Hello, It looks there is an issue in the flightgear-data package. When launching the simulator, runaway and grass is replaced by a grey display.We can see the plane dashboard but it is almost the only thing we can see. Also when program is launched there are a lot of complains about textures issues and missing files (at least this one, which looks to present in the fgdata repository. One thing to correct is to modify the line 377 /usr/share/games/flightgear/Compositor/Effects/terrain-default.eff that looks to contains an orphan </or> tag (seems to be a closing tag, but i didn't found the opening tag. When the </or> is put in comments, the track is visible and grass is green and no longer grey.
Thank you for the report. Can we take it that the simple correction you found in /usr/share/games/flightgear/Compositor/Effects/terrain-default.eff to suppress the unpaired </or>, restores the program to working properly? Or does that still leave the other issues "there are a lot of complaints about textures issues and missing files"? Apart from annoyance, do these messages actually matter? Assigning anyway to Chris who maintains the Flightgear packages.
Assignee: bugsquad => eatdirt
Hello, Sorry for late answer didn't see your answer. Yes the removal of the unpaired </or> resolving the issue on my computer ( and appears also for users that apply the change). Another user on MLO ( Helijah) report it on the Flighgear forum in this post. https://forum.flightgear.org/viewtopic.php?f=68&t=41686 ( correction is to add the missing <or> instead of removing/comments the </or> tag/ Regarding the missing files, the .wav file seems to be relative to file of the used aircraft, so not relative to Flightger but the aicraft i think. For the texture file, it looks like Compositor/Effects/water-inland.eff make a reference to Textures/Globe/ocean_depth_1.jpg which is not existing ( only the png file exists). Unfortunately, i can't tell if it is annoyance or not ( don't know how to go somewhere the texture file is used but i suppose it might cause graphic issue where the texture file is used). But looks like more a bug in the flighgear file ( and don't think a bug report done for this in Flightgear forum / bug tracker). https://www.mageialinux-online.org/forum/topic-30823-3+ou-est-passe-le-sole.php#m315343
I'll fix that. Indeed, preparing for the next major bump of FG, the version we ship on mga9 has been built with the "compositor" enabled, so bugs can be there. PS: However, guys, a FG's bug should be reported first on the FG's bug report website, they fix it, we import the fix :) https://sourceforge.net/p/flightgear/codetickets/
No need to report, it is already fixed upstream. I'll backport patches and push a new fgdata tonight.
Simgear, flightgear and flightgear-data bug fix release 2020.3.18 landing in updates_testing for mga9 (sync to latest snaphots) To test, install them, and, in a terminal enter: fgfs --compositor=Compositor/als --launcher NB: Note the new option compared to mga8 (--compositor=Compositor/als). Forgetting this option, it should work on lower end machine without a good rendering quality. Play around. If you want to fly a plane, don't forget to allow "enabling terrasync", which downloads the scenery where you start. If you don't, you'll be in water. Suggested advisory: ======================== Updated simgear, flightgear and flightgear-data packages to version 2020.3.18.1-1 fixing rendering errors under the new compositor pipeline. ======================== Updated packages in core/updates_testing: ======================== simgear-2020.3.18-1.1.mga9 flightgear-2020.3.18-1.1.mga9 flightgear-data-2020.3.18-1.1.mga9 Source RPMs: simgear-2020.3.18-1.1.mga9.src.rpm flightgear-2020.3.18-1.1.mga9.src.rpm flightgear-data-2020.3.18-1.1.mga9.src.rpm
>the version we ship on mga9 has been built with the "compositor" enabled, so bugs can be there. Forgot to say that if this creates too much issues, we can revert the built to non-compositing (but the rendering won't be as nice).
Corrections, I have also fixed the shader effect reference to ocean_depth_1.png in the following, these should be used: flightgear-data-2020.3.18-1.2.mga9 flightgear-data-2020.3.18-1.2.mga9.src.rpm
Assignee: eatdirt => qa-bugsCC: (none) => eatdirt
Corrections^2, after reporting the effect file problem upstream, the correct reference is that water-inland.eff points to globe_colors.jpg Here the last push for fgdata, still in updates_testing flightgear-data-2020.3.18-1.3.mga9 flightgear-data-2020.3.18-1.3.mga9.src.rpm NB: all flightgear-data-2020.3.18-1.*.mga9 files are fixing the reported bug, but please, check the latest flightgear-data-2020.3.18-1.3 to be sure I did not add spurious bug in. thanks!
Hello, Installed the last fgdata and earth texture looks ok.Could not see anymore the error message that appears earlier ( except the one for the aircraft but it is not related to FG). I also tried to check how the water looks, and it seems it is blue ! So i think the reported bugs have dissapears after the FGdata fixes. Thanks
The following 3 packages are going to be installed: - flightgear-2020.3.18-1.1.mga9.x86_64 - flightgear-data-2020.3.18-1.3.mga9.noarch - simgear-2020.3.18-1.1.mga9.x86_64 No installation issues. Played around a bit, again making myself a menace to the airport before crashing. But, the scenery looked good, as far as I could see. Giving this an OK, and validating. Since the issue is with the noarch data rpm, one arch should be sufficient. I don't think anybody is likely to use this with 32-bit hardware, anyway. Suggested advisory in comment 5, except that the data rpm version in that text needs to be corrected.
Whiteboard: (none) => MGA9-64-OKCC: (none) => andrewsfarm, sysadmin-bugsKeywords: (none) => validated_update
Keywords: (none) => advisoryCC: (none) => marja11
I had missed that flightgear-data-2020.3.18-1.1.mga9.src.rpm from the suggested advisory had become flightgear-data-2020.3.18-1.3.mga9.src.rpm So the uploaded advisory didn't allow flightgear* / simgear to be pushed to updates. The uploaded advisory has just been corrected, on next run of the script to push updates it should work better.
An update for this issue has been pushed to the Mageia Updates repository. https://advisories.mageia.org/MGAA-2023-0084.html
Resolution: (none) => FIXEDStatus: NEW => RESOLVED