Open Source Train Sim (OSTS2?)

Suppose Microsoft really has abandoned MSTS2.Anyone interested in doing an Open Source train sim?There seem to be lots of Open Source game engines out there, possibly what's needed is a terrain engine (there is at least one book out there on writing terrain engines in C++). Compared to a more traditional action game, I would think a terrain engine would not be such a demanding problem.After all, BVE, written in Visual Basic, not exactly a high-performance language, is a respectable simulator.

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

RE: Open Source Train Sim (OSTS2?), Bill Hobbs, Feb-13-04, 09:31 PM, (1) RE: Open Source Train Sim (OSTS2?), SeHenley, Feb-16-04, 04:10 PM, (3) RE: Open Source Train Sim (OSTS2?), Silverliner, Mar-03-04, 00:44 AM, (9) RE: Open Source Train Sim (OSTS2?), SeHenley, Mar-03-04, 12:09 PM, (11) RE: Open Source Train Sim (OSTS2?), stationmaster, Feb-15-04, 10:44 PM, (2) RE: Open Source Train Sim (OSTS2?), decapod, Feb-18-04, 12:08 PM, (4) RE: Open Source Train Sim (OSTS2?), lirc201, Feb-21-04, 10:04 AM, (5) RE: Open Source Train Sim (OSTS2?), bap70, Feb-27-04, 07:11 AM, (8) RE: Open Source Train Sim (OSTS2?), Engineer, Feb-22-04, 06:51 PM, (6) RE: Open Source Train Sim (OSTS2?), wdq4587_2, Feb-22-04, 11:05 PM, (7) RE: Open Source Train Sim (OSTS2?), Hobart, Mar-03-04, 09:05 AM, (10) RE: Open Source Train Sim (OSTS2?), lirc201, Mar-05-04, 10:15 PM, (12) RE: Open Source Train Sim (OSTS2?), dziesig, Mar-07-04, 09:04 PM, (13)

Conferences | Topics | Previous Topic | Next Topic

Hey, you're talking about some real work here. I just like to sit here and complain, not do any real work.
Seriously, I have done some work on steam locomotive simulation that I'd contribute if this project ever gets off the ground.

Bill Hobbs

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

It seems to me that there are really 3 train simulators - the activities people - who would like to log on one night week with their favorite group and run a railroad in real-time, the engineers who like to see how many cars they can drag up an impossibly steep hill. The car builders love to build things and show them off. If this project goes anywhere we should first determine what we are going to build - Microsoft build for all 3 groups - commercial decision - A steam simulation is a minor problem - I was assigned to a steam plant simulation for a frigate in the late 70's we use the steam tables - some filters equations - this is much faster then obscure equations and much more accurate in a simulate use for the human bandwidth. It is also easy to use - adjust a curve in a table is simple - results are instant.

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

I can think of at least one more - the route builders, who want to build the Trans-Siberian Railroad in its entirety, and then duplicate the American Interurbans, all of them, circa 1910. This is my perspective at least. However, your message brings up a very valid point, in that a train simulator is a large problem which can be compartmentalized. I recall reading something on Kuju's site a while ago which suggested that MSTS1 was a 10 man year effort. (This number is not particularily enlightening considering Kuju invariably had plenty of code which could be reused, plus 10 man years probably would not include the supplied routes?)So the question is how to break up the overall train sim problem and hand pieces of it over to interested developers. Also, how to manage the development.Personally, I rather like the Microsoft OCX / ActiveX / Automation approach, as this is an answer to how to combine components created by individual developers in different languages. The challenge would be to duplicate that on Linux.

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

COM (OCX ActiveX etc.) may be slow? But there is an article in Game Programming Gems 3, "An Object-Composition Game Framework", by Scott Patterson it uses the Object Composition Design Pattern. You add a task with a fixed interface to be executed every frame or executed one time. It has the advantage that the task modifies a system object and the component task added can makeup any application - game, route editor etc. The demo shows Direct X and OpenGL. The problem with multi-platform system is that games also have sound, input devices and some network system. It would be good to have voice communication from dispatcher to cab. If you use an existing game engine you lose one of the game's advantages in collision detection - the train usually follows the track. Something could be done for the route builders - first terrain can only be seen from the track, and the track profile is known - the terrain should be modeled around the track - not track lay on the terrain. I also think that the terrain should be a model like the engines, cars and buildings - there is only one path - LOD can be preprocessed and that may leave time for animating turn tables and 16 drive wheels.

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Along those lines there was the Black 5 project. A freeware project, but I do not think it was open source. The web site is still there, but work on the project seems to be suspended.http://www.black5.co.uk/index.htmRon
http://www.railwaystation.com

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

If you would like to import MSTS shapes into the project, I have just put up a beta version of my shape viewer OCX which can be used in Visual Basic or C++ projects to load and display.It's not exactly open source, but it should also work with other game engines if you ignore the viewport part.It's a bit primitive at the moment, but you could probably write a reasonable model railroad sim with it.http://www31.brinkster.com/decapod , under shape viewer downloads.
Limited bandwidth so it probably won't take too many hits - it resets at midnight EST.

Paul
UK GWR Highworth Branch
http://www.highworth.freeuk.com

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

There was a similar situation that happened in the Flight Simulator community. A group of interested folks broke off and started an open source version called Flight Gear. Their website is http://www.flightgear.org. Go ahead and email Curtis L. Olson and ask about what it takes to pull off a project like this.Also, you can use the free services of Source Forge to do your version control, bug tracking, and project information. http://www.sourceforge.netThere is also hundreds of graphics engines available on the net and some are very very good. This way you would not have to re-invent the wheel.You may want to keep in mind an additional object of a multi-platform train simulator that runs on Linux as well.Let me know if you need help in writing up requirements documentation for such a project. I would be willing to help with the project management aspect.

Brian LaVelle
Cincinnati Terminal Route
http://belcommunications.net/msts

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Hi!I think it's a quite complicated project, but as programmer and train fan, I would join for an OSTS development.
I've also some ideas, such as "roles" like traffic manager o a simple driver, generate random traffic based on pre-selected consists, etc,etc...

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Sounds good if MS abandons MSTS, but seems a waste of time if they don't. I'd like to see a Linux Train-Sim myself, it would probably be faster. I've all but abandoned MS for everything else, to much excess baggage. I don't mean to be bashing MS but they are becoming more irrelevant everyday. They seem to have abandon the PC strategy in favor of the net strategy. Think they bought into all that Al-Bore BS. The net is to slow for any practical serious usage.

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

It seems like a good idea, but it's impossible!The simple reason is there are no enough programmer among MSTS fan. All of you can see, MSTS fan are less than 1/10 of MSFS fan, and may be less than 1/50 of all Flight Simulator (include combat) fan. So you can not hope such things like in Flacon 4, Mig Ally, BoB or flightgear and other projects. And these new freeware projects like flightgear still can not attract enough users as commercial software.Another reason is we can assume Microsoft will not open MSTS code even they cancel MSTS2. If they can, there are some possibility to improve it and repair bugs and flaws. But a second version still near impossible.

------------
Formerly called "wdq4587" until forum software or someone let me not Activate!

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Nothing is impossible, Deal me in

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Alright. I have started the discussion here with an initial proposal. More documentation plus a development committee will need to be formed in order to start the direction of the project.http://www.train-sim.com/dcforum/DCForumID3/22512.html

Brian LaVelle
Cincinnati Terminal Route
http://belcommunications.net/msts

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Count me in! When M$ "abandoned" MSTS2, I started investigating what it would take to do a train sim open source. I have the time (retired), some of the background (Electrical Engineer; been programming since 1961), the tools (lotsa Linux and Windows boxen), and the desire (Model Railroader since 1952, train fan for as long as I can remember)!My primary interest is steam locomotives - especially how to get appropriate sounds synchronized with the drivers, but I would be willing to contribute in any way needed.If one of the earlier posts is correct and MSTS took 10 person-years, we could have a beta in 6 months or less. My guess is that the supposition that it was built on an existing code-base is accurate, but there are many good open-source libraries that we can use as our foundation so we wouldn't be starting from scratch either.

Paranoid Don - It IS as bad as you think, and they ARE out to get you.

Alert | IP

[ Home | MSTS Forum | MSTS FAQ | File Library ] Copyright © 2004 by Nels Anderson. All Rights Reserved.