On 03/11/2007, Melchior FRANZ <[EMAIL PROTECTED]> wrote:
> Not sure if you are talking about "dynamic view", but this *isn't*
> supposed to change the view based on physical forces. The first
> three lines in $FG_ROOT/Nasal/dynamic_view.nas are:
>
> # Dynamic Cockpit View manager. Tries to simulate the pilot's most likely
> # deliberate view direction. Doesn't consider forced view changes due to
> # acceleration.
I had been looking at dampEyeData in src/Main/viewer.cxx, assuming
that was causing the problem. I wasn't looking at any NASAL code. Is
the NASAL code enabled by default?
> And that's how I wanted it to be. I'm not a RL pilot, but I
> consider it quite unlikely that a pilot flies a left turn and
> doesn't look left, or strongly pitches up and doesn't look
> up. Usually we want to know in which objects we are about to
> crash. :-)
I can't speak for every pilot, but that doesn't jibe at all with my
real-life experience. I hadn't noticed the problem so much in VFR
flights, but flying FlightGear for an IFR approach, I actually lost
control of the Warrior twice before I figured out what was going on.
When you're flying IFR in actual IMC, you try to keep your head as
still as possible no matter which way you move the controls. Even
flying VFR in good day VMC, it's common (for me, at least) to look
*before* I start a maneuver then return my head to the
straight-forward position while I actually fly it, so that I can use
outside references properly.
Finally, the position you move the controls isn't necessarily the
direction the plane will move. For example, on approach you'll be
pitching the nose up but the plane's moving down.
Forces give a more realistic head movement, I think, because the head
moves only if something unusual is happening, like an uncoordinated
turn or high/low G forces. It's OK to disrupt the pilot in that case.
> There are some implementations of forced head changes used in
> various aircraft ("head shaking"), but those are IMHO too
> unrealistic, as they treat the head only as a mass on a spring,
> and completely disregard the physiology of the
> eye/equilibrium organ/brain system. If we could agree on one
> that *feels* realistic, then I would be willing to integrate
> it in dynamic view. But so far we didn't.
A spring's not quite right. I think any movement should be very
slight (max 2 cm), and could track the forces on the pilot (maybe
using a squared function so that minor forces have virtually no effect
and major forces are obvious). The head should be perfectly centred
when the pilot feels 1G in the (plane-referenced) vertical axis and 0G
in the forward and side axes. No matter what, though, the head should
not swivel unless the user decides to swivel it.
All the best,
David
-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
Flightgear-devel mailing list
Flightgear-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/flightgear-devel
