Re: [Flightgear-devel] keepingup with OSG(was Who turned the sun off?)
-- From: ThorstenB bre...@gmail.com Sent: Saturday, May 07, 2011 10:40 AM To: FlightGear developers discussions flightgear-devel@lists.sourceforge.net Subject: Re: [Flightgear-devel] Who turned the sun off? On 07.05.2011 10:20, Alan Teeder wrote: Ever since the atmospheric scatter patches went in I have had black sky, instead of blue. If I turn on material shaders, then there is a small area of blue close to the horizon, but that is the best I an achieve. The attached is from my TSR2 cockpit with shaders off. Which OSG revision? Saw the same effect when I experimented with latest OSG trunk last week. Sky turned black when attempting to render fog. Since there were also other issues (text display and clouds not working) I reverted back to OSG 2.9.9. Not seeing any issues now. cheers, Thorsten Which begs some questions. How is FG going to progress if it is stuck in a time warp with an old version of its key graphics library? Are the FG code developers liasing with their OSG counterparts, especially in the area of backwards compatibility? Alan -- Achieve unprecedented app performance and reliability What every C/C++ and Fortran developer should know. Learn how Intel has extended the reach of its next-generation tools to help boost performance applications - inlcuding clusters. http://p.sf.net/sfu/intel-dev2devmay ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
Re: [Flightgear-devel] RFC: fgdata merge request 76: Improvedairport Textures
Ryan M -Original Message- From: Ryan M [mailto:tpbspamm...@gmail.com] Sent: 10 May 2011 03:46 To: vivian.mea...@lineone.net; FlightGear developers discussions Subject: Re: [Flightgear-devel] RFC: fgdata merge request 76: Improvedairport Textures On Sun, 2011-05-08 at 23:46 +0100, Vivian Meazza wrote: I'm afraid that the improved textures really aren't an improvement. Not only do they not work for taxiways, we have also lost the chevrons at the threshold that were quite recently added: Really? I certainly did *not* touch the stopway textures- I just copied the high-resolution stopway textures, scaled them down, and put them into the low-resolution folder. If they were removed, I'm not sure what happened. I didn't do that in my commit. Stopway textures have been around for quite some time, but they were missing from materials.xml. I was the one who made that fix. :) As for them not looking good at all, most people agree that the grass is an improvement. Perhaps you could tell me what, specifically, is wrong with it. Is it too bright? Too bland? Too repetitive? The main problem is that the taxiway textures expose the workaround that we use because we don't (yet) have curved taxiways. The concrete colour does not blend with the old texture, which is still used for aprons etc., and the edge and centre lines also serve to emphasize the problem. Here are a couple of examples. ftp://abbeytheatre2.org.uk:2121/flightgear/Terrain/EGMH-textures.jpg ftp://abbeytheatre2.org.uk:2121/flightgear/Terrain/EGMH-textures-2.jpg Note the lack of stopways. Presumably something went wrong with the merge. See how every segment of the taxiways is obvious. The colour of the grass is probably a matter of opinion, but note how it fails to merge with the surrounding textures, exposing the cut-in edge of the airfield. I'm very sorry, but IMO this all looks rather unprofessional. I would not wish to release this as it stands. Vivian -- Achieve unprecedented app performance and reliability What every C/C++ and Fortran developer should know. Learn how Intel has extended the reach of its next-generation tools to help boost performance applications - inlcuding clusters. http://p.sf.net/sfu/intel-dev2devmay ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
Re: [Flightgear-devel] keepingup with OSG(was Who turned the sun off?)
On 10.05.2011 10:36, Alan Teeder wrote: How is FG going to progress if it is stuck in a time warp with an old version of its key graphics library? Are the FG code developers liasing with their OSG counterparts, especially in the area of backwards compatibility? First of all, 2.9.x are OSG developer releases (unstable). The stable OSG release was 2.8.3 for a long time - now there's 2.8.4 (since April). I'm not aware of major issues concerning stable OSG. We're checking compile-compatibility against OSG trunk regularly (automatic Hudson build). When necessary, we're adapting to OSG changes - though that will take a few weeks. Sometimes bugs are also introduced on the OSG-side - like http://code.google.com/p/flightgear-bugs/issues/detail?id=268 Incidentally, I have just invested a lot of time this weekend, tracing the specific OSG commit causing the problem and debugging it. The problem is now identified (OSG forgot to update one of their font plugins) and thankfully the OSG maintainer is now looking into this. There's also a temporary patch in our tracker. If the black sky issue persists for some time (i.e. OSG isn't fixing the issue themselves - which they often do, since most bugs aren't FlightGear-specific and there are a lot of active OSG devs/testers), then someone of us will also need to look into this. So, we're not stuck in a time warp just because there are occasional OSG issues. But it's tiring to be a constant OSG beta-tester. Seems we have enough own issues to take care of. Anyone volunteering though is welcome to test OSG trunk regularly, check for new issues, trace conflicting changes and report problems - preferably on a level which can be reported to OSG directly. Indeed, I guess Tim would be happy for any help in maintaining the FG/OSG code. And remember: you're not getting any new features just by installing a fancy new OSG library. We'd also need someone to add new code to FG in order to take advantage of available OSG features. cheers, Thorsten -- Achieve unprecedented app performance and reliability What every C/C++ and Fortran developer should know. Learn how Intel has extended the reach of its next-generation tools to help boost performance applications - inlcuding clusters. http://p.sf.net/sfu/intel-dev2devmay ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
Re: [Flightgear-devel] A proposal for a Nasal based Honeywell/Boeing-style CDU for FG
Hi Claus, adding support for more complex cockpit instruments, like the CDU, indeed sounds very exciting! And I think we have a lot of users here who enjoy advanced and realistic cockpits. So any help in this area is highly welcome! I'll hopefully find some time during the next days to test and look at your CDU. The only thing which generally concerns me about larger Nasal additions are some unresolved performance issues. Will be interesting to see what impact the new CDU module has. Anyway, performance issues might also be topic for the next few days at LinuxTag. I hope to see a lot of active FGers in Berlin. Anyone else: don't forget, it starts tomorrow/today - and a visit is worth it: http://www.flightgear.org/forums/viewtopic.php?f=42t=11881 cheers, Thorsten On 09.05.2011 23:50, Claus Christmann wrote: Good Day everybody, first of all, let me introduce myself: My name is Claus, I am an aerospace student at GeorgiaTech and recently got into FlightGear in relation to one of my research projects. I have been not so good on reading the FG-devel email list, but have had quite some interactions on the FG IRC chat with other FG enthusiasts. For my research project I need(ed) a somewhat functional Boeing style CDU. So I got into contact with Gijs de Roy and started to learn about his CDU project, then still in this git repo for the 744. I started to make a wiki writeup for Gijs CDU (http://wiki.flightgear.org/Howto:_Development_of_the_CDU, not up to date anymore) to learn about how Gijs implemented the CDU. As I needed the CDU to be fairly flexible - particularly in respect to [ages outside the currently implementation - I decided to change from a XML approach to a more Nasal based approach. I have finished a first draft of my code and started to write some documentation for it in the FG wiki. I would like to ask for some dev's to check out my code and make any suggestions for improvements. As I have the clearance to release the code under GPL v2, I also would like to offer to merge this into the main FG repository. You can find my code at https://gitorious.org/~hcc23/fg/hcc23s-fgdata/commits/cdu-development The docu is contained in these two wiki pages (and will be expanded) http://wiki.flightgear.org/Nasal_CDU_Framework http://wiki.flightgear.org/Howto:_Coding_a_Boeing_CDU I am very much looking forward to hearing back from you, particularly with comments and improvement suggestions. Regrads, Claus PS: for those of you active in IRC, my nick there is hcc23 -- Achieve unprecedented app performance and reliability What every C/C++ and Fortran developer should know. Learn how Intel has extended the reach of its next-generation tools to help boost performance applications - inlcuding clusters. http://p.sf.net/sfu/intel-dev2devmay ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
Re: [Flightgear-devel] keepingup with OSG(was Who turned the sun off?)
-- From: ThorstenB bre...@gmail.com Sent: Tuesday, May 10, 2011 11:01 PM First of all, 2.9.x are OSG developer releases (unstable). The stable OSG release was 2.8.3 for a long time - now there's 2.8.4 (since April). I'm not aware of major issues concerning stable OSG. We're checking compile-compatibility against OSG trunk regularly (automatic Hudson build). When necessary, we're adapting to OSG changes - though that will take a few weeks. Sometimes bugs are also introduced on the OSG-side - like http://code.google.com/p/flightgear-bugs/issues/detail?id=268 Incidentally, I have just invested a lot of time this weekend, tracing the specific OSG commit causing the problem and debugging it. The problem is now identified (OSG forgot to update one of their font plugins) and thankfully the OSG maintainer is now looking into this. There's also a temporary patch in our tracker. If the black sky issue persists for some time (i.e. OSG isn't fixing the issue themselves - which they often do, since most bugs aren't FlightGear-specific and there are a lot of active OSG devs/testers), then someone of us will also need to look into this. So, we're not stuck in a time warp just because there are occasional OSG issues. But it's tiring to be a constant OSG beta-tester. Seems we have enough own issues to take care of. Anyone volunteering though is welcome to test OSG trunk regularly, check for new issues, trace conflicting changes and report problems - preferably on a level which can be reported to OSG directly. Indeed, I guess Tim would be happy for any help in maintaining the FG/OSG code. And remember: you're not getting any new features just by installing a fancy new OSG library. We'd also need someone to add new code to FG in order to take advantage of available OSG features. cheers, Thorsten Thanks for the response, which has allayed my fears. My posting this was because there have been several postings about problems with recent versions of OSG and I had not seen much progress on this forum. I don´t have the time. or expertise, to consider getting involved with the OSG group. I appreciate that with a project with FG, which in turn depends upon other projects such as OSG etc etc there will be problems such as this. My attitude as a would be aircraft developer is that keeping up with bleeding edge has the advantage of being more future-proof, but is bound to produce such distractions. Regards Alan -- Achieve unprecedented app performance and reliability What every C/C++ and Fortran developer should know. Learn how Intel has extended the reach of its next-generation tools to help boost performance applications - inlcuding clusters. http://p.sf.net/sfu/intel-dev2devmay ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
