hyperion

Agreed, might also be the kernel (KMS). Maybe try a newer or older one.

Works for me (tm). If it only works sometimes maybe some sleep before launching 0ad might work. For sound maybe you use sound card by index in your config instead of identifier and some reordering happens? Borderless fullscreen doesn't matter here either.

So to make the issue worse, we move some of the few leftover useful bits of p2 to p1. A player should be forced to phase up with around 50 pop to be competitive, current p1 is simply broken.

Looking at the code it's a bug on non-windows, nothing you can do here

No need to restart X if you have the randr extension which you are almost guaranteed to have, just write a small script that sets mode, runs 0ad, resets mode. Then use that script to launch 0ad instead. Apart from some flickering when launching or closing 0ad there won't be any side effect. Might also depend on wm/dm and sdl version, so not all that sure either that this is always true.

@andy5995 A quick test suggest ActorEditor is working now, just a nit 23:57:04: can't open file 'ICON_ActorEditor' (error 2: No such file or directory) 23:57:04: Failed to load image from file "ICON_ActorEditor".

xres/yres are for windowed mode, in fullscreen mode 0ad renders at display resolution. You could change display mode or scale using X, see "man xrandr".

@andy5995 According to ldd it was the only one missing.

There is no shoe that fits all. Some do well with a fixed release schedule like the linux kernel or ubuntu for instance, others fare well with rolling releases, even others just are fine with scm only. For 0ad 2-3 releases a year I say would probably work best, more creates issues with having people on the same release, less and contributions take to long to get into a release. Anyway, the only thing to never do is silent / in-place releases.

Hi @PlayerMat first welcome to the forums. About your issue, maybe not the only one but to my knowledge the first to report it. Please attach your log files and hardware report files so people may guess what is going on. See https://trac.wildfiregames.com/wiki/GameDataPaths

@rm -rf, you don't have to extract the zip, just place it into ~/.local/share/0ad/mods/feldmap I agree this isn't "as expected". People shouldn't offer zip files in the first place, only pyromod files.

The above try was made on a Gentoo machine and as I don't speak Thai there is no libthai.

Might not be the same as #2, you only define pyrogenesis as executable and not ActorEditor, so no coping might happen for that reason. Just guesswork. Also if the base disto offers wxwidgets 3.2 you might want to opt for that instead, I faintly recall there being a bug / bugs with ActorEditor and 3.0. PS: The multi-call solution is a nice idea

$ ./ActorEditor /tmp/.mount_ActorEge8Yqx/usr/bin/ActorEditor: error while loading shared libraries: libthai.so.0: cannot open shared object file: No such file or directory $ ./0ad-0.0.26-alpha-2210010553-x86_64.AppImage -editor TIMER| LoadDLL: 53.6057 ms Unable to open crashlog.txt for writing (please ensure the log directory is writable) Location: debug.cpp:195 (debug_WriteCrashlog) Call stack: (0x5623a5de2b5e) /tmp/.mount_0ad-0.UaI40J/usr/bin/pyrogenesis(+0x5e2b5e) [0x5623a5de2b5e] (0x5623a5d97361) /tmp/.mount_0ad-0.UaI40J/usr/bin/pyrogenesis(+0x597361) [0x5623a5d97361] (0x5623a5d988eb) /tmp/.mount_0ad-0.UaI40J/usr/bin/pyrogenesis(+0x5988eb) [0x5623a5d988eb] (0x5623a5d98569) /tmp/.mount_0ad-0.UaI40J/usr/bin/pyrogenesis(+0x598569) [0x5623a5d98569] (0x5623a5d988fd) /tmp/.mount_0ad-0.UaI40J/usr/bin/pyrogenesis(+0x5988fd) [0x5623a5d988fd] (0x5623a5a9b44b) /tmp/.mount_0ad-0.UaI40J/usr/bin/pyrogenesis(+0x29b44b) [0x5623a5a9b44b] (0x5623a5a9b52a) /tmp/.mount_0ad-0.UaI40J/usr/bin/pyrogenesis(+0x29b52a) [0x5623a5a9b52a] (0x5623a58a664c) /tmp/.mount_0ad-0.UaI40J/usr/bin/pyrogenesis(+0xa664c) [0x5623a58a664c] (0x5623a5893817) /tmp/.mount_0ad-0.UaI40J/usr/bin/pyrogenesis(+0x93817) [0x5623a5893817] (0x7f268c81934a) /lib64/libc.so.6(+0x2934a) [0x7f268c81934a] (0x7f268c8193fc) /lib64/libc.so.6(__libc_start_main+0x7c) [0x7f268c8193fc] (0x5623a58a4bba) /tmp/.mount_0ad-0.UaI40J/usr/bin/pyrogenesis(+0xa4bba) [0x5623a58a4bba] errno = 30 (?) OS error = ? Btw, the game runs fine, no performance issues either.

true, but there is an additional 3 or so turns command delay for networked games. So the effective delay didn't get reduced by much, it just looks much better.

https://launchpad.net/~wfg/+archive/ubuntu/0ad already has a26

Which should be more computationally expensive than adding the "platform" to the model itself, right?

Finally proper paving stones on Acropolis Bay! Yay! Looks so much better. Congratulations on the release!

just press h for halt.

Not to long ago the forge techs were simplified to make it easier to switch rooster, now you want the opposite. Care to elaborate why this was fundamentally the wrong direction?

But props are actors, right?

Variants can also be actors, so one level of indirection might do the trick for you.

@ivicok, grep for DamageVariants

Gorgeous and with fantastic sound https://www.youtube.com/c/play0ad

Right in-game scenario editor is Atlas (-atlas option). Well, appimages have exactly one payload application but that could also be a shell script passing args to pyrogenesis unless there is an option -editor in which case it would run ActorEditor or in case of -h, --help print the readme.txt in binaries/system.