On samedi 04 octobre 2008, Stuart Buchanan wrote: > --- On Fri, 3/10/08, Martin Spott wrote: > > gerard robin wrote: > > > And others JSBSim updates within PRE_OSG_PLIB_20061029 > > > > > > Updating that branch , is not going backward :) > > > > Maintaining this branch for such a long time, including the > > pile of work that had been spent to get the 1.0 release out, has > > been the biggest distraction and waste of developer resources since > > I'm monitoring this project (which counts approx. 9 years now), > > Gerard: - What I am about to say is not aimed at you personally - it is a > general comment. I apologize for not speaking enough French that I can > provide an accurate translation. > > I've avoided commenting on this thread, and the many similar threads that > have appeared over the past year because I don't feel they are productive, > or to the benefit of the project. Nevertheless, as one of the people who > argued for the initial move to OSG and a minor contributor of OSG code > (random objects, trees) I feel it is necessary to speak. > > I know that everyone on this list has the best interests of the FlightGear > project at heart, but the almost constant harking back to PLIB, and sniping > on the progress of OSG have had a corrosive influence on the project and > have dissuaded contributions. > > To give some background to the following comments: I am certainly not a > good coder - just ask Melchior about some of my Nasal code ;). Prior to my > OSG contributions, I had done some small Nasal enhancements, a couple of > (rather rough) aircraft, and some weather interpolation code (for which I > required significant help - thanks once again). > > However, it only took me around 3 man-days of effort to implement random > objects in OSG, having done no OSG coding, and little C++ previously. The > shader-based trees took about another 5 man-days. I got a _lot_ of help > from Tim Moore, but the bottom line is this: > > _In a very short time, an average coder was able to implement a useful > enhancement to FG using OSG_ > > Coding in OSG is much easier than in PLIB, and means that anyone with some > C++ knowledge can do something useful. You don't need to be an OpenGL guru. > > There is absolutely no way that I could have contributed equivalent > graphics code to FG using PLIB. For example, Harald's PLIB 3D cloud code > implemented Impostors (effectively painting a texture at run-time) from > scratch, while OSG provides these for free. The only difference is that his > code worked, while the OSG code crashed ;) > > For those that are not familiar with the graphics internals (which may > account for most of the people complaing about OSG progress!), PLIB > requires writing very low level OpenGL code to achieve anything useful.To > be able to make a significant graphics contribution to OSG FG requires an > order of magnitude less skill than PLIB. To use an FG metaphor - imagine > writing all your Nasal scripts in C. > > If we had remained using PLIB, as well as the supportability issues (PLIB > being effectively dead) the pool of graphics coders would have been even > more restrictive than it is today. > > With less sniping from the side-lines and more encouragement, I am sure > that more contributors would have appeared, and we would have already > released v2.0, with 3D clouds, shadows, heat-haze and more animations than > you can shake a (OSG-shader) stick at. On a personal level, I have been > dissuaded by some of the comments and expectations of people for 3D clouds > on- and off- list. > > If I may provide a second bottom line: If people really want to maintain > FlightGear on PLIB they are welcome to - fork the project. Otherwise, > contribute what you can, and realize that promoting a positive atmosphere > on this list is the easiest way to encouraging FlightGear's continual > development, especially in the field of graphics. > > Everyone is well aware that OSG-FG is behind PLIB-FG in eye-candy. > Criticizing progress is inappropriate if you are not prepared to put in the > work to fix it. Excuses such as "I don't know C, otherwise I'd..." are just > that - excuses. Leaning C++ and OSG is just a step up from Nasal ;) > > -Stuart > > PS: Despite all of this, I'm going to be looking again at 3D clouds using > Shaders instead of Impostors. > > PPS: The above is not meant to be a personal attack on anyone. I realize > that from an aircraft-developers perspective we don't appear to have made > much progress, and the differences between PLIB and OSG are frustrating.
Hello, Stuart Thanks for your comment, and the work you are doing on FG. I hope that nobody here has understood that i was criticizing progress. I know that everybody who love FG, here, give the best of themself. The work which has been done on FG is wonderfull. I like both versions FG 1.0.0 and FG CVS. I know that the choice of OSG is the best choice for the future. Looking at the existing simulators which are using it, ( i have just built CSP ) we can conclude that FG with OSG is a big step, it is very promising. As you said, PLIB is dead and limited, everything which could be done with it has been done. Probably my analysis of the forecasted schedule is too much provocative. Estimations are only estimations :( If i have vexed somebody , i apologize. Cheers -- Gérard http://pagesperso-orange.fr/GRTux/ J'ai décidé d'être heureux parce que c'est bon pour la santé. Voltaire ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ _______________________________________________ Flightgear-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/flightgear-devel
