Searched and not found. Sorry if already posted.
Link: http://www.pcgamer.com/why-pc-games-should-never-become-universal-apps/#page-2
-----------------------------------------------------------------
Peter "Durante" Thoman is the creator of PC downsampling tool GeDoSaTo and the modder behind Dark Souls' DSfix. He has previously analyzed PC ports like The Witcher 3, written about why broken PC ports are unacceptable, and written an open letter to game developers about the features PC gamers want.
Microsoft’s vision for the future of application development and distribution on Windows, dubbed “Universal Windows Platform” (UWP) and “Universal Windows Apps” (UWA) is currently the most controversial subject in PC gaming. Universal apps first came to our attention with the Windows Store version of Rise of the Tomb Raider missing display settings available in the Steam version, and escalated with Epic Games co-founder Tim Sweeney criticizing universal Windows apps as a closed platform that “can, should, must and will, die as a result of industry backlash” if Microsoft doesn’t open it up.
This article aims to clarify what UWA is and and provide my own perspective on the matter. The first part will deal with the current state of the platform and all the problems and limitations you should be aware of before buying a game on it. After this objective analysis, I’ll provide my more subjective commentary as a long-term PC gaming enthusiast and modder on the potential impact of Universal Windows apps.
The state of UWA and the Windows Store
As of now, the only way to purchase UWAs is through the Windows Store, so the two are closely intertwined. While they each come with their own set of restrictions and limitations, the choice that can currently be made for a game like Rise of the Tomb Raider is either buying a traditional Windows executable by any of the common means such as Steam, or buying a UWA on the Windows Store. The drawbacks of UWAs over the former can be associated with four overall categories:
- Lack of features
- Lack of interoperability
- Modding restrictions
- Ecosystem drawbacks
Before going into details on each of these categories, we should clearly define what a few of the terms we’ll be discussing here mean. Win32 is a set of programming interfaces (APIs) as well as an executable format which has existed for 20-odd years and continues to form the basis of how programs for Windows are written and distributed. Even though today it’s convenient to download games through clients like Steam or GOG Galaxy and launch them without ever directly clicking on a .exe file, that good old Win32 executable is still there in a folder waiting for you to use or modify it as you wish.
Microsoft’s vision for UWP. Not pictured: restrictions and limitations.
UWP is a set of standards and restrictions as well as programming interfaces which Microsoft proposes in order to target various devices running Windows 10, including PCs, tablets and cell phones. UWAs are applications making use of these standards and APIs. A crucial difference between UWAs and Win32 applications, beyond the API restrictions, is that interaction of the former with arbitrary programs—and even the user themselves, like in terms of accessing the program’s files—is limited in many ways. UWAs cannot be distributed and installed easily without going through Microsoft as the gatekeeper.
Think of the difference as a bit like a traditional PC application compared to an iPhone app. One splays out its files in front of you in folders if you want to dig through them. The other merely installs and lives its life as an icon you click to launch; you’re not going to be digging around in its file structure.
So, what does running a game distributed as a Universal Windows Platform App mean for you compared to a Win32 game?
No Exclusive Fullscreen Mode
One of the most immediately obvious limitations imposed by UWP is the lack of any support for exclusive fullscreen mode, aka the way we’ve been running most PC games for decades. This is clearly an intentional change documented in MSDN here: “If a Windows Store app calls CreateSwapChain with full screen specified, CreateSwapChain fails.”
The Eizo Foris FS2735 is one of many interesting, recently released Freesync monitors which suffers from its selling point being rendered useless by UWP restrictions.
What this means for actually playing games depends on the individual game and hardware setup, but generally includes a small performance impact, and often also less reliable frame pacing. I observed and documented the latter e.g. for The Witcher 3.
It also means that other features depending on exclusive fullscreen will simply not work with UWAs. This prominently applies to AMD’s Freesync, while Nvidia G-Sync will work but is restricted to its less reliable desktop mode.
Lack of interoperability
Interoperability is a big word, and unlikely to be the first thing many gamers—even enthusiasts—think of when playing games on PC. What I mean by it in this context is that in Win32, it is relatively easy and straightforward for any program to integrate itself with another arbitrary program, even if the latter knows nothing of the former. This facet of the execution model is the foundation for many of the conveniences gamers take for granted on PC, and you’ve probably used it many times without even knowing it.
A Rivatuner Statistics Server overlay displayed in The Talos Principle.
One of the most common uses for this type of interoperability are performance monitoring overlays and screenshot tools such as the popular FRAPS and RTSS, or the more involved and accurate FCAT, the lack of which presents obstacles for in-depth performance analysis.
However, these are far from the only applications of interoperability in games. For example, voice chat programs such as Mumble might display an overlay or even generate positional sound based on game data, and input mapping utilities often hijack API calls in order to provide more configuration options and broader hardware compatibility.
Another very common use of this type of interoperability: the overlays of various game distribution and social services, with the most popular one being Steam. Both the desktop and the Big Picture mode Steam overlays depend on interoperability to function. This in turn even affects the Steam controller, which I’ve personally come to love for anything which requires first or third person camera control. It’s unusable for UWAs.
Gamers with SLI graphics card setups who bought the Windows Store version of Rise of the Tomb Raider discovered to their chagrin that Nvidia Inspector cannot support setting profiles for UWAs. Doing so was required to fix SLI issues in the initial version of the game. In a similar vein, external tools like RTSS are often used to enable consistent frame pacing in games, yet another very convenient and often necessary enhancement impossible for games distributed as UWAs.
Modding Restrictions
There are two types of modding common on PC. One is publisher- and developer-approved and supported modding, such as adding new maps to a game with Steam workshop support or developing a new quest in Skyrim. The other is modding done without the knowledge or support of a game’s publisher, and this type depends on the accessibility of a game’s execution state and data.
A screenshot of Dragon’s Dogma: Dark Arisen with injected ambient occlusion, software downsampling and post-processing. Not happening with a UWA app.
It is the latter type which is primarily endangered by UWP, and it encompasses a broad swathe of modifications with varying degrees of popularity. For example, it is very common for enthusiasts these days to inject anti-aliasing or additional post processing into games with tools such as ReShade or my GeDoSaTo. These tools depend on intercepting API calls of applications, which is easy enough to accomplish in Win32 but a practice which UWP specifically sets out to eliminate.
Of course, it’s not just these visual modifications which are affected. Game-specific mods like the recent Tales of Zestiria 60 FPS fix or my own DSfix would equally be rendered impossible to accomplish – or at the very least far more difficult both to develop and for the user to apply.
There are myriad other examples of this type of modding: greatly extending the functionality of existing modding interfaces, adopting older games to new display standards, even polishing games for over a decade after their original developers dissolved. Every single one of them is enabled by the easy access to game files and relative ease of changing executables without asking permission of anyone other than the user. And as such, every single one of them is restricted by the UWP model.
Ecosystem Drawbacks
While the previous section looked at issues that were related to the UWP execution model, there are also some comparative drawbacks when buying on the Windows Store specifically. Obviously, it will mean giving up on features provided by other platforms, such as in-home streaming on Steam or DRM-free games on GOG, but these are hardly surprising.
What could, however, be a very nasty surprise is that the Windows store does not appear to have a clear policy regarding perpetual access to downloads of purchased games when they become delisted, a problem we’ve encountered with other digital stores in the past. As such, if a publisher becomes defunct or simply decides not to offer a game on the Windows store anymore, your purchase might essentially vanish (Microsoft’s website is vague on the permanence of delisted apps). This is a really bad look so early into the store’s life, particularly given Microsoft’s history with Games for Windows Live (GWFL).
Additionally, updates on Windows Store Apps have so far trailed behind compared to their Win32 equivalents, likely due to certification delays. This is doubly problematic if, as in the case of Rise of the Tomb Raider’s SLI problems, the primary issue an update addresses can’t be fixed by the user themselves because of UWP’s interoperability limitations.
A more holistic view
Up to this point, this article is an objective description of exactly what you can—and more importantly cannot—expect today if you buy a game as a UWA on the Windows store. However, that alone is not what fuels the ongoing discussion about this topic: there are also concerns about what the widespread adoption of the Windows Store and UWP would mean for the future of PC gaming, both as a medium and as a marketplace. This debate is inherently more subjective and based on speculation, but I believe it is an important one to have.
It actually started a few years back when Gabe Newell called Windows 8 a “catastrophe” for the PC space—as it turned out, he was generally right, and Windows 10 was in a large part designed to recover from that catastrophe. This time we have Tim Sweeney leading the charge and being suspicious of another attempt to change the way Windows programs are distributed.
Some younger readers, or people who haven’t been following the PC market for long, might find it difficult to understand why such well-established industry figures who have often worked successfully with Microsoft (or even worked at Microsoft) in the past would take such a strong stance. But with some historical context, it’s not so surprising after all.
History and Context
Microsoft as a company has been the subject of many investigations for anticompetitive behavior in the past, both in the US and the EU. One particular strategy uncovered and named during these investigations is Embrace, Extend, and Extinguish: it involves entering a market with an open and standards-compliant platform, subsequently adding proprietary features and using those to disadvantage the competition.
How does this relate to the current discussion about UWP and games? Microsoft, ever since the release of the original Xbox, only seem to truly remember PC gaming in one of two scenarios:
- When they need it to prop up an ailing project or product; or
- When they see a new way to monetize it.
Some examples of this pattern include the DirectX 10 exclusivity push with Windows Vista, the abortive attempt to introduce paid GFWL subscriptions and subsequent gradual abandonment of that platform, and most recently the restriction of DirectX 12 to Windows 10.
At this point in time, their efforts at establishing a presence in the phone and tablet spaces are clearly not bearing fruit (Surface revenue is on the rise, but phone sales have fallen off a cliff), and their console is struggling to hold its own against its primary competition. As a company-wide countermeasure, they are designing a uniform platform, clearly taking cues from the mobile app store model—and what company wouldn’t be impressed by Apple’s revenue streams?
However, a unified programming and distribution platform is no help at all to your corporate strategy when no one is using it, which brings us all the way back to gaming: why not use some of your investment in console games to try and increase the popularity of your new platform?
Potential consequences and the future
The reason I went into some detail about Microsoft’s history, circumstances and outlook is to make it clear that their primary strategic objective is to make sure that UWP and the Windows store is a success, not that gaming on PC is as good for enthusiasts as it can be. Once we recognize this, many of the design decisions which result in the limitations outlined in the first part of this article make sense.
What also makes sense is the concern of very smart and experienced people involved with the PC gaming market such as Gabe Newell and Tim Sweeney. Because the ultimate corporate goal of the Windows Store and UWP is clearly an Apple App Store-like situation where every piece of software sold for Windows is a Universal Windows App which generates some profit for Microsoft. Of course, Newell and Sweeney could well be primarily concerned about the impact of such a scenario on their own profits, but I believe that any gaming enthusiast should also be concerned to some extent, for reasons I will now try to explain.
If this future—one in which Microsoft gradually incentivizes UWP and the Windows Store at the OS level, and gradually disincentivizes Win32 and competing stores—ever came to pass, many of the wonderful things we love about PC gaming would be imperiled. The preservation advantages of perpetual backward compatibility. The large set of third-party tools free to interact with any game and enhancing or customizing our experiences. The competitive distribution market which results in the lowest prices and most frequent deals compared to other core gaming platforms.
And most near and dear to me, the end to a platform on which enthusiastic fans of a game can continuously improve and polish it, and ensure that it is easily accessible to future generations, even decades after the original publisher lost interest or ceased existing.
Vampire: The Masquerade - Bloodlines is one of the best first person RPGs of all time. While released in a sorry state, years upon years of hard work by fans has ironed most of the bugs.
Handing Microsoft—or any other company, but given Microsoft’s history it’s particularly egregious—the metaphorical keys to the castle and giving them the possibility to enact such change, regardless of the likelihood of them actually implementing it, is something I can never countenance.
Conclusion
There was a lot of ground to cover in this article, and there is yet more I’d like to say. However, I feel like what is really necessary is a summary that makes it very easy to understand what I consider to be missing in the UWP ecosystem. I’ve boiled it all down to two questions, one from the user and one from the developer perspective:
- Can I, as the administrator of my PC, grant any application—regardless of its source—the ability to do anything it damn well pleases on the entire system—including to other applications and UWAs—without either myself or the developer of the application having to interact with Microsoft at all or overcome unnecessary hurdles?
- Can I, as an application developer, freely distribute my UWA to users by any means I deem adequate, without going through Microsoft and without any disadvantages in terms of features or user experience compared to selling them on their store?
The answer to both of these question is currently a resounding “No.” Only if this changes to “Yes” for both of them—and in a well-documented, implemented, technically solid way, not just vague promises—can I even start to consider UWP as a future platform for PC gaming on equal footing with Win32.
Microsoft’s replies so far have been inconsistent. For some obvious implementation issues, like lack of V-sync settings, a quick solution has been promised, and I see no reason to doubt that claim. When asked about modding, the conversation is immediately steered towards the publisher-approved and supported kind, with no word so far about the more common (and, arguably, important) case. On interoperability, there are only the most vague of promises.
Whether we will see solid technical solutions at Microsoft's upcoming Build conference or more promises remains to be seen.
At a recent Microsoft event, Xbox head Phil Spencer told PC Gamer “for most people that just want to go play a great game the UWA version doesn’t keep them from playing a great game.” Regarding developer concerns, all the answers point towards Microsoft’s upcoming Build conference at the end of March. Whether we will see solid technical solutions there or more promises remains to be seen.
So, am I telling you to not ever buy a game on the Windows Store? Not necessarily. It seems like Microsoft is going to publish all their first party games exclusively as UWAs, and if you really want to play one of those then your only choices are between Xbox One and UWA. In such a case, all the limitations and concerns outlined above don’t really weigh on the decision, and if you have a fast PC it will probably provide a better experience than the console.
However, one fact should be clear. If you buy a game as a UWA then, in many aspects such as user control, interoperability, moddability and the overall ecosystem, what you are getting is closer to a console game running on a PC than what we traditionally consider a PC game.
Currently Online 
