Open Source 3D Modelling Tool for FSX/P3D

Hi Guys,

As some of you may know, I am the former Principal Engineer for Lockheed Martin Prepar3D. What you may not know is that while at another company that I ran, we developed a 3D modeling tool for the creation of real-time simulation objects. i.e. low polygon count OpenFlight format mainly. When we started talking with the Microsoft ACES team during the development of ESP (my company was one of the original launch partners), we added in FSX/ESP model export support. We originally sold the product for $995 per license and initially had some very good success with it with large defence organizations around the world. Of course it can also support P3D models.

Long story short, when I left the original company, I retained the IP for the modelling tool and it has languished on a shelf doing nothing but gather dust. So, I would like to make the product open source and provide a tool that can be used by the flight simulator community. Here is a screen shot to give you a flavour:



There is a little bit of work required to re-brand it and get rid of the current license code. It also needs to go into SourceForge or GitHub and a developer or developers need to own it. A current limitation is that it does not currently support the advanced textures of FSX, such as chrome, but it isn't that critical. This would need to be added at some stage, but it is still a full modeling tool. I have rebrand documents and assets and a list of tasks that need to be completed. I also have some modeling tutorials, manual and so on. The application can also import and export 3ds, ac, dae, flt, dxf and obj files. MDL is export only.

Now Open Source doesn't mean that there isn't a cost. I am currently involved with MakerPlane (www.makerplane.org), an Open Source Aviation organization and it costs money to run. So what I was thinking putting together an Indiegogo campaign and getting contributions to allow us to launch this so that I can pay a lead developer to manage the effort and pay for any extras that might be needed. Any funds left over I want to donate back to MakerPlane..... don't worry there are seriously cool things that we are working on that can be used in the flight simulation community as well.

So, this is a dip of the toe in the water to see if it is warm enough to take the plunge, or if there are toe-eating eels. ;) Any thoughts on this? Do you think that there are enough people in the community that might be interested?

Regards,
John
 
Last edited:
Sounds interesting and could be exciting, but you need a selling proposition, even for freeware. Why did people pay $995 for it? What can it do that FSDS, Gmax, Blender, Max, Sketchup, ..., cannot do, or cannot do as cheaply, easily or quickly? Why would it be worth one's while to learn it? Would it do texture baking? Would it export large poly-count models (seems not - what are the limitations?)?
 
Last edited:
Hmm, does it work with scenery objects as well?
Sounds interesting and could be exciting, but you need a selling proposition, even for freeware. Why did people pay $995 for it? What can it do that FSDS, Gmax, Blender, Max, Sketchup, ..., cannot do, or cannot do as cheaply, easily or quickly? Why would it be worth one's while to learn it? Would it do texture baking? Would it export large poly-count models (seems not - what are the limitations?)?
Right off the top of my head:

  • It can import and export high poly count models.
  • Toolsets for DOF articulation
  • Flip-book animation
  • Level of Detail Control
  • Conditional Switching
  • Light Point Creation and Animation
  • Shader Integration
  • Boolean Operations
  • OpenFlight support
  • Easier to use than Blender
  • Easy export wizard for MDL models
  • UV texture mapping in 2D and 3D
  • Materials palette

I realize that there are free tools out there such as Blender and Sketchup, however this tool is specifically designed and built as a 3D object modeling tool for simulation and game objects. It grew up from the OpenFlight world, which is a standard model format for military simulators where performance is key. The $995 price point was the cheapest for an OpenFlight modeling tool on the market at the time. Even plug-ins for free tools did not fully support the standard.

This tool will not appeal to everyone, but many new to the modelling world should find it relatively easy to use with lots of depth for more advanced techniques such as mesh splitting, warping and so on. The Export wizard built into the product steps you through the process of creating a model for FSX. Various dialog boxes in the wizard prompt you to choose an external model, an internal model, a flight model, thumbnail image and so on and automatically populate a SimObject subdirectory. I could be wrong, but I believe no other tool does that. So it makes it really easy for a novice to create a model, particularly for FSX. For FSX/ESP/P3D specific tasks, I should post up some images of the dialog boxes as well. Built in functions include:

  • GUID generator
  • Visibility tag assignment
  • Platform assignment to mesh/ object (asphalt, grass etc)
  • Effect assignment to mesh/ object(smoke, fire etc)
  • FSX Animation support
 
Last edited:

Heretic

Resource contributor
WOAH! :eek:

You might be able to draw more people in if the tool supports more than one sim.
I was thinking in the direction of FlightGear...

The financial kickstarting idea sounds good if it really is supposed to be open source in the end.
 
WOAH! :eek:

You might be able to draw more people in if the tool supports more than one sim.
I was thinking in the direction of FlightGear...

The financial kickstarting idea sounds good if it really is supposed to be open source in the end.
Flightgear can use AC3D, which is already a supported import/export format in this software.
 

n4gix

Resource contributor
John, you wrote in your original post the "Open Source doesn't mean that there isn't a cost." Later you mentioned that the commercial license was $995/seat.

Now I'm certain that everyone reading will understand that while there is a cost to create and maintain an Open Source project, what isn't clear is what cost there would be for the end users of such a project.

Presumably anyone investing via Crowd Funding would receive some benefit from doing so, but what about J.Q. Public?

Having already seen the demo version of the program several years ago, I already know for sure and certain that it is far more powerful than either FSDS or GMax, and without question less powerful than Max. It does have the virtue of supporting a much wider range of modeling types than any of them, which is a great plus, and being Open Source it would allow for continuous enhancement via plugins and so forth.

There is an opportunity here to position such a program somewhere inbetween "Free" and "Horribly Expensive," the real question is where in that range could anyone reasonsably expect it to land? :twocents:
 
Last edited:
John, you wrote in your original post the "Open Source doesn't mean that there isn't a cost." Later you mentioned that the commercial license was $995/seat.

Now I'm certain that everyone reading will understand that while there is a cost to create and maintain an Open Source project, what isn't clear is what cost there would be for the end users of such a project.

Presumably anyone investing via Crowd Funding would receive some benefit from doing so, but what about J.Q. Public?

Having already seen the demo version of the program several years ago, I already know for sure and certain that it is far more powerful than either FSDS or GMax, and without question less powerful than Max. It does have the virtue of supporting a much wider range of modeling types than any of them, which is a great plus, and being Open Source it would allow for continuous enhancement via plugins and so forth.

There is an opportunity here to position such a program somewhere inbetween "Free" and "Horribly Expensive," the real question is where in that range could anyone reasonsably expect it to land? :twocents:
Hi Bill,

These are great questions. I am certainly open to suggestions and interested in any feedback. I have no agenda or opinion apart from looking at ways to get it out there and alive. It was my thought that there would be no cost at all to the end user and make it completely free and open. I just wanted to illustrate with the license cost that this is a high-end tool for a specific purpose, so there is value there. I want this to be sustained by a part-time paid developer (that isn't me -because I am not clever enough!) for at least the first year to get it stabilized within an open source environment and get the initial feature requests sorted. The reason that so many open source projects fail is lack of care and feeding despite all the best intentions in the world starting out and everyone wanting it. I have found that there is no motivation to get things done when people really need it.

I haven't given rewards too much thought as I am still at the early stages of asking you guys if this is in general a good idea or not. Certainly interested in knowing if folks would contribute a few dollars to get it off the ground with the end-state in mind, or if there would be a WIFM component that would need to be more concrete, like a t-shirt or mug etc? Would people be more interested in say a $20 software product and a crowdfunding campaign gave licenses of the software in return for $10 contributions? Thoughts?

John
 

arno

Administrator
Staff member
FSDevConf team
Resource contributor
Hi John,

What features would this tool offer that gmax or a tool like FSDS does not have? I mean if it does not have unique features people might not want to switch so quickly.

Also does it depend on xtomdl to export the mdl or not? If not some of the limitations of xtomdl might be overcome.
 
Hi John,

What features would this tool offer that gmax or a tool like FSDS does not have? I mean if it does not have unique features people might not want to switch so quickly.

Also does it depend on xtomdl to export the mdl or not? If not some of the limitations of xtomdl might be overcome.

Let me turn this around, I am not familiar too much with FSDS or gmax. I have used them in the past, but not that much. What are the perceived or real limitations of these packages? What are the features that people would like to see or wish they had? I know that gmax is no longer supported, so is that important or not? There are a subset of people that will want to use this for P3D development in a simulation environment, so importing OpenFlight and exporting mdl for use in P3D is a real capability and unique, but may not matter to the majority. Do people want another open source free modeling tool specific to FSX and P3D or are they happy to stick with the tools they have? If that is the case, then that is great, no problem. What we might need to do is get a couple of people familiar with FSDS and gmax to try this one out and see if it will do the job. Bill has tried a very early version of the software (pre-version 1.0 I believe) and this was developed up to v2.5 before taking it off the market.

Yes, it uses xtomdl. Any reverse engineering of the mdl format was not going to be condoned by Microsoft at all and yes I asked. We had an MDL import function working and we had our hands slapped. What are the current limitations of xtomdl that you would like to see fixed? If there are real issues, I can send a note to Lockheed to see if we can get these onto the rather large list of things to get done.....

From my own experience, I have not had too many issues with xtomdl, although it can be finicky sometimes with not too much feedback in the UI when things go mushroom shaped.

John
 
Last edited:

arno

Administrator
Staff member
FSDevConf team
Resource contributor
Hi John,

For some reason I missed some earlier posts, so part of my question was answered already in the discussion above.

Speaking from the scenery point of view, if you could mix the fun and ease of use of SketchUp, with integrated support of FSX features like animations, FSX material settings, etc, that would be a big plus. Plus you would not need to use an additional tool in between to go to FS (although I like to make that intermediate tool :)).

Another important point is always how easy is it to model things. In SketchUp that is quite easy for example, but not all tools make manipulating the geometry very easy. It should not take people too long to learn it I think.

And ideally it would make it easy for people to model efficient for FSX. So a good texture mapping editor, so that you can easily work with one texture sheet for the model (in SketchUp that can be a PITA for example).

About XtoMDL, I know there are people who have issues with exporting complex models. And the latest P3D version seems to have issues as well for complex models.

But interesting idea to make such a tool open source. I think I would support it :).
 

n4gix

Resource contributor
As a point of order Arno, xtomdl.exe is a compiler, not an exporter...

As long as the input .x and .xanim files are well-formed and licit, xtomdl.exe never gives a problem of which I'm aware.

On the other hand, the Prepar3D export .dle module for Max2012 does have some serious issues that need to be addressed.

Oh, to tie this post in with the thread, were you just lucky not to get your "hands slapped" with MCX's .mdl import ability Arno? :duck:

Of course, I can see where MS/ACES might care about such in a commercial program, but I've noticed that they always turned pretty much a blind eye at any freeware or open sourced non-commercial programs.
 
Last edited:

rhumbaflappy

Moderator
Staff member
Resource contributor
Hi Bill.

Microsoft kindly turned a blind eye to many of our "explorations" into their coding. If they had not, there would be no scenery developers at all, and the number of hobbyists would be much smaller than it is now. It was actually a good business decision. But Microsoft has gone in another direction, and different people are making the decisions.

Dick
 

arno

Administrator
Staff member
FSDevConf team
Resource contributor
Hi Bill,

As a point of order Arno, xtomdl.exe is a compiler, not an exporter...

As long as the input .x and .xanim files are well-formed and licit, xtomdl.exe never gives a problem of which I'm aware.

On the other hand, the Prepar3D export .dle module for Max2012 does have some serious issues that need to be addressed.

Oh, to tie this post in with the thread, were you just lucky not to get your "hands slapped" with MCX's .mdl import ability Arno? :duck:

Of course, I can see where MS/ACES might care about such in a commercial program, but I've noticed that they always turned pretty much a blind eye at any freeware or open sourced non-commercial programs.
True, the exporter problems are not related to xtomdl. I only believe some people wanted a 64 bit version of xtomdl because they run out of memory. But with modelconverterx I haven't had much trouble with xtomdl either, as long as I debug my x files correctly :)

Direct mdl export would be interesting if fs2004 was also supported, because then we directly export with features that makemdl does not support and require tweaking.
 
3D Modeller

Oh, and I did not mention that we also figured out a way to have damage and destruction modelling on FSX models as well within the current framework of MDL models. This is something that can be done within this modeling tool, although it is not documented anywhere as it was part of our secret sauce. :D
 
Oh, yes, as arno mentioned - fs2004 mdl export possibility would be nice, with or without ASM intermediate (or integrated asmtweak ability, but I think that would be beyond your scope - maybe for someone talented to add later)
 
Oh, yes, as arno mentioned - fs2004 mdl export possibility would be nice, with or without ASM intermediate (or integrated asmtweak ability, but I think that would be beyond your scope - maybe for someone talented to add later)
Being open, the code would be available for someone to add that feature in if required.
 
Top