|Summary:||Empire Total War - Graphical Corruption|
|Component:||Drivers/Gallium/radeonsi||Assignee:||Default DRI bug account <dri-devel>|
|Status:||RESOLVED FIXED||QA Contact:||Default DRI bug account <dri-devel>|
|i915 platform:||i915 features:|
|Bug Depends on:|
glxinfo | grep strings
steam system info
Description andrew.m.mcmahon 2019-01-12 15:04:58 UTC
Created attachment 143078 [details] glxinfo | grep strings 4.20.0-arch1-1-ARCH mesa 18.3.1-1 steam 126.96.36.199-1 R9 285 (Tonga) In game menus and other game play elements aren't rendered correctly by default: https://imgur.com/a/AvclRkS There are no special steps required to replicate the issue. Affects other Linux distributions and other cards: https://steamcommunity.com/app/10500/discussions/0/1319962244123571976/ A known workaround for the problem is to edit: ~/.local/share/feral-interactive/Empire/preferences Changing the value of the following line to zero: <value name="UsePBOSurfaces" type="integer">1</value> There doesn't appear to be any downside to this. Medieval 2 Total War doesn't appear to be affected on my setup. Unable to test Shogun 2 or Fall of the Samurai as I don't have access to those.
Comment 1 andrew.m.mcmahon 2019-01-12 15:05:38 UTC
Created attachment 143079 [details] steam system info
Comment 5 andrew.m.mcmahon 2019-01-13 13:12:00 UTC
I've used apitrace (DEBUGGER="apitrace32 trace" steam steam://rungameid/10500) to capture some trace files. I'm able to replay them using apitrace replay [file] so I hope others are too. I've uploaded a 7z archive containing a "bugged" trace and a "workaround" trace with their corresponding terminal output: https://drive.google.com/file/d/1BAQFnrJqujU7dYGQTfheABTRjj-_a8Ys/view?usp=sharing By complete coincidence the archive happens to be 666MB. The bugged trace is several gigabytes in size whilst the workaround is a mere 90MB so there must be something funny happening there...
Comment 6 andrew.m.mcmahon 2019-01-13 13:53:27 UTC
One last comment before I leave it to the experts. Another bug report with Unity 18.04: https://steamcommunity.com/app/10500/discussions/0/3441214221460524307/ If anyone wants to go above and beyond the call of duty then I've uploaded some traces for Medieval 2 here: https://drive.google.com/open?id=17LUb2EsS3oqYjEyZ-yxIB_C_DBnS0Wz8 I've used the same trick of disabling/enabling PBO surfaces. Finally what exactly are PBO surfaces? Do I need them? What's the penalty for not using them? The game seems to trace faster with them off than on. Are the Linux ports by Feral a bit wonky and/or prone to strange behaviour - or are they generally decent?
Comment 7 andrew.m.mcmahon 2019-01-13 16:35:57 UTC
Created attachment 143088 [details] mesa-git glxinfo As a last ditch effort I thought I'd try the latest mesa/llvm that are provided by [llvm-git] and [mesa-git] unofficial repo's. Unfortunately the issue still occurs.
Comment 8 andrew.m.mcmahon 2019-03-06 19:27:31 UTC
My system has changed somewhat since writing this bug report. As all Feral games on STEAM officially support either STEAMOS or Ubuntu; I thought I'd check out xubuntu 18.04 LTS - specifically: https://pastebin.com/f8cTGLSY The same glitch appears with the current Mesa drivers provided by the distribution (18.2.8) But using oibaf's ppa: https://launchpad.net/~oibaf/+archive/ubuntu/graphics-drivers Mesa related packages are updated to: Mesa 19.1.0-devel (git-54522d0 2019-03-06 bionic-oibaf-ppa) And I can no longer reproduce the issue! https://i.imgur.com/VhIylSp.jpg As I've made no other changes other than switching Mesa packages I can only assume that this has been fixed by an update/commit recently. Many thanks to everyone who works on Mesa. In just a few years (since I last tried Debian) my R9 285 has gone from having no support whatever in Linux to being very capable indeed to the point where I no longer require Windows: https://imgur.com/a/keUM4VS https://imgur.com/a/awddkUg