I just created a small Benchmark running for about 1 Minute in Lion's Arch with best appearance settings on 1080p.https://flightlessmango.com/games/9691/logs/821I might create longer benchmarks later, probably actively playing 1hr Dragonfall each. I compare the Build by @"ArmoredVehicle.2849" against a custom setup in Lutris using GloriousEggroll Proton-5.9-GE-5-ST with FSYNC.All other settings in exports are the same. Both tests even use the same Wineprefix and shader cache.The results are nearly the same and differences are well withing margins of error. I feel like the results are quite bad considering my hardware. It's playable, but even on lower graphics settings the FPS don't get much higher.Currently I'm the only one who has uploaded benchmarks from MangoHUD. It would be great if other people could create benchmarks on flightlessmango.com too, for easier comparison. You can get MangoHUD here:https://github.com/flightlessmango/MangoHudIf you are running Arch, it's available from AUR. To run MangoHUD with the ArmoredVehicle scripts I simply created the following file in the main folder, next to the play.sh: You will need to create a config file to be able to save the benchmarks and later upload them for comparision: In Game press L-Shift + F2 to start recording. Afterwards go to your output folder and upload it to https://flightlessmango.com/games/9691 .Thank's for pitching in! Another feedback I would like to leave is, that Deltaconnected never changed the URL of arcdps.I simply added a little script in the arcdps_easysetup_tool folder: Whenever GW2 get's an update and arcdps needs to be updated too, I run this and than run arcdps_setup.sh.This is a tiny thing, but that way I never have to google for arcdps, search the d3d9.dll, download it and manually move it to the folder.
Oh, yes, Thanks for the correct link. From my tests I've had the same results , for this game Esync is the better choice, however I found Fsync to help in others. Some games are even unstable with Esync whereas with Fsync work just fine while still retaining better performance. I did a longer run, actively playing Dragonfall Meta for about a hour each.Each run contains a full Meta. Having many people in one place during the end (killing Kralk and hunting the Bosses afterwards) having a huge impact on performance while building the camps goes easier, because the action is less centered and distributed on the map.I expected, that recordings a longer timeframe including the full meta run would even out unfair advantages based on what was happening on the map for either of the builds. My results were so close (FSync being slightly better, about 3%), that all differences are well within margin of error.I disabled supersampling to get playable performance.https://flightlessmango.com/games/9692/logs/848