Quoting: MohandevirLooking at this:

https://www.phoronix.com/scan.php?page=news_item&px=Gallium3D-NIR-Link-Opts [External Link]

Valve still cares and is still investing in Linux. They wouldn't bother with that otherwise.

Something will happen regarding this, I'm not going to put all my faith in that, but it smells like an AMD based Steam Machine or something similar is on the way... Maybe... Hopefully...

How I would like to see Valve release it's own Steam Machine (no third party involved) with only one or two hardware specs (like PS4 Slim and PS4 Pro) along with a complete SteamOS overhaul.

I would not completely rule out a Valve-produced device, but it would be against their general reasoning regarding the value to gaming of an open market of upgradeable hardware.

To produce something compelling at a similar price point ( c.f. XBox / PS4 ), Valve would likely have to follow a similar route, contracting AMD to produce semi-custom designs. They could yet do that; it depends on where the market pressures fall.

Valve support Linux because it is the leading platform amongst those that are unowned and uncontrolled. Valve have clearly helped make Linux a more viable platform for gamers, but I think it would be a mistake to assume this means they view Linux the same way evangelical Linux supporters do.

View this comment - View article - View full comments

Quoting: liamdawe

Quoting: subSorry, but I don't see the point always mentioning that the absolute number of Linux Steam users "still" grows. It sounds like saying "It's not that bad".

I'll be honest, really struggled with replying to this. How can you not think that's important? It's extremely important to know when dealing with percentages if the overall number you're dealing with is bigger and where that change has come from. It would be a completely different story if the Linux percentage dropped as much as it has, without seeing the trends behind the percentages. That's the entire point of the article, to attempt to explain how it's not actually as bad as some people keep saying it is.

Again, see my other reply with PC World as the example of why such things are important to be clear about.

Quoting: subEven if some dev plans on absolute numbers, it wouldn't be clever from an economical point.
as every man hour spent to improve experience for the potentially 100 times larger user base is most likely far better spent.

As I said in the article "nothing has actually changed". This has always been the case and even if we went up to 2%, it would still be the case.

I get this is a hot topic for some people, but it is important to highlight and look into the actual reasons behind change and understand them.

Yes, this is all true, and any developer that can produce a Linux or Mac OS game with minimal cost ( perhaps because they already use a x-platform games engine ) will probably do so almost regardless of user base gyrations.

But games companies are still businesses, and many will still follow the money. That is why we have seen much free-to-play, fragmented DLC rather than coherent expansions, on-line competitive multi-player, pay-to-win, random ( often bad ) content purchasable loot crates, and so on.

These changes directly improve revenue, so they make their way into games, even though most players dislike at least some of them ( most of them in my case ). My main concern with these stats ( as per your analysis ) would be how a massive additional player base from China would skew the games marketplace? I'm certain that likes/dislikes and accepted behaviour will be different, and suspect that will affect future decisions in the industry.

View this comment - View article - View full comments

SDL is a great basis for starting cross-platform code. It's very helpful for SDL to provide access to an initialised Vulkan drawing context, mainly because you can also use SDL input, event and device handling which is a good multi-platform abstraction. The threading and utility APIs are good too.

But SDL doesn't really help learn Vulkan ( or any other graphics API ), since it has drawing abstractions only for limited 2D and sprite animation. The complex 3D and compute operations generally used in simulations and games still need to be learned for each graphics API you want to support. In other words, you need to provide your own abstraction over all the graphics APIs you want to support, but you don't need to be bothered about which OS platform you are running on.

View this comment - View article - View full comments

Quoting: Purple Library GuyI think mostly what annoys people about Stallman is that, while his actual analysis and positions are cogent, consistent and rigorous, he goes around insisting on actually taking them seriously even when it's inconvenient, and suggesting that other people should do the same. It makes us uncomfortable since most of us have no intention of giving up any convenience in the pursuit of liberty or any other ethical good, whether in the arena of computing or any other.

Richard was a very important catalyst in the 1980s in getting people to think differently, and in kick-starting the platforms we prefer to use today.

However, the FSF view is essentially that ALL software, without exception, should be available as source, gratis, and removing all rights of the software author to determine financial beneficiaries.

To many people in the software industry, FSF dogma was/is not much better than the being controlled by the dominant companies. By failing to balance interests among all producers and consumers, the FSF view becomes not so much uncomfortable, but impractical.

The FSF view remains a radical outlier, which is useful to have for comparison. However, much of the important open source software is developed by paid-for developers working on behalf of producers that collaborate on mutually useful code that they probably could not justify individually. Still open source, but not free in the hard-core GNU sense.

View this comment - View article - View full comments

Quoting: pb

Quoting: razing32

Quoting: etonbearsIt is notable that any "philosophical advantage" of open development does not yet seem to have produced a significant body of high-quality games.

If I may , wouldn't this have more to do with lack of other professions joining in also ?
Sure we have coders who can script a great game.
But without artists , musicians , composers , art directors , voice actors and so on , how far can we truly get towards a great game.
Just my two cents.

It's easy to contribute code under permissive licence, because it's usually only useful in that one place anyway. Artists can have a bigger dilemma, because once they create a sprite or a tune for an open-source game, they're likely to see it floating around the web, reused 100 times without as much as attribution.

Not sure I quite understand. You are either giving stuff away for the public good, or you are not. It doesn't really matter whether it is code, images, video or audio. To my mind, the only properly permissive open source "licence" is Public Domain. By contrast BSD and MIT are permissive with attribution, and GPL is ideologically restrictive.

View this comment - View article - View full comments

Quoting: razing32

Quoting: etonbearsIt is notable that any "philosophical advantage" of open development does not yet seem to have produced a significant body of high-quality games.

If I may , wouldn't this have more to do with lack of other professions joining in also ?
Sure we have coders who can script a great game.
But without artists , musicians , composers , art directors , voice actors and so on , how far can we truly get towards a great game.
Just my two cents.

Yes, that's true. For the type of game you have in mind the artistic personnel greatly outnumber the coding team ( if you assume coding is not artisic ). However, it is entirely possible to produce high quality games where most or all game assets are created procedurally by the coding team.

The real reason there are few significant open source games is simply the cost in time to organise and create them. Doing it for free in your spare time becomes a never-ending slog which, if it ever does complete, produces something that is generally out-of-date. Many people start games, but few are finished.

View this comment - View article - View full comments

Quoting: Plintslîcho

Quoting: jensNo matter of Steam, DRM, always online only, Nvidia closed source only, I don't care what.

I don't want to sound rude but a genuine question: why are you using Linux then? Wouldn't it be easier to simply run Windows, where all the cool game are anyway?

Quoting: jensI wish that people would only want to play some cool games[...]

The way I see it a lot of people don't use Linux primarily because it's free (as in free beer) but because there is also a certain philosophy to it (open source, Linux being free as in free speech an all that). Naturally, a lot of people are also very vocal when it comes to defend those values.

The Linux ecosystem supports more than one philosophy, and consequently there's more than one reason that people choose to use Linux.

If you have used Linux long enough to remember, you will recollect that the term "Open Source" was specifically coined ( by Eric Raymond and Bruce Perrens, I think ) to advocate the benefits of the "bazaar" open development model. However, it was also coined to advance an alternative "open" narrative to the GPL, which is a restrictive licence, and was seen as quite poisonous by some.

For myself, I find the Richard Stallman/FSF worldview very unappealing, and it is the self-righteous moths that flutter in Richard's candlelight that I usually find most irritating ( I suspect these are the ones that annoy Jens as well).

My personal reasons for prefering Linux are that choice and control are vested in me, and not in the provider of the OS. Although I am comfortable with C/ASM programming, in 20+ years I have had almost no reason to modify any Linux ecosystem code ( ironically, when I have modified code, it is in order to compile it with MinGW for Windows ).

I do value the ability to read the source of open code libraries, since I can better judge whether I want to use their implementations, or would be better off creating my own; but this is entirely secondary to the issues of choice and control.

Beyond the basic availability of an operating system under my control, I am entirely pragmatic. Anyone can offer whatever they like, under whatever conditions they like, and it it then my choice as to whether I buy or not.

I would be the first to agree there have been many distastful decisions made by games developers/publishers, but they have the right to make those choices ( assuming they are legal ), in order to ( at least ) break even to continue making games. No-one is forced to buy games.

It is notable that any "philosophical advantage" of open development does not yet seem to have produced a significant body of high-quality games.

View this comment - View article - View full comments

Quoting: Sir_DiealotI don't know why you are listing reasons and excuses for companies to not bring out stuff for Linux. Where there's a will... And as others have said, EA could do it easily if they wanted to. They might not be the worst games company in the US anymore since WB is giving them a run for the money, but I'm still not going to buy any newish titles from them. Not as long as they require players to be online and similar BS.

Of course EA could do something similar to Valve and build a fixed Linux target to port Origin and then their Frostbite titles. It is certainly not beyond them, and from past comments, Johan probably wants to support Linux.

But EA don't need excuses; unless they see clear financial motivation, Linux support it is not likely to happen. Evidence from this thread suggests that quite a few Linux gamers would shun EA Linux games on points of principle, which will not encourage EA to change their view.

Even publishers with existing Linux support cannot be guaranteed to do so in the future. Egosoft/Deep Silver were an early supporter of Steamplay and ported all 4 of their "current" games to both Linux and Mac, but they have recently shown off their X-Rebirth VR beta, and upcoming X4 engines, both of which are Vulkan only, and both are on Windows only. When asked about other platforms Egosoft were supportive, but suggested that it was not their decision. Make of that what you will.

View this comment - View article - View full comments

Quoting: Kimyrielle

Quoting: iiariI'm actually more optimistic for Linux gaming in the future than even today, as I think eventually gaming will largely be streaming, and as long as the streaming clients exist for Linux, then we'll be good. That way, if anything, we'll have an even bigger selection than we do today...

I don't think streaming games per se has much of a future. In contrast to music (tiny files) and movies (you usually watch them only once) there is no compelling reason to stream games over installing them. You -still- need a gaming capable PC to run streamed games, and the performance of locally rendered games will ALWAYS be better compared to sending data to the other side of Earth and back and keeping everything in sync. But we will probably see more client/server online games like MMOs and online FPS games. The days of pure offline games might actually end sooner rather than later.

If streaming games become the norm ( and I actually think they will, but not for a while ), it will be because the companies producing games choose to do it that way, not because gamers necessarily think it a good idea. The driving forces are fairly strong - elimination of piracy, greater control of revenue generation, reduced support and development costs. But it will need a better global infrastructure than we have now, and there will likely always be a niche market for installable titles that someone will service.

View this comment - View article - View full comments

Quoting: Purple Library GuyProgramming for all those things? Linux. And given that, a lot of education related to programming also teaches Linux. Game designers are part of that community and came through that education, so they too are likely to be familiar with and involved in Linux. There's a mindshare there independent of the actual Linux desktop market; a lot of these people want an excuse to make games available for Linux.

There are still a lot of PC game developers that their gained their experience between 1990 and 2005 when Windows was almost unchallenged. But since then Linux/Unix have certainly been ascendant in the technical community.

However, the money people in gaming, and business in general, remain primarily Windows-oriented, as are PC gamers themselves. With gaming ( and other software ) increasingly bypassing the desktop to mobile platforms and streamed services, and PC/Laptop sales seemingly in a long slow decline, I still struggle to see where the growth of Linux on the Desktop will come from.

View this comment - View article - View full comments