Slashdot Mirror
X-Plane - An Obsession For Realism
caseih writes "Popular Science is running an article on Austin Meyer, the creator of the popular X-Plane flight simulator. Although not an open source project, X-Plane has a devoted community of flight enthusiasts and developers who are striving to make it the most realistic flight simulator ever. In fact, flight characteristics are calculated in real time from aircraft design data, not static tables like MS Flight Simulator. PopSci has a neat picture showing X-Plane calculating the lift-drag vectors in real-time across an aircraft. Meyer's quest for realism in his simulations dominates the development and use of X-Plane."
As any animal becomes used to something, it inevitably ecomes less affected by it. By providing people with simulations closer imitating reality, you raise expectation for future simulations.
I'm just waiting for the cybernetic implants which allow simulations indistinguishable from reality. They already have brain cells growing on silicon (not made out of), it will just probably take a while for science to catch up with years and years of evolution.
Contact Me (got tired of viruses emailing me).
Alot of other flight sims are games, X-Plane is not. This is a serious piece of software used by alot of professionals to model and simulate prospective aerospace designs. I can't count the times it has been emphasized to me that this is not a game. ...that said, it's damn fun sometimes.
"The worst tyrannies were the ones where a governance required its own logic on every embedded node." - Vernor Vinge
I can already imagine all the "great, now terrorists won't even have to go to flight school!" comments.
My advice: forget about it. If we want to prevent a repeat of 9/11, the solution is common sense initiatives such as locked cockpit doors and military quick response procedures... NOT by restricting basic technical information.
I love X-Plane, precisely for the reasons posted: it's very cool that the simulation strives to be as accurate as possible, and gets better with each release. On the other hand, at least one journalist disagrees...
Yesterday, Salon had a ridiculous article(might be restricted to subscribers only, sorry) that claims that modern consumer flight simulators are too realistic, and implies that they should be banned or restricted somehow. And of course, as the headline promises, the article does indeed place some of the blame for 9/11 on such flight simlulators!
Bad Salon, bad. What is it with the media hating video games, anyways?
I am not familiar with X plane , but I wonder how it compares to the Flightgear project. One of the advantages of that project being open source was extensibilty. The project has all sorts of modding potential.
apt-get install flightgear for all you deb heads.
runs on win32 also
For a moment there, I thought this was a story on Mike Meyers and his next Austin Powers movie, featuring a futuristic X-Plane thingy that spoofs Moonraker.
I'm honestly not sure whether to be disappointed or not...
I'm sure the author would like nothing better than to give away X-Plane for free. The trouble is that there are some applications that require the bulk of one person's time, for years on end, if they are to be amazing. And, like anybody else, the creator of this work needs to eat.
I think we're certain to see a greater variety of open source freeware apps in the years ahead. But there just aren't enough people out there with serious expertise in both aeronautics and coding, who can pitch in and build an open source X-Plane in their free time.
You want 1001 small apps, from hard drive erasers to science calculators, the open source movement has you covered. You're also covered with the handful of huge apps, that everybody needs, since there are sufficient coders out there to recognize the supreme significance of these apps and donate bits of their time (take Mozilla and OpenOffice.org as two examples.)
But if you want something as niche as a world class flight simulator, sorry, you're probably not gonna get it open source. It's gonna take a huge effort from a tiny group of people, and they need to do it to the exclusion of other things. Like anyone else, they've gotta eat. And to call their code "useless" because it's not open source, that's unfair, mean spirited, and ignorant.
I'm generally "Interesting," "Insightful," and even "Funny" here. What the hell happens to me at parties?
At the end of page one:
"I have a moral duty to make it fly as realistically as I can."
Now consider: if every programmer was able and willing to make a similar statement about their code, what would our software "ecosystem" (as MS likes to phrase it) look like?
Why do that when there is a perfectly good open source flight simulator available in the guise of FlightGear?
If these people contributed that passionately to FlightGear then it would be awesome.
Free Gamer - Free games list and commentary
The only reason that OpenOffice and Mozilla exist as open source is that big companies bought the source, and released it but continued to pay developers to make it better in an effort to kill Microsoft's power bases. I don't think those sorts of efforts could be duplicated by a team of coders who never get paid.
Degaussing scares the bad magnetism out of the monitor and fills it with good karma.
you know, one gets really tired of this kind of attitude. Did you read the article? This is his life's work, and I think he deserves compensation for it. X-plane (versions 6 and 7) is the only software I've paid for in the last 2 or 3 years, and I was happy to do it, because I'm supporting a real person, and getting a (albeit quirky) quality piece of software for it. I'm certainly a believer in the open source movement, I use it whenever possible, and I've contributed some of my work to the community. Anyway, why don't you put your money where your mouth is and code your own flight sim. (Or quit making inflammatory statements like code hoarder)
I played it alongside FS2002, if I remember correctly, about 18 months ago.
It was not too bad, but it had some serious physics problems. There was a lot of incorrect behavior at extremes. It may have been technically more realistic in terms of calculation, but it was also incorrect in enough cases to make it not fun for me.
FS behaved more consistently, and close enough to correct to make it superior, in my opinion.
I'd love to give some detail about the problems I encountered, but I really don't remember anymore. I just remember buying it, giving it some good play, and deciding it had fundamental problems.
Now thinking back, the best flight sim I recall playing was the one by the guys who made System Shock. It was an aerobatics focused sim, and the physics seemed very accurate.
.sigs are for post^Hers.
Apparently not, if M$ bet the farm on it before creating DirectX...
YFI
-uso.
Dreams, dreams, don't doubt dreams, dreaming children's dreaming dreams. Sailor Moon SS
very cool but can I make my own plane and then play with the wingfoil design (and see how quickly I crash)?
Anyone know the format or what is used to create the 'aircraft design data' that gets processed real-time? I'm assuming it requires a high-res model of the plane and a fancy windtunnel simulator.
A motion has been filed to ban "The Sims" because it is "too realistic and offers insight on human life and behavior that can be misused in various different ways"
Sigs for Nerds. Sigs that Matter.
A friend of mine who flies ultralights put the details of his recently bought machine into xplane. While it flew sort of well in xplane, it had a few problems in some situations that just seemed to defy logic, and needed a bizarre set of movements (relative to the logic he'd been taught) to get him out of a dangerous situation.
Sure enough, after time flying the real ultralight, he got into very similar situations. The xplane practice paid off and he's still alive today, where if he hadn't he'd at least be badly injured with a broken plane.
I don't know how much longer that luck will keep up. I wouldn't call ultralights a long term hobby - but it gives an idea of the benefits of xplane. You get to customise the machines you fly in any way you choose, and it'll do a damned good job of emulating reality.
You could always try Orbiter. http://www.orbitersim.com
"Aren't all the data known in advance?"
Short answer - no.
Medium answer - a lot of pre-computing
Long answer is a course in fluid dynamics.
The Singularity is closer than you think
Quant
Passion isn't enough. This type of programming is not "write a text editor" or write a replacement for Notepad.
Its scientific, requires loads of specific realm based knowledge, and eons of refinement and highly technical skills.
Its a simulation engine that is precise and accurate. It's not just a toy.
OSS is great. But not for everything. Somethings are too narrow for a sufficently wide pool of programmers to latch onto and program on thier own. The number of OSS-comitted programmers, with aero-engineering skills, with 3D programming knowlegdge, with sufficent free time and sufficent drive to replicate this level of work is very, very, very small.
In relation to FS games, This guy here has got da bomb setup 9 PCs, 13 monitors, one plane. Since some idiot mod put me to -1 me down as troll for replying to an AC with the subject line of "eat shit and die", I'll try this again and wish a speedy death to the mod in question.
Eat recycled food - it's good for the environment, and OK for you.
I agree that real time calculations of forces and stuff is really cool, but can you really tell a difference from that and pre-calculated data?
Martin
I guess I can't really comment on the accuracy, having not actually flown a plane. Also, on the scale of "Flight sim hardcore rankings", I'm really just a dilettante. (OK, that rarely stops people on the forums, but it's just not my style.) But I can comment on the processing power needed to run the dynamic model, and so can you.
The demo is free, runs on Windows and OSX, which is Austin's preferred development platform. Since it is a demo, it only lets you control the plane for a few minutes, but it will let you see how quickly he can do the aerodynamic calculations and experience the realism for yourself.
My opinion: Performance is ok but not great on my 500Mhz G3 ibook (although it is quite playable). It is pretty spiffy on my Athlon 2100. The video cards probably have a lot to do with that, too, not just the aerodynamic model.
It was almost open source
"But there are disadvantages to being a one-man show. It's hard to imagine Microsoft coming home drunk one night from a party and accidentally uploading its entire source code, as Meyer did a few years back. "I woke up the next morning and found an e-mail from a friend alerting me to what I'd done. My heart stopped. I had basically given away 12 years of work. I thought my life was over." He was able to remove the files before anyone could spread them around, but to this day he feels like he dodged a bullet. "I don't drink anymore," he says."
Those of you who have visited the X-Plane site have no doubt found the article, but here's a quick bit for everyone else...
So what sort of planes can fly on Mars? Not anything from Earth, that's for sure. Not enough lift or thrust. A Cessna or Boeing will just sit there on the ground without even moving. Put them in the air and they drop like beveled bricks with no wings. Both of my Mars-plane concepts are much like the U-2 Spyplane (designed to operate at around 100,000 ft, in simlar density air) one with a HUGE high-bypass jet engine built AROUND THE FUSELAGE, and another with a smaller rocket engine in the tail, like the X-15. The rocket plane has a lower-thrust engine, with plenty of fuel, for about 30 minutes of flight or so... the JET plane can fly for hours!
Article link (you'll have to try to ignore the excessive use of ALL CAPS)
I've always thought X-Plane was cool, but after reading this article I was convinced... and that's when I read the article well over a year ago!
I believe you are talking about Flight Unlimited... The company was named Looking Glass Studios.
The screenshots are unbelievable! Look at the little people!
http://pcblues.com - Digits and Wood
> But if you want something as niche as a world class flight simulator, sorry, you're probably not gonna get it open source.
"The FlightGear flight simulator project is an open-source, multi-platform, cooperative flight simulator development project. Source code for the entire project is available and licensed under the GNU General Public License.
The goal of the FlightGear project is to create a sophisticated flight simulator framework for use in research or academic environments, for the development and pursuit of other interesting flight simulation ideas, and as an end-user application. We are developing a sophisticated, open simulation framework that can be expanded and improved upon by anyone interested in contributing."
-- <http://www.flightgear.org/>
But accurate: I can't use it and, given his attitude, I never will be able to. So "useless" pretty well covers it for me.
The odd thing is that this is a very good candidate for open-sourcing without cutting the programmer's throat. Flight simulations (particularly this one from the looks of it) are complex and users really need and want a good manual.
Give away the code; sell the manual! In this case I think the sales would probably go up, not down.
TWW
"Encyclopedia" is to "Wikipedia" what "Library" is to "Some people at a bus stop"
X-plane already simulates both Space Shuttles and flying experimental planes on Mars, as well as other space-going crafts. Heck you can even take off into space with a Rocket!!
"I tend to think of OS X as Linux with QA and Taste", James Gosling, creator of Java
Unfortunately, flightgear seems to be years behind X-Plane and falling away.
I'm quite annoyed by the parent argument. I know of more than one aviation lab that has approached Meyer and said "we would like this change or alteration made. how much?" and they have recieved by return of email, absolutely for free, modified versions of the software. He could of made thousands each time. Meyer is thought of very highly in these circles. Its an almost unique attitude towards customers that I find incredibly impressive. Puts a different spin on it huh. You can't argue the guy is some sort of profiteer; he's doing *free* bespoke programming for no reward. Why begrudge him the means to earn a living?
Plays violent online games as: Nerfherder76
Yes, dynamic modelling is better than static tables. Not only is the performance more in line with reality, but it allows people to design airfoils and aircraft with the included software, and test their performance before any parts are fabricated. No guess work. Real engineering.
:)
It's a hackers dream, because ALL of the flight controls and flight data can be imported/exported over a network. It also has a very sensible plugin system, and the author encourages people to come up with new and cool tools without any licensing restrictions. It might not be open source, but the architecture is very open.
X-Plane is the flight simulator of choice for many companies, including Scaled Composites, the builders of Spaceship One. It's also FAA approved for training towards commercial, transport, and instrument certificates.
Not only is the flight model incredibly accurate, but you also have to deal with differences in traction between tires on a wet runway, damaged windscreens from hail, and more equipment failures than you can shake a stick at.
It's amazingly beautiful with a reasonable graphics card and the latest scenery plugins, and it can use real-time weather information from NOAA.
It's not a toy or a game, even though it may be fun. It's as close to flying as you can get on your PC. I could go on and on, but it's probably better that you head to the web site.
http://www.x-plane.com/
I have been using X-Plane for a few years and I and my pilot friends use it. My only real dissappointment was when Austin canceled the Linux port, but 6.x will run under wine now, haven't tried 7.x betas.
X-Plane is the only FAA training approved consumer package available. Read the front page on the web site, people have been using this flight program for a couple of years now to model aircraft behaviour during development.
I think you may be confusing ease of flight or level of fun with realistic physics. I had a couple of programs for my bro-in-law to practice with while he was taking flight lessons and he too latched onto X-Plane.
Kindness is the language which the deaf can hear and the blind can see. - Mark Twain
It's absolutly amazing that Austin could achieve this, but the project is getting at it's limits. Why?
Even the 4th release candidate still has quality issues, or, with a less friendly word: bugs. The user interface is really, really bad. Not only does it use custom widgets, but the widgets do not follow the usual expectations. The dialogs behave strangely (exit buttons), and, for example, if you increase the rendering quality, the system drops you down to the nearest airport, which comes handy if you're flying a 747 and you end up on a helipad.
People also develop flight models and (photo)realistic landscapes (e.g. the Global Scenery Project or, e.g., Cormac Shaw's high-detail scenery for Ireland and his Aer Lingus Jets at the Irish Hub.) Stuff like that generally works much better, and there is a great variety to choose from!
I also tried to evaluate FlightGear. This project is not anywhere near X-Plane. If I'm not mistaken, they only accurately simulate piston engines (other engines are a weak approximation). Besides, FlightGear doesn't compile if you don't have certain libraries installed, which turned out to be a pain on OS X...
That said, I believe that FlightGear may outperform X-Plane in a couple of years. Until then, I'll stick with X-Plane...
I used to support MS Flight Simulator over the telephone, so ive had to play with it quite a bit. Although I never really had to figure out how the game.. excuse me simulator, calculates flight data (ive had people yell at me on the phone for calling MS Flight Simulator a game.. you know, those people that dont play games and its a "simulator").
Anyway, to the point.
With MS Flight Simulator, when you build a new plane for it, you need to build a file that has all the information about how the plane reacts if it were to fly in the air. (it may be a bit different then this I haven't worked there for a couple years). I made a new plane in GMax, but used a different planes files and pretty much changed the model of the aircraft. When you go fly, it uses this file to figure out how winds, altitude, etc are going to react with the plane.
With X-Plane, when I build a model and fly it within the 'simulator', its going to use the actual model itself to calculate how the plane is going to fly. This seems pretty useful to someone who wants to design planes. They called it a 'virtual wind-tunnel' in the article (I only read the first couple pages).
So, the aircraft design data in MS Flight Simulator isn't based off the model of the plane itself, whereas X-Plane uses the model of the plane to figure out how it flies. I would say that for accuracy sakes, your better off testing and simulating stuff with X-Plane vs. Flight Simulator.
As a side note, Flight Simulator is the biggest pain in the ass to support. I would take a few calls a week from some ex-pilot who doesn't think a certain plane is flying correctly (it should be able to go this speed, or the panel doesn't have the gauges like they are in real-life). I really wish those kinds of people could come to my work and find out what kinds of things I can and cannot do.. If everyones plane flies the same way, I cannot really fix that.. Of course ive also talked to a guy who put a bunch of monitors into an old 737 cockpit for MSFS, and another guy who had 9 monitors hooked up, mainly using them for Flight Simulator. That product has some hardcore fans attached to it... (warning im a bit drunk and i tend to ramble, ill stop now..)
I spent perhaps half my life playing flight simulators of all kinds. I loved it. About 2 years ago I started playing X-plane, investing in the CH sim yoke and everything. I flew Cessnas all the time.
Finally, about a year ago, I decided that 15 years of simulators was quite enough, and I started to take flying lessons in real airplanes. The same aircraft - Cessna 172. I was expecting that with my many years of simulator experience, including 2 years with mega-realistic X-plane, I'd be able to waltz right in and fly the plane as if I was an expert.
Guess what? The plane kicked my ass. Flying it felt *nothing* like the simulator. Although x-plane may accurately simulate how the aircraft moves through the air, the air itself is in motion in very complex ways that aren't simulated. The plane moves around in ways the simulator never prepared me for. I couldn't land for shit until I'd done it in the real plane maybe 100 times, and I didn't get really good at it until about 300 times. I've taken some of my other X-plane addict friends up flying with me and let them take the controls, and they always say "It doesn't feel anything like the simulator".
And, completey separately from the actual aircraft control feeling unrealistic, no simulator I've ever played has done a good job of simulating the stress of a real flight. X-Plane does nothing to prepare you for trying to fly the plane while a controller is continuously talking in your ear to you and the other 10 airplanes in his airspace. X-Plane does not put you in a panic that you just intruded on a class B airspace boundary. X-plane does not wait until you're on short final, then tell you to start a climb, do a 360 and then reestablish yourself on final because a jet just got his IFR release. X-plane does not ask you to keep 3 other targets in the pattern in sight while landing. X-plane does not make you try to listen to the ATIS recorded weather information and controller simultaneously while also trying to fly through clouds on instruments. All of these things happen to me regularly in real planes.
(Admittedly, I do fly in the most complex airspace in the world - the LA basin - so maybe this is an extreme example.)
On the positive side, simulators do an excellent job of giving you an understanding about how navigation works (e.g. how to track VORs, when they're reversed, how to form a mental picture of where you are based on navaid information, etc.)
Yeah, pondering the implications of nanotech for years can do that to you. Of course, it's easier for people like you to be pessimistic and overly cynical about the future because you didn't get the impractical flying car that was promised you for The Year 2000.
--
Power to the Peaceful
Nope, OpenGL is not covered by the GPL, it is covered by an X style license now and was covered by several commercial licenses from SGI in the past. See This
There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
That's simply not true: I'm currently studying for my PPL in England, and one of the things you are taught is how to deal with spiral dives and stalls.
/.'ers seem to love it.
Now, no matter what I do in MS Flight Simulator, in whatever plane, have I ever been able to enduce a spiral dive or spin. (Trust me, it's real easy in a plane. That's why they teach you how to recover.)
And the low speed stall, particularly in the Cessna model, is a lot more benign than the real thing.
That said, MSFS is great for learning instrument work. And if you want to practise a hundred circuits, or the effects of cross-wind on landings then its good.
But I shall try X-Plane, certainly,
--- My dad's political betting
I had a run-in with the X-plane people (Basically Austin and his buddy Mike Brown on sales) over their "guaranteed upgrades for life." In a word, they lied. Furthermore, they continue to lie, and are absolute ASSHOLES about it.
To be specific here, they will always make the latest patch available for the current version. That is, if they're developing X-P 6.x, then the latest 6.x patch is available. However...
1) You cannot get any older patches. This is a problem because several times the current versions has been buggy, unstable, or broken.
2) Once the initial 7.x release is out, you are absolutely SOL on downloading the final 6.x patch. He will NOT provide it under any circumstances, once he's decided to get rid of 'free support for life' on a previous version.
I'm sorry to have to post this. I think that X-P is a really cool program. I'm utterly amazed at how far he's gotten with it. However, his code review (poor), attitude ("fuck you!"), and flat out lying on support all lead to something that I'll never drop money on again.
Pity, really. If he lost his ego, he'd write better software.
You can read more about it here.
"People who do stupid things with hazardous materials often die." -- Jim Davidson on alt.folklore.urban
I'm a 2-year user of X-Plane, although I confess I've not been following the v7 beta stuff (I've stuck with 6.60 for now). I've logged hundreds of hours on the sim, and tinkered with the other tools available (e.g. I've modified aircraft etc).
X-Plane is a fantastic piece of work for a single person to have to his name. Probably the highest praise I can think to give it, is that it's not only the best at what it tries to do - simulate flight of all kinds - but it's also usable.
Having said all that, usable is about as far as it goes. It's not, and has never set out to be, a polished application with a glitzy UI. The interface for the sim and the tools is good enough and no better. If you need to get out and see what'll happen to an NF-104 at 100K feet when the control surfaces fail on you, then X-Plane is for you. But be prepared to adapt to it's interface, rather than have it teach you.
I've been looking at FlightGear recently too. And about all I can say about this right now is that it's clearly got promise, it looks good, but there's a looonnggg way to go. At least X-Plane lets me choose aircraft from a file selector dialog; I have to shut down FlightGear and use a different command line switch to load a new aircraft. It's clearly still very much for the Geeks for now, wheras I know there are professional pilots using X-plane.
--
I'd rather have a bottle in front of me than a frontal lobotomy
It's important to note though that building a 3d model for X-Plane is not the same as building one for MSFS. In X-Plane, the 3d model's shape is defined by the specifications set using the Plane Maker tool. For example, you type in a value for wingspan, and the wingspan of the model changes. In this way is the 3d shape each aircraft defined.
This is much more difficult (in my opinion) than modeling an aircraft in a 3d graphics package such as Maya or 3ds max, though it is necessary for the way X-Plane calculates flight models.
FlightGear is designed to simulate more than just piston aircraft "accurately", though accuracy is infinitely arguable. FlightGear supports multiple types of aerodynamics models, including LaRCsim (developed at the NASA Langley Research Center), JSBSim, and YASim, the latter of which simulates performance based on aircraft 3d geometry, much like X-Plane's flight model.
A flight model developed by UIUC is also supported which is capable of modeling nonlinear aerodynamics.
I own a motorcycle. A fast motorcycle (a Suzuki Hayabusa).
Anybody telling me that piloting a fast sports bike in whatever simulation comes anywhere near the real thing with wind, rush and G-forces tickling your fear and adrenaline buttons as you push the bike to the tarmac in sharp curves at 300 km/h (175 mph) have probably never been on two wheels, just in their little safe steel cage on four.
Saying that it's even close to the same thing to be on one of those arcade-style simulators, even one where you lean the bike (which is about as close as you can get in bike sim), is a ridiculous statement.
That's my first point: a sim is nowhere close to the real thing. You can get useful lessons for the real thing from a sim if you are actively doing both, but otherwise, forget it.
Now for my second point: An aggressive low-flying sportsbike has 2 (two) important gauges. Speed and RPM. You can ignore the speedometer. Most of the time, you can ignore the RPM gauge too (you get that info from the engine noise anyway).
How many gauges are there on a 747, and how many of these can you routinely ignore?
Subscribe to the xplane-tech list at yahoo and see what his customers really think.
If we could somehow get Austin Meyer and Sid Meyer to work on it together, we'd have the world's finest airport management simulator ever.
Then if we also enlisted the help of Oscar Meyer, we'd have winged weenies waging war.
Moreover, FlightGear, because it is open source, can be tweaked, modified, optimized. .
This sig is a test. If this had been an actual sig, you would be reading something quite a bit wittier than this now.
Is the reason you can't use it truly because it's not Open Source? If so, you are the one ideologically cutting yourself off from a good piece of software.
If the reason you can't use it is because it's written for Windows, that has nothing to do with Open Source. Don't confuse the two. Open source software exists for Windows, and closed source exists for *nix, but if it's not written for your platform, chances are* you're SOL.
*Yes, programs can be ported.
Every so often, salon publishes an article that can only be described as "slashdot bait." Mildly interesting-- but tarted up to the point where they're practically begging to get a slashdotting, and all the ad revenue that presumably comes with such a distributed denial of service.
The basic theme of that salon article was that a skyjacker of today would have an easier time learnng the ins and outs of a modern jet instrument panel with a computer program then he would four or five years ago. Apparently, earlier flight simulators took certain liberties with instrument layout. Additionally, many of the modern simulators also simulate the flight management sytems fairly well, so if a terrorist wanted to automate portions of his flight plan, he would be more prepared to do so.
It's slightly more sophisticated than the "Doom trains students to kill" articles of a few years ago, but not by much. (And I say this as a loyal subscriber of Salon.) It might well get debunked by salon's "Ask the Pilot" column.