Support us on Patreon to keep GamingOnLinux alive. This ensures all of our main content remains free for everyone. Just good, fresh content! Alternatively, you can donate through PayPal. You can also buy games using our partner links for GOG and Humble Store.
We do often include affiliate links to earn us some pennies. See more here.

There seems to be some confusion brewing on what games will actually run on the Steam Deck, so let this serve as a reminder on keeping expectations in check. Here's a quick refresher of how things are right now.

Some of the confusion seems to appear from an IGN interview, where Valve developer Pierre-Loup Griffais mentioned how "we haven't really found something that we could throw at this device that it couldn't handle" that we covered before. Here's the thing though: even though the Steam Deck will ship by default with the Arch Linux based SteamOS 3, they have also been testing Windows 10 and they've been working with AMD to get it supported on Windows 11 too. So speaking from a hardware standpoint, yes it probably will work with the vast majority of games on a performance level.

For regular readers and regular Linux gamers, knowing that the Steam Deck won't run everything is a given and this won't be news but there's plenty of people out there seemingly expecting a bit too much from it. Part of the problem though is Valve's marketing too, with it repeatedly mentioning your entire Steam library. They obviously want every game to work but that's simply not going to be the reality - at least not for a while.

For people who stick with SteamOS 3 this is where it doesn't quite match up. There's a good number of native Linux games (those actually built for Linux) and for everything else there's Steam Play Proton to run Windows games. A big majority of games work on Linux already between native and Proton but there's caveats.

Currently, Proton does not work with games that have the likes of Easy Anti-Cheat and BattlEye which is used in some of the most popular online titles. When it comes to EAC and BattlEye, we do know that Valve are working on it as they said in the developer documentation FAQ that they're "working with BattlEye and EAC to get support for Proton ahead of launch".

Another issue is Microsoft Media Foundation for videos, it's something of a nuisance and they don't play in Proton either. Currently Valve appear to be re-encoding the videos into a playable format which downloads with your game when run through Proton. That's a lot of work though too, there's a lot of games on Steam.

Then there's certain DRM too. Valve's older CEG (custom executable-generation) DRM also doesn't work in Proton, and there's no doubt a few other DRM/anti-tamper solutions that also don't work with Proton. Launchers can be a problem too, with some developers using .NET / Windows Presentation Foundation (WPF). For launchers Valve recommends that developers use something cross-platform and standalone like Qt or just skip launchers completely, which would be vastly better for gamepad/controller support too.

Compatibility will improve over time though as more developers hopefully look to support it directly and as Proton continues maturing. So even if your favourite or the latest AAA doesn't work right away, it might do later. There's still plenty of time until the Steam Deck releases and Valve has opened up requests for developer kits too. Valve also stated in their Steamworks video how "our goal is for every game to work by the time we ship Steam Deck" and that "there is a lot of work that has been done that doesn't yet affect the public version of Proton" so we are expecting the situation to improve. Until we see this special Proton release though, this article sums up the current situation.

If you're looking to try out Linux gaming and you're confused with Proton, be sure to check out our guide.

Article taken from GamingOnLinux.com.
34 Likes
About the author -
author picture
I am the owner of GamingOnLinux. After discovering Linux back in the days of Mandrake in 2003, I constantly checked on the progress of Linux until Ubuntu appeared on the scene and it helped me to really love it. You can reach me easily by emailing GamingOnLinux directly. You can also follow my personal adventures on Bluesky.
See more from me
The comments on this article are closed.
All posts need to follow our rules. For users logged in: please hit the Report Flag icon on any post that breaks the rules or contains illegal / harmful content. Guest readers can email us for any issues.
80 comments
Page: «3/4»
  Go to:

Purple Library Guy Sep 8, 2021
So, four categories of games not working:
1. Anti-cheat
2. Microsoft Media Foundation video bits
3. Old DRM, particularly Valve's own CEG, and launchers
4. Games that just have problems--various odds and ends and weirdities

Of those,
1. Is plausibly going to be fixed by launch. It better be. Time is running a bit short though, so they'd better have some internal stuff happening that they're not releasing yet.
2. Depends how fast their procedure is. A quick google suggests that there are around 50,000 games currently being sold on Steam. Let's imagine that 20,000 of those have MMF clips. If they can do their thing with 100 games per day, that's 200 days to fix them all. But probably only 50 days to fix the ones more than a handful of people ever currently play. That's pretty workable. If they can only fix 10 games per day, that's not so great. Mind you I suspect this vastly overestimates the amount of MMF clips or Proton would work on far fewer games than it already does.
3. You'd think Valve can fix letting its own technology work on Proton. So like, CEG, plausibly will work by launch. But there's a cluster of other technologies here, each pretty small, and the chance they'll get them all seems slim. But it may not matter that much if it's all old stuff that few people play.
4. Obviously they're not going to be able to fix all the little corner case problems by launch. If they can clean up the popular stuff that'll probably be good enough.

What I'm thinking is that some of the posts are a bit too pessimistic. Sure, I don't see Valve getting all games working on Proton by launch. But what we're ignoring is the vast gulf between the sales of Cyberpunk 2077 and the sales of Shovelware 0.8. There are masses of games on Steam that hardly anyone ever bought, or which are old and not one of the relatively few considered classics, that hardly anyone ever plays any more. Even games that are in their Indie way solid little successes, or were that three years ago, are nonexistent in terms of market share. It is plausible that 90% of people could find all the games they want to play on a Steam Deck work perfectly even if Valve fixed less than the top half of games on Steam, especially since a whole lot of games do already just work.

Bottom line: If they have anti-cheat nailed and do the videos for a lot of games including all the most popular and squash a good amount of general bugs, particularly in top-tier games, most people may not notice a problem even if technically there are still a lot of games that don't work well in Proton.


Last edited by Purple Library Guy on 8 September 2021 at 4:59 pm UTC
I wonder why all the news sites jump on the assumption of James Ramey which he based on his interpretation of Pierre Griffais comment and simply take it as a fact.

I won't bet that every single game will work flawlessly on day one, but Valve seems to have confidence that their steamOS will handle mostly every important game.
Pierre Griffais was not the only one stating, that the whole library will be playable, which you can see at another IGN video about the deck:
https://youtu.be/hJoUs0pM4GU?t=34

Additionally we know that Valve has a different fork of Proton which has not been published yet and James Ramey also stated in the podcast, that CrossOver was not involved in the development of the deck.

We will have to wait until SteamOS is released and I really, really hope that this will be done before the deck is shipping.
Linuxwarper Sep 8, 2021
I believe Valve will achieve a level of compatibility to make most of people's library playable, assuming anti-cheat support comes. They don't have to reach 100%, but the percentage treshold that is acceptable at launch. With developers entering the fray, Proton compatibility will accelerate much faster in 2022 than it has the years before. Media spreading FUD about compatibility isn't helping.

Seems Boiling Steam's Ramey interview is making the buzz...

https://www.pcgamer.com/if-the-steam-deck-doesnt-run-your-entire-library-at-launch-valve-sees-that-as-a-bug/
Seriously..?


Last edited by Linuxwarper on 8 September 2021 at 7:51 pm UTC
Mohandevir Sep 8, 2021
I believe Valve will achieve a level of compatibility to make most of people's library playable, assuming anti-cheat support comes. They don't have to reach 100%, but the percentage treshold that is acceptable at launch. With developers entering the fray, Proton compatibility will accelerate much faster in 2022 than it has the years before. Media spreading FUD about compatibility isn't helping.

Seems Boiling Steam's Ramey interview is making the buzz...

https://www.pcgamer.com/if-the-steam-deck-doesnt-run-your-entire-library-at-launch-valve-sees-that-as-a-bug/
Seriously..?

I just meant that PCGamer felt the need to correct Ramey's opinion... Making the buzz is probably exagerated, but it had echoes at PCGamer:

That's not what Griffais was referring to though. When we interviewed the Steam Deck developers last month, and asked how Proton was shaping up, they were still very much of the idea that it should just all work come launch.
whizse Sep 8, 2021
View PC info
  • Supporter
I believe Valve will achieve a level of compatibility to make most of people's library playable, assuming anti-cheat support comes. They don't have to reach 100%, but the percentage treshold that is acceptable at launch. With developers entering the fray, Proton compatibility will accelerate much faster in 2022 than it has the years before. Media spreading FUD about compatibility isn't helping.
Yeah, I think I mentioned this before, but it will be interesting to follow media reports. Going by ProtonDB* 82% of games tested work. That's pretty good. But if you focus on the top ten, only 50% work. Is the glass half full or half empty?

Most of the top ten games with problems are most likely issues with AC/DRM so hopefully more will work on launch day, but for most users and reviewers the success of the Deck might come down to how well it runs one particular game.

Interesting times ahead!

* Not the most reliable metric, I know.
1xok Sep 8, 2021
Ok. If I get it right, the module could be installed with the Steam runtime(?) for the specific game? I mean included into the game's specific "container system" that Valve is starting to implement?

If Valve solves the AC problem with a kernel module (we don't know), this would be effective for all containers. The Valve Soldier containers with the Steam runtime are based on Linux Namespaces and thus all share the same host system kernel.

It is also possible that Valve solves this in a completely different way. We just don't know. But I don't see why they shouldn't be able to solve it.

If it goes in the direction of TPM, it will probably only work with Steam Decks. At least for now. Whether normal Linux clients will also be allowed to participate at some point could vary from publisher to publisher.

But I think it's impossible that the publishers lock Valve out. They will work with them. Just like with Microsoft, Sony and Nintendo.

And the Media Foundation will either be reimplemented or licensed from Microsoft. For reasons of antitrust law, certainly at reasonable conditions.
Linuxwarper Sep 8, 2021
I just meant that PCGamer felt the need to correct Ramey's opinion... Making the buzz is probably exagerated, but it had echoes at PCGamer:
That's not what I am reacting to. Ramney said Pierre's statement was that Deck can handle almost any game from a hardware pov. Meaning that Proton (software side) isn't necessarily going to run 100% of your Steam library, which is such a ambiguous statement as it spreads FUD around Proton's compatibility..and we know it's really good, and going to be much better by time Deck launches.

Then PC gamer is combining an answer they got from Lawrence Yang when asking a question and combining that answer with what they got from podcast with Ramney to make a headline as if it's a direct statement from Valve. Valve never said "If Deck doesn't run your entire library we see it as a bug". PC gamer combined statements from two places into one! They are glancing over actual statements from Valve which is that Proton works with tons games and that they are working diligently towards making 100% of games compatible. Not a promise but a goal.

Yeah, I think I mentioned this before, but it will be interesting to follow media reports. Going by ProtonDB* 82% of games tested work. That's pretty good. But if you focus on the top ten, only 50% work. Is the glass half full or half empty?

Most of the top ten games with problems are most likely issues with AC/DRM so hopefully more will work on launch day, but for most users and reviewers the success of the Deck might come down to how well it runs one particular game.

Interesting times ahead!

* Not the most reliable metric, I know.
It's not reliable, and Deck hasn't even launched and media is using ProtonDB as if it's accurate. It's not accurate. Deck has the same motherboard, CPU, PSU, ram etc across all SKUs. ProtonDB gets reports (and the reports can be faulty) from PC gamers with varying hardware and software. I think it's safe to derive that if a game's compatibility is silver on ProtonDB, it will likely be gold or better for Deck (with heavy work Valve is putting towards it). This is another thing that's glanced over, where ProtonDB is treated as if it's accurate for Deck's compatibility. It's not. And when they report on compatibility they barely mention actual issues, which Valve has made clear is anti cheat among others. Not a single word about anti cheat in that PC gamer article, and as we all know anti cheat is a huge compatibility blocker.


Last edited by Linuxwarper on 8 September 2021 at 8:53 pm UTC
BielFPs Sep 8, 2021
Ok. If I get it right, the module could be installed with the Steam runtime(?) for the specific game? I mean included into the game's specific "container system" that Valve is starting to implement?
Legally yes

But then the AC won't work because it requires high level access on the system (windows on this case) and because you're running inside a container, it can't escalate outside of it so will fail to execute.

So you have two possible ways

1 - Explicitly install the AC software by the users, with root access so the AC software can verify if you're not using any malicious software (and do everything else he wants because of the root permissions)

2 - The game you've asked to install will also ask to install the AC software as some kind of dependency (similar of what it does on windows), and you also have to conceive root access. It also would need the game to detect you're using a linux system and install a specific AC version for linux, because AC software for linux and windows are naturally different, due to the way they work.

What Valve can't do is bundle together with the blob of steam application, in this case they would have to do some kind of "blob inside the blob" and would be a risk for them (legal and practical) because they don't own the AC software in question.
whizse Sep 8, 2021
View PC info
  • Supporter
A quick google suggests that there are around 50,000 games
That's an interesting tidbit.

There's about 5000 (open and closed) game specific issues on the Proton bug tracker, and around 20 000 game reports on ProtonDB. So half the games on Steam have never been put through the ringer?

On a similar note, I have never seen anything that suggests that Valve themselves are doing any sort of large scale testing. Given that they likely sit on the worlds largest repository of Windows apps they could do regression testing on a VERY large scale. What fun that project would be to work on!
CatKiller Sep 8, 2021
View PC info
  • Supporter Plus
I wonder why all the news sites jump on...
Because they want the clicks.

The Steam Deck is a hot topic, but there's no actual news till December, when it's in testers' hands. Maybe Valve will release SteamOS 3, or a significant update for Proton, before then, for early testing, but they might not. Maybe a game dev will show their game running on a dev kit, but they'll probably keep that internal. Websites want to say something.

Misleadingly overhyping something, and then tearing it down later through something else misleading gives them controversy and drama, which they can turn into clicks. I'm really not surprised that it was the NME that was one of the first outlets to jump on a not-particularly-groundbreaking interview on a small Steam-enthusiast website: that cycle was NME's Standard Operating Practice back in the day.
Purple Library Guy Sep 8, 2021
I wonder why all the news sites jump on...
Because they want the clicks.

The Steam Deck is a hot topic, but there's no actual news till December, when it's in testers' hands. Maybe Valve will release SteamOS 3, or a significant update for Proton, before then, for early testing, but they might not. Maybe a game dev will show their game running on a dev kit, but they'll probably keep that internal. Websites want to say something.

Misleadingly overhyping something, and then tearing it down later through something else misleading gives them controversy and drama, which they can turn into clicks. I'm really not surprised that it was the NME that was one of the first outlets to jump on a not-particularly-groundbreaking interview on a small Steam-enthusiast website: that cycle was NME's Standard Operating Practice back in the day.
All of which is probably fine from a Steam Deck marketing perspective. I mean, what do these two article topics have in common?
1. There's a shiny new thing coming that might be great!
2. There's a shiny new thing coming that might have problems!

--> There's a shiny new thing coming!
Win.
Purple Library Guy Sep 8, 2021
A quick google suggests that there are around 50,000 games
That's an interesting tidbit.
Which, just so we're clear, I do not stand behind in any way.

There's about 5000 (open and closed) game specific issues on the Proton bug tracker, and around 20 000 game reports on ProtonDB. So half the games on Steam have never been put through the ringer?
Well, yeah, but specifically the games nobody cares about.
Grazen Sep 8, 2021
Sharing the link to the source story that is being referred to by most mainstream gaming sites but oddly enough not GOL.

https://boilingsteam.com/steam-deck-the-start-of-a-golden-age-for-linux-gaming/
whizse Sep 8, 2021
View PC info
  • Supporter
Well, yeah, but specifically the games nobody cares about.
Or games people are reluctant to admit to playing? No Proton reports for titles like The Seduction of Shaqeera!?
Derheim Sep 9, 2021
I used to be preoccupied with "how are we going make Linux as widely used as windows". But after quite sometime of going full Linux user/gamer I no longer see it as good thing.

First of all I think Linux will never be a mainstream OS, that is if we want Linux to conserve it's current identity (which in my opinion is what pulls people to the current niche market that it has, I'm ignoring server market here).

I think achieving a full "plug & play" Linux is not gonna be enough to make it go full bombastic.

At this point, I think the PC user all it know is, to install Discord they have to go to discord[dot]com and download a binary blob then double click it and install it, for them, "that's it". Doesn't matter that we know that typing "apt/pamac/yum install discord" is many times faster and simpler.

so I think Steam Deck will be a step up but is not gonna be it in terms of the Linux adoption that we are looking for.

And honestly, I'm OK with the current status, I think more than this will have to be achieved with sacrifices that many of us won't like.
Shmerl Sep 9, 2021
With state of the art free codecs, no one should be using Microsoft Media Foundation junk.

Besides, didn't MS back AOM? They should start using free codecs as well.


Last edited by Shmerl on 9 September 2021 at 3:23 am UTC
TheRiddick Sep 9, 2021
I heard VAC doesn't work either via proton, but I don't know which games use that. (counterstrike?)
ElectricPrism Sep 9, 2021
Expectations had to be checked sooner or later.

At this time of writing ProtonDB.com reports Gold++ on

Single Player:

Top 10: 90%
Top 100: 89%
Top 1000: 82%

Indie Games:

Top 10: 90%
Top 100: 80%
Top 1000: 76%

Local Multiplayer:

Top 10: 90%
Top 100: 82%
Top 1000: 48%

I suspect the lowest number in Local Multiplayer is 100% due to DRM tech which should be ready around launch time.

To to summarize the numbers, as of September 2021 it's already at 80-90% -- my estimation is that it'll be generally at least 90% or more by launch.

Taking into account Pierre-Loup Griffais' generalized speech, I think his statement is still appropriate as VIRTUALLY all games will work out of box, or at the very least the hardware can do it.

Of course, no one really expects to be playing VR titles like Half Life Alyx, or Beat Saber on their Valve Index Steam Deck -- so if you take what he said with that in context -- it's obvious that it was meant as a generalization or summary of their compatibility & targets. They said themselves that if a game doesn't run, they see it as a "bug" which needs fixing so the future is bright especially considering not even Windows plays 100% of windows games -- many Windows games from the old times don't even run on new Windows.

The only question I have is were we better off grounding ourselves in reality & putting the anchor down now, or popping the bubble later after people had these expectations for a longer period of time.

I think most reasonable and not hype-train people already correctly interpreted reality -- I think we're likely better off since I'd rather have these news articles come out now instead of at launch or post-launch to sour the party punch.

It's sad that we live in a day of clickbait and fearporn to sell news as entertainment, another round of hype videos like the initial ones from VIP before launch could help a lot too.

I think most people's expectations will be met and they will like the device a lot.


Last edited by ElectricPrism on 9 September 2021 at 6:41 pm UTC
ElectricPrism Sep 9, 2021
[quote=BielFPs]
All I know is: If this statement is just PR then SteamOS is over, because a lot of people already think that this is just steam machines all over again, and I'm worry that if Steam deck fails, then it'll have a huge negative impact in the progress of games running on linux.

Bruh, you got it all backwards, if people literally make the Steam Machines connection they will immediately have lower-expectations to begin with which will lead to pleasant surprise when it kicks ass.

The Doom & Gloom expectation problem is the same problem as the Hypetrain Choo Choo MF problem -- both are extremes not really grounded in reality.

The Deck is gonna be awesome, just not perfect -- does that clarify. They literally already sold out pre-orders for nearly an entire year. Can you imagine the infusion of developers, gamers, and average users into our community and what that will mean for all kinds of new github projects and the increase volume of gamers demanding support?

We good, homez. 2022 is gonna be lit.

Edit: Seeing some other sentiments in the comments I wanted to comment on

Yeah, I think I mentioned this before, but it will be interesting to follow media reports. Going by ProtonDB* 82% of games tested work. That's pretty good. But if you focus on the top ten, only 50% work.

See ProtonDB stats 1 comment above and toggle the category drop-down on ProtonDB.com -- both of those numbers are not representative of Proton's capability because they are obviously specifically lower in Multiplayer which have very high levels of DRM. If EAC or BattlEye didn't exist those numbers would be in the 80s and 90s as is seen in other categories like Single Player.

Note to Valve: If you're reading this, please for the love of yourself, DO NOT put any Epic Games blobs in my kernel. It doesn't make sense to give a competitor that kind of power, and it's a potential vector for Epic Games, Timmy Tencent & China to have powers over me as a gamer -- I would appreciate some clarification on this at some point in the future and expect to not literally have a Chinese Root-kit running on my Steam Deck. Thank You.

Also, as we saw when Wacom made a Companion & Mobile Studio that competed with their ex-partner Microsoft Surface Pro (Gen 4?), we saw YEARS of driver issues -- I honestly would not be surprised if Microsoft was purposely sabotaging Wacom devices as they competed with Surface Pro -- we saw Apple sabotage iPhone 6 users and degrade performance for no good reason and then get sued and loose for that exact kind of anti-user thing. I really wouldn't bat an eye if Microsoft found "oopsies" for potential Steam Deck games who might choose to run Windows 10 or 11 -- they need to have the expectation that their Steam Deck experience is not as guaranteed as the official Valve Steam Operating System 3.0

Also, I would appreciate if someday there was a voice interactive AI that we have the GlaDOS announcer do it lol


Last edited by ElectricPrism on 9 September 2021 at 6:48 am UTC
While you're here, please consider supporting GamingOnLinux on:

Reward Tiers: Patreon. Plain Donations: PayPal.

This ensures all of our main content remains totally free for everyone! Patreon supporters can also remove all adverts and sponsors! Supporting us helps bring good, fresh content. Without your continued support, we simply could not continue!

You can find even more ways to support us on this dedicated page any time. If you already are, thank you!
The comments on this article are closed.