>> I get to see a seamless >> and plausible match between sky and terrain from ground level to low >> Earth orbit, at all times of the day and under any weather condition. >> Scenery and models are now rendered correctly with the sky at all locations >> and >> all times. > > Really? Look at the trees- Does they blend correct here? > http://www.hoerbird.net/Treeblend.jpg
Heiko, you really walked into this... I wrote 'scenery and models', I did not write 'trees'. I explained earlier: quote myself: " http://www.hoerbird.net/fgfs-screen-764.jpg illustrates that the random vegetation shader isn't converted - trees are fogged too weakly as compared with the rest of the scene. " Which of these two statements did you find hard to understand? > We have shaders. We decided once ago to haven them. > And when a user is also able to have them, as his computer allows them > to have, he wants to have them all of course. And what's the relevance of this? I want plenty of things... Someone with GLSL knowledge to explain things to me so that I can get rid of trial and error based development for starters. Someone who learned digital image processing to help me with cloud texture extraction. Someone who flies his own plane to take GPL compatible semi-ortho-images to create textures and to take aerial images of some rare cloud types. I want people to pick up some of the space capabilities of Flightgear, create orbiting AI model code and some more spacecraft so that we could fly to ISS. Do I get any of these things? No, I don't, and I'm guessing there is a reason every no. I know pretty well what what users want - you can get a good impression here for instance: http://www.flightgear.org/forums/viewtopic.php?f=68&t=16800 Everything should be as realistic as possible, but under no circumstances should it lead to a loss of framerate, and every idea needs to be implemented yesterday. So, yes, users may want to use all shaders at the same time, but as I (and Emilian) have said a few times by now, it is not simple to do and takes development time, and time is a rare commodity. So people may wish for it all they want, but that doesn't give me the time to do it, and as long as my time is the limiting factor, it gets done when I have time, end of story. You can pick it up and do it if you want it done faster. Besides, I'll take a 50 Euro bet that once all shaders are working together, we get frequent complaints that the framerates are very low from users which max out all quality sliders. So then the users simply want different things. Anyone wants to take that bet? > The user doesn't want to check everytime which of the shaders is > compatible with others. That's one BIG secret behind the success of some > flightsimulators and even a requirement there. If a shader can be > deselected, than just for perfomance reasons, but not due to being not > compatible with other shaders. Well, he doesn't have to check anything - incompatible shaders are de-selected automatically. In essence, your suggestion is to 'dumb down' flightgear and not to offer any features which are not completely integrated with everything else because the user may not be willing to accept that not everything works with everything else. Sounds like a bad idea to me. We are not a commercial enterprise, and we play by different rules. Trying to adopt commercial development strategies will only make us fail. > And that's exactly the problem here. We had no consistently way in > developing and adding shaders. I beg to differ - you are not in shader development. I am, and I have talked on and off-list with Emilian, Fred and Vivian who also are (sometimes a lot), and I would say that we have a pretty clear picture of what the issues are and a general agreement what the development goals are (we also have some disagreements over details how to solve particular problems). Nobody disputes that eventually aircraft shaders should work with Rembrandt and atmospheric scattering. > 1.)Zan's shader was known that it gave artifacts on the horizon. But it > wasn't plain wrong. Show me the screenshot from 2.6 with visibility set to 3000 m, skydome shader on and repeat your claim (I'll also take the sunrise picture from cruise altitude). > It was included in the hope that this issue can be solved. > Now you introduced a shader which doesn't have problems with the > horizon, but other shaders doesn't work. If you don't like the solution, then show me a better one. Anyone who understands shader development agrees on how to solve the horizon problem. There is no known solution which does not require to re-write all other shaders to some degree, which means that other shaders don't work till this is done. > It is not really an improvement- unless you just look at the horizon and > nothing else. Common sentiment seems to be that atmospheric scattering does quite spectacular things, but I take note of your different opinion. Personally, I like both the volumetric fog and the Mie scattering in clouds very much. Also, please note that in a standard flight position you're looking at the horizon all the time whereas you're hardly seeing any aircraft exterior. > 3.)And somehow myself as Aircraft Author feel ignored and mistreated > when you say you develope a shader just for looking out the window and > not at the aircraft. Why the hell spending a lot of times in trying to > create an accurate exterior model? A simple box should be enough. Yes, a box with a good 3d cockpit and realistic FDM would do just fine for me. I honestly spend 99% of my time looking out from the cockpit view, and my enjoyment would not be reduced much if the remaining 1% would show me a box. Look, this is silly. You can't seriously take offence because someone doesn't enjoy the same things as you do. I would hope that you spend a lot of time to create a good 3d model because you enjoy it and because there are others who enjoy it. Why should you feel mistreated because there are users who don't? Should I feel offended because not every single user uses Advanced Weather? Some don't like it for various reasons and could do without it, and that's cool with me. > The big problem behind is, as I already said above: the > non-compatibility between the shaders. This is *not* a compatibility problem. A shader can not, even in principle, be compatible with the default scheme and atmospheric light scattering simultaneously, the same way as you cannot write a sentence which means the same thing in English and Japanese. Here you really need to do two different sets of computations. A shader can be designed in a way that the computations which differ between these two schemes are factored out and the computations which are common are in the same file. Emilian favours this design, as it is maintenance friendly, my own ideas are more along the lines that each shader should be optimized, because that may give another 20-30% performance we'll see where this ends. > As I heard, some people are planning to create an own Viewer, outside > but compatible with FG - exactly due to this reason. Let's get this straight - some people are creating an own viewer outside Flightgear because I haven't converted shaders fast enough to the atmospheric scattering scheme and you can't have your favourite shader along with the scheme? I didn't know I was that influential... And creating a 4th scheme is supposed to make anything better? Really, please let's get this discussion grounded. There's a problem, you (and others) want a quick and simple solution, but there is none, but it takes a while to understand the problem enough to appreciate the problems. Please read again the explanations that have been given so far, perhaps you will then come to think in a different way. Cheers, * Thorsten ------------------------------------------------------------------------------ Live Security Virtual Conference Exclusive live event will cover all the ways today's security and threat landscape has changed and how IT managers can respond. Discussions will include endpoint security, mobile security and the latest in malware threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ _______________________________________________ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
