Steam for Linux

Steam for Linux

[GNU/LINUX] Norbert 2022 年 7 月 17 日 下午 1:03
The negative trend of native linux games (with data)
Forgive me for opening a new thread
I think that this deserves public attention.

Few days ago I was browsing Steam's "new releases" page,
this one: https://store.steampowered.com/search/?sort_by=Released_DESC&category1=998

and at a glance
I had the gut feeling that there were less native linux games than usual.

Today I collected some data and produced some stats,
here they are:

Year / Percentage of native linux games compared to the total releases
2017 - 22.4 %
2018 - 17.9 %
2019 - 15.4 %
2020 - 13.4 %
2021 - 12.0 %
2022 - 12.0 %

It's a negative trend
and now you know it.

I am painfully aware that less than 1%
of the already 1%
of the userbase here on Steam
care about having software which is independent from another OS
instead of being 100% vendor locked by microsoft.


Data is directly from steam
from the releases page
only games (dlc, software and other types are excluded)

txt by month, year, mysql dump and csv dump:
https://we.tl/t-b7fUq17Phs
for you to double check and play around

Sorry about that,
have a nice day.
< >
目前顯示第 61-75 則留言,共 90
Zyro 2022 年 7 月 25 日 上午 5:53 
引用自 GNU/LINUX Norbert
引用自 Zyro
Prediction is hard, especially regarding the future, they say.

That's why you use statistics to make a projection.

And statistics is, by the way, a science.

Your projection seems to predict as well though that at 90 or 99 or 100% Linux users, the percentage of native would be very low, probably zero. So I think there's something more to it than some unlimited linear extrapolation.

Mind that I don't disagree with your conclusions. (I think I even said something similar in this thread.) It's just not as cogent a conclusion you seem to believe.
KrazyOne 2022 年 7 月 26 日 下午 5:19 
Introduction to Proton has resulted in the decrease of native games. Why? You ask, because the financial burden of converting lines of codes to correctly function on Linux is no longer necessary with Proton. Since Linux already has proven for most titles to cost more in maintenance than profit, clear to understand what is happening.

Why Mac has saw virtually all their game titles disappear over the years.

Cost effective business practice leads to profit not lost. So long as Windows has 80%+ user base advantage developers aren't going to risk profit lost on a small market share.

Hence my assumption of 20% user base means their a market share that no longer loses money, but may break even or profit depending the titles. Regardless of how you feel the current data shows, 2025 introduces the next Window Exodus crisis. We don't know if Windows caves and changes their Windows 11 requirements. If they don't you are going to have a 10 to 20% reduction of Windows users. Most will use Linux as it free and simple. Others will likely upgrade or buy a new pc.

Low income gamers aren't going to buy a new pc to use Windows 11 to play a game if they use current pc with Proton on linux. Thus a new wave of users will flood into Linux. Once this wave grows large enough developers are going to have to acknowledge the OS monopoly called windows is no longer the market giant.

Or Linux never gets adoption that can happen too.

Either way everything is driven by Money in business. Fact of life will never change.
最後修改者:KrazyOne; 2022 年 7 月 26 日 下午 5:22
[GNU/LINUX] Norbert 2022 年 7 月 27 日 下午 3:32 
引用自 Zyro
引用自 GNU/LINUX Norbert

That's why you use statistics to make a projection.

And statistics is, by the way, a science.

Your projection seems to predict as well though that at 90 or 99 or 100% Linux users, the percentage of native would be very low, probably zero. So I think there's something more to it than some unlimited linear extrapolation.

Mind that I don't disagree with your conclusions. (I think I even said something similar in this thread.) It's just not as cogent a conclusion you seem to believe.

With Nintendo having sold 107M Switch units in 5 years
and the Steam userbase being 135M active monthy users,
Valve could very well sell 27M Steam Deck units in few years
and so the GNU/Linux userbase would be 20%
(well, 1% linux and 19% steam deck)

Could this increase the number of native games ?
I don't think so.
They would just optimize for proton and the steam deck.


You all seem to forget that most games dev hated us
with ardent passion.

They hated everything about us:
the OS
the distros
the maintenance
the community

From industry veterans to indies
from platforms to publishers.

They hate everything about linux.
i_nive 2022 年 7 月 27 日 下午 11:53 
引用自 GNU/LINUX Norbert
...
From industry veterans to indies
from platforms to publishers.

They hate everything about linux.
// Promo time
I must intervene to say a few good words about:
- Paradox (itself hated by many for their DLC-as-a-service policy) having a game engine of their own (Clausewitz[en.wikipedia.org]); they have a day one Linux build. But see https://www.gamingonlinux.com/2019/06/paradox-interactive-on-linux-support-its-being-done-on-a-case-by-case-basis/ about latest development (Paradox putting the publisher's hat)
- Illwinter[en.wikipedia.org], the guys famous with Dominions series (also available on Steam) are developing on Linux(!!!) and then porting to the rest of OS-es.

So there it is, not a black/white world. There will always be rebel scum, https://www.youtube.com/watch?v=MRoBv-9ueMo
KrazyOne 2022 年 7 月 28 日 下午 6:06 
引用自 GNU/LINUX Norbert
引用自 Zyro

Your projection seems to predict as well though that at 90 or 99 or 100% Linux users, the percentage of native would be very low, probably zero. So I think there's something more to it than some unlimited linear extrapolation.

Mind that I don't disagree with your conclusions. (I think I even said something similar in this thread.) It's just not as cogent a conclusion you seem to believe.

With Nintendo having sold 107M Switch units in 5 years
and the Steam userbase being 135M active monthy users,
Valve could very well sell 27M Steam Deck units in few years
and so the GNU/Linux userbase would be 20%
(well, 1% linux and 19% steam deck)

Could this increase the number of native games ?
I don't think so.
They would just optimize for proton and the steam deck.


You all seem to forget that most games dev hated us
with ardent passion.

They hated everything about us:
the OS
the distros
the maintenance
the community

From industry veterans to indies
from platforms to publishers.

They hate everything about linux.

I think you are missing the point most company will do the minimal work for the most profit. Why EA jump on board with Proton to allow more money with lest work. Nothing changes until Linux user base becomes a Market share worthy of investing for. Not to say there wont be outliers to the every data point.

20% market share might seem like an arbitrary number, but it has a huge impact on Financial burdens. I mean look at MAC users once gaming community died most went back to windows. As weird as it is Gamer Industry is driving force in the PC/Console markets. They out weigh most other technology base profit margins.

Why I said adoption is key to native development. Until Linux can become a threat financial to Windows Adoption alone wont change Native vs Port production.

Proton is allowing Adoption to happen sooner than projections would have forecast. After all I still believe key to mass adoption of Linux is adoption of Gaming 100% on Linux. Corporations have no need for gaming thus are fine with status quo of Linux.
[GNU/LINUX] Norbert 2022 年 7 月 29 日 下午 3:08 
meanwhile

on Steam's "Top Releases of June 2022"
of the Top 20 releases, only 1 has a native port
(a 2d action game)

That is 5% of the total.
Enigmatic 2022 年 7 月 29 日 下午 6:25 
The less native games the less I'll spend on my Desktop. Less reasons to hop on steam less chance I buy stuff from the store.
最後修改者:Enigmatic; 2022 年 7 月 29 日 下午 6:27
CobraA1 2022 年 7 月 30 日 上午 7:31 
引用自 GNU Eklypse
The less native games the less I'll spend on my Desktop. Less reasons to hop on steam less chance I buy stuff from the store.

You think developers and publishers care?
KrazyOne 2022 年 7 月 30 日 下午 6:48 
Native will never be profitable until adoption of Linux reaches critical mass. Until then you are going to see push for Proton over Native.
[GNU/LINUX] Norbert 2022 年 7 月 30 日 下午 9:05 
引用自 KrazyOne
Native will never be profitable until adoption of Linux reaches critical mass. Until then you are going to see push for Proton over Native.

Proton is 4 years old and the GNU/Linux userbase on Steam didn't improve in any surprising way above the usual linear and boring way.

Instead of repeating your unsubstantiated theory like a broken record
why don't you spice up the discussion and calculate how long would it take for Valve to produce enough Steam Decks to reach your famous 20% linux share ?

Don't forget to add variables like Taiwan having 'political' problems
(I'm sure that you are well educated on this topic as well)
CobraA1 2022 年 7 月 30 日 下午 9:28 
引用自 GNU/LINUX Norbert
引用自 KrazyOne
Native will never be profitable until adoption of Linux reaches critical mass. Until then you are going to see push for Proton over Native.

Proton is 4 years old and the GNU/Linux userbase on Steam didn't improve in any surprising way above the usual linear and boring way.

Instead of repeating your unsubstantiated theory like a broken record
why don't you spice up the discussion and calculate how long would it take for Valve to produce enough Steam Decks to reach your famous 20% linux share ?

Don't forget to add variables like Taiwan having 'political' problems
(I'm sure that you are well educated on this topic as well)

We've all got access to the data you presented. Do you think you will change minds by repeating the same tired arguments?
[GNU/LINUX] Norbert 2022 年 7 月 30 日 下午 11:35 
引用自 CobraA1
引用自 GNU/LINUX Norbert

Proton is 4 years old and the GNU/Linux userbase on Steam didn't improve in any surprising way above the usual linear and boring way.

Instead of repeating your unsubstantiated theory like a broken record
why don't you spice up the discussion and calculate how long would it take for Valve to produce enough Steam Decks to reach your famous 20% linux share ?

Don't forget to add variables like Taiwan having 'political' problems
(I'm sure that you are well educated on this topic as well)

We've all got access to the data you presented. Do you think you will change minds by repeating the same tired arguments?


Yeah God forbid your bring something new to the table ...
KrazyOne 2022 年 7 月 31 日 下午 5:20 
Has nothing to do with new, your date shows exactly what was expected. Developers will side with easy Money over less profits. Until user base reaches critical mass you wont see native development go up.
[GNU/LINUX] Norbert 2022 年 7 月 31 日 下午 5:53 
引用自 KrazyOne
Has nothing to do with new, your date shows exactly what was expected. Developers will side with easy Money over less profits. Until user base reaches critical mass you wont see native development go up.

You keep repeating that
without providing the *how* and the *why*


*how*
can the Steam Deck push toward critical mass adoption
if at this pace, it'd take 27 years to produce 27M units
(20% of the steam total userbase)
assuming there are even 27M customers that wants to buy it.

Steam Deck (AMD's SoC)
is competing for TSMC's wafers production with Apple, Tesla, Sony, AMD Epyc, Nvidia
which have products with much bigger markets and higher profit margins
than a toy.


*why*
should games developers build native ports
when the linux userbase on steam hits your magical 20% ?

Because of performance ?
Proton works just as well, if not better, than native ports.

Because the community asks for it ?
If anything, this thread confirms that nobody cares about native ports.

Because of such great standards among linux distributions ?
Sure, write once, fix it at every major upgrade.
(i don't know if the Steam Runtime fixed this problem, hope so but the reputation damage remains. Too little too late)

Because of brownie points ?
For a community that send death threats to the devs if you dare not to deliver your product ?
(remember the Witcher 3 drama)
CobraA1 2022 年 7 月 31 日 下午 6:26 
If a target platform becomes large enough, it does make sense to start targeting it. The tools work better, you can take advantage of platform-specific features, the performance can be tuned better.

Proton doesn't do anything magical that can't be done in a native build. For the most part, it's just taking advantage of the fact that many developers didn't make a great port to begin with.
< >
目前顯示第 61-75 則留言,共 90
每頁顯示: 1530 50

張貼日期: 2022 年 7 月 17 日 下午 1:03
回覆: 90