Steam's top releases of May show why Steam Play is needed for Linux

Quoting: GuestInstead, the correct message to send is we will not send you money until you support us like every other gamer gets to enjoy. We demand Linux support. That's how you get more Linux support.

Hate to break it to you so directly, but that is wishful thinking without any base in reality.
Uhhh, about 1% of gamers demand Linux support. Yes, that will sure convince developers. As a 1% group demanding anything has always been very successful, I'm sure.
Seriously, it convinces barely anyone even here and we're all linux gamers ;)

No, Linux gamers will be ignored by most developers without a personal interest in the platform itself - until there are enough linux gamers to actually listen to them, to make the effort viable.
And that number can only be increased by growing game support. Which - since a larger percentage of native ports won't happen without a larger target audience first - can only be provided by Wine/etc.

Though, of course, the increased support of Linux by engines naturally also helps here, as it lowers the perceived barrier of having to maintain another platform.

Obviously, preferably buying games with Linux support has a positive influence, as has (nicely!) asking for future Linux support of game X. But beyond that, buying a Windows-only game on Linux and playing it there will show a developer that there's an audience here which has the one thing any developer desires most: $$$.
Which might convince a developer to give the platform a try with a proper release. Or not.
But at least it is a chance, in contrast to a handful of people screeching "No tux no bux!" in their ear, which only leads to the "ignore those Linux freaks"-reaction you can regularly see wherever the topic comes up.


Last edited by TheSHEEEP on 28 June 2019 at 4:49 pm UTC

Quoting: GuestInstead, the correct message to send is we will not send you money until you support us like every other gamer gets to enjoy. We demand Linux support. That's how you get more Linux support.

In "supply and demand", the "demand" doesn't stand for "users sending demands". It stands for the "quantity demanded", which in this context is most closely associated with "market size".

I agree that (eventually) Linux gamers should prefer Linux native games, but I don't think we're at a point where it matters yet. Proton, if anything, might be helping that (or at least not making it worse) by increasing Linux's gamers market size and (in the future, potentially) increasing their demand power. IMO, if nothing else, it's not making things worse.


Last edited by GermainZ on 28 June 2019 at 5:20 pm UTC

The situation with Wine has always been 2 sided, it is both beneficial and damaging to Linux, which in many cases has been the former. While Steam Play/Wine helps users switch to Linux without sacrificing their favorite games, meanwhile from the devs end, they get to piggyback and profit from a platform they don't support.

In all fairness, I can sympathize with small companies where they lack the resources or the experience for Linux and Wine becomes the better than nothing solution. The large companies on the other hand, they have the funds but don't want to invest in Linux and id Software is a prime example of this.


Last edited by Avehicle7887 on 28 June 2019 at 5:08 pm UTC

Quoting: GuestIt's been almost a year since Proton came out (August 2018).
Where is this fabled growth in market share? Did we even reach 1% of Steam users?

Well, it certainly wasn't growing before Proton, and the AAA ports that had been a feature of the Steam Machine days had long since dried up (discounting Aspyr and Feral of course).

We're holding steady over the last few months, and the lack of increase can be explained by reading this article.

Why would vast numbers of Windows users switch when only a tiny fraction of new games work OOTB?

That's why improvements in Wine/Proton/EAC compatibility are so important. Once gaming on Linux isn't as much of a sacrifice, there'll be less barrier to switching for those who'd rather be on Linux but for their games.

Quoting: GuestThats not how economics works. Why would any business create a product for a market that does not exist or is irrelevant in terms of the size of that market? Businesses always look at the size of the market and how much they could possibly get in sales vs the cost of servicing that market. The truth is for many game devs, the outgoings vs projected revenue isn't worth it. We have to show them that the market exists. The only metrics that really matter are sales metrics

Imagine you and a small group of people want chairs. All that's produced is tables. Then half of the group starts going to the retailer, buying tables, but tell everybody "I will use it as a chair! A chair, understand!"
What would be my reaction as a table manufacturer?
I'd say think: "Fine, have fun."
It's no incentive at all for me to produce chairs when I know people will still buy my stuff if I wont - quite the opposite.
What they are showing is not a market for chairs - it's a market for "whatever four-legged you make".

I understand that you're saying that less than a percent is not incentive enough to get chairs many Linux games - it seems you're right. You're hoping for an increasing Linux percentage - I don't see it. But either way, if your long term goal is native Linux games, people must demand native Linux games sooner or later.


Last edited by Eike on 28 June 2019 at 5:22 pm UTC

This is going in circles. Kind of like Linux gaming before Proton :P

1. Conflating two meanings of the word "demand" isn't helping anyone.
2. Yes, it's clear that if publishers can get money from Linux gamers without supporting Linux, they will do so.
3. It's also obvious and has been shown that the current number of Linux gamers doesn't matter to most publishers. They don't really care either way if we don't buy their games. Crying about it gets us nowhere.
4. We're not going to break out of that paradigm without a larger userbase. Proton can help grow our userbase by removing barriers to adoption, i.e. making Linux more attractive as an OS for gamers who are on the fence.
5. Once we have bigger numbers, maybe we can start mattering to publishers and convince them to make the effort. Or maybe our own compatibility efforts will have gotten so good that it won't really matter anymore in most cases but publishers will pay attention.

I think there's a middle road we can take as advocates here. First, assuming a limited gaming budget, I do prefer to prioritize games that have actual Linux support. Put my money where my mouth is. But also, what we can do to take advantage from Proton is incite developers to support us through Proton. Everyone likes to take the path of least effort and make more money. If a developer can't be bothered to port to Linux but can be shown that a few fixes to enable Proton would generate enough sales to be profitable, that's a step in the right direction for us.

We're dealing with proprietary software here anyway. Whether it's native or not makes little difference as long as it's properly supported. If a dev wants to support a compatibility layer rather than a full port, I say go for it. As long as we have a promise of support, we're not so much of an afterthought anymore. And then if our numbers continue to grow, there will be incentive to make sure our experience is on par with other platforms, and if that implies an actual port or changing middleware, we'll be more likely to get it.

I agree more or less with Swiftpaw.

Proton and wine in general is good to play games you already have that only have a Windows version but no Linux version.
But the demand for Linux games is shown by us buying and playing games on Linux, and if we want support then we should buy games that support Linux.

If a developer/publisher states that they officially support Proton, similar to that we also have seen wine wrapped games in the past, then I would maybe buy the game if I also am interested in it.

But buying Windows games with Steam on proton and then playing it without support is questionable. There have been numerous example of games that work for a while and then suddenly not any more. Most of the times because of anticheats.

So I will not buy any new game that does not have a Linux version, sometimes I make an exception of old games that work well in wine/proton or that have a open source engine reimplementation. So I am not 100% no Tux no bux. But I have mainly bought Linux games for over 5 years now, getting some windows games due to bundles or as giveaways.

I am a big fan of Dirt Rally, and have not bought Dirt Rally 2 yet since I still hope that there will be a Feral port, but who knows if that ever will come. For now I still enjoy Dirt Rally, and also Dirt 4 but its different and I like Dirt Rally more.
Sooner or later I will get Dirt Rally 2, but if its the windows version then first after a year or more and on a steap discount.

Regarding the market demand try to see it as this:

We are approximately 1 million Linux Steam users (I do not have en exact number but these where put out previously based on the percentage and overall Steam users reported by Valve.)

Just to have example number lets say that each user buys for 20€ a year on Steam, some buy for a lot more, but there are also people that only play free to play games.

That would be a market value of 20 million €

Lets say we have got 200k more users due to proton and that is good, but proton probably also has an effect on users already gaming on Linux. So now Linux users spend 10€ for Linux games (games with Linux supported on the store) and 10€ for Windows games. So Windows sales will be reported as Linux sales, but overall the market value for Linux supported games will then be 12 million € so a big drop. (I have no idea how much an average Linux gamer spends, so those numbers are made up, I myself spend a lot more.)

This is why we fear that Proton will lower demand for Linux games. Getting some more numbers will help to understand where we stand.

Quoting: Sir_DiealotOr there are fewer games for Linux released now that SteamPlay is around. Who can say with certainty? Historical data and statistics probably can. Who is going to dig in?

I've been repeating this mantra for 26 years now. What good it has done. I've missed out on a lot and I've come to the conclusion that it will not work for Linux users until the % of users is high enough.

Right now, Linux users are no where near high enough to make anyone care and anything that helps bring people over to Linux is a good idea.


Last edited by KohlyKohl on 28 June 2019 at 5:51 pm UTC

Quoting: SalvatosThis is going in circles. Kind of like Linux gaming before Proton :P

Snip ---

We're dealing with proprietary software here anyway. Whether it's native or not makes little difference as long as it's properly supported. If a dev wants to support a compatibility layer rather than a full port, I say go for it. As long as we have a promise of support, we're not so much of an afterthought anymore. And then if our numbers continue to grow, there will be incentive to make sure our experience is on par with other platforms, and if that implies an actual port or changing middleware, we'll be more likely to get it.

Is there actually any game where the developer/publisher have said they will support Proton? I have not read about that, but seen some examples of Developers saying that Linux users just should use Proton since other users reported it works well.
I do not consider that support, but if they actually say they will test any patches under Proton to make sure it still works then I would consider it as support.

Quoting: GuestIt's been almost a year since Proton came out (August 2018).
Where is this fabled growth in market share? Did we even reach 1% of Steam users?
The only thing Proton is doing is strengthening the position of Direct X as the only viable APIs in PC games development, no need for SDL, Vulkan and so on... there is Proton after all!

A few things to note, Proton has been "out" for nearly a year, but it's still very much hidden on Steam. Unless you're going looking for it, you don't really know it exists. It still has a lot of work ahead of it right now, it's early days. If anyone seriously expected a shift in behaviour in less than a year, they're not being realistic in any way.

As for your other bits, well, Vulkan is needed for a big part of Proton remember, so that makes no sense. SDL is also used a lot outside of Linux FYI.