Quoting: dubigrasuNaive question, what's with the name Proton? Must be some reason behind this name. Just curious.

Whatever the reason, it's a pretty positive name.

View this comment - View article - View full comments

Quoting: aFoxNamedMorrisLiam, I'm pretty sure that only whitelisted Windows titles count as Linux purchases. Could be wrong, but I was under that impression from their initial announcement.

I'm pretty sure no. What they said to Liam was, "Playing on Proton counts as playing on Linux". Seems pretty unambiguous. I don't see why it would matter whether the thing you played on Proton was whitelisted or not, you're playing on Proton.

View this comment - View article - View full comments

Quoting: silmethRegarding the Linux ports… I find it pretty ingenious that Valve released the Proton-for-Steam-Play now. If they had done this – provided their own Wine for launching non-native games – when they launched SteamOS, I think what would have happened would be lack of native Linux releases – most developers would be happy they are on SteamOS thanks to translation layer, perhaps they would put some effort to ensure their Windows game is Wine-friendly, but that’s all. Perhaps SteamOS could gain a bit more popularity that it did in reality, but I don’t think it would be a great success in this scenario either…

Instead, Valve kept Steam on Linux without an easy play-Windows-games-on-Linux mode for 5–6 years, during that time they’ve released SteamOS which did not provide any easy way to install non-native games (one had to manually install Wine and separate Windows version of Steam for that), and that forced quite a few developers who did not want to be left out of Valve’s new platform to actually start properly supporting Linux. It also made Feral and Aspyr focus on porting to Linux.

And only now, after releasing native Linux version became a pretty standard practice (at least for some game publishers) and stopped being a weird exotic exception to overwhelming number of Windows-only releases, only after over 3 thousand games on Steam have native versions, they release built-in Proton that allows one to install and play the remaining Win-only games.

Valve already brought some publishers to Linux, in doing so proved that porting to Linux is technically feasible, that games do work on Linux, and that distros diversity isn’t really a problem.

Only now, when porting to Linux is an established practice, thanks to this feature – which seemed obvious to me already around 2013–2014 that it’d be needed to convince more people to use Linux / Steam Machines – Valve, I believe, will actually bring new players to Linux. How many will be convinced, though, remains to be seen. But I think the future is bright.

I think the timing isn't quite so ingenious as you suggest. This may be good timing . . . but it's more making a virtue of necessity. Back then it simply wouldn't have worked. It's not just that Wine wasn't ready, graphics drivers weren't ready, Vulkan wasn't there. Pieces have been coming together gradually, some of them with strong encouragement from Valve, and now they're in place--or, close enough that getting this thing going is less a disappointment and more an impetus to making it more solid.

I still think there could be a re-release of Steam Machines in the works--although given Valve time, probably not right away. Imagine if Steam Machines had had this Steam Play feature and games running Vulkan--there might have been quite a bit more uptake, although there were other issues.

View this comment - View article - View full comments

The googly eyes make them look a bit like Kerbals.

View this comment - View article - View full comments

Quoting: GuestI'm sure I will get attacked for this, but here goes. As per usual, disclaimer, my thoughts are my own and not that of VP, etc..

A big issue here is, does Proton really mean "Linux support" ? I don't think it does. Here's why.

When you get a port to Linux, even if it's by a third party, a lot of work has been done to bring that game over to the platform... even if a D3D wrapper has been used etc (i'm not going to go into the "lazy wrapper" argument). The porter is selling you a product that is certified by them to work with Linux, and has official backing of the original developer/publisher. If something doesn't work, or breaks, there is an official support method available. It is someones job to provide help on getting it working, and to fix that if it doesnt.

With Proton, the original developer and/or publisher doesn't have to do anything - that includes support the game running in any way on Linux. It is effectively the same situation has it has been for years with Wine - if it runs, great.. if not, it's not the original dev/publishers problem. They wont support you. As far as they are concerned, even if you bought on Linux, and it shows up on their ticker that it was a Linux sale, you bought a Windows product, and you are not running it via an officially supported method.

Nor will it be Valve's problem. There is no way they are going to provide support for every issue running every game on Linux. Nor will they be any more able to deal with bugs in those games on Proton than the Wine developers are. It could be years, or never, before you get a fix for a particular game not working. Or the next build of Proton could break a previously working game.

So, while I can see how Valve thinks this is a good thing for getting games on Linux, and getting gamers over to Linux, it in turn has a big impact to those of us who were bringing games over to Linux officially. I don't see how Proton is going to help us with the big issue of getting publishers interested in Linux as a platform... in fact, I see it doing the opposite.

You're absolutely right, but none of it matters. The question, the only real question, is whether this maneuver (and perhaps other related maneuvers) can actually drive an increase in Linux use. Network effects ultimately rule over the sort of issue you bring up--if Linux has a strong desktop marketshare, publishers will ultimately release on, or want to port to, Linux. Even if there's a solution of sorts through Wine, the sales will be stronger with a real Linux release. If Linux does not have a strong desktop marketshare, publishers will ultimately not release on Linux. Bigger marketshares get paid attention to, smaller marketshares do not.
The only reason desktop/gaming Linux gets as much attention as it does is its very large marketshare in most computing areas other than the desktop. It is a niche gaming/desktop OS, but not a niche OS overall. But that only takes us so far.

View this comment - View article - View full comments

Quoting: GuestOh absolutely, it's a good thing, and I don't mean to diminish the work Valve is doing. But Valve isn't doing quite as much as people think - for example, Valve isn't making any game work through wine directly. Well, that can be argued via dxvk, but the point I guess I'm trying to make is that while the impact might be large to users, the effort itself from Valve is actually not equivalent.

Hrm . . . more generally, it's an interesting question of value: Who deserves more kudos, someone who spends a lot of effort for a result that has little impact, or someone who spends little effort for a result that has a lot of impact?
Of course in this case it's more like, group A spend a lot of effort building most of what makes things work and outfit B puts on some finishing touches that allow it to be more broadly and easily used. On the other hand, one might ask if the Wine community on its own had shown many signs of ever turning Wine into something that worked transparently? And my answer would be "no", and that includes Lutris as far as I can tell from some of the other conversations around here in recent months. I even wonder if one problem might have been, if Wine started to Just Work, Codeweavers would be out of a job. An outsider coming and shaking things up may have been more important than we know.

View this comment - View article - View full comments

Quoting: mirowell I am not pessimistic about this, surely it will do lots and lots of good.
I am just afraid that too many people will blame "linux" for the less performance and more bugs, not understanding that this is not a native build, that there is dx->vulkan translation and .dll emulation aka at least one additional layer which makes things slower and more buggy.

That is a point, actually one of the most solid "con" points I've seen. Something like this happened with the Steam Machines only it was about bad ports.

View this comment - View article - View full comments

Quoting: Alm888

Quoting: stretch611Consumers are more likely to be ignorant of the lawsuit…

Precisely!

Quoting: stretch611…and even if they know about it they are more than likely to side with Valve…

Guess what will happen when Steam refuses to launch after the Windows™ Update? My bet they will run towards Microsoft® and Valve® with their sticks and stones and demands to "fix it back". And what happens next? Microsoft® won't budge and Valve will be unable to do anything.

If the US were the only significant venue that existed, you might be right. Valve would sue in the EU. And China, probably. The EU courts have a LOT more teeth with respect to this kind of anticompetitive behaviour, and would right now probably gleefully nuke MS just as a side amuse-bouche to the trade war.

View this comment - View article - View full comments

Quoting: GuestSo now that I've had time to think about this news more, I thought I'd try add more discussion points.

Disclaimer: I personally find it odd that people are praising Valve so much here. Valve didn't make wine. Or dxvk. Or actually any of the tools that make this possible. They're just packaging it into Steam.

Uh, I think they did make dxvk. Didn't the main DXVK guy just turn out to have been paid by Valve all along? Your general point is taken though. Most of the software this relies on (like Linux itself) is a community effort and we owe, and have owed all along, our thanks to the hardworking members of that community.

Anyway, so the first point for me is that nothing here couldn't be done before, and wasn't already being done before. Wine could already be used, and what whitelisted games I have already ran completely fine anyway on a vanilla wine with no extra patches. Furthermore, some games that run on that vanilla wine don't run with Steam Play, likely because the latter is based off an older version of wine - and this I consider a "con" of Steam Play. It's not very likely to get updates in a hurry, and a lot of games benefit from those updates.

That's not so great. But this is after all a beta, and they do seem to be paying people to make improvements. I expect it to get a bit more solid before they move this to main Steam from beta Steam. I don't get the impression that it's based on a very old version--Wine-related development is just moving at such lightspeed right now that the bleeding edge is significantly ahead of even quite recent stuff.

As a "pro", this does package everything nicely into Steam and avoids a lot of hassle for people who just want to push a button. I'm really not the target audience for this - I run gentoo, I like bleeding edge, I like customisation.

You may not be the target audience for this, but I don't think that's relevant. The point is that the target audience for this is a lot bigger than the audience of people like you--and, the audience of people like you are mostly already using Linux anyway. People are stoked about this because there is a perception that it is tactically significant in the struggle to increase Linux desktop/gaming market share, and similarly a perception that in the longer term if we want games we need that market share. I'm a fan of this as a tactical move helpful to Linux adoption. Of course it may not work, plenty of moves to promote Linux use have failed in the past. But it's a good shot, and seems likely not to be an isolated one. It also at the same time promotes Vulkan, and IMO increased Vulkan use is good for, well, every platform that isn't Windows--even MacOS, whether they want to realize it or not.

Another "con" however, is that something like this already exists! Lutris, for example. And many games really only work properly with different wine versions, different overrides, that sort of thing. Wine can be a bit of a temperamental beast, and needs coaxing to get it running sometimes, which Lutris handles far better (and doesn't need Steam to do it). appdb.winehq.org already has what I consider a "whitelist" - a community run lit of games that work nicely. Maybe there are plans for something similar here - it's still early days to know what the longer term plans are for Steam Play.

Doesn't need Steam to do it . . . no, you don't get it. Look, I'm not the target audience for this either, because I don't really have any non-Linux games and I have too many Linux games owned or wishlisted to ever really worry about the titles I can't get natively on Linux. But I'm a lot closer. Here's the thing: My games are on Steam. I'm going to be using it anyway. Having to fiddle with Lutris, much less vanilla Wine and different Wine versions and go find some whitelist half-ass maintained by some community I've never heard of and on and on when all I want is to relax and play a game . . . not interested. If I can buy a game and run it on Steam just like with any other game, I'm there--or I would be if I wanted any Windows games.

On the matter of support, I personally would prefer eON over wine.

What's eON?

To end on a "pro", the real news for me is apparent financial backing for some development effort. The wording makes me think that's what most of the effort from Valve really is (other than Steam packaging): financial backing, so if Valve employees are doing a lot more then I do apologise! This means that hopefully development will generically improve on things like dxvk and vkd3d, which helps the wine experience in general, for everybody, not just Steam users.

I for one would be far less approving if they were doing some kind of sequestering thing where they made their own private fork or whatever and weren't passing changes upstream. But they seem to be acting pretty responsible about it and like good open source citizens.

View this comment - View article - View full comments

Quoting: Alm888

Quoting: GuestYou might as well be a Microsoft employee at this point...

And, considering how you are propagating this Proton™ thing, you may be Valve®'s employee… :S:

See, those two things are not equivalent insults. Because, like, Valve is on our side (more or less), while Microsoft is our enemy (not more or less, but quite specifically). So, "might as well be employee of our ally" is different from "might as well be employee of our enemy".
This is perhaps only tangentially relevant, but if you don't (want to) grasp that basic point, I don't trust what else you have to say on the topic.

View this comment - View article - View full comments