Fall Guys (1097150)

[kode54]

Whitelist Request

• Name of the game to be whitelisted: Fall Guys
• Steam AppID of the game: 1097150

System Information

• GPU: AMD RX 480 8GB
• Driver/LLVM version: mesa-tkg-git 20.2.0_devel.126859.71572ebb32e-1 / llvm 10.0.1-1
• Distro version: Arch Linux
• Link to full system information report as Gist: https://gist.github.com/kode54/3ff952f39b407d6f806a6779838f5335
• Proton version: 5.0-9

I confirm:

• [x] that pressing the Play button in the Steam client is sufficient.
• [ ] that runtime config options are necessary to run the game.
• [x] that no workarounds other than PROTON_* env variables are necessary.

Issues

• [x] I haven't experienced any issues.
• [x] There are no issues left open for this game.
• [x] Although I consider the gaming experience equal to Windows there are remaining issues:

1. Only one issue: If you are unfortunate enough to try running the game with Proton-GE version 5.9-5-ST instead of trying the official Proton releases first, and have a PS4 controller, it will not report axes correctly, and will report some axes that shouldn't exist, rendering the game unusable. Works perfectly with official proton 5.0-9.

[kisak-valve]

Hello @kode54, please clarify what runtime config options are needed since you indicated that something is needed in that regard.

[kode54]

Sorry, I misread that, I though it said "no runtime options are needed". The only option I needed was manually opening the game's properties and setting it to the latest official Proton, 5.0-9.

[kode54]

Can't say it's totally perfect, I did experience some disappearing polygons during a spectator portion, but that may have been the game itself and not my drivers.

[braiam]

@kode54 Have you noticed any soft lock when qualifying for the next round?

[tarcisioe]

I am consistently getting "Failed to login, please check your connection" after a sometimes very long loading screen. On Windows, on the same PC and network, I am able to play with no issue.

[kode54]

I notice a soft lock only occasionally, and same for failed login. This may be due to an update? I'm not sure.

Either way, I've earned two crowns so far, one from winning my first match, another from reaching a high enough level to win another as a level reward.

[RyuzakiKK]

@kode54 Have you noticed any soft lock when qualifying for the next round?

I have. At first I thought it was related to an issue with a specific map (Jinxed in particular) but apparently it seems this isn't the problem.

I'll try to gather some logs the next time it gets stuck and report back if I find something.

[RyuzakiKK]

steam-1097150-partial.log

The soft lock just happened. So far it always happens after 2-3 completed games and after you qualify for the next round.

I cut down the log that I attached because it was 120MB and Github has a limit of 10MB (if necessary I can upload it somewhere else). After the soft lock I waited 1-2 minutes and then I closed the game with alt+f4.

[minionprocyk]

can also confirm soft lock after a few games at the qualify screen. Though some of my windows friends have experienced it, it happens much more frequently using proton.

[katk0smos]

No clue if this is specific to my machine, but I do think it's notable as I haven't yet found a fix (trying other Proton versions, updating graphics drivers, etc.): characters don't render at all. Every other aspect of the game renders and works completely fine, it's just the characters themselves.

I'm running Fedora 27 (I know it's long out of date, I've been slowly working on backing things up so I can update, I just need to be sure my system won't break), and the game runs perfectly fine on my Windows 10 install on the same machine.

[braiam]

@TheWaffleDimension can you show a screenshot?

[kisak-valve]

Hello @TheWaffleDimension, please copy your system information from Steam (Steam -> Help -> System Information) and put it in a gist, then include a link to the gist in this issue report.

[alosarjos]

There seems to be a problem where after X matches, game is broken on the qualified screen (The graphics are not frozen, looks more like a connection problem). Have seen more of these reports on ProtonDB. Any idea of what could it be or how to fix it?

[braiam]

@alosarjos I've seen streamers also affected with the issue, and they use Windows. It seems that the developers should be contacted for these cases.

[alosarjos]

@alosarjos I've seen streamers also affected with the issue, and they use Windows. It seems that the developers should be contacted for these cases.

I've been playing with friends for a couple of hours, my game had the softlock 4 times, them none. So seems like maybe Proton is handling this error worse or something... Kinda annoying when playing with friends.

UPDATE: Have played for some hours on a Windows boot. Not happening a single time. There is definitely a problem in Proton.

[NikoUY]

I can also confirm that it's an issue with Proton, I play with some friends and the one that's using Windows does not have this issue while my other friend and I have to restart it after ~10 games to avoid the problem, I have played for 43h and I always get soflocked after ~10 games if I don't restart the game but the friend that's using Windows hasn't experienced this after playing for around 30h, I'm not saying it's not possible but it might only happen very very rarely on Windows while it's really consistent with Proton.

Also there is another very minor issue that only happens in Block Party and Roll Out, every so often a black outline of the rollers (they probably use invisible rollers or something similar in Block Party) will flash on the screen for a frame, it's not a big issue and rarely happens, the game works perfectly otherwise.

[Guy1524]

I've noticed this too, it appears to coincide with recv failed for host {15}, error {676}. Host will be closed maximum hosts cannot exceed {16}

[katk0smos]

@TheWaffleDimension can you show a screenshot?

Hello @TheWaffleDimension, please copy your system information from Steam (Steam -> Help -> System Information) and put it in a gist, then include a link to the gist in this issue report.

I've gone ahead and finished backing up files, then updated to Fedora 28. I still have yet to get time to upgrade the rest of th way up, but everything is now rendering correctly. I'm not sure what the issue was, maybe there was a package update available in 28 that wasn't in the 27 repos?

Either way, issue is resolved, so you can ignore that rendering problem.

[SergioGlorias]

I am also having these softlock They only happen when you show the classifieds (to happen (more often) more when I literally go to the final round and everything was going well (I haven't won yet ;-; )) or stop finding a match The animations continue without any problem but it is consecrated in that area And only decide if Alt + F4

My System: https://gist.github.com/SergioGlorias/3dbd4efc4dd9ae660e21632eff61a734

I also played until I gave the softlock (when I was going to the last round) using the active DEBUG but I can't send the log to the gist because it's too heavy and the browser breaks the page (I still can't read logs :/ )

[kode54]

I can't say I've experienced any lengthy soft lock, I've also won three crowns from Linux. But I usually have poor luck these days, I've played at least 30 other episodes without winning more than participation prizes. I have also leveled up to 18 with the fame thus far.

[SergioGlorias]

I can't say I've experienced any lengthy soft lock, I've also won three crowns from Linux. But I usually have poor luck these days, I've played at least 30 other episodes without winning more than participation prizes. I have also leveled up to 18 with the fame thus far.

I am also not lucky but as I am in the middle of the match and I am in solftlock I am not lucky enough to move on to the next stage and I think I also lose XP and the battel coins goes

If it is a softlock to find a match, I can leave the game in normal mode, but if the way to start is with Alt + F4

this solftlock seems to me as if the game lost its connection to the system and does not notify

[Tyfuzzle]

I can also confirm soft locks on the "qualified screen". For me, it seems to happen around every 40 minutes game time. I have to Alt + F4 like others.

Here's my system info

Here's the last 99999 log lines

[SergioGlorias]

I had testing the proton in the beta version "next" and continued the same problem smooth wheel game taking out the softlock

[blazingkin]

I get freezes when alt tabbing in and out of the game, bringing up the steam overlay. I'm on Kubuntu 20.04 with the latest stable Nvidia drivers

[kode54]

Now, after today's annoying mandatory maintenance period, the game seems to have stopped working.

[NikoUY]

The game it’s working fine for me with the exception I mentioned earlier.

[braiam]

@kode54 can't reproduce that.

[kode54]

Okay, I hear from someone else who had the same issue, they temporarily fixed it by forcing WineD3D, and then by scanning their logs, they tracked it down to SteamVR, which they promptly removed. I have no idea why I'd even have SteamVR installed, or how it would have gotten installed, if that's the case.

[NTMan]

Now, after today's annoying mandatory maintenance period, the game seems to have stopped working.

same crash here, I only upgraded mesa...

steam-1097150.log

[kode54]

It appears to be specific to the Vulkan driver installed. Are you running Mesa Git? I didn't feel like bisecting this one, so I'll just report it to them and let someone there bisect it instead.

[NTMan]

It appears to be specific to the Vulkan driver installed. Are you running Mesa Git? I didn't feel like bisecting this one, so I'll just report it to them and let someone there bisect it instead.

$ git bisect log
git bisect start
# good: [865a2ad0866e271adda3cd18c46f74514131da11] clover/nir/spirv: Use uniform rather than shader_in for kernel inputs
git bisect good 865a2ad0866e271adda3cd18c46f74514131da11
# bad: [5c45eaf9b3578094c36911cbbd085097642d3b13] clover/spirv: fix vec3 alignment
git bisect bad 5c45eaf9b3578094c36911cbbd085097642d3b13
# bad: [a6291b1b1177f5728e2e1998225f0b8676c6e710] freedreno/ir3: rework setup_{input,output} to make struct varyings work
git bisect bad a6291b1b1177f5728e2e1998225f0b8676c6e710
# bad: [d383133ce1ddd16e3099615fcd31e6f28c11b17b] docs: Stop claiming to implement OpenVG
git bisect bad d383133ce1ddd16e3099615fcd31e6f28c11b17b
# bad: [50e14c3d258f5be887b4101cc3b01bb16ad50caa] radv/winsys: Fix memory leak.
git bisect bad 50e14c3d258f5be887b4101cc3b01bb16ad50caa
# bad: [701cb9d60c0350b0134e52cb1b51eefbbf27bd22] nir/vtn: Handle integer sampling coordinates
git bisect bad 701cb9d60c0350b0134e52cb1b51eefbbf27bd22
# good: [de36b5b805be1732e6798afd1a12743df8d32718] nir/vtn: Add support for kernel images to SPIRV-to-NIR.
git bisect good de36b5b805be1732e6798afd1a12743df8d32718
# bad: [a196f05fc22620be5809f7b28ce7f2671b1ef32c] nir/vtn: Use return type rather than image type for tex ops
git bisect bad a196f05fc22620be5809f7b28ce7f2671b1ef32c
# first bad commit: [a196f05fc22620be5809f7b28ce7f2671b1ef32c] nir/vtn: Use return type rather than image type for tex ops

[kisak-valve]

Hello @NTMan, please test mesa git from today, specifically after https://gitlab.freedesktop.org/mesa/mesa/-/commit/abbfc5b8db7412fa2e6f8c1b66ff97b886c4ac70.

[NTMan]

Hello @NTMan, please test mesa git from today, specifically after https://gitlab.freedesktop.org/mesa/mesa/-/commit/abbfc5b8db7412fa2e6f8c1b66ff97b886c4ac70.

Thanks, at least at commit https://gitlab.freedesktop.org/mesa/mesa/-/commit/612ef74190ca6c4d036e014e5fc228b671513771 the game works as intended again.

[kode54]

Thanks for bisecting, I was going to do it tonight if nobody else got around to it sooner. I mentioned the range of commit hashes on Mesa's tracker. Oh well, nice work.

[abhishek-rawal]

The issue was observed in Nvidia drivers as well. With vulkan-icd-loader 1.2.151.r14 I am observing the same dialogue box as mentioned by other users and game crashes, reverting back to vulkan-icd-loader 1.2.151.r10 fixed it for me.

[kisak-valve]

Hello @0x221e, the previously discussed crash was a single day regression in bleeding edge mesa drivers. Since the regression you're seeing has the same upstream version, you've encountered a distro-specific issue that should be reported to your distro's package maintainer for vulkan-icd-loader.

[abhishek-rawal]

@kisak-valve Didn't knew that. Thanks for the information. Will update at required forum.

[kddlb]

The game might stop working, because of (a proper) anti-cheat system being implemented:

We're really sorry about the cheating problem!

We're expanding the current detection system this week to improve things

We also have a BIG update in the next couple of weeks that adds the same anti-cheat used by games such as Fortnite

Thanks for bearing with us!

[Not BeanBot]

-- https://twitter.com/FallGuysGame/status/1302680927605338113

[rezzafr33]

If they decide to use EAC, i hope they activate support for wine64/wine32 just like Squad does. Or perhaps someone from valve can convince them to use wine friendly anticheat solution. My son love this game, i don't want to install windows just for Fall Guys :(

[NikoUY]

They could release it on Linux and use the Linux version of EAC, I doubt that they are going to do it but you never know, it’s not impossible to do using Unity, not sure if they use any problematic middleware.

[SergioGlorias]

I know I can’t do anything, but when I knew I opened a ticket on the support They responded with a generic message: https://user-images.githubusercontent.com/9739913/92455053-21691500-f1b9-11ea-806a-79ddfbdbf331.jpg

But who knows they hover

[exentio]

I confirm that today's update, which added Easy Anti-Cheat, made the game stop working on Linux. If there's any useful log I can give you let me know

[rezzafr33]

[sonic2kk]

This is extremely unfortunate. I praised this game for its lack of Anti-Cheat Spyware solution and it's seemingly Linux-friendly developers. Shame.

[kisak-valve]

fall guys ultimate knockout starting ok but first person player not showing

Issue transferred from https://github.com/ValveSoftware/Proton/issues/4207. @thsnielsen posted on 2020-09-19T10:01:45:

In short i am on opensuse tumbleweed 5.8.7-1-default on a lenovo x220 16g ram much ssd disk free I have the same problem on 2 different equal machines I had to add this option to make the game launch, which it does fine now my son even completed a game with this invisible player!!: This option makes the game run : PROTON_USE_WINED3D=1 %command% This screenshot says it all - there should be the player in the middle of the screen, but he is missing ...???...

There are many errors below - not sure where to begin, and then again - the game mainly works?? Well except i do not see the players :-)

<System information omitted, please see #4207>

[tronje]

So @thsnielsen you are able to play the game despite the new anti-cheat update which broke the game for most players?

[thsnielsen]

@tronje I only tested on your remark above, and after the easy anti cheat was installed, i get "Anti-cheat is not initialized" when i press play. I never tested play after anti cheat was installed, because of the missing graphics of the player. So though it looks the same to me, it looks like i have one more problem now - the anti cheat :-/ Have a good day

[sfjuocekr]

If you contact MediaTonic support they will tell you they will bring up the issue internally.

So I suggest, everyone who wants to play this game -again- sent them a support request! This game was perfectly playable, I did not face the missing puppet at the start screen and since EAC was implemented the game just does "not" work.

[kode54]

Yes, but you're still talking about the difference between eliminating a lot of mass cheating from the large Windows audience, and the relatively small user base of Linux users being collateral damage.

Though, it may be worth their effort to try pushing a Linux native client, since EAC does sort of work in Linux. Or doing like one other game I can't think of the name of off hand that whitelisted Wine with their EAC system, which either sort of works or doesn't work? I'm not really sure.

[sfjuocekr]

There are better ways to achieve "anti-cheat" solutions than to police in the users end.