A new Flight Simulator

[IAHM-COL]

No idea what you mean by modular FDM.... Every airplane is different, and you can't just plug in wings and such,

Yes. I heard this from Richard as well. I don't disagree.
But also I dont see why/how.

Per example:

you can't just plug in wings and such,

Why not? I mean every aircraft has wings, and the general properties of those wings seem to me kind of shared. Now how about, importing the wing, and then modify it to its specific differences? Very simliarly as I've shown you with the 727 Engine in the python example above. I had first created Engine, then modify it to have reverser (see FS_PropertyTree). But that was not enough, as your nasal shown, you needed to overwrite the defaults and so something else, so I took the Engine_reverser, thru inheritance and simply coded what was different. And ended up with a new version of engine that now matched specs.

Why do you think FlightGear is so bloated?

I'll give you an example. How do you (or at least I) implement liveries in an aircraft? Usually we have a very brief nasal code that implements such livery system here:
https://github.com/FGMEMBERS/737-800YV/ ... veries.nas
But then a new aircraft, now here
https://github.com/FGMEMBERS/IDG-A33X/b ... ery-rr.nas
and here
https://github.com/FGMEMBERS/IDG-A33X/b ... ery-pw.nas
and here
https://github.com/FGMEMBERS/IDG-A33X/b ... ry-frr.nas
or in other aircraft now the code can appear in any other haphazard location but again, basic same code just copied again
https://github.com/FGMEMBERS/c182s/blob ... /c182s.nas

Expand that to maybe 500 of the 830 aircrafts that already have livery implanted.

But if you want to implemented in another aircraft, there it goes again. Copy more times.

That' I'd dare to call bloating.

Instead of contributing and helping the FG project

Agreed. This would be preferable. Do you dare to propose I could bring some idea to that team?

PS
python is free and opensourced. No sale necessary.

[N3266G]

You really going to create a new sim? PM me when you got a stable version and I'll test it out.

[Octal450]

Hi Israel,
This is also a bit of tiny worhtless code rather than more advanced code, it could easily be compressed down into one line which loads the liverys based on the which -set is loading, but again, back when I added that, I didn't know any better, and couldn't be arsed to rewrite it later down the line. You don't need to switch programming languages and break compatibility to sort it.

If all aircraft used the same livery system, the simulator could do it without needing the line, but that breaks modularity and customiztability.

JSBsim also just has a wing.... you don't tell it that, it's not like Yasim. You specify things like chord dihedral etc... and you can then make your own aerodynamics from scratch or modify a aeromatic based on data, or such. It's already very generic and modular. (I'm sure jwocky would agree with me here, JSBsim doesn't need modules...)

Further, maybe you can share things like system files, but then they wouldn't be realistic, as they wouldn't be customized for the aircraft.

Also, the aircraft aren't part of the sim, they are an addon for it. Further, it's ONE LINE who is executed ONCE at the very beginning and takes a TINY amount of space -- doesn't seem bloated to me...

If you are judging nasal based on these simply lines -- that is waste. You think just changing the format is worth breaking the tons of things that rely on nasal (including Canvas) that you will need to learn about, and completely rewrite from scratch to work?

Hi N3266G,
I doubt it.

Kind Regards,
Josh

RE: PS, joke

[bomber]

Josh I don't hate you..... But I do find you insincere.

[Octal450]

But I do find you insincere.

How so? Why would I write things if I did not believe them? I am giving my 100% honest opinion here, and I don't see how me having a different opinion makes me insincere?

Kind Regards,
Josh

[IAHM-COL]

Hi Josh
Yes. I am of the opinion that it's better to live without that canvas thingy as well. So no concern if canvas goes when nasal goes.

Per example, you could update svg patterning as well which can do the canvas work in a more efficient and streamlined manner. I think,

In a way, I think these FG solutions are ineficient niche solutions (and are not fun to work with). So let me say that beyond being a bloated software, it seems to me that on every step of the way the development core group have discarded the very essentials of good programming

1. Keep it simple
2. Dont repeat yourself.
3. reuse code.
4. create namespaces

I don't say I hate FG. But when I think how do you/I/anyone else make a new aircraft or an advance aircraft, I feel every step is following the same old wrong choice-set.

[IAHM-COL]

You really going to create a new sim? PM me when you got a stable version and I'll test it out.

I would like to get a team of people interested to get it started. Trying to gain some ideas floating here, and see who could be like: we can do this. And jump in a boat like that. I am not saying that I am creating it myself, nor that this is jumping from conception to reality overnight.

In other words, I think having an alternative FS would be a nice thing to have. And it will be great to thing it could happen. And if it happens, how would it be nice it works? how would it be nice that new content gets added? etc.

[Octal450]

@Israel,
Sorry that I spent over a year to make what is one of the most realistic airliners for FlightGear (was told to me by others, not told by myself), sorry that I spend so much time doing it, I must be doing it all the wrong way.

If you think of getting rid of Canvas because it isn't important, you completely loose all credibility. How else can you make realistic screens without a drawing API? For example, NAV displays and such to draw paths and items which cannot be done simply need it. To make any realistic looking displays, you need an efficient and customizable 2D drawing API. All major sims have one, here is the old PFD in my Airbus, compared to the Canvas one, which uses an SVG I drew, animated by nasal. I can modify this in ways you could never do without a drawing API.

Old 3D/XML PFD:


New Canvas PFD:


The Canvas PFD is looking close to the real one, wheras the 3D/XML one looks terrible.

If you can't see what I'm saying now, then you don't really want realism do you... it means you care more about programming language than realism...

Kind Regards,
Josh

PS: It may not be legal to import pythons, some species of which are endangered, as pets into many countries.

[IAHM-COL]

you misinterpret me.
I am saying: Canvas is the hard way of achieving that. (maybe the only way that FG would allow, but not the logical programming way of doing that.
In other words, yes, you need a drawing API, but canvas/nasal is not the only solution to that problem. My philosophical proposal here is: actually canvas/nasal is a wrong choice of a drawing API.

Now, to go pythonic again,
http://pythonhosted.org/svgwrite/

If python could be a functional way of coding the aircraft you can find a couple of functionable modules, let aside you could modify them to suit you if you wish. Above just a good example

[Octal450]

Now I am convinced you are a Python salesman.

Canvas is actually pretty easy. I don't know why you are so pissed off about Nasal, but hey. What I do know, I'm just an aircraft developed who did everything wrong when programming my aircraft

Kind Regards,
Josh