Evaluating Realism
Robin J. Lee
One of the most widely held values in this flight simulation community
is what we loosely term the "realism" of the product. But too often a
sim is labeled "realistic" or "unrealistic" without a critical
examination of what it really means to say that. This often leads to
discussions that are not especially useful, because there is essentially
no common set of understandings upon which a useful discussion can be
based. The result is commonly a series of disconnected posts in which
individual posters single out isolated factors which may or may not be
"realistic," and judge the product on that basis. What I hope to do is
to identify and explore different aspects of "realism" that may help us
to refine future discussion.
SYSTEMS REALISM. This measures the success of a sim at rendering the
complex technical aspects of an aircraft, including but not limited to
avionics representation, pilot displays, cockpit layout, and other
mechanical functions such as simulating the correct number of stages of
afterburner, the correct number of independent flap settings, or the
manual option to switch anticollision lights on and off. In short, this
is simulating the physical aspects of an airplane at the lowest level;
this is the representation of an airplane as an airplane. Current sims
that have a high degree of systems realism include MSI's Back to
Baghdad, SSI's Su-27 Flanker, and MSC's F/A-18 Hornet 3.0.
PILOT-PERSPECTIVE REALISM. This addresses the capability of a sim to
capture the experience of a human pilot in a cockpit setting, which we
can all agree differs greatly from the experience of sitting in front of
a 15" monitor in a 1-G environment. Most of this is connected with
being able to achieve situational awareness, and includes such factors
as a workable viewing system that simulates the natural head motion of a
helmeted pilot, the effects of G-induced loss of consciousness, being able
o impart a sense of relative motion at all distances, and other "human
factors" which go into flying an aircraft. Sound, and the quality
thereof, is also included in this category. This is the fantastic realm
of VR headgear, force-feedback sticks, full-cockpit-simulators, and
creative padlock/viewing systems. Pilot-perspective realism focuses
less on simulating the aircraft and more on simulating the
player-as-pilot (and only as pilot, with considerations restricted to
those physical elements involved in actually flying the airplane - it
does not include combat elements beyond the purely physical.)
AERODYNAMIC-PHYSICAL REALISM. This one should be obvious to everybody.
Briefly, it evaluates the manner in which a sim presents the complex
physical interaction between wing and surrounding air. This is the realm
of turn rates and turn radii, acceleration and T/W ratios, wing loading
and inertia. There are certain non-aerodynamic elements in this
calculation as well, which explains the addition of "physical" to the
category - for instance, the physics of falling bombs or bullets or
pilots. The accurate portrayal of inertia plays a big role in this
category. In short, this factor concentrates on simulating the behavior
of the airplane, player, and other units in the sim as physical objects.
ATMOSPHERIC REALISM. This encompasses all of the "soft factors" that too
often are dismissed by hardened sim veterans as "eye candy."
Nevertheless, this is the element which appeals directly to the senses in
order to convince the player that the simulated world is real. Natural
elements include realistic looking terrain; sunrises and sunsets which
occur at the right times and in the right directions; weather effects;
and even wildlife. Non-natural elements include buildings, moving
non-combatant vehicles, airspace controllers, and airfield activity. As
important as the mere existence of these elements is the player's level of
interaction with them. Looking at the sun produces glare that washes out
vision. Interfering with traffic (both ground and airborne) stops it.
Landing or taking off without permission produces a certain reaction.
All of these aim at suspending the player's disbelief by creating a rich
world in which there are active units beyond the player's immediate
aircraft. This is the business of world simulation, mostly restricted to
those elements not directly involved in combat AI.
TACTICAL REALISM. This is combat simulation at the most immediate level,
and includes all modes of AI that involve the control of individual units
or small groups of units; virtually everything pertaining to weapons
employment, PK, and effects; fight-or-flight calculations; and
limited-intelligence filters. Doing this for a single-ship engagement
is doable; the challenge today is getting it right for multi-ship
engagements, with all of the additional variables that entails. Gauging
the level of tactical realism is hard, particularly with modern sims,
because there often do not exist enough real-world examples to
meaningfully compare sim results with real world outcomes. Still,
indications of a high degree of tactical realism are a variety of
tactical options that are not hopelessly executed, AI behavior that
does not demonstrate a consistently high degree of unforgivable
stupidity, weapons performance that does not clearly depart from commonly
accepted, published sources, and fight-or-flight calculations that
cause reasonable, prudent retreat under certain circumstances. A very
important part of tactical realism is the presence of other friendly
units in the theater along with the player's aircraft, carrying out both
related and fully independent missions; combat is no longer a clash of
individuals, but rather a clash of coordinated teams.
OPERATIONAL-STRATEGIC REALISM. This includes the vaunted "dynamic
campaign" and involves those combat calculations that take place at a
level once-removed from that of individual vehicles. The main purpose of
the operational-strategic factors is to link together discrete tactical
engagements into a coherent, logical series of outcomes that reasonably
portrays a larger operation, campaign, or war. It is at this level that
the long-term consequences of tactical action become apparent; for
instance, closing an airfield or losing an AWACS shifts the balance in
subsequent missions. There is also a tension between the need for
logical predictability and the desire to include the unforeseen; we want
preparations and outcomes that are militarily sound, but, within reason,
we want to be surprised, as well. Needless to say, this is exceedingly
difficult to do well, because of the sheer number of variables. It has
also very difficult to tell when it is being done well, since game
events can be interpreted in many ways. In the end, it is the
perception of a logical system, which drives the player to engage in
certain forms of strategic behavior, which is important.
SCENARIO REALISM. The plausibility of the scenario in which the combat
takes place is often overlooked as a factor in sim environments.
However, it may become increasingly important as players become more
sophisticated and look beyond the classic "Desert Storm" or "Second
Korean War" scenarios for realistic portrayals of future conflicts. In
many cases the scenario hurts more than it helps; rarely will a
brilliantly drawn scenario improve the sim very much, but a ridiculous
setting can seriously hurt player suspension of disbelief.
Next, putting this stuff into context.
--
____________________________________________________________________________
Robin J. Lee amr...@netcom.com
Vulture's Row Worldwide Web Page URL: http://www.webcom.com/amraam/
Robin J. Lee
[Apologies if this appears twice; Netcom appears to have had my first
send for lunch. Again.]
causes reasonable, prudent retreat under certain circumstances. A very
Next, looking at this stuff in context.
pap...@ix.netcom.com
You apologized for posting it twice but I wont apologize for reposting
it a third time....great stuff.
PAPA DOC
Ben Chiu
Hi Robin,
A>One of the most widely held values in this flight simulation community
A>is what we loosely term the "realism" of the product. But too often a
A>sim is labeled "realistic" or "unrealistic" without a critical
A>examination of what it really means to say that. This often leads to
A>discussions that are not especially useful, because there is
A>essentially no common set of understandings upon which a useful
A>discussion can be based. The result is commonly a series of
A>disconnected posts in which individual posters single out isolated
A>factors which may or may not be "realistic," and judge the product on
A>to that basis. What I hope to do is to identify and explore different
A>aspects of "realism" that may help us to refine future discussion.
Great post! I was working on a column similar to what you've posted - I
suppose "great minds think alike." <vbg>
Anyway, if I may, please allow me to reveal some of what I was working
on as it adds to your sentiments. I won't go into too much detail as I'd
still like my work to be printed, but this should suffice.
While ultimate realism is a commendable goal for a sim, the reality
(pardon the pun) of the sim market place doesn't seem to be so much
about what is absolutely real, but what it really seems to boil down to
is the amount of realism in relation to the players perspective. This is
the old argument of what players perceive to be real and how much
realism is one ready to *tolerate*. As a pilot, I'll estimate that 70-
80% of what goes on in the cockpit and before you climb aboard wouldn't
be of much interest/fun to the vast majority of sim'ers.
This is one of the major reasons why it takes more than a real pilot to
make a good sim. To quote Crunch's bud, George Wargo (retired F-15 III
bus driver), "it takes more than expertise in real world flying to
understand the terms and conditions necessary to enjoy a PC-based flight
sim. If military expertise were all we needed to have great flight sims,
they would've been available 6 years ago already."
He's absolutely right. There needs to be a fine balance struck between
realism, playability, and hardware limitations (and budget of course if
you're a developer). As you know, like it or not, due to hardware
limitations, we're unable to reach true, unobtrusive, transparent
illusion of realism (not that I'm against anyone trying). So, instead of
just breaking realism down into components alone, I believe one needs to
balance these realism components not only in relation to the real
vehicles modeled (if it exists), but also to in our case, as end users -
pilot/user perspective. After all, realism just for the sake of realism
can not only make for really, really boring sims, but unrealistic sims
as well. Let me explain.
From my experience, sim's are rated (and bashed) on basically 5 areas.
If I were to relate these as types of players, I'd list them as:
Sightseers
Arcade-meisters
Combat-crazies
Number-crunchers
Procedure-hounds
Of course, trying to pigeon any sim'er into any one category would be
silly because most of us would classify ourselves as combinations of the
above - but that's the point. While some go on and on about flight
models or weapons performance, etc., what it all relates back to is
cockpit work load - that's right - our Procedure-hound category. Cockpit
workload is really what most realism buffs are after. Real performance
numbers require greater realistic cockpit management skills/work load.
Real world flying really is all about following procedures. (That's why
there are so many checklists, plates, and charts found in a cockpit.)
However, as I mentioned earlier, having too much emphasis on pilot
workload would be really boring for most people. So therein, lies the
rub. There can only be so much realism before it works against you.
To give you a simple example of this phenomenon, if a developer left out
an "up" view from their air combat sim because in real life, your helmet
wouldn't allow you to look straight up. That would be "realistic," no?
However, when it comes down to it, the lack of motion cues, full 360
degree spacial audio and visual perspective and acuity, you'd be placed
in an un-realistic environment which would actually place you at a
disadvantage (read less-realistic) than if you were actually flying in
the real plane - effectively raising pilot workload.
Realism may be the Holy Grail of the flight sim world, but placing too
much emphasis on realism with out balancing the consequences or the
limitations of current hardware technology may lead to incarnations that
are really not what we as sim'ers (even us hardcore types) want or can
use. What is that saying? Be careful of what you wish for... ;^)
Regards,
Ben
Internet: ben....@patchbay.com
IamFritz
I believe another element in the discussion of "realism" is the actual
flight parameters to the plane of the sim in question. Way WAY too many
desktop fighter jocks say "the flight model of this sim is RIGHT ON". Yet,
all they have ever flown themselves is seat A on Row 32 of a McD L10-11.
They have no idea what the actual flight model to an airplane should be.
It just matches their imagination.
Now, I have done some work to verify flight models in the past. I got all
the stats once, for the F-14B, which is portrayed in F-14: FLEET DEFENDER,
(I think I did it for Falcon 3.0 too), including acceleration at different
altitudes, roll and pitch rates at alt, G rate, wingloading (in
otherwords, everything), and, presenting my findings to a couple fighter
pilots at an airshow at Hill AFB (F-14 pilots) came to the conclusion that
the sim was very, very close. Case closed.
At another Airshow, there were some F-15E crews (2 planes) there, and my
friend and I asked them (when we finally got thru the line to the
cockpit), what they thought of Strike Eagle III. Let me tell you: Their
faces lit up to stage 5 (first informed conversation they'd had all day,
he said!) and they began expounding to us the realism and accuracy of
Strike Eagle III. The B/N even abandoned the line of people and gave us a
personal tour of the plane, plus some neat stories about Operation
Southern Watch, which he had been flying the previous week. The pilot
joined us shortly after that and added his fifty cents. Another case
closed.
On the other hand, the few pilots I've talked to about USNF/ATF, say
they're great sims for the EXPERIENCE of air combat, but ". . . don't
really reflect how the actual planes drive."
So, next computer-pilot-wana-be who says "ATF's ASTOVL has the most
realistic flight model I've ever seen", or, "The AV-8B is easier to land
than in ATF:NATO" will just get my golden raspberry award" "PPPPPPT!"
Robin J. Lee
In article <8609...@patchbay.com>, Ben Chiu <ben....@patchbay.com> wrote:
<good stuff snipped>
>Realism may be the Holy Grail of the flight sim world, but placing too
>much emphasis on realism with out balancing the consequences or the
>limitations of current hardware technology may lead to incarnations that
>are really not what we as sim'ers (even us hardcore types) want or can
>use. What is that saying? Be careful of what you wish for... ;^)
I agree completely. To clarify my initial post, I regarded the question
of how much realism is actually desirable as an issue separate from the
question of what actually constitutes realism. The desirability issue is
extremely complex, and I'm sure that the views of developers will differ
from the views of consumers (who won't be able to agree on an answer
themselves).
The example of whether simulating an up view is realistic is an
interesting one. Like you, I don't regard it as an unrealistic
concession to playability, but rather one which enhances the overall
realism of the model -- something like attitude cues in the padlock
view. I guess the point here is that there are many ways to implement
realism, many of which simply reflect design choices and personal
preferences.
Components, levels, implementations...this realism business can be tricky
for what really amounts to a recreational product. ;)
Robin J. Lee
In article <19970413080...@ladder01.news.aol.com>,
IamFritz <iamf...@aol.com> wrote:
>Now, I have done some work to verify flight models in the past. I got all
>the stats once, for the F-14B, which is portrayed in F-14: FLEET DEFENDER,
>(I think I did it for Falcon 3.0 too), including acceleration at different
>altitudes, roll and pitch rates at alt, G rate, wingloading (in
>otherwords, everything), and, presenting my findings to a couple fighter
>pilots at an airshow at Hill AFB (F-14 pilots) came to the conclusion that
>the sim was very, very close. Case closed.
>At another Airshow, there were some F-15E crews (2 planes) there, and my
>friend and I asked them (when we finally got thru the line to the
>cockpit), what they thought of Strike Eagle III. Let me tell you: Their
>faces lit up to stage 5 (first informed conversation they'd had all day,
>he said!) and they began expounding to us the realism and accuracy of
>Strike Eagle III. The B/N even abandoned the line of people and gave us a
>personal tour of the plane, plus some neat stories about Operation
>Southern Watch, which he had been flying the previous week. The pilot
>joined us shortly after that and added his fifty cents. Another case
>closed.
Well, but there's another factor in play here. I don't doubt that real
life aircrews are the best qualified for judging flight models, but it's
important to find out whether they are impressed with the flight sim for
being a true representation of the aircraft's handling, or impressed
with it for being a PC simulation of the aircraft's handling. The
standards are different; "that's realistic" is not the same as "that's
realistic, for a PC simulation."
I suspect that pilots with little experience with this hobby (and the
level of detail that it has evolved) would be impressed that a PC
flight-sim models varying aircraft performance with varying aircraft
weight at all, regardless of how well it was actually done. Likewise, I
think that some pilots who had no idea that current sims are regularly using
real HUD symbology would be quite impressed with that fact alone, even if
there were some errors in detail.
In some sense, their assessment of how "realistic" it is all depends on
their expectations of what today's PCs can do. Those expectations are not
easily standardized across aircrews; even people with considerable flight
sim experience may differ as to how capable a PC is of simulating the
airplane. That subjective evaluation sets the standard of measure, which
in turn leads to the thumbs-up/thumbs-down. In sum, that a pilot was
impressed with a sim doesn't necessarily mean that the sim is
"realistic" by all objective standards.
CJ Martin
In article <amraamE8...@netcom.com>, amr...@netcom.com (Robin J. Lee) wrote:
[snip]
>Components, levels, implementations...this realism business can be tricky
>for what really amounts to a recreational product. ;)
Tell me about it...
<G>
BTW, excellent posts, guys.
CJ
-----
CJ Martin, Associate Designer
Origin Skunk Works - Baltimore
Ben Chiu
Hi Robin,
A>I agree completely. To clarify my initial post, I regarded the
A>question of how much realism is actually desirable as an issue
A>separate from the question of what actually constitutes realism.
Understood and agreed, however, what constitutes realism compared to
what? That is one of the things I was getting at. Sorry my original post
wasn't very clear. (I'm currently in the middle of a couple big projects
and I tried to whip out a couple of quick thoughts - which only seemed
to complicate matters. My apologies to everyone.)
A>The desirability issue is extremely complex, and I'm sure that the
A>views of developers will differ from the views of consumers (who won't
A>be able to agree on an answer themselves).
Yes, desirability is a separate issue. But on top of desirability, the
other aspect one needs to take in account when evaluating realism (and
*almost* everything else) is a frame of reference.
A>The example of whether simulating an up view is realistic is an
A>interesting one. Like you, I don't regard it as an unrealistic
A>concession to playability, but rather one which enhances the overall
A>realism of the model -- something like attitude cues in the padlock
A>view. I guess the point here is that there are many ways to implement
A>realism, many of which simply reflect design choices and personal
A>preferences.
You've expounded one of my points exactly. However, to clarify another
point, some aspects of realism are very cut and dried so they can be
discussed as isolated examples as you propose (an absolute G loading
where the wings will rip off your Spad for instance). However, for other
aspects of realism, they must be real in reference to *someone*. The
problem now becomes unless you specify what your reference is/was when
making those evaluations, you're going to create
confusion/misunderstandings/new arguments. (I won't go as far as to say
everyones' realism reference is different, but they do vary. I think it
probably sums it up to say that reality - or realism in this discussion
is in the eye of the beholder.)
Okay, this isn't a great example, but here goes. Everyone votes for the
absolute best terrain ever found in a sim.
The Sightseer says, "Sim X has the most realistic terrain. See how the
light bounces off the accurate colors and how the surface textures
create depth."
The Game-meister says, "Sim Y has the most realistic terrain. There's
gravity and the bad guys blow up when they crash into it." (Okay, that
one is a bit of a stretch. Fill in your own.)
The combat-crazy says, "Sim Z has the most realistic terrain. You can
hide behind the hills to avoid being spotted by the enemy."
The number-cruncher says, "Sim A has the most realistic terrain. The
actual elevations and distances are accurate to within 1mm."
The proceedure-hound says,"Sim B has the most realistic terrain. The
placements of landmarks and nav aids are accurate. I can navigate to all
the way to Bill Gates' home."
Then you have the hardware-hacker, "The terrain in all of those sims
suck! There is no perspective or anti-aliasing, I'm seeing it through a
little crappy window at a freakin' 2 frames a second!"
Who is right? IMHO, it all depends on your background/experience, which
usually indicates what you think is important.
A>Components, levels, implementations...this realism business can be
A>tricky for what really amounts to a recreational product. ;)
Agreed! But IMHO, it can account for why some products despite all of
the best intentions turn out the way they do, and also why guys like
Andy Hollis get paid the big bucks.
Best!
Ben
Internet: ben....@patchbay.com
Internet: ben....@patchbay.com
Ben Chiu
Hi Robin,
Sorry for the additional reply.
A>Well, but there's another factor in play here. I don't doubt that
A>real life aircrews are the best qualified for judging flight models,
A>but it's important to find out whether they are impressed with the
A>flight sim for being a true representation of the aircraft's
A>handling, or impressed with it for being a PC simulation of the
A>aircraft's handling. The standards are different; "that's realistic"
A>is not the same as "that's realistic, for a PC simulation."
This is a great example of the point of reference aspect I've been on
about.
A>In some sense, their assessment of how "realistic" it is all depends
A>on their expectations of what today's PCs can do. Those expectations
A>are not easily standardized across aircrews; even people with
A>considerable flight sim experience may differ as to how capable a PC
A>is of simulating the airplane. That subjective evaluation sets the
A>standard of measure, which in turn leads to the thumbs-up/thumbs-down.
A>In sum, that a pilot was impressed with a sim doesn't necessarily mean
A>that the sim is "realistic" by all objective standards.
Hehe, now you're gettin' it! Just as an aside, this is the main reason
why I was against star ratings for reviews. Giving a rating of X stars
means what? While the marketing depts are all hot to quote their star
ratings, IMHO, unless the consumer reads the actual review or quotes
from it, it really doesn't mean much because it doesn't tell you much
about the game other than you might want to take a closer look at it.
Then again, maybe that was the point.
Cheers!
Ben
Internet: ben....@patchbay.com
Robin J. Lee
In article <8610...@patchbay.com>, Ben Chiu <ben....@patchbay.com> wrote:
>
>Understood and agreed, however, what constitutes realism compared to
>what? That is one of the things I was getting at. Sorry my original post
>wasn't very clear. (I'm currently in the middle of a couple big projects
>and I tried to whip out a couple of quick thoughts - which only seemed
>to complicate matters. My apologies to everyone.)
>Yes, desirability is a separate issue. But on top of desirability, the
>other aspect one needs to take in account when evaluating realism (and
>*almost* everything else) is a frame of reference.
>You've expounded one of my points exactly. However, to clarify another
>point, some aspects of realism are very cut and dried so they can be
>discussed as isolated examples as you propose (an absolute G loading
>where the wings will rip off your Spad for instance). However, for other
>aspects of realism, they must be real in reference to *someone*. The
>problem now becomes unless you specify what your reference is/was when
>making those evaluations, you're going to create
>confusion/misunderstandings/new arguments. (I won't go as far as to say
>everyones' realism reference is different, but they do vary. I think it
>probably sums it up to say that reality - or realism in this discussion
>is in the eye of the beholder.)
Interesting observation. In the abstract where we have unlimited
resources to drive the simulation, we can imagine a sim where we have a
total convergence of all the varying perspectives by addressing every
one of them. I suppose we could derive an objective "reasonable man"
kind of test under those circumstances (where we have sufficient
homogeneity of perception, due to the richness of the simulation, to
make a single standard at all meaningful), and call that "realism." I
guess in my abstract schema, this is where the standard would fall, and
all partial solutions would be on separate pathways to this finish line.
Whether it's actually possible to tell at any given time which
implementation is closest is another matter. But you do bring up an
intriging point -- one I have to admit that I hadn't thought enough
about before you brought it up.
But that's in the hypothetical world of frictionless ropes, etc. In the
real world of limited resources, where realism can really only address a
few of those frames of reference, I think the place to look for guidance
is the design goal. For this product, do we want tactically functional
realism, so that terrain is convincing only insofar as it encourages and
facilitates its proper use in combat? (That seems to be the goal of
"real" simulators.) Do we want experience-based (er, there's a better
term for that but it's been a long day) realism, where the terrain is
convincing enough to elicit a visceral response but lacks features that
facilitate the tactical aspect? Design choice is all about choosing a
frame of reference and sticking to it.
By categorizing aspects of realism I don't mean to suggest that
subjective debate about preferences will go away. It's just that I think
some systemic way of thinking about the problem will make the debate
tidier -- at least flight model people won't be arguing aerodynamics with
radar people, and so on. People's arguments may still be missing one
another, but at least the CEP will be smaller. ;)
Ben Chiu
Hi Robin,
A>Interesting observation. In the abstract where we have unlimited
A>resources to drive the simulation, we can imagine a sim where we have
A>a total convergence of all the varying perspectives by addressing
A>every one of them.
Okay, please allow me to "spill the beans" here (then I promise to go
away and get back to work). Yes, you can imagine such a beast, but as I
see it, there are problems. Aside from physical realism, we're basically
dealing with 3 types of realism - one that can be absolutely be
recreated, one that can't, and the third is based on the sum of the
first two. (Hopefully it'll make sense in a minute.)
Things such as AOA stall and structural failures (what I call conceptual
object realism) can be absolutely recreated. However, when it comes to
things such as terrain (what I call virtual object realism), the "scale"
or "measurement" for evaluation purposes for these items is ambiguous.
On the high end, unless technology moves to something like brain jack
interfaces, the ultimate goal is impossible to reach, therefore all
points in between become ambiguous. Because you're dealing with virtual
realism, you're dealing with representations of the real things and
compromises are made, by nature/definition, you're not actually creating
an identical copy of the real thing. That's why we're not dealing with
an subjective scale here.
In such circumstances, we resort to comparisons made with other similar
products - which as we've seen has its problems as well due to the frame
of reference variations. That's why it's pointless to discuss realism in
any concrete terms unless a form of reference is also included. You
cannot deal with a subjective topic without objective references.
This leads us to the third form of realism (what I call cumulative
perceived realism - kind of redundant I suppose). This is the pilot
workload thing I eluded to in one of my original posts. The sum of
conceptual objects and physical objects create perceived realism. While
perceived realism is judged on a subjective scale (as we've discussed,
your experiences and backgrounds), IMHO, this is the only real true
scale to judge physical object realism as it imposes a obtainable high
end limit in addition to being a form of realism itself.
Okay, a lot to grasp. First things first. To get back to our terrain
example, you need to ask yourself, does this terrain increase or
decrease my workload? If it increases your cumulative workload beyond
what is real (or based on you perception of workload should be if you've
never done it for real) then you've reached the upper limit causing the
effect to be deemed unrealistic. While this high end limit is a sliding
scale because it is cumulative to a point, it is perceived realism -
which really is what "seat of the pants" impressions are based on even
though most fail to realize this. (Of course, not everyone wants
absolute realism and this is where the gameplay issues come in, but
let's stay on the realism topic for now.)
A>I suppose we could derive an objective "reasonable man" kind of test
A>under those circumstances (where we have sufficient homogeneity of
A>perception, due to the richness of the simulation, to make a single
A>standard at all meaningful), and call that "realism."
Nice idea, but the frame of reference of this "reasonable man" needs to
be based on some ideal. Would that be average market share, or average
tastes of this newsgroup? See how that would cause problems as well?
A>I guess in my abstract schema, this is where the standard would fall,
A>and all partial solutions would be on separate pathways to this
A>finish line. Whether it's actually possible to tell at any given time
A>which implementation is closest is another matter. But you do bring up
A>an intriging point -- one I have to admit that I hadn't thought enough
A>about before you brought it up.
This is why I favor the user/pilot centralized form of reference model.
It's where, as some say, "the buck stops."
A>But that's in the hypothetical world of frictionless ropes, etc. In
A>the real world of limited resources, where realism can really only
A>address a few of those frames of reference,
I don't think so. But that's my opinion. ;) While you can't please all
of the people all of the time, I believe (and most idealistic developers
will probably agree - note I am not a developer) you can please a good
portion of them.
A>I think the place to look for guidance is the design goal. For this
A>product, do we want tactically functional realism, so that terrain is
A>convincing only insofar as it encourages and facilitates its proper
A>use in combat?
Agreed. As I've explained above, this is the only way that I know of to
objectively discuss subjective criteria. Real life pilots may know how X
radar works in real life, but IMHO, it's the cumulative effect of that
implementation on the sim as a whole that makes sim designers like
Jonathan Baron irreplaceable.
A>By categorizing aspects of realism I don't mean to suggest that
A>subjective debate about preferences will go away. It's just that I
A>think some systemic way of thinking about the problem will make the
A>debate tidier -- at least flight model people won't be arguing
A>aerodynamics with radar people, and so on. People's arguments may
A>still be missing one another, but at least the CEP will be smaller.
A>;)
Agreed and a commendable goal. I hope I haven't come across as
trivializing your efforts with my (as some put it) "babbling." That was
not my intention. I do see your efforts as a service to the newsgroup,
and to some extent the industry. And as such, might I suggest that you
retain the concept of realism categories, but further break them down
into conceptual, virtual, and cumulative objects as well, then propose
to discuss them as they relate to a particular frame of reference.
Golly, I think I just gave away my whole article! <vbg>
Well, there's always my physical realism article... ;)
At the very least I hope this helps or gives everyone something else to
consider.
Best!
Ben
CGS+
Internet: ben....@patchbay.com
Edward G. Kinateder
Previously Robin J. Lee wrote:
<snip of Robin's excellent essay on categories of realism>
:SYSTEMS REALISM.
:PILOT-PERSPECTIVE REALISM.
:AERODYNAMIC-PHYSICAL REALISM.
:ATMOSPHERIC REALISM.
:TACTICAL REALISM.
:OPERATIONAL-STRATEGIC REALISM.
:SCENARIO REALISM.
This is the one I'm interested with (more than the others).
:PILOT-PERSPECTIVE REALISM. This addresses the capability of a sim to
:capture the experience of a human pilot in a cockpit setting, which we
:can all agree differs greatly from the experience of sitting in front of
:a 15" monitor in a 1-G environment. Most of this is connected with
:being able to achieve situational awareness, and includes such factors
:as a workable viewing system that simulates the natural head motion of a
:helmeted pilot, the effects of G-induced loss of consciousness, being able
:o impart a sense of relative motion at all distances, and other "human
:factors" which go into flying an aircraft. Sound, and the quality
:thereof, is also included in this category. This is the fantastic realm
:of VR headgear, force-feedback sticks, full-cockpit-simulators, and
:creative padlock/viewing systems. Pilot-perspective realism focuses
:less on simulating the aircraft and more on simulating the
:player-as-pilot (and only as pilot, with considerations restricted to
:those physical elements involved in actually flying the airplane - it
:does not include combat elements beyond the purely physical.)
I've not had a Coke is quite a while until today.....sorry if my comments appear
as caffeine induced rantings.
As articulated by Robin Lee, our resident NG academic and fellow sim addict, I
wanted to extract one of his "realism" categories, PILOT-PERSPECTIVE REALSIM,
and venture off on this subject which is of particular interest to me. My
thanks go to Robin for providing a wonderful base from which to expound upon.
Also I would like to thank those individuals with their subsequent posts,
comments, arguments and contributions.
Additionally, I will focus my discussion on a very specific aspect of PPR
(Pilot Perspective Realism), that being the use of padlock views and fixed views
as they perform in contrast to VR headgear (VRH).
My basic premise is that fixed viewing systems (FVS) are not the offspring of
contemporary innovation but more a methodology grandfathered into our PC gaming
industry from days past. Further, padlock viewing systems (PVS) being the
primary evolutionary step designed to overcome the limited SA from PC gaming,
were created through a logic that pilots can freely go outside (look out of the
cockpit) and generally maintain a visual lock on objects even as the aircraft
rolls and pitches because they are free to move their head, neck and eyes
without conscious thought. I would contend that until VRH is standard in the
industry and designed such to allow fluid head tracking in a high resolution and
high detailed landscape and object environment, that PPR will be grossly limited
for PC immersion and the true feeling of flight. Finally, I also believe that
FVS and PVS will always remain as half solutions to an essentially unrealistic
PC flight viewing environment.
The FVS requires the pilot to locate his targets and various outside world
objects without aid from the simulation. This is the essential realism feature
which makes FVS quite appealing. FVS creates the need for the pilot to actively
LOOK in specific directions. However, with every significant pitch and roll
change of his AC, the pilot must often select a new view perspective. The
object of viewing interest is only rarely centered in the pilots view. It is
impossible to maintain a padlock on an object or AC during violent or radical
maneuvering. That is, although general tally can be maintained (with high
skill), it requires several successive view changes. I include within FVS, the
scroll type view which is a FVS with incremental scrolling capability. These
scrolling features generally prove to be adequate only when slow sweeps are
required and when not under significant combat tasking or duress. The also
become less effective when the pilot is changing the AC pitch or roll.
The PVS achieves the fluid scan capability of a real pilot, irrespective of AC
bank, roll, turn rate or attitude without requiring successive unnatural
conscious view commands. The essential flaw is that it allows the pilot a
wholly unrealistic ability to ACHIEVE and maintain tally on the outside objects
and AC irrespective of his own scan ability, SA, AC pitch, roll or G. Many PVS
will impose HUD targeting boxes over the padlocked object, further reducing
realism. It is said that a fighter pilot's ability to visually identify and
sort objects is possibly his most important skill that he possesses. This has
arguably become less a factor (although I don't think so) though as BVR
engagements become ever more relevant in today's air combat.
VRH offers the only true dynamic and intuitive viewing system that is practical
for the home PC system (at this point). Total surround view systems being too
costly and none being written for the home PC market. VRH allows the pilot to
move his head freely and unconsciously to the direction of his intent. VRH
requires the pilot to locate the objects and AC without any artificial software
assistance. VRH requires that the pilot have the ability to sort and visually
identify outside objects and respond accordingly. VRH requires that once the
pilot tallys an enemy AC, that he maintain his own visual padlock throughout the
engagement. VRH pilots are subject to more realistic spacial disorientation and
loss of SA through violent AC maneuvers (sometimes inducing real nausea). The
primary weakness with VRH is it's implementation and support at the software
level and the present level of resolution available with the various VRH (to my
knowledge). Software companies must design total virtual interactive and
READABLE cockpits that function as effectively as the static, 1/2 screen
cockpits. We are already seeing some of these now which allow mouse to cockpit
interaction. A VRH pilot must be able to glance to 10 O clock then glance to
his virtual HUD, then look down at his virtual radar and be able to evaluate his
situation based on this information. Additionally he must then be able to
interact with this virtual cockpit through either virtual gloves which sense his
cockpit touches or some other interactive device such as a cockpit pad or mouse.
TAKING IT A STEP FURTHER. Imagine VRH becoming the standard and internet
bandwidth capabilities becoming 10 fold greater than what it is now. Imagine
having voice comm capability and flying from your desk. You have VRH strapped
onto you head, feeling the equivalent of a fighter pilot's helmet. You adjust
your boom mike, check the volume levels and do a quick radio check with your
wingie who lives in Toledo.
"Snake One radio check"
"Snake Two Lima Charlie". All is good in virtual fighter pilot land.
Your VR gloves are on snug but you give them a tug just the same. You take grip
of your throttle and stick. For grins you let go of the stick with your right
hand and turn to wave at your wingie. You wave wildly for five seconds and he
does not respond. NO WONDER -, he is looking down in his cockpit, parked next
to you in position for takeoff. You see the rendering of his helmeted head as
it looks down. He is either doing system checks or playing with himself.
"Snake one zero one is Kilo" you press the push-to-talk on your throttle and
speak into the boom mike, the signal you are mission ready.
"Click" your wingman acknowledges.
With VRH strapped on and VR gloves in place you glance down left and see your
virtual gloved hand driving the throttle forward. You hear and feel the engines
spool up as you keep your brakes applied. POWER CLIMBS - you glance at the
pressure gauges. All is green and your RPMs increase. Satisfied that engines
are good, you release the breaks and your jet jerks forward as the runway
beneath you rolls by at a quickening pace. YOU GLANCE RIGHT - noticing the E2
holding short of the runway intersection. As you speed forward you search the
HUD and note triple digits on the airspeed so you know you will go flying. Your
seat is vibrating and you feel the ka-thunking with each concrete runway seam
that you roll over (love those thunder seats). Rotate speed and you are up.
GEARS UP - you glance at the VSI, positive rate of climb, FLAPS UP. You are
clean and mean.
Now, except for the high whir of the engines, it is most peaceful. At 800 feet
you crank your head left to look back at the base and the beautiful coast. You
swing your head back to the right as the view of the outside world, your cockpit
and the front of the AC sweep by quickly. You look over your right shoulder to
scan for your wingie. At first you see nothing but STARBOARD WING and SKY and
GROUND clutter. THERE HE IS - your wingie closing onto your starboard side,
just off your wing. As he gets within 50 feet you see his helmeted head turn
towards you within his cockpit. "I don't remember his face being that ugly",
you think to yourself jokingly. He gives you a thumbs up as he eases back on
the power. Then he gives you the bird and you can't help but smile. He is the
best virtual wingman you have ever had. You are glad you both own the VR
gloves. You are glad that he is with you on this mission.
You swing your head forward, CHECK YOUR ALTITUDE, SPEED, and HEADING - then left
to scan your nine O clock and the ocean coastline. No traffic in site. Looking
back to your HUD, you note your heading and pull some power.
"Check left ninety" you call to number two on the flight frequency. Pausing a
moment, waiting to hear wingie's radio click, then rolling to port, adding some
power and putting in some lift vector. Watching the HUD until your aircraft is
banked at 45, you then turn your VRH head left to catch the view that you love
the most. Looking down at the white caps and the big ocean, with your wing in
the foreground and the AIM9 on the tip....almost wishing it could fly. You turn
your head back to the HUD and begin your roll out. A few air bumps and your HUD
view is unsettled. CHECK ALTITUDE, SPEED and HEADING. You strain your neck to
look over your right high six O clock and see the belly of number two as he
pushes to say tight. Then he begins pulling up and away from you suddenly.
WHAT! You swing your head forward quickly, at first overshooting your cockpit
view. You are disoriented for a moment, you note that you should not snap your
head until you are used to the speed. Suddenly you realize you accidentally
nosed the stick forward and to the right. You crank your head up and around
again to get a visual on your wingie. 100 FEET VERTICLE SEPARATION - I can pull
up now. LOOKING FORWARD AGAIN - You pull back and level out.
Wings level now, you turn your head right and aft and see number two nudging
into place. Then he comes on the freq and says "Was I briefed on that one, or
are you just expressing your individuality".
"Click" one press on your push-to-talk button and all is forgiven.
END OF STORY
Does anyone know or has anyone heard what the future state of VR gear is? What
can we expect and when can we expect it? Most of all, what should we expect to
pay for it? What about you software companies? What are you working on to
bring this next, and most immersive level to the flight combat sim enthusiast?
Andy, Brett, Carl, Gabe, No 19? Please comment, we would love to hear what you
see for the future as it relates to your products!
*caffeine surge over*
T-Man
Michael Irvine
On 25 Apr 1997 02:11:46 GMT, MJes...@aol.com wrote:
>Note:
>
> FVS = fixed view system
> PVS = padlock view system
> VRH = Virtual Reality Helmet
>
>byth...@n2.net (Edward G. Kinateder) wrote:
>
<snip>
>>The PVS achieves the fluid scan capability of a real pilot, irrespective of AC
>>bank, roll, turn rate or attitude without requiring successive unnatural
>>conscious view commands.
>
>I disagree, believing that the view changes are more natural than a computer
>controlled pan. That is because it simulates the pilot's sense of
>where his head is turned, that you don't have with a pan.
>
>>The essential flaw is that it allows the pilot a
>>wholly unrealistic ability to ACHIEVE and maintain tally on the outside objects
>>and AC irrespective of his own scan ability, SA, AC pitch, roll or G. Many PVS
>>will impose HUD targeting boxes over the padlocked object, further reducing
>>realism.
>
>Yes, but a bigger flaw is it's disorienting (my comment above).
>
I had no problem with the F3 PVS once I had enough practice. I agree
that it is initially disoreintating but once you learn to interpret
and use the mini-HUD and the rest of the info at the top of the screen
there is no problem.
You may like to check the thread on Su-27 and other view systems where
some points have been made on both sides (by myself and others).
Regards,
Michael
MJes...@aol.com
Note:
FVS = fixed view system
PVS = padlock view system
VRH = Virtual Reality Helmet
byth...@n2.net (Edward G. Kinateder) wrote:
>The FVS requires the pilot to locate his targets and various outside world
>objects without aid from the simulation. This is the essential realism feature
>which makes FVS quite appealing. FVS creates the need for the pilot to actively
>LOOK in specific directions. However, with every significant pitch and roll
>change of his AC, the pilot must often select a new view perspective. The
>object of viewing interest is only rarely centered in the pilots view.
So? ;)
Actually, in WarBirds (WB), I am often maneuvering the aircraft with the
object of looking. Rather than hindering SA, in this case, maneuvers
(quick rolls, etc) are enabling SA.
>It is impossible to maintain a padlock on an object or AC during violent or radical
>maneuvering.
IMO, this is not true, given practice with a high quality FVS
like AW/WB/FD's.
> That is, although general tally can be maintained (with high
>skill), it requires several successive view changes.
If the FVS is well designed, this isn't a problem, and actually
becomes a virtue.
>The PVS achieves the fluid scan capability of a real pilot, irrespective of AC
>bank, roll, turn rate or attitude without requiring successive unnatural
>conscious view commands.
I disagree, believing that the view changes are more natural than a computer
controlled pan. That is because it simulates the pilot's sense of
where his head is turned, that you don't have with a pan.
>The essential flaw is that it allows the pilot a
>wholly unrealistic ability to ACHIEVE and maintain tally on the outside objects
>and AC irrespective of his own scan ability, SA, AC pitch, roll or G. Many PVS
>will impose HUD targeting boxes over the padlocked object, further reducing
>realism.
Yes, but a bigger flaw is it's disorienting (my comment above).
Even the tiniest bit of disorientation means that you lose any benefit
of glimpses of other (non-padlocked) aircraft in the area. In WB, I
can maintain decent SA of other planes than the one I'm attacking
as they enter and leave the views I'm watching my target with.
That's because I know the maneuvers I'm putting in, and I
know my head position. If I have to expend great mental effort
just interpreting the relative position/orientation/velocity
of the locked bogey (or if I lose understanding of his
relative state through "padlock disorientation", I lose any
mental track on the other planes as well.)
>VRH offers the only true dynamic and intuitive viewing system that is practical
>for the home PC system (at this point).
> VRH pilots are subject to more realistic spacial disorientation and
>loss of SA through violent AC maneuvers (sometimes inducing real nausea).
What I wonder about (and the reason for my comments above) is whether
the VRH system described by the previous poster will present new problems
with maintaining SA. I wonder if the way the eye works inside the headset
will be the same. Will it be as natural to move the head and track with the
eye without losing "lock", or will it merely add another marginally
stable loop to the system. Will frame rate need to be higher than usual
nowadays to let the eye smoothly track as the head turns?
It seems to me that VRH would need excellant periferal vision
modeling, otherwise the "wide angle views" common today might
become disorienting themselves if used with VRH as the
mind/eye gets confused... "It looks real, but something's wrong"
??
- Matt
WB: para
Edward G. Kinateder
Previously MJes...@aol.com wrote:
:Note:
:
: FVS = fixed view system
: PVS = padlock view system
: VRH = Virtual Reality Helmet
:
:byth...@n2.net (Edward G. Kinateder) wrote:
:>The FVS requires the pilot to locate his targets and various outside world
:>objects without aid from the simulation. This is the essential realism feature
:>which makes FVS quite appealing. FVS creates the need for the pilot to actively
:>LOOK in specific directions. However, with every significant pitch and roll
:>change of his AC, the pilot must often select a new view perspective. The
:>object of viewing interest is only rarely centered in the pilots view.
:
:So? ;)
:Actually, in WarBirds (WB), I am often maneuvering the aircraft with the
:object of looking. Rather than hindering SA, in this case, maneuvers
:(quick rolls, etc) are enabling SA.
Ahh, good point. I suspect that it is something you get used to, even expect
and integrate into your scan style.
:
:>It is impossible to maintain a padlock on an object or AC during violent or radical
:>maneuvering.
:
:IMO, this is not true, given practice with a high quality FVS
:like AW/WB/FD's.
We'll, let me explain. When I tally a real aircraft or object, my eyes
initially look directly at it, generally. Meaning that the object is generally
in the center of my field of view, independent of my ac bank, roll or pitch.
This allows for consistent peripheral vision in all directions of the object
(unless my AC obscures that). This is not the case with FVS, such that the
object will always tend to be cross tracking in that particular view selected.
Not a BIG issue but still present.
:> That is, although general tally can be maintained (with high
:>skill), it requires several successive view changes.
:
:If the FVS is well designed, this isn't a problem, and actually
:becomes a virtue.
I suspect that you have a good point. I'm sure that AW/WB/FD's have superior
FVS compared to most other sims. Point still is that it is not the same as real
life, i.e., turning your head and looking.
:>The PVS achieves the fluid scan capability of a real pilot, irrespective of AC
:>bank, roll, turn rate or attitude without requiring successive unnatural
:>conscious view commands.
:
:I disagree, believing that the view changes are more natural than a computer
:controlled pan. That is because it simulates the pilot's sense of
:where his head is turned, that you don't have with a pan.
Another good point. With the FVS the pilot must initiate the "looking" action
and input his direction of intent. This is quite real, as I have already
stated. Where it differs is that FVSs do not emulate how a pilot padlocks
visually, in that his head does not snap from one fixed view to the next, while
tracking a bogey. It is fluid and continuous. With FVSs, even masterful users
like yourself, must switch view, then scan, then visually lock again. Although
we are only talking about scanning 12 to 15" screens, you still get the point.
I'm not suggesting that it is inadequate, just that it is not as good as VRH for
previously stated reasons.
:>The essential flaw is that it allows the pilot a
:>wholly unrealistic ability to ACHIEVE and maintain tally on the outside objects
:>and AC irrespective of his own scan ability, SA, AC pitch, roll or G. Many PVS
:>will impose HUD targeting boxes over the padlocked object, further reducing
:>realism.
:
:Yes, but a bigger flaw is it's disorienting (my comment above).
Yes indeed. I have experienced this all too many times with past sims. I have
found that I will frequently alternate between forward view and padlock, to
cross check my ac orientation. Actually, I do this when in real flight also,
but then again, in real flight I can lose tally when I do this. It's harder
than people realize, maintaining sight in real life.
On a side note, about 4 weeks ago I was northbound in s.w. Iowa, 1600 feet,
above the Mississippi. I went head down to check my board. One one thousand,
two one thousand, three one thousand, four one thousand, head up. There it was,
a two seat experimental, low aspect, 100 feet off port, CO-ALTITUDE. I rolled
right and pulled. He never even saw me, or at lease he never maneuvered. We
would not have collided but the scary part is that if we had been destined for
the same airspace, nothing would have prevented a collision at THAT stage of the
closure.
With experiences like that, I appreciate sims that make it tough to tally. But,
alas, customers get upset when it is too difficult.
:Even the tiniest bit of disorientation means that you lose any benefit
:of glimpses of other (non-padlocked) aircraft in the area. In WB, I
:can maintain decent SA of other planes than the one I'm attacking
:as they enter and leave the views I'm watching my target with.
Yes, I see your point and I can understand that advantage. That is another
strong arguments for FVS, which I agree with.
:That's because I know the maneuvers I'm putting in, and I
:know my head position. If I have to expend great mental effort
:just interpreting the relative position/orientation/velocity
:of the locked bogey (or if I lose understanding of his
:relative state through "padlock disorientation", I lose any
:mental track on the other planes as well.)
Makes sense to me. I almost suggested that a sim pilot can also become adjusted
to PVS and achieve the same peripheral scan end result. But in this area, I
don't think he can. One does become dependant on the PVS to some degree. It
can become a lazy man's view system, if not properly modeled.
:>VRH offers the only true dynamic and intuitive viewing system that is practical
:>for the home PC system (at this point).
:> VRH pilots are subject to more realistic spacial disorientation and
:>loss of SA through violent AC maneuvers (sometimes inducing real nausea).
:
:What I wonder about (and the reason for my comments above) is whether
:the VRH system described by the previous poster will present new problems
Me, being the previous poster (in case anyone was wondering).
:with maintaining SA. I wonder if the way the eye works inside the headset
:will be the same. Will it be as natural to move the head and track with the
:eye without losing "lock", or will it merely add another marginally
:stable loop to the system. Will frame rate need to be higher than usual
:nowadays to let the eye smoothly track as the head turns?
You raise some very good points. It is my understanding that the eye can not
distinguish between anything higher than 30 FPS. I'm not sure I believe this,
or at least I don't believe all people's sensitivity is the same. I'm sure
there is an average. I personally can perceive the 60 cycle strobe of
fluorescent lighting and Sylvania's Halogen bulbs which use a diode to chop the
current. It creates a faint strobe effect visually. Some people can not detect
it. So do they design games for my eyes or yours?
:It seems to me that VRH would need excellant periferal vision
:modeling, otherwise the "wide angle views" common today might
:become disorienting themselves if used with VRH as the
:mind/eye gets confused... "It looks real, but something's wrong"
:??
:
:- Matt
:WB: para
I suspect that design should begin by using human/real world benchmarks to
establish design criteria. If the average person with 20/20 can FIRST detect a
60' wingspan (plan form) AC traveling at 250 knots, at 3 miles, when in direct
center of is Field of View (FOV) than that is a beginning reference point, of
sorts, for resolution. Aside from issues of background contrast, atmospheric
obscuration and light source.
Also worth consideration is the issue of detecting relative motion, which your
peripheral is superior at doing for large amounts of sky. Then what are the
image rendering rules?
WBs were the first (I may be wrong) to implement a dramatic wide angle viewing
system. All now use some degree of view distortion, but none as much as WB (I
believe). VRH has only been in the hands of the average combat sim developers
for a few years (maybe more). This is bound to impact the design of the
finished product.
T-Man
Michael P. Smith
While I generally agree, I want to bring up an interesting
experience I've had. I play Warbirds (alot....:>) and have gotten to
the point that the viewing system is second nature. In my job, I
occasionally get to fly our F/A-18 flight simulator (with a 270 deg
dome) and was AMAZED that the transition from the WB viewing system to
the actual head movement was very natural and I was able to perform
manuevers without any noticable transition period. I was truley
amazed! Don't get me wrong.....the advent of High-res, low-latency
VRH will be a true boon to simmers, but until then, the WB stlye view
system is pretty decent kludge.
--Vila
The Flying Pigs
"The goal is defeat the enemy aircraft, however you
accomplish that, and to return to base. Some of the
best dogfighters in history got killed because they did not
understand this simple rule." - CDR Bob Christiansen, USN
Simon Chua
Hey all,
Instead of arguing about the merits and demerits of the Fixed View System
(FVS) Vs the Padlock View System (PVS) perhaps a combination of both would
be a good solution. Briefly stated, the FVS gives the simmer a better idea
of where he is looking, but the "snapping" action is not reflective of
actual head motions. PVS offers a more realistic head tracking motions but
can be disorientating.
I heard that the upcoming Korean War sim Sabre Ace has a new padlock
system that tracks the target across traditional fixed view quadrants. In
this system, a bandit on your 12 heading to your left would appear on your
straight ahead view first. As it moves across to the left, you view would
'snap' over to the left front view. Thus solving to a certain extend the
disorientation found in PVS. However, the 'snapping' is still there.
Thus, I have this to suggest. Use a scrolling PVS ala EF2000 and Flanker
but divide the sky up into quadrants as per FVS. The padlock would scroll
to track a bandit but when it moves from one quadrant to the next, a small
dialogue box would tell you which quadrant you are currently in. So instead
of a snapping action, when the bandit moves out of my 12, I get the 'front'
dialogue box at first then the 'left front' box; all while scrolling
smoothly. Best of both worlds don't you think?
Still, the 'automagic' tracking makes it too easy to tally your target. So
perhaps a system like Flanker's which requires you to first find the target
via the FVS and then again whenever the bandit moves into your blind spots
would be good. But this should be an option, as beginners often find
'automagic' tracking PVS easier to deal with.
Comments?
Simon :)