Hello [Again!] - New Project in Development: 777 Advanced

[AnglaisEnUzes]

Hi everyone,

I few weeks ago I introduced myself and the "generic" cockpit I was planning. Since then I have had a rather major change of circumstances and have started developing a new project, which at the moment is dubbed "777 Advanced" (although the open source nature of the software is intended to be able to plug any aircraft into it).

At the recommendation of Andy on the simFlight forums, I am posting here. I am hoping that there will be all sorts of answers to help me define the precise "feature list" of this project (although bear in mind that I personally want to stick with Boeing; maybe when the framework is there others will adapt it for other 'pits).

The problems I keep mentally coming across when thinking about simulator 'pits, is that they rely on such a wide variety of interfaces and software, that to have everything properly integrated is very difficult - every piece of hardware from different manufacturers has a different driver and setup procedure, and all the various pieces of software need separately seting up. Also, the problem that every addon comes across is the limitations of the FS SDK.

So, I have decided to develop a "hybrid" system of soft- and hardware which includes:

a) an FS aircraft addon (model/FDE)

b) software simulation system (see below)

c) hardware ("traditional" cockpit building!)

The "software system" mentioned above will run on a second computer and includes several modules:

- system logic (the system logic will be completely external to FS - i.e. not programmed as gauges, but as a separate Windows application). This will overcome the limitations of the SDK, and will allow for things like sounds being triggered from a multichannel sound card.

- glass cockpit main and standby displays: programmed using XAML using the WPF (CLR 3.5).

- Quasi "Class III" Electronic Flight Bag (obviously excluding the video surveillance system, but including Moving Map, Moving Airport Diagrams, Electronic Documentation, Charts etc.).

- Aircraft Configuration Module: configures the aircraft and displays - all the settings will be made here. In addition, differnet ops procedures can be selected for different airlines (the choice of airline can be made in this module).

- Flight Planning / Ops - planning the flight, printing the routes and details etc, setting the aircraft loading and fuel etc etc etc...

- Hardware I/O interface: an interface to simplify the connection of hardware to the system (incidentally, the hardware will connect to the system PC, NOT the FS PC. The system logic will communicate over the network with FS).

Another part of the EFB will be defining new XML schemas for vector enroute charts and electronic documentation.

Now, while a lot of the above has already been achieved for FS, there is currently no completely "integrated" system dedicated to a certain aircraft.

As for the aircraft itself: 777-200LR. Why? Firstly because I adore the 777, and secondly, because it is the only aircraft I have an almost full set of documentation for (Ops manuals, MMEL, Training Manual, AMM etc). Oh, and an ex 777 captain is a very good friend of mine

If anyone has any comments or ideas for "features" that they would like to see in system like this, then please let me know! I am looking for constructive ideas for features. This is a HUGE project, and could well be 2 to 3 years very hard work (as well as a couple of FS iterations down the line), but everything has to have a beginning

Hopefully in a couple of weeks I will have a website with a far more structured description of the project.

As I mentioned before, the complete source code will be released under the GPL. The development environment I am using is Visual C#/C++ 2008. A lot of the system logic will be definied in MatLaB/SimuLink, and the code generated from that.

Thoughts and ideas welcome!!

Cheers,

Martin

[leonpo]

Martin, Hi

Looking forward for your project and for sure I will contribute if you will make it open via GPL

Regards

Leon

[Trevor Hale]

Wow, thats one heck of a goal Martin.

You have successfully stumbled across the hardest part of development. The main reason that people like Project magenta have "issues" is because so many people have so many different setup's it is almost impossible for the software to work exactly the same on everyone's platforms.

This is truly one of the major setbacks in our hobby.

I look forward to seeing progress of your project.

Best regards,

Trevor Hale
Mycockpit - Admin

[AnglaisEnUzes]

Yes I think that is the main sticking point.

Although it may not be the most flexible way of doing things, a completely integrated system is the best way of achieving as fully integrated a cockpit as possible.

[AndyT]

Will you be sticking with Microsoft or will you add the option for X-Plane?
I've actually been thinking about this myself but I tend to shy from such a huge project.
This would be a massive undertaking you realize.

[AnglaisEnUzes]

Personally I will be sticking with MSFS, but the idea is that as much as possible of the system logic and functionality will be external to the simulator application, so to "port" the system to another simulator will only (only being a relative term... ) require the connection to the simulator to be programmed, and the aircraft model and flight dynamics created.

One limitation is that the software systems will only run on the Windows x86 (and possibly x64) platform with CLR 3.5 (the graphics subsystem used will be WPF). I suppose the simulator could theoretically be on another platform, as it communicates over ethernet, but my experience of anything other than Windows is very limited...

I am at the moment considering whether to "modularise" a lot of the hardware - ie, program systems like the CDU/FMC as a completely separate hardware entity (plugged into the network), perhaps using Windows XP Embedded (I shy away from Windows CE), or whether to have everything except the simulation itself on a single computer.

I am aware of the hugeness of the undertaking, but hey, if it comes off it will be very much worthwhile! Bear in mind that once the basic system is programmed, it can be modified far more easily than programming from scratch (I would really like to do the 747-8, which will share commonality with the 777/787, but I am not going down that road until I have accurate systems documentation, which will be some time after the aircraft is in revenue service!).

[AnglaisEnUzes]

Incdentally, something else I am thinking of is programming the EFB for a tablet PC - they have touchcreens and are of a small form factor, and would be ideal for an EFB.

Edit: Having thought about the CDU issue a little more, I have decided NOT to use an embedded OS. I *will* be having it as a separate hardware unit (a single board computer), running XP Pro SP2, but the software will be programmed as a replacement Windows shell (so the Windows GUI never shows). For startup/shutdown and other functions (like updating the FMS database, IP address etc) I will implement another page on the CDU.

M