Hermann
28.04.2009, 19:11
X-Plane 9.30 Beta-8 is here for Mac, Windows, and Linux!
I am really refining the flight model here!
As well, we are really focusing on making sure 9.30 is really bug-
free... if you can find any bugs, report them to me at this email addy!
austin
Here is the new Stuff for 9.30 Beta 8:
FLIGHT-MODEL REFINEMENTS:
Further flight-dynamics refinements... I think we can actually,
consistently, pull off a pretty decent approximation of a SPIN now!
The secret: No longer does the airplane really have any idea of a
flight path!
Instead, each PART of the airplane has it's OWN flight path.
So now, an airplane is NOT really a single entity: It is a whole bunch
of parts all moving in formation!
As odd as this seems, THIS is the way to understand how EACH bit of
the airplane has it's OWN flight path (different for each bit of the
airplane if the plane is pitching, yawing, or rolling), and THIS is
the secret that really allows us to get a good SPIN: Each bit of the
wing, stabilizer, etc all have their own flight path angle and
speed... this builds up the forces in the dynamic cases of hi spin-
rates: Most noticeable, not surprisingly, in a spin. We have had an
approximation of this since version 1.00, but only now, for 9.30
Beta-8, have we really completely overhauled the physics to really
have no concept at all of the flight path of the 'airplane', and
instead handled it totally independently for each ELEMENT of each
surface of the plane. This, combined with the new wind model that has
no concept of the wind on the 'airplane', but instead knows the wind
on each BIT of the airplane. THIS is really where we start to get into
serious computations to learn new things about how the plane will fly.
Just a HAIR more refinement the lift, drag, and moment lookup
calculations... just enough to make the propeller efficiency model
just a hair more accurate, in theory. A little more refinement in the
transonic flight-model as well: That tricky speed between Mach 0.85
and 1.00. It's a very minor refinement, but we may see a hair more
accuracy here. Remember: If you mean to break Mach-1, you really will
want THIN wings to make that happen! (And plenty of power, and
altitude, and a jet engine inlet efficiency speed of more than Mach-1).
To give you an idea of the accuracy we are seeing here, if you design
a wing with a bit of extra incidence near the root, or an airfoil that
produces more lift near the root, and less so at the tip, then X-Plane
will see, in flight, that there is more lift near the ROOT of the
wing, and INCREASE the effective aspect ratio of the wing in flight,
reducing the spillover of air around the wingtip, and increasing the
efficiency of the wing. Warning: As in reality, the reciprocal is also
true!
Improvements in the propulsive-efficiency indication: It USED to
indicate the transmissions losses as well, which was really sort of
WRONG, since most people would not lump the transmission-losses into
the propulsive efficiency of the propeller... so now the propulsive
efficiency of the propeller is really showing the ratio of power GOING
TO THE PROP that is making it to the airstream as thrust, not making
any reference to losses in the transmission. This also now properly
indicates propulsive efficiency for unusual configurations, such as
where multiple engines drive one prop, or vice-versa. This is only a
difference in INDICATION on the data output screen, and should not
actually effect the flight model, but there you have it.
Further electrical system modelling... We should now handle the
various cases properly where the APU, ground power, air-driven
generator, and regular generators are all overloaded, going into a
voltage sag and passing the load down to the battery, all as in reality!
Flap-Overspeed protection system (Plane-Maker, Special Controls screen),
Flap-Overspeed Indicator (EFIS airspeed indicator as moving red barber-
pole),
and Flap-Overspeed flaps-removal (Operations and Warnings screen) are
all now scaled to run the max allowable flaps smoothly between Vfe1
and Vfe2 based on all intermediate flap deflections and speeds.
A bit more tuning on the FADEC system, to give finer throttle control
and make overboost a bit less likely.
GENERAL FLYING FEATURES:
Towplane should circle to the left on climbout, as desired.
EFIS-App Avidyne now has ADF-1 and option for ADF-2, as requested.
GENERAL SIM FEATURES:
If you put the mouse over the lower-right corner of the screen in X-
Plane, then a small rectangle appears there to remind you that you can
just drag the corner to re-size the X-Plane window now!
Tuned mouse-control input now follows the non-linear power-curves that
you can set for the joystick, which is kind of nice for mouse-users.
Generate OBJ from ACF gives a much more detailed OBJ for you to use as
a starting point for your aircraft OBJs for the plane or custom scenery.
New commands for transponder and fuel-transfer.
Sound volume sliders show %, as requested.
SLIDERS can go to a power curve as specified in plane-maker, now.
Mag compass and clock should function in FDR replay mode. No change in
fdr format at all, we are just extracting more data from it to display!
MINOR TWEAKS OR BUG-FIXES:
Proper HUD-colors restored.
The custom setting of prop incidence in plane-maker now auto-sets for
all subsequent props when you enter data for the first one.
The parachute should not catch wind from the propwash now... typically
this will be a bit more accurate.
Old ARGUS nav-instrument back working again!
Tip-rockets operational again... they were out for the earlier betas..
sorry!
Bezier lines can now use the LAYER_GROUP command - see http://scenery.x-plane.com/library.php?doc=linspec.php
WARNING: ATTR_light_level has a syntax change - see http://scenery.x-plane.com/library.php?doc=obj8spec.php
New apt.dat (from April-09) cycle included
Newer version of the C172 and Cirrus Jet
Generic rotary brightness fixed. It will now match the EFIS when set
to "glass" mode.
Panel texture does not cut a hole in the airplane shadow.
Planet map is much faster.
3-D flight path with FDR recorder files is faster.
Translucent panel mode should work.
Fixed: newly created key binding won't be "stuck" down.
HUD won't turn black when brightness is turned down.
LOD handling tuned for large monitors.
Cockpit object texture is now uncompressed for better visual quality.
Generic trigger instruments won't double-press commands when combined
with ATTR_cockpit_region
"Draw outside of fuselage" checkboxes in Plane-Maker should work as
expected.
Gamma level on windows defaults to 2.5
"make_panel_previews" command makes one panel texture even when
subregions are used.
Lots of new datarefs - documentation will be posted shortly.
I am really refining the flight model here!
As well, we are really focusing on making sure 9.30 is really bug-
free... if you can find any bugs, report them to me at this email addy!
austin
Here is the new Stuff for 9.30 Beta 8:
FLIGHT-MODEL REFINEMENTS:
Further flight-dynamics refinements... I think we can actually,
consistently, pull off a pretty decent approximation of a SPIN now!
The secret: No longer does the airplane really have any idea of a
flight path!
Instead, each PART of the airplane has it's OWN flight path.
So now, an airplane is NOT really a single entity: It is a whole bunch
of parts all moving in formation!
As odd as this seems, THIS is the way to understand how EACH bit of
the airplane has it's OWN flight path (different for each bit of the
airplane if the plane is pitching, yawing, or rolling), and THIS is
the secret that really allows us to get a good SPIN: Each bit of the
wing, stabilizer, etc all have their own flight path angle and
speed... this builds up the forces in the dynamic cases of hi spin-
rates: Most noticeable, not surprisingly, in a spin. We have had an
approximation of this since version 1.00, but only now, for 9.30
Beta-8, have we really completely overhauled the physics to really
have no concept at all of the flight path of the 'airplane', and
instead handled it totally independently for each ELEMENT of each
surface of the plane. This, combined with the new wind model that has
no concept of the wind on the 'airplane', but instead knows the wind
on each BIT of the airplane. THIS is really where we start to get into
serious computations to learn new things about how the plane will fly.
Just a HAIR more refinement the lift, drag, and moment lookup
calculations... just enough to make the propeller efficiency model
just a hair more accurate, in theory. A little more refinement in the
transonic flight-model as well: That tricky speed between Mach 0.85
and 1.00. It's a very minor refinement, but we may see a hair more
accuracy here. Remember: If you mean to break Mach-1, you really will
want THIN wings to make that happen! (And plenty of power, and
altitude, and a jet engine inlet efficiency speed of more than Mach-1).
To give you an idea of the accuracy we are seeing here, if you design
a wing with a bit of extra incidence near the root, or an airfoil that
produces more lift near the root, and less so at the tip, then X-Plane
will see, in flight, that there is more lift near the ROOT of the
wing, and INCREASE the effective aspect ratio of the wing in flight,
reducing the spillover of air around the wingtip, and increasing the
efficiency of the wing. Warning: As in reality, the reciprocal is also
true!
Improvements in the propulsive-efficiency indication: It USED to
indicate the transmissions losses as well, which was really sort of
WRONG, since most people would not lump the transmission-losses into
the propulsive efficiency of the propeller... so now the propulsive
efficiency of the propeller is really showing the ratio of power GOING
TO THE PROP that is making it to the airstream as thrust, not making
any reference to losses in the transmission. This also now properly
indicates propulsive efficiency for unusual configurations, such as
where multiple engines drive one prop, or vice-versa. This is only a
difference in INDICATION on the data output screen, and should not
actually effect the flight model, but there you have it.
Further electrical system modelling... We should now handle the
various cases properly where the APU, ground power, air-driven
generator, and regular generators are all overloaded, going into a
voltage sag and passing the load down to the battery, all as in reality!
Flap-Overspeed protection system (Plane-Maker, Special Controls screen),
Flap-Overspeed Indicator (EFIS airspeed indicator as moving red barber-
pole),
and Flap-Overspeed flaps-removal (Operations and Warnings screen) are
all now scaled to run the max allowable flaps smoothly between Vfe1
and Vfe2 based on all intermediate flap deflections and speeds.
A bit more tuning on the FADEC system, to give finer throttle control
and make overboost a bit less likely.
GENERAL FLYING FEATURES:
Towplane should circle to the left on climbout, as desired.
EFIS-App Avidyne now has ADF-1 and option for ADF-2, as requested.
GENERAL SIM FEATURES:
If you put the mouse over the lower-right corner of the screen in X-
Plane, then a small rectangle appears there to remind you that you can
just drag the corner to re-size the X-Plane window now!
Tuned mouse-control input now follows the non-linear power-curves that
you can set for the joystick, which is kind of nice for mouse-users.
Generate OBJ from ACF gives a much more detailed OBJ for you to use as
a starting point for your aircraft OBJs for the plane or custom scenery.
New commands for transponder and fuel-transfer.
Sound volume sliders show %, as requested.
SLIDERS can go to a power curve as specified in plane-maker, now.
Mag compass and clock should function in FDR replay mode. No change in
fdr format at all, we are just extracting more data from it to display!
MINOR TWEAKS OR BUG-FIXES:
Proper HUD-colors restored.
The custom setting of prop incidence in plane-maker now auto-sets for
all subsequent props when you enter data for the first one.
The parachute should not catch wind from the propwash now... typically
this will be a bit more accurate.
Old ARGUS nav-instrument back working again!
Tip-rockets operational again... they were out for the earlier betas..
sorry!
Bezier lines can now use the LAYER_GROUP command - see http://scenery.x-plane.com/library.php?doc=linspec.php
WARNING: ATTR_light_level has a syntax change - see http://scenery.x-plane.com/library.php?doc=obj8spec.php
New apt.dat (from April-09) cycle included
Newer version of the C172 and Cirrus Jet
Generic rotary brightness fixed. It will now match the EFIS when set
to "glass" mode.
Panel texture does not cut a hole in the airplane shadow.
Planet map is much faster.
3-D flight path with FDR recorder files is faster.
Translucent panel mode should work.
Fixed: newly created key binding won't be "stuck" down.
HUD won't turn black when brightness is turned down.
LOD handling tuned for large monitors.
Cockpit object texture is now uncompressed for better visual quality.
Generic trigger instruments won't double-press commands when combined
with ATTR_cockpit_region
"Draw outside of fuselage" checkboxes in Plane-Maker should work as
expected.
Gamma level on windows defaults to 2.5
"make_panel_previews" command makes one panel texture even when
subregions are used.
Lots of new datarefs - documentation will be posted shortly.