How many times have we waited in anticipation for the new
simulation where advertisements boast claims of "ultra-realistic"
flight-models that don't stand up under scrutiny?
Fighter Squadron: Screaming Demons Over Europe is one of the most recent simulations to suffer the indignity of exagerrated claims.
Before you write off FSSD as a failure, however, all is not
lost. In an unprecedented effort, Michael Harrison (Senior Software
Engineer for Parsoft) has undertaken the daunting task of rebuilding
the flight models of the complete list of shipped aircraft. To further
exemplify his passion, he is doing this primarily on his own time. In
the following article, I will pass along what I've since learned from
Michael regarding what is missing, what is being done and how it will
affect this simulation, as we know it.
What's the big deal about accurate flight models?
To many this is an obvious question with an even more obvious
answer. But the truth is there are surprisingly quite a few first-time
sim'ers, who simply echo what they hear online and in forums without
fully understanding what they're hearing.
One of the most important aspects of ACM (air-combat maneuvering) is
knowing the limitations of your aircraft and your opponent's aircraft.
This knowledge dictates how you will engage, attack and defend yourself
from your opponent. If a simulation's flight-model does not properly
depict a given aircraft's performance accurately, you've effectively
removed virtually all the strategy from the simulation. Doing this to a
sim flight model is the equivalent of changing all the pieces of a
chess game into queens, or playing the game without any rules.
What's really wrong with the FSSD flight-models?
Most of the complaints and observations from the flight-sim
community focus primarily on the lack of proper departure from flight.
In actuality, stalls in FS are already present; it's the repercussions
of exiting the stall that are missing.
Currently when you do get into a stall situation you are welcomed with
a soft and very manageable recovery. Other issues, such as lack of
proper bleeding of speed during climbs or flat turns, were corrected by
the release of the 1.5 patch in which the viscosity of the air was
increased to better facilitate drag.
The reasoning for the flight departure problem is best answered by
a paragraph I received from Michael Harrison. "The airfoil data that
was generated had spikes in it (presumably to simulate the turbulence
that is typical during stalls) and the lift did not drop off fast
enough in the stall region. This meant that it was difficult to get
into a true stall.
"Furthermore, the weight distribution on the planes was off, causing
the "center of gravity" to be too far forward. This would aid the
aircraft in stall recovery, which is unrealistic."
To simplify... the generated data that was manipulated to induce
stalling did not have a long enough "occurrence" to allow the aircraft
to lose enough lift to enter a proper stall condition. That coupled
with the fact that the "center of gravity" was off, allowed the planes
to automatically "nose-over", causing an overly easy stall recovery.
Fine … How to fix it?
To better answer that question, let's discuss Fighter
Squadron's flight-model and how it works. Fighter Squadron's
flight-model derives its data from a CFD program Parsoft licensed from
NASA. With this program, airfoils (simple aerodynamic shapes that
develop lift) are placed within a liquid-dynamic environment and fluid
is run over the surfaces to determine their lift values. Tables of
lift, drag and moment (with optional modifiers for leading and trailing
edge devices) are used to parameterize each airfoil.
This data then is inserted into a specific airplane's LOD file, (a file
that contains all the aircraft's specific data) which in turn provides
us with the given flight-model, as we know it. Obviously once this raw
data is obtained, much of it can be manipulated and "tweaked" to
accurately match known aircraft specifications.
In case you are new to Fighter Squadron, all of the sims aircraft are
made up of multiple objects. Each of these objects independently has
lift, drag, moment, and damage values. It's the combination of these
objects that make the aircraft and its flight model as a whole. In
laymans terms … each individual object (pieces of the wing, fuselage,
propeller) have specific flight data that contribute to the entire
flight model of the aircraft. So when you shoot a portion of a wing off
an aircraft, the flight-model is affected in a real and dynamic way.
Let's go back to Michael Harrison:
"All the known texture mapping problems have been fixed and I've been
systematically going through all the airfoil files and fixing the curve
data for lift, drag and moment so that they are as correct as they can
be. Without a Cray, they won't be "perfect" but they will be quite a
lot better than "good enough" for a PC simulation.
"I'm currently going through all the aircraft and adjusting the weight
distribution so that it's all as correct as can be. (Which can be
difficult at times. Just how much * does* the outer wing section
weigh?) This particular modification is a tedious process that we
didn't go through for the initial release version of Fighter Squadron,
due mostly to time constraints and the fact that the aircraft flew
fairly close to "correct" as they were. When facing a deadline, it's
very difficult to get some people to understand that they may not be
correct, just "good enough.
Cool … So how will the changes affect FS?
One of the obvious answers is that people will have to get
used to planes stalling under certain conditions. No longer will the
aircraft allow people to simply "yank and bank" without any real
repercussions. Veterans should have little problem understanding and
adapting to the changes. However, "newbies" will have the difficult
tasks of learning how to keep the planes in the air.
A flyable B-17.
The "global" weight (the default weight) for an aircraft will remain
the same. But when one of the accrued objects that make up the aircraft
is removed (blown / broken off) … the new weight and inertia settings
will more realistically affect the performance.
The instruments are legible even out here.
What and when?
The important thing to keep in mind is that Michael Harrison
has undertaken this project on his own, to be completed largely on his
own time. To compound the difficulty in this undertaking, he is also in
the process of moving himself and his family to a new home.
As of our last discussion, the flight models were coming along nicely.
He has been prepping the P-51 and the Fw-190 and should have those out
shortly. Once those have had some limited testing, the rest should
follow shortly afterward. Suffice it to say that because Michael is
doing this on his own time, "real life" has precedence. Michael's best
guess for an ETA on the whole plane set is, "when they are done".
It is my personal opinion that thanks to Michael Harrison's
personal sacrifice and Parsoft's understanding Fighter Squadron will
rise out of the ashes of condemned simulations like a newborn Phoenix.
I expect it will easily meet the hyped expectations that had been
initially set out.
Unofficial FM Patch
Don't want to wait? Download Etienne Labuschagne's Flight Model patch
right now. This patch can be used with or without the 1.5 patch since
it only affects the aircraft models themselves. Changes/Fixes:
1) All the aircrafts' wing loadings have been increased to be
historically correct (as historically correct as my sources are!).
Those who's wing loadings I could not get, was changed via and educated
guess and some thumb sucking.
2) The Lanc, B17 and Me262 should not lose their engines anymore due to damage.
