Say what you will about PC gaming―there’s no denying that software prices can be almost ludicrously cheap, especially if you’re willing to be adventurous and trawl charity shops and second-hand stores. This weekend I picked up a copy of Monolith’s horror-themed shooter opus F.E.A.R., widely regarded as one of the last great FPS titles the storied developer put out and a technical tour de force in its time, and all for a pauperly sum of 20p.
After a little bit of canoodling, patching, and singing to it softly, I had the game up and running, maxed out at an ungodly 3840x2160 and easily pushing northwards of 90fps. Benchmarks checked out great so I jumped into a new game and was greeted with this screen:
There's supposed to be something written in the center. Look really carefully.
Being as bespectacled as I am, I’m not going to be fooling anybody into thinking I have particularly good eyesight. But the above image, resized to 1280x720 and boxed within your browser, should give you some idea of how the text appeared on my 27" monitor that was barely more than a metre from my face (as well as prevent this page from taking an eternity to open). It should be noted that this picture comparison isn’t a perfect representation of the UI shrinking effect as in actual use the higher density of pixels on the 3180x2160 display would keep text clear even if it took up a small portion of the screen. But depending on the game and resolution you might find yourself squinting anyway.
In any case, I couldn’t for the life of me make out what the hell I was looking at, even with my face pressed to screen in a painful squint. I’m guessing this isn’t an abstraction designed to get you to use your imagination to fill in the blanks. And keep in mind that this was with the HUD text scaling option set to "Large" (the only other option being "Small"); for whatever reason, this particular option appears to only affect in-game text such as your Journal and Objective details sheet.
I couldn’t help but feel that I was missing out, that the text might include some life-changing revelation for me if I could only just make out the wisdom scrawled upon my screen. I dropped back down to 1920x1080 and ran the same scene again:
Slightly easier on the eyes in person, but still tough to read. An improvement over the 3840x2160 image, however.
Okay, not exactly earth-shattering stuff―deep breath right there. But the text was still tiny (although it didn’t appear quite this harsh in person) and wasn’t as comfortable a viewing experience as I might have liked. With a little effort I could make everything out fine, but I had to question the practicality of playing PC games well below resolutions I was easily able to achieve just to get on-screen elements comfortably visible. It’s a design oversight that far too many developers behind PC games overlook, and one that’s gone by with hardly any protest.
And I find it strange that so little has been said on the topic, given that scalability and options are, for many people, the whole point of PC gaming in the first place. There’s more than simple vanity at play when PC gamers ask for an open approach to game settings. A game capped and V-synced at 60fps might look great on a 60Hz screen, but on higher refresh panels the effect can be an unpleasant judder as the monitor and graphics card refresh rates clash: uncapped framerates and options for various fps caps (including user-defined ones) are a forward-thinking solution (let’s keep any talk of variable refresh to an aside for now―I’m trying to make a point here).
Many gamers make similar demands of scalability with resolution as well. 1920x1080 might seem like a high enough resolution to be fixed at, but the interpolation effect when moving up to resolutions that aren’t neat multiples of 1920x1080 make for a blurry, unflattering image, something which any 2560x1440 panel owner can attest. Open resolution support for every resolution possible within the monitor's aspect ratio gets around that as well as any similar quagmires that might re-emerge when monitors with native resolutions that aren’t neat multiples of 1920x1080 (or whatever resolution a game's UI is optimised for) start becoming common. Yet we have no standard (or awareness that we should have one) for UI size scaling itself.
Every gamer's rig is different and games have to be made to take not only all those variations into account, but more importantly, to take into account set-ups that don’t even exist yet. Accounting for every potential variable is an impossible feat, but developers can go a long way towards addressing potential issues down the line by making their games malleable to change. That is, designing them to take advantage of the scalability of higher-end hardware that might appear.
Many games already do this today with resolution and frame, and after playing around with UI scaling settings (here the term "UI" is used to cover both UI and GUI) and finding some games nigh on unplayable at 3840x2160―and not for performance issues―I’ve started thinking that UI scaling should be added to the list of things we demand from PC game developers alongside open, scalable, future-proofed resolution support and framerate standards. Out of the holy trinity of framerate, resolution, and UI size, UI size is arguably the most important to enjoying a game because a game's playability can hinge entirely on it, whereas it might only be swayed by a matter of degrees by the other two (though granted, resolution and UI size tend to go hand in hand―but that’s precisely the problem here).
Dragon Age’s main menu at 3840x2160. Modders shouldn't have to fix this... nor should your local opticians.
But there’s little of the furore surrounding UI size that there is around framerates and resolution, perhaps because there’s no console or competing game box to single out as the root-cause-of-all-evil. Or perhaps it’s because UI size issues are almost entirely the domain of PC gaming in a videogames coverage media that is console-oriented. Or maybe it’s because there’s a misconception that the UI issue has largely been solved as we never hear about it―making it a thing from which only crusty retro games suffer.
But that line of thinking fails to take into account why UI shrinking occurs in many games in the first place, as well as ignoring that the games of today will be the retro games of tomorrow. They might not show symptoms of poor scaling now, but we might be having the same issues I’ve shown above come to the fore a decade from now on our treble-wide holoscreens. Sadly, the solution isn’t quite as simple as designing UIs to scale 1:1 at both the base resolution as well as any potential higher resolution, nor is it to include UI size sliders―a combination of both is going to be key.
1:1 UI scaling of the kind seen here in Sonic Adventure 2 works well for console conversions, but isn't a perfect solution for all PC games.
Depending on the type of game you’re playing, UI shrinking can actually be beneficial, giving you more (non-UI) screen space to work with or, conversely, more UI screen space total. It'd be great for MMO players who want to load their screens with oodles of spell macros for convenient access. Some games at certain levels of competitive play practically demand spare pixels be repurposed for practical means. A 1:1 UI upscaling for these games wouldn’t work and I’d imagine the majority of players would rather fiddle a bit with UI scale sliders to offset UI shrink rather than have 1:1 UI scaling forced on them. But even this is an imperfect solution, as evidenced by Guild Wars 2, a game that’s only a few years old yet tough on the eyes at 3840x2160, even with the UI scaling cranked all the way up.
An example of UI scaling failing to overcome resolution-tied UI shrink.
Or at least, far tougher than it needs to be: including several sets of UI art assets for future resolution support (drawn at twice, three times, or even four times that of the base GUI assets) and a wider range of UI scaling to toy around with would help stave off UI shrink, but these are temporary solutions to a more permanent problem. Allowing the player to both scale a UI as it would appear on a lower resolution (allowing them to include as many or as few menu items as they’d like on their skill bars) and then lock that setting and have the game increase that up 1:1 until the appropriate amount of screen space is filled would all but solve the creeping issue of UI shrinking.
Of course, there would still be the issue of low-resolution UI elements being 1:1’ed up to a higher resolution, ending up looking ugly from the stretching process, but at least by giving players these options they can choose between a pristine, sharp UI and one they can read easily (or at all). The issue of UI shrinking hasn’t been solved just yet. Developers need to get into a habit of future-proofing the readability of their UIs, one of the most important aspects of any game. Either that or years from now we’ll be squinting at our 16K screens complaining about how things were better in the good old days.
But there needs to be an awareness on the part of gamers that this is indeed an issue their games may encounter in the future when they move up to higher-resolution displays. In recent years, concerns over digital content ownership has pushed gamers to look more critically at preserving games so that they’re always available to play, either for their personal use (through DRM-free digital storefronts such as GOG) or for consumers generally as a matter of posterity (the recent P.T. takedown comes to mind). But access means little without accessibility; to preserve games on the one hand so that everyone can experience them and not preserve the experience itself strikes me as self-defeating―honoring the letter of videogame preservation but not the spirit.
I didn't ask for this... I don't think...?
It’s not uncommon for games of old to make their way onto GOG complete with adjustments and tweaks to make them playable on modern OSs, a design philosophy that speaks to that spirit of videogame preservation. It may well be time to extend that same philosophy for game UIs themselves, ensuring that they’re playable as well as bootable. Because, you know, that’s kind of the point of videogames.
Like what you read? Please help spread the word and share our stuff on Facebook, Twitter, Reddit, or Google+ using the buttons at the top of the article!
Back to AUTOMATON
