Front-page
Updated
Emulation, game quality, and the value of "intended play"
Zelda games have always been some of the highest-profile titles on each Nintendo system onto which they’re released. Breath of the Wild was both the Wii U’s swansong as well as one of the most highly praised Zeldas to come out in… good lord, it’s been almost a decade since Skyward Sword. Does that game count as “highly praised”? Well, with Zelda being the uber franchise that it is, it’s no surprise that a Wii U emulator would focus so much on compatibility with BotW, and focus it certainly has. Through the use of community graphics packs, Cemu gives BotW; an extremely comprehensive list of all kinds of odd internal rendering resolutions, the ability to adjust level of detail bias, a toggle for depth of field, enhanced reflections, settings for shadow resolutions (up to 400% base resolutions), and many more aesthetic tweaks. For myself, though, the big cake underneath the delicious icing is the ability to blast that framerate beyond 30fps, beyond 60, even beyond 120, and up to 165fps. I’d have loved a 240fps option, but I have to be frank with myself, my PC would never hit those numbers.
I’m entirely CPU bottlenecked with my RTX 2070 and an i7-4790k, but even on my rig I can maintain 60+ fps @1920x1080 in the open world, and by the sweet sweat of Satan, it both feels and looks like a dream. This is to say nothing of the framerate inside shrines, which is consistently above 120fps, where much less needs to be rendered. At 60fps, actions like sweeping the camera to admire the sprawling landscape, or keeping track of enemy movements to time a parry no longer give me the eye strain it used to on the Switch, and the reduced input latency means that I’m taking less damage as a result of my directions not registering quickly enough. It really is indescribable how it feels to play the game on PC at 120fps as compared to 30, to watch things glide across the screen, and I imagine it’s only something I can show rather than tell. More impressive is how this is all accomplished while also enhancing the quality of the visual assets, as high quality textures can be loaded at extreme distances, the boosted resolution gives everything new clarity, and reflection tweaks give wet or metallic objects a gorgeous, realistic sheen in a sunrise. Combined with the higher resolution, these make the rolling hills of Hyrule look so much more alive, and the act of standing atop a tower becomes more of a marvel than it was on console.
Comparison pic provided by user "Abdulrahman" on Resetera.
Interestingly, there are a few unintentional side-effects of the framerate which also end up enhancing the experience in subtle ways, one of which being that the game’s menus don’t exactly know how to deal with the sped up framerate, and as a result they themselves begin running more quickly. This “glitch” actually makes menu management so much less painful by speeding up the act of sorting through items, putting them in your hands for cooking, eating multiples of the same item, and swapping weapons via the d-pad. So many management actions are made less intrusive, and I’m no longer so reticent to swap weapons mid-combat, making what once broke the flow of gameplay now more integrated into it. As a result, I was finding myself less exasperated on average during casual play. It hadn't really occurred to me just how clunky the original game’s menu system was until the problem was accidentally rectified. The game’s input latency is also reduced significantly by the higher framerate and ability to disable v-sync, making techs like bomb impact launches (or “boomy zoomies”, if you prefer the scientific term) much more consistent to perform, though even the most uncomplicated actions like running and jumping are improved by the lowered latency. Granted, not all glitches introduced by the higher framerate are positive, as the physics of some objects tend to break, especially in regards to gravity/buoyancy. Hanging ropes holding lanterns or platforms jiggle continually, items in water behave strangely if thrown from high up by continually bobbing deep into and high out of water, and loosing a chest from dirt using magnesis causes it to rocket much too high into the sky, only falling many seconds later. They don’t impede puzzle-solving, at least as much as I’ve experienced, so I find them more amusing than anything, but it’s still clearly not how objects were intended to interact. It’s a mixed bag, but for me, the benefits outweigh the negatives by many times.
Now, it’s no secret that companies like Nintendo aren’t exactly fond of the option of emulation being available to the consumer, and understandably so. Competent, accessible console emulation does inherently more easily facilitate software piracy as an alternative to buying games/consoles, even though that very much isn’t the intended effect. Indeed, most high-profile emulators like Cemu specifically and emphatically do not condone the act of piracy, and are more concerned with the act of games preservation. Even still, emulation sits in somewhat of a gray area when it comes to the games industry, not at all illegal (given no copyright is being infringed, code stolen, etc.) but still on the moral/legal fringe. In many circles, emulation is seen as a de-facto less legitimate method of play than official hardware, with some more extreme groups decrying it as nothing but a piracy enabler. That being said, I just can’t deny that what emulation does in service of preservation and accessibility long-term is necessary for games as an artform, despite, though not to minimize, whatever hypothetical damage it might or might not enable to the immediate industry.
What I’ve come to understand more clearly in my time using Cemu, however, is how the act of emulation can actually provide what I consider to be a more “pure” experience for games than even the hardware they were on which they were originally published can provide. In my own case, because of emulation, I was able to engage with a form of Breath of the Wild which I found much more entertaining than it was at launch, and that was solely due to the enhancements provided by said emulation. I’ve seen it repeated often that emulators should do nothing more than provide an experience which endeavors to exactly mimic that of the original hardware, with no exceptions, with anything enabling otherwise being called a detriment to gaming as a whole. There’s a question someone posted in 2013 regarding a mod to potentially increase the framerate of Ocarina of Time from 20fps (NTSC) to 30, or even 60, to which one user replied saying that people who request and utilize emulation-based enhancements are a detriment to "proper" emulation.
Similarly, in 2019, a small debate regarding how much responsibility an emulator should have with representing the exact state in which a game originally played spawning as a result of the popular SNES emulator, bsnes, allowing for the removal of console slowdown in some games, with many similar arguments made against the feature. From the perspective purely of preservation, I can immediately see why retaining artifacts of the limitations of a console during emulation is extremely important. For example, there currently exists a 60fps mod for Super Mario 64, which is a game I certainly wouldn’t bring myself to play at any other framerate, but while it’s the only viable way for me to personally enjoy the game, I recognize that it’s important that this doesn’t somehow become the only way for people to play it, despite the fact that I regard it as an objective upgrade. Reviews of the game from back in 1996 can be readily dug up online detailing the thoughts people had on the game at the time they were thinking them, but while these contain mostly subjective assessments and takes on the game reflective of the ideals of the gaming community at the time, emulation gives us a unique method of experiencing the game as it existed in the past; a past which could easily be skewed with misinformation.
What if, hypothetically, the 60fps version of SM64 were the only version in circulation? How many decades would it be until the information regarding how the game originally played was lost? Can’t you just picture some edgy 10-year-olds, 50 years from now, saying things like “gosh, games were so much better back in the old days” based off a flawed understanding of the originals due to enhancements in emulation? I mean, if microtransactions become the future then they’d still be right, but I digress. This is something we’ve empirically seen in regards to many things in history, where the negative gets forgotten and the positive gets idealized. For something as quantifiable as the framerate, it’s easy to imagine that that information could be preserved via text alone, but it’s one thing to imagine the game playing at 30fps, and another to understand how it felt. Moreover, what about those aspects less quantifiable? Nobody has, or will ever, document every single position Mario can stand in, or every direction the camera can look in which causes framedrops, or the exact severity of the drops. If emulation doesn’t allow you to preserve those dips in framerate accurately, then in an eventuality where emulation is the only means by which one can access these games (already a truth for a great many games presently), the zeitgeist’s idea of the original experience will have already, if subtly, differed from the reality. These perceptions are important, as they’re frequently drawn upon for contemporary comparisons and discussions, and it affects what we think we understand of the past.
I understand how fragile this situation can be, as I’ve already experienced the phenomenon personally. Having never owned a Super Nintendo as a child, my introduction to the gems of the SNES was via emulation, somewhere around the mid-late 2000s. As I played the seminal classics via the emulators of the time, I found myself not having very much fun at all, and stopped before I even got halfway through any game. I kinda hated the experience, thinking that the games felt extremely clunky and hard to control, which I assumed was because of their age. After all, these games had come out so many years before, and it seemed reasonable to expect some rough patches and growing pains before game developers mastered the minutiae of how games should control. My experience playing an NES emulator made for similar results, and it only solidified this idea for me. What I had yet to understand until much later, as I played a real SNES on a CRT, was that the clunkiness I had just assumed was a product of the games’ ages was actually the input latency brought on by emulation, and had nothing to do with the games at all. In this way, in a microcosm, I had been unintentionally mislead as to the nature of those old games just because of mildly inaccurate emulation. It left an impression of just how easily anyone's perception of the past can be skewed, with no notice or warning. In the near future, there’ll be a generation who will have never played a real N64. Heck, I’m sure there’s already a generation that has no access to a SNES or NES, and whose only experiences of those games have come from emulation, whether official or otherwise. Were it not for Retroarch’s “runahead” feature bringing input latency in line with console hardware (amusingly something which could be called an “enhancement” depending on who you ask), many others would likely have come to the same erroneous conclusions as I did. For those who have only played these retro games via Nintendo’s own, more laggy emulation, their perspectives on these old titles are almost certainly at least somewhat perverted as well, and I imagine they’re likely more disposed to be critical of these older games in discussions than they would be otherwise, just as I was.
It’s definitely important for historical accuracy, then, that once most consumer N64s have ceased to work, we don’t look back upon landmark titles like Super Mario 64 through a 1080p 60fps lens. The critical acclaim that game received upon launch takes on a very different context when you can play the game at its native resolution of 320x240, with a framerate cap of 30fps which frequently dipped into 20fps. I’m not trying to slag the original release of Super Mario 64, mind you, only to point out that just how beloved this game was in its time gives insight into the standards, expectations, and desires of the gaming community as a whole during the late 90s. Without a way to preserve all its jaggies, all its slowdowns, all its original input lag, we lose key information necessary to understand the industry and the consumers at the time. With 1:1 console emulation, or at least as accurate as we can get, we attempt to achieve a unique tool by which we can essentially go back in time to see and feel exactly what caused such a sensation to the best of our abilities, instead of being limited to inference via text descriptions. For perspective, cultural and art historians would kill to have access to a machine which could accurately reproduce the works of Michelangelo, or the plays of Shakespeare, exactly as they were seen in their time, and for good reason. Emulation is that tool for gaming.
That being said, the mere existence of this 60fps SM64 mod alone is not capable of compromising historical preservation, as some would imply, in the same way that a digital edit of Van Gogh’s Starry Night doesn’t change the original. So long as the original experience is maintained/labeled as the original and made readily accessible, then we’ll keep our metaphorical time-machine to see things through the eyes of the consumers of the past. In fact, by completely denying the enhancements that emulators provide, we would strip them of, what is in my strong opinion, one of their most powerful abilities: the means to give us better, more accessible versions of games which we wouldn’t get otherwise.
Though “improved” is not how many would describe the emulation of these games. A common argument I’ve repeatedly seen against not only the personal implementation of these enhancements, but their existence in general, is that “the games were not intended to be played at higher than 30fps” or “this game was designed around the input lag”. Among others, my biggest issue with these arguments is that they show a critical misunderstanding of the concept of developer intentions. Breath of the Wild was “intended” to be played at 30fps, only insofar as the team made the decision to release it knowing the framerate at which it ran. Yes, the designers deliberately published the game with the understanding that consumers will be playing it at 30fps, but what the argument fails to understand is that this does not mean they would have preferred it to be 30fps, only that it ended up this way. BotW’s designers didn’t implement a 30fps cap because it made the game feel/look the best it could; it was a compromise of hardware with respect to the kind of game they wanted to make. An open world with wide areas full of weather effects, enemies, and many distant objects takes quite a bit of processing, after all, and if the choice was between their entire artistic vision for the gameplay and maintaining 60fps, they naturally prioritized the former. The game has certainly not been made more enjoyable by way of having higher input latency, or by making it more difficult to interpolate and react to objects in motion, or (if you’re like me) giving the player a headache as they try to focus their eyes on 30fps; these were necessities to make it run on the Switch in the form that they wanted. Similarly, the development staff making Ocarina of Time didn’t make the gameplay 20fps purely as a means to facilitate their artistic vision or enhance the player’s experience, it was in order to maintain a relatively high polygon count on screen without the N64 catching fire. Super Mario 64 doesn’t swap out Mario’s default model for a low-poly one when the screen zooms out because the developers thought it looked cool, it was so that they could try their best to maintain the framerate while more and more objects come on screen. These are not elements that serve the game’s or the developers’ intentions, they’re strategic compromises in order to make the best game they can.
This is actually why I see these enhancements, like those I’m using in Breath of the Wild, as a more accurate representation of the experience that the developers would have wanted to provide by circumventing the technical limitations of the hardware. For Breath of the Wild, lowering input latency, making the game run at as high a framerate as your machine can handle, and loading high-quality textures at longer distances serve to remove some of the barriers to immersion that the hardware brought on, making for a game which better achieves the core feelings the developers wanted to evoke. Of course, it’s always important that these enhancements are opt-in, so as not to breed the misunderstanding that the game always ran this way.
In this way, emulators have become not just a powerful tool for artistic conservation, but also a means to give us better, more accessible versions of the games we currently play. Many games have had elements compromised by the hardware to which they’re beholden, and emulation has evolved with the ability to remove those barriers. Due to console exclusivity, we will almost certainly never get a PC version of Breath of the Wild, but this also means that the game will never be as fun as it could be, or feel as good as it could feel, or look as gorgeous as it might on PC. Experiencing the impact of Cemu’s enhancements has already proven that much to me. I mean, if just a few user enhancements could change the game so much, imagine what an official port with a budget would look like. Emulation can serve as a means to not only show us just how much games can be held back by the concept of artificial exclusivity, but also just how the final quality would have been affected had certain design elements been changed. After all, it’s one thing to guess that a game would have been better if it had less input lag, a higher framerate, a higher resolution, less motion blur, etc., but it’s entirely more useful to see as much empirically, and be able to draw conclusions via observation rather than supposition. Emulation’s enhancements serve to clear up the ambiguity in these kinds of arguments, which are important as a means to give critical feedback and influence how games evolve.
So it's by emulating Zelda on Cemu that I've come to realize how much of a difference emulation can make in gaming. It’s become evident to me, at least, that the conversation should focus less on whether or not emulators should provide enhancements to the games they emulate, and instead on methods by which we can maintain both hardware-accurate emulation and playability enhancements in tandem. Emulation is both the gateway to experiencing the enhanced PC ports of games that we would have never been able to play otherwise, while also a way to preserve the history of gaming in a form which is readily accessible to anyone willing to spend a few minutes. This accessibility is a critical point, mind, as just having original games preserved in a museum somewhere drastically limits the amount of people who will be able to personally educate themselves with it. These two functions are not mutually exclusive, both serve critical roles in today's and the future's gaming cultures, and should both be actively pursued if we want to make the most out of our tools.
Last edited by Meteor7,
