Albert

By Andrew Herd

It isn't often that I am driven to write op-ed, but generally, it is the Space Cadets who inspire me to do it. If you recall, I introduced them in the Reviewer's Tale. My fellow reviewers will no doubt cringe in unison and clutch their garlic and wooden stakes when I mention the Cadets, because these people are the bane of our lives, but first of all, I am going to make a declaration on behalf of all of us.

We don't get paid for writing this stuff.

Not one cent has crossed my palm for writing a review on this website and as far as I know, 99% of reviewers can say the same. There may be some exceptions, where Nels has paid folk for specific pieces of copy, or where members of FlightSim.Com's vanishingly small professional staff have written articles, but in general you can pretty much rely on the fact that you are reading freeware. It gives me a nice, warm feeling to write it and I hope it gives you a nice, warm feeling to read it.

I guess we ought to put a danged great notice up before every review saying, 'Look, don't confuse Flight Simulation with the real world, the result could be dangerous,' but we don't, because the majority of you instinctively know that. After every review I get all kinds of people writing in and telling me interesting stuff, or just passing the time of day, or even just saying thank you for taking the time to point out this or that product. Yet there are always a few who can't believe that we passed some thirty buck imitation of a million dollar airplane when the doodang switch in the ion generator section of the warp drive panel goes the wrong way, and at that stage, I guess all reviewers do a double take and think, 'What?' and then we shake our heads and forget about it until the next weirdo comes along.

There is a particularly dangerous type of Space Cadet who poses as an airline pilot and for all I know, some of them may actually be airline pilots, but when they come at me complaining about the doodang switch my first and immediate emotion is: 'If you are a real Airbus/727/DC10 driver, just what the hell are you doing fooling about with a thirty dollar add-on for a computer game?' After all, I practice medicine and I don't go home and play Theme Hospital all night long, and all the real airline pilots I know would run a mile from trying to simulate badly what they can do at work all day long. OK, I realise that the weakness in my argument is that I am a PPL and I do play Flight Simulator, but flying is my hobby, and I absolutely promise that if someone does a simulation of a Rallye 110ST, I will not complain about the doodang switch, regardless of its orientation, color, or creed. That is the Rallye up there, my very own plane, if you allow for the people I share it with.

Maybe I had better introduce you to them. On the left you have Grahame, Chief Executive of our local air ambulance service, in the middle is George, another medic, and on the right is John, our accountant. Between us we own Juliet Delta and while she may not be the sexiest airplane ever made, she sits in a hangar twenty minutes from where most of the group live and I can go down there more or less any time I want and fly as soon as I have won the inevitable game of hangar chess. Juliet Delta costs £30 an hour to fly ($45 to readers west of the pond) and though she is old and slow, we love her, because there is nothing better than taking off up the hill on 26, with the smell of new-mown grass in your nostrils and climbing away at a dizzy 300 feet per minute into the sky purely and simply for the purpose of killing a cloud.

But if someone did produce a simulation of a Rallye, I doubt I would use it. Why? Well, would you, if you could drive ten miles and take to the air in the real thing? And more than that, it takes forever to get somewhere in Juliet Delta compared to a MiG 21 and when I use Flight Simulator, I am looking for some stimulation, not simulation of something I can do in real life. For this reason, if a developer every produces an add-on for FS2008 which recreates the problems of trying to get a heavy old airplane out of a hangar with three too many aircraft in it, it isn't going on my review list. This is the same thinking as I apply to Theme Hospital. All right, all right, I admit it, I have played the game, I enjoyed the epidemic bloaty head stuff too, but the reason I enjoy flying a real Rallye and wouldn't enjoy flying a simulated one is that the real one lets me smell the grass. Neither do I fly the default 172 much, for much the same reason.

Which takes me to the point of this piece. There are people out there who clearly believe that Flight Simulator is so close to reality that it is possible to develop an add-on which is realistic in every detail, for thirty dollars or less. We get them writing in after every review saying that this, or that isn't right and the third doodang from the left is the wrong color and why didn't we point that out? Given that the Space Cadets have the same opportunity to look behind the back of their monitors and work out that there isn't real sky there as we do, I find their attitude surreal, but it doesn't stop them coming. The DreamFleet 737 was a case in point, I had half a dozen emails complaining about the lack of IRUs when most simmers never got any deeper into the sim than hand flying it using the 2D panel. If you don't know what an IRU is, by the way, please don't ask, because you do not want to know.

So let me introduce Fishburn International Airport. We usually call it Fishburn for short and the pic shows it on a particularly un-British day last summer, when you could see forever and there was hardly a cloud in the sky. You won't find Fishburn in Flight Simulator - for some unknown reason Microsoft didn't see fit to put it in. Maybe they didn't think that a 600 meter farm strip merited inclusion, but if they had done, it wouldn't have looked realistic, because as we all know, runways are level in Flight Simulator and the one you are staring at right now slopes towards the east at what many flyers regard as an alarming angle. The actual figure is given as 1.6%, but it is much steeper in places and if anyone lands too deep on 08 (that's coming in from the right of the picture), they sometimes go all the way into the field at the far end there, an exercise normally reserved for combine harvesters.

The other thing about Fishburn is that it is grass, which means that it is skiddy when it is wet, sticky when it is muddy and draggy when it is uncut. None of these problems can affect you in FS, and while we are on the subject of the weather, when is the last time you loaded FS2002 for a flight, downloaded 'real' weather and then decided it was below your limits and you couldn't go flying? Real pilots experience this all the time, particularly if they live in the UK. Today is a good example of what Atlantic frontal systems can do to the UK, try this for a METAR:

EGNT 261020Z 12008KT 2000 BR -RA SCT001 BKN002 08/08 Q0994

If you want to know what a 200 foot ceiling looks like - there it is in the shot below. I took that standing in front of our hangar. I suspect we won't be flying for a few days, because the runway is like a sponge right now, with a small lake forming at one end. You never see water on the runway in Flight Simulator, nor do you run the risk of electrocuting yourself on wires - there is a set one mile to the east of Fishburn that loops around to the north - of hitting a partridge on take-off, smacking into a tree during a crosswind approach, sticking a wheel in the long grass at the runway edge, bogging the plane while you are completing the vital actions, colliding with a non-radio microlight, or having to answer a complaint from Mr. Angry in Bishop Middleham if you forget and overfly his house in the village. Compared to real flight, Flight Simulator is a safe sandbox and that is one of the reasons I enjoy playing in it. Besides, you can always make the sky blue.

I said Juliet Delta isn't quick. This is a vast understatement. In a strong wind, it is one of the few airplanes I know in which it is possible to be blown backwards. She climbs at 70 mph - not knots - cruises as 90, and we fly the approach with full flap at a mere 60. Miles per hour. Not infrequently, when we follow line features like motorways, we get overtaken by the faster traffic. She has leading edge slats and she doesn't so much stall as begin to sink rapidly, an event which happens at about 45 mph and has led to the Rallye's nickname of 'Tin Parachute'. This means she has great short field performance, but with any kind of load you learn to watch the airspeed like a hawk, because speed is life.

At this point I am going to introduce you to Juliet Delta's panel. I suggest that nervous readers avert their eyes, because FS panels never look like ours, but that is another story. We really should do something about those wires, now the RNAV is dead. Anyway, the reason I include this shocking bit of photography is that I can take in the whole of that piece of brown wreckage with one brief downward glance of my eyes, while I watch out for other traffic and try and hold my own with Albert, who always seems to be manning the radio when I take off. Albert is famous because he has been flying since Pontius was a pilot and has been known to do orbits in the vertical axis. He addresses everyone as 'kidder', including Teesside radar, leading to the memorable line, 'Listen, kidder, I just want permission to fly through your zone, I 'aven't time to read you out my life story,' which is rumored to be his response to 'X-Ray, Bravo, pass your message' on at least one occasion. Albert's Rallye is even more underpowered than ours, but that doesn't mean that there is hardly a field in the country that hasn't seen it.

The reason the cockpit view is taken from the right seat is that I am flying the plane and haven't worked out how to take internal photographs and stay insurable, but I hope it gives you the idea. You don't have to pan around with a hat control to see what you want, the basic flight instruments are not only big enough to see, but they are always in easy view and once the plane is in cruise, you can see forever, without having to click in the top of the glareshield to load the VFR panel. A view like this will not work in FS, because screen resolutions make the instruments unreadable - which, for those of you who remember it, was the problem with an otherwise excellent 747 released some years ago - it tried to be too real, and in the process failed as a sim. Flight Simulator is a compromise built on other compromises.

Now you won't credit this, but not all FS enthusiasts appreciate that there is any kind of gap between simulation and reality. On one of my rare visits to the forums, I read a post from a guy who had gone up for an hour's trial flight consisting of straight and level and a few gentle turns, and had concluded that there was nothing to flying a plane that couldn't be learned from Flight Simulator. He was right, there isn't, except for a few detailed things, like how to navigate VFR and how to land the plane without breaking it and how to deal with bad weather and oh, a couple of thousand other things, not all of which have I learned yet (you won't find many pilots who claim that they know it all and if you do, my advice is, don't get in an airplane with them). I would just love to get that guy, stick him in the left seat of someone else's airplane and leave him on short final in a left-hand crosswind approach onto 08 at Fishburn.

To give you another example, one of the exercises which always impresses upon me how different Flight Simulator is to reality is when I do a circuit. When you approach to land, you almost always end up doing some kind of circuit, be it an abbreviated one. Straight in approaches are comparatively rare, even in airliners - that is what all that surveillance radar approaches are all about, ATC are making you fly a kind of circuit, just to keep you from thinking you are grown up enough to do it on your own. Anyway, the moment where reality and simulation diverge is at the point where I turn from the base leg onto final. In a real plane, once you have got the hang of landing it, timing this turn is trivially easy, assuming you have the aircraft correctly configured, but in FS, I consistently get the location and duration of the turn wrong, because it is almost impossible to keep the runway in sight and flying the plane at the same time.

Yes, I know there are work-arounds, you can open a second window, or even rig up another monitor, but that isn't my point, in real life I just keep half an eye on the threshold, another half on the airspeed, and split what is left of my attention between the altimeter, the engine and the radio, and I get it right pretty much all the time, whereas in Flight Simulator I never quite know where I am going to end up when I turn final.

Same with landing. They say that once you get the hang of landing a real airplane, the same principles apply to all of them, and it may be so. But compare landing even something as basic as a 172 or the Rallye with the same process in Flight Simulator and you might be forgiven for wondering if the same physics are involved. One reason why Flight Simulator is so different in this respect is the lack of control forces, but I have reached the inescapable conclusion that the general flight model in the game has been fixed to make it easier to land. Why? Well it takes the average student hours to learn to land a real plane without breaking it and yet I have talked people down to almost perfect landings in Flight Simulator when they have had no prior experience.

Other things that I notice are that braking conditions are always good on FS runways, you don't shit yourself with fear when ice appears on the leading edges, the ADF always points determinedly at the beacon instead of rotating uselessly round and round before directing you straight to the nearest thunderstorm, the VSI never shows you are going down when the altimeter says you are going up, the windshield is never dirty, the fuel gauges are capable of showing readings other than completely full or completely empty, it doesn't matter if you get lost or run out of fuel, the weather (and particularly the weather transitions) is a joke, ATC don't speak American outside the US, and that if real aircraft taxied like Microsoft ones, we would all be dead.

So if the sim is not like reality, and it has all these dreadful, completely unforgivable faults, why do I bother using it? Well, the short answer is that it is fun. I enjoy playing with it and I can live with its deficiencies. I like blasting off the runway in a MiG, I love landing the DreamFleet 737 at Heathrow, and I can indulge my fantasy of soaring over the Rockies in a glider. What do all these things have in common? I can't do them in real life - and that, I am pretty convinced, is why most people use Flight Simulator - it lets you do stuff you couldn't do under normal circumstances. You and I make allowances for the fact we know it isn't real, because it lets us dream.

But there are some people who aren't like us. For them, every flight has to be made in conditions of the greatest possible realism; they spend days pouring over manuals and books, and in a few cases, the border between simulation and reality gradually blurs and dims for them. Each flight starts in a cold cockpit, with full ATC, and ends with them turning out the lights many hours later. Most of they end these sessions tired, but happy, but sooner or later, the fact that they are looking at a rectangular screen in a nice warm room, rather than a lowering sky with a reducing cloud base escapes some of them. In a sense, there's nothing wrong with that, apart from the problem that it can lead to raised expectations of what flight simulation can provide and particularly, of what is achievable at current prices. But getting too deep into what is, at the end of the day, marketed by Microsoft as a game, isn't necessarily a good idea, particularly when it leads you to excess worry about the correct function of obscure doodangs.

Does it matter that these people want everything to work in an FS airliner panel exactly as it does in real life? The more I think about it, the more I think the answer is yes, because when they start posting in the forums they introduce enough fear, uncertainty and doubt to spoil everyone else's enjoyment. I feel for these guys, because they are rarely likely to get what they want, unless they are prepared to spend a lot more money.

A typical FS add-on retails for thirty dollars and it has long been held that it is financial suicide to charge any more, because the market won't stand it. The logic behind this statement is moderately well developed - the average simmer is a turn and burn guy or gal and many are young and impecunious, so developing a product that takes four thousand man hours to code, is completely accurate and costs a hundred bucks on the shelf is wasted effort, because the buyers won't pay for it. There is a tiny market for products like this at the Space Cadet end, for about three people, none of whom will ever be happy, because one of the doodangs will be found to go the wrong way. Where developers haven't helped themselves is that although five years ago, it was perfectly acceptable for an FS panel to consist of a single 2D forward view, with the instruments in more or less the correct style and positions, nowadays we have to have eighteen auxiliary panels or you can't sell the product. As teams have vied with each other to code packages to ever higher standards, usually because they are getting noise off the Cadets, they have gradually been trapped into making out that they can develop completely accurate sims which sell for thirty dollars. Inter-developer competition doesn't help in this respect, particularly when you consider that there must be few markets where the participants get on so badly as they do in Flight Simulation - at times it seems that the only thing that prevents blood being shed is the oceans separating one developer from one another.

The trouble is that there isn't enough money in the system to do what some users want. I realise that often it has been as much the development community as the Cadets which has been responsible for pushing up expectations, but in the process we have reached a point where it is hardly makes economic sense to code products to the standard we are being led to expect. Just to put some figures to this, the average add-on is doing really well if it shifts more than a few thousand copies. Five thousand sales is wildest dream territory and most sell far less than that and end up as labors of love, rather than financial successes, because the whole process is so incredibly time intensive. When you have to do a visual model, flight model, 2D panel, 3D panel, sound set and code a dozen sub-panels, a six month product development time looks tight and it doesn't surprise me that some releases slip badly. Add to this the danger that someone else might be developing a similar package and it isn't surprising that venture capitalists don't list flight simulation at the top of their list of safe investments. Users have little appreciation of this, which accounts for why the doodang question occasionally pops up - the people who ask it presumably assume either that FS products can be developed in a flash, or that they sell in such vast numbers that it makes economic sense to thoroughly research the habits of doodangs. One doodang is OK, but stick several dozen in a panel and their code can interact in all kinds of unpredictible ways, leading to unexpected behavior and the dreaded, but routine, patch. You may laugh, but I had an email the other day from some guy who was enraged that the operation of the pneumatic pressure system on some sim of an obsolescent airliner didn't work exactly the same way as the real thing did, especially (he added) given that package cost a whole twenty five bucks. Purrrleease.

Despite the pressure to produce what amount to procedural simulations, financial considerations force most developers to compromise somewhere along the line, purely and simply because you can't duplicate something that costs tens millions of dollars to develop on a budget of a few thousands. This means that while they try to capture the spirit of the way a plane works, they don't put in all the doodangs and if they do, some of them will be non-working bitmaps, a sad, but inevitable consequence of kow-towing to the thirty dollar bar. The rot sets in when developers don't make the fact that there are shortcuts in the panels explicit, because it leads to cerebrovascular events among the more excitable Cadets and a general feeling of dissatisfaction among the people who read their musings in the forums. The interesting thing about the airliner I mentioned above was that it did feature such a disclaimer in the manual, in red, but my correspondent hadn't read it. I've made a note of his name - if he does turn out to be a real ATPL, I'm not going on any of his flights.

One exception to this rule is Wilco's outstanding 767 PIC, a procedural sim that was a one of a kind project. I think it is fair to say that it became a glorious obsession for the highly talented team that developed it and it took tens of thousands of hours to code; because of which we may never see anything quite like it again. 767 PIC is without question the best attempt to simulate a passenger jet cockpit of its kind and kudos to the team that developed it. But why no Gmax FS2002 version?

I would guess that the reason is purely financial. I would assume that Wilco's dilemma right now is that with the release of the next version of Flight Simulator less than a year away, the sales life of a redeveloped version of such a complex and relatively low margin product is becoming seriously constrained. Logic dictates that it is more sensible to produce and sell two lower level simulations than to maintain the code of one highly sophisticated one, if you end up having to charge the same price for each package, though Wilco are developing a new version of the 767, we would all be delighted to see it, because there is nothing else like it.

I mentioned financial considerations, so perhaps I had better explain how the add-on market works. Most FS aircraft have to be written and released far more quickly than the developers would like, because the FS product cycle is only two years. Functionally, there is even less time than that to create and sell a product, because it takes around six months for Microsoft to release all the software development kits necessary to code stable add-ons, so when you add on another six months for product development and testing, it isn't totally surprising that there is a glut of commercial releases for FS2002 right now. Shortly, this flood will mysteriously dry up, because FS-ACOF will be on the horizon and who wants to buy a package that may have a limited shelf life?

Now, if we were prepared to pay more for our add-ons and the Flight Simulator version cycle was longer than a couple of years, developers could devote much more time to making their add-ons more realistic, but as it is, we won't, the cycle isn't and they don't. Even if we were happy to cough up several times what we do now and Microsoft were prepared to take a 50% hit on their FS profits by lengthening the cycle and simmers were prepared to be patient and wait longer for thier next fix of eye candy; Flight Simulator aircraft still wouldn't be completely realistic, because they are chimera created on a PC and not real planes that have to gain a Public Transport Certficate of Airworthiness. You, I and most other simmers out there instinctively know this, as we enjoy the thrill of flying inverted under the Golden Gate bridge in the 747, but there are one or two people out there who haven't got the message.

If all the billions of dollars behind Microsoft can't make their 172 panel look and behave like the real thing, what chance have third party developers got? Well, there is another way out of this maze, but it would involve a radically different approach. If an add-on developer stuck with a product in the long term and upgraded it each time a new version of Flight Simulator came out, and if users were prepared to pay for those upgrades, then as the programmers gained experience of the way the plane worked, they could increase its functionality and realism with each upgrade. Effectively the packages would get better and better as time went by, which theoretically ought to please everyone, even the Cadets. But the payback would be that we would get fewer new planes to play with, because 99% of the coding effort would end up being invested in old products and ordinary simmers like you and I would raise the justifiable complaint that our dreaming was being curtailed.

Quite what Albert would think of this, I hesitate to imagine. But I imagine he would say something like, 'You can't have it all ways, kidder.'