The Player Experience How this will translate into player experience would (I hope) be something like this: You would treat it much the same as either of the existing modes: you press trim, then you re-center your stick and then continue giving control inputs. Initially everything is at physical (0,0). You apply stick (or rudder) inputs, and that translates directly to Effective Control inputs, exactly as you expect. You get the helo flying in trim with the stick at some position - call it (X,Y), and you press & release the "trim" button; your stick is at (X,Y); the Trimmed Control Position is set to (X,Y), AND the (joystick reference (0,0) position) is set to (X,Y). If you don't move the stick, the calculated Effective Control input is unchanged - (X,Y) - and there is no "bobble". As you then relax the stick back to physical (0,0), the (joystick reference (0,0) position) follows your action; Effective control input remains (X,Y) and there is still no "bobble". If - before you reach physical (0,0) - you suddenly find a need to add a little more input, the algorithm detects that and applies it proportionally from wherever (joystick reference (0,0) position) was when you started re-applying input. As you do this, (joystick reference (0,0) position) doesn't change unless you re-trim, or you again start relaxing the stick back to physical (0,0) and pass the current (joystick reference (0,0) position). If conversely you suddenly find the need to apply opposite input, you will experience a slight "deadzone" effect as you move the stick back past physical (0,0), but as soon as you move past physical (0,0) - at which point (joystick reference (0,0) position) hits physical (0,0) - your inputs immediately start applying as Effective Control inputs - you never get permanently locked out. So the slight "deadzone" effect as you move back to (0,0) is the only quirk you would experience, and if you are following the habit of "trim then re-center", I think you would barely ever notice it. On the up side, there should be no "bobble" effect, and you can never be unexpectedly locked out of your controls. I hope this idea is of some interest! 2
FalcoGer Posted March 24, 2022 Posted March 24, 2022 (edited) Doesn't seem impossible to implement, and also seems like it might work. from me. What should happen if you trim, then move the stick halfway back towards (0,0), effectively not adding any input on the helicopter, and then hit trim again? I would suggest that nothing should happen at that time, aside from whatever the helicopter thinks should happen when you press the trim button without moving the controls. Edited March 24, 2022 by FalcoGer
Recommended Posts