FlightGear Newsletter November 2010
We would like to emphasize that the monthly newsletter can not live without the contributions of FlightGear users and developers. Everyone with a wiki account (free to register) can edit the newsletter and every contribution is welcome. So if you know about any FlightGear related projects such as for example updated scenery or aircraft, please do feel invited to add such news to the newsletter.
Call for volunteers
The OpenRadar project is looking for a new maintainer.
Anders Gidenstam has continued to develop Walkview which opens up all sorts of exciting possibilites. Here is a short video of Walkview being used to give a different perspective on starting the Sopwith Camel . In addition to the rectangular plane that we have had for some time, we can now define the area in which we can "walk" as a line, polyline, or circle.
We use the new definitions in Multi Player Carrier Carl Vinson to move around the Goofer's deck while we are goofing, or to seek a more advantageous view from PriFly, or to inspect the Flight deck as the Flight Deck Officer, as seen in this video.
Meanwhile, in other news, Tim Moore has fixed the condition tag in the material animation. We can combine these two facilites to generate some more interesting views - . Alexis Bory has said that he will see if he can produce some more models with which to populate the hangar. Let's hope he finds something!
Though working perfectly for years, the "live weather" feature in FlightGear 1.9.x and 2.0 is now broken. Unfortunately, the FlightGear code of past releases contains minor problem which so far remained undetected and without effect. Since early November, probably caused by a server upgrade at the weather services, the FlightGear weather requests are now being rejected due to this problem. To revive live weather you have the following options:
- As a workaround, configure a proxy-server on the fgfs command-line. Relaying the request through a proxy solves the issue (see ).
- A description on "How To Fix (temporarily)" you can find under http://www.emmerich-j.de/FGFS/METAR-fix.html
- Experts being able to patch and recompile FlightGear may try to apply a patch (also see forum).
- Developers can use FlightGear's latest developer version from GIT, which already contains the required bugfix.
- Otherwise please stick with manual weather configuration and wait for the next FlightGear release (won't be too long).
In the hangar
Boeing 737NG Series
Three out of four models from Innis Cunningham's Boeing 737NG series have been included in FlightGear's GIT repository. Despite being an initial release, the models are characterized by great attention to detail in the 3D cockpit, with some of the more advanced systems such as the autopilot and autobrake systems already fully operational. The models are further characterized by excellent Flight Dynamics Models (FDMs), co-developed and tested by real-life 737 pilot David Culp, and excellent exterior texturing, provided by FlightGear's very own texture master Gooneybird. Included are the 600, 800, and 900 variants. The 737-700 is expected to follow soon.
The ZF Navy free balloon has been updated with a custom shader effect that changes the shape of the balloon envelope according to its gas content.
Boeing 757 and 717
Skyop has picked up development of the Boeing 757-200 and the Boeing 717. New features included generic autostart systems, improved models, new instruments and new YASim FDMs. New and improved cockpits and aircraft systems are planned. Skyop is requesting livery makers to start painting their favorite airlines and submitting their textures to the official FlightGear livery database.
Armchair Ace has been developing the DC-8-73's cockpit, which has also been used in the new DC-8-63.
The 777-200ER received a cockpit and systems upgrade. Extensions by Syd Adams and ThorstenB include better GPWS integration, improvements for cockpit instruments, sounds, textures, key/mouse bindings and many bug fixes. The upgrade also contains a FlightGear's first: brake temperature monitoring.
Pilots are advised to use spoilers and thrust-reversers for deceleration to avoid overheated brakes - especially when landing a heavily loaded plane. The new auto spoilers, automatically deploying on touch-down when armed, provide some assistance here.
New Bangkok International Airport
Mike (aka D-Sky1) is busy again. This time he is working on Suvarnabhumi Airport (VTBD, also known as New Bangkok International Airport) in Thailand) as the architecture of the airport buildings is very beautiful. It will sure take some time but a first glance at the work in progress is already possible. Discuss the progress with us in . After the airport is finished, placed and available, AI traffic will follow (mainly Thai Airways).
AI traffic arrives at RPLL (Manila, Philippines): Mike created a comprehensive list of AI traffic of Cebu Pacific Airlines. Soon available via GIT, Cebu Pacific Airlines will provide domestic and international flights to the most important destinations. AI for the second big carrier of the Philippines (Philippine Airways) will follow later.
Land cover updates
Detailed land cover data for Vermont has been added to the map server. This state in the northeast United States will be a part of the next world scenery release.
Work continues on hand-created, detailed land cover data for northern Switzerland. Photos of landmark buildings or airport buildings in Switzerland would be welcome as we try to model this beautiful country.
NASA Interactive Launch Complexes
Jack Mermod has released the first of many NASA Interactive Launch Complexes. This first release is based on Launch Complex 5 and the Mercury-Redstone 3 launch. The model should be in the official database soon, for now you can download the scenery here.
After attempts to provide a new World Scenery build in sync with the 2.0 software release failed miserably (due to software deficiencies), some effort has been expended into investigations, experiments (some of which ended up in filing bug-reports for related software) and large-scale tests for replacing the entire polygon processing in TerraGear by the use of GRASS GIS. Just to give an idea about the term "large scale": The procedure to turn the Europe-wide CORINE CLC2000 vector data set into a topologically consistent coverage is taking several weeks on a 3 GHz Xeon machine.
The "*-decode" commands and old-style work directories are affected by this change as well as all polygon-mangling routines in "construct". The GRASS toolbox provides all the features we need for reliable, topologically correct polygon handling (cleaning, clipping and the like). Those (early) steps of the new toolchain which have proven to be rather stable are available here (GRASS7 required for certain features).
Jet d'Eau in Geneva
The Jet d'Eau in Geneva is with 140m height a recognisable landmark. Ot-666 created the fountain and the pier it is located on in the Lake Geneva. The model features night illumination for the fountain.
Looking for new airports to visit? Try a Dublin (EIDW) to Aalesund (ENSD) flight! These two airports, along with Sandane airport in Anda, Norway (ENVA), have recently had models added to the database to make them a more interesting place to fly.
In an effort to improve the cities in FlightGear, Daniel Strawn (aka "Intel-Qube") has modeled a simple "one size fits all" city. Affectionately codenamed "ICEJAW", the city, still in development, features tangible helipads to land on, billboards and snow cover for heavy snow-fall. Follow development in the .
Amateur built collimated display
Contributed by Gene Buckle
A couple of months ago, my friend Wayne and I embarked on the development of a fully functional collimated display system for our DIY flight simulators. A collimated display has been the holy grail of DIY cockpit builders because of the image it generates. Collimated displays are also known as "infinity" displays. It's described that way because the focus point to your eye becomes "infinity" when you're located at the right spot when viewing the mirror. The exact details of how they work are beyond this article, but you can learn more at Wikipedia; see the section marked "Collimated Cross-Cockpit Displays".
Building the display successfully is a pretty big deal. Up until we completed the proof-of-concept project, collimated displays were the sole domain of multi-million dollar flight simulators from the likes of CAE, FlightSafety and others. It was declared too complex and too costly by just about everyone that ever approached the problem. Fortunately for us, all the "experts" were wrong. Not only did we build a fully functional collimating display, but we did it with off the shelf materials and very, very little expenditure. Total material costs for the mirror and projection screen is under $100.
The test mirror was built using an 8 foot diameter spherical shape. The framework itself describes a 60 degree wide by 40 degree high "slice" of that shape. The mirror itself is made of a 48" wide, 2mm thickness of Mylar sheet. In order to properly stretch the mirror, a set of "ears" had to be added to the framework.
This shows the frame prior to the mirror being glued into place:
The framework itself is made from 12mm Baltic Birch and is under 36" tall. The arcs defined by the "inside" ribs show where the spherical section mirror will form. In order to form the mirror, vacuum must be applied. This will stretch the mirror into a spherical section shape.
Attaching the Mylar:
The Mylar is attached to the oval ring by a high-tack spray on adhesive made by 3M.
In order to test the visual properties of the system, we set it up on a platform I'd built some time ago to use with racing games:
The mirror just reflects what is on the screen above it - the test screen in this case is just a bit less than half the radius of the mirror and is 17" tall. In order to start testing as soon as possible, Wayne skinned the screen with kraft paper. It's very wrinkly, but it did an adequate job of being a screen. It's also what we had handy at the time. :)
Here's the complete test configuration:
We used a Fein Turbo III dust collector as our vac source and it pulls the mirror down very well! It's pulling about 5 PSI in order to force the Mylar into the correct shape. Below is a picture of the mirror after being pulled down.
Photographs and video cannot do this justice! It's just incredible to look at!
Here's a pic of Wayne performing the first test flight:
...and thus was born the world's very first amateur built fully collimated display for DIY flight simulators. I can't stress enough how thankful I am for Wayne and his skills. Were it not for his deep understanding of the math involved, this would have never been built.
Here are a few YouTube videos of the system in operation:
- First draw down of the mirror
- My badly narrated test flight
- A walk-around of the setup taken the next day
The project site has more pictures and details on the current state of the project. Work has already begun on a 180 degree version!
The Beechcraft B1900D is a twin-engine turboprop aircraft, designed to be an all-weather regional airliner. It's been used by U.S. military and many other governments. In total, 695 very produced until 2002, and it's one of the default aircrafts in FlightGear and has been available since version 0.9.8.
With a well-implemented cockpit and fully functional GPS, it is one of the better developed aircraft within FlightGear.
We have added a new page to the wiki titled shader requests. This is intended to provide a place for people to post ideas related to new shaders for use in FlightGear. But this is also meant to be place for aspiring shader developers to find interesting ideas for new shaders. In the past, these ideas were often spread over many different places, such as forum or mailing list discussions, and often ended up getting hidden rather quickly. So if you have ideas related to improving existing shaders or creating completely new ones, please add your ideas to the Shader requests page. And if you are aware of any ongoing shader development efforts, please also add those there. Consequently, if you want to get started creating shaders for FlightGear, please make sure to check out this page.
As in previous year's FlightGear organized a booth at the annual FSweekend event in Lelystad, the Netherlands. This year's edition of FSweekend came and went by before we realized it. Although there was initially skepticism as to whether we would be able to top last year's presentation, we did so, by quite a large margin. Main attractions this year were our Thomas-Krenn sponsored server, driving 10 monitors, and our near continuous use of anaglyphic 3D. Although we had considered it to be a neat gadget, running flightgear in 3D analglyphic mode turned out to be a brilliant move. Not only did it draw a lot of people to our booth who wanted to check it out, handing out 3D glasses also worked as an extremely good vehicle to start a conversation with them.
Some assorted highlights include:
- Our Thomas-Krenn AG machine has been running for 8 hours straight, on both days, without a single glitch. At one point Torsten ran out of fuel, because he hadn't realized that FlightGear had been running for so long continuously. The machine ran in three 3D mode most of the time, and the way the screens were set up gave the impression that the were just holes in a wall though which you could see a (virtual) reality known as FlightGear. Very nicely done.
- Very positive comments on the Lockheed Constellation's artwork, particularly because that model is really shining when presented in 3D (probably the best cockpit we have for a 3D demo).
- Also very postive comments on the advance copy of the 737NG (800) that Innis Cunningham sent me. Also in particular due to the fact that the switches on the overhead panel look very real in 3D.
- Booth members Gijs and Jorg trying to outperform each other in terms of their helicopter flying skills (trying to land it on a chimney, once they found the helipad to easy).
- A number of people claiming that our booth was the best one at FSWeekend this year.
- Our aerotowing demo involving all of our major demonstration machines (my machine as the glider, Jorg's computer as the tow plane, and our sponsor's machine as a joint (3D) model viewer.
- Some people just not believing that a single machine could drive 10 monitors and actively checking where all the monitor cables actually went to.
- Another opportunity to meet FlightGear developers (one expected encounter: Geoff McLane, and one totally unexpected meeting with Thorsten Brehm). Thanks guys for showing up!
And last, but not least, everybody involved in getting this year's booth organized. It's been a lot of fun. Some pictures of the event can be found here
FlightGear on youtube
has made an excellent video showcasing the great scenery at LOWI.
And finally ...
A christmas present for you
Do not forget to pay a visit to the Gerbrandytoren in the Netherlands. With a height of 366.8m and 120 lights, it is the world's largest Christmas tree between December 11 and January 6. The tower is located 21 nm South East (HDG 120h) of Amsterdam Airport Schiphol (location on MPMap).
One of the regular thoughts expressed on the FlightGear forum is "I'd like to contribute but I don't know how to program, and I don't have the time". Unfortunately, there is a common mis-conception that contributing requires programming and lots of free time. In fact, there are a huge range of ways to contribute to the project without needing to write code or spending days working on something.
For ideas on starting to contribute to FlightGear, you may want to check out: Volunteer.
Reminder: Google's Summer of Code 2011
We would like to remind all readers that the FlightGear project is planning to participate in GSoC 2011. However, doing that really requires a fair amount of work, planning and organizing. This is not something that can be done by a single person. It really needs a coordinated team effort, or otherwise FlightGear won't be able to apply/participate at all.