Heavy Virtual

The gaming community has looked on in horror at image after image of GPUs grafted onto cryptocurrency torture racks, shackled up inside industrial power plants hashing away at computationally cruel tasks, twinkling in soothing RGB patterns1. There is no coincidence that the shortage of GPUs, computation, and other core gaming components would carry with it a crisis at the very fabric of what gaming is for, and what it does. First, an outsider group to video games has come to the table with different assumptions on virtual value, governance, and the telos of play2. Second, creatives and consumers are frustrated at the new scarcity of the GPU, raw computation, but computation is the sea we all swim in now. 

Far from its stigmatic origins, the figure of the gamer is exposed from all sides, unsure if player or worker, content-creator or consumer, professional or amateur. Every ludic individual now makes an impossible decision about where the game begins and ends in their life. I worry most that the game designer does not want to address this ambiguity, and probably works to maintain it. It is in the best interest of the industry to keep play and virtual assets deregulated and in their dark pools. Worst case scenario: the game designer continues to exist in the world at most as a cultural agent that does good by symbolically manufacturing game worlds that show good (but do not exist). We are floundering inside game worlds that picture positive futures that do not acknowledge the technologies and energy underwriting simulation. If what we do when we play is compute, what we compute might be worth reconsidering. 

The RGB now twinkles a little menacingly now, like the last few decades were the tutorial and we are about to reach the first level of computation. Level 1: the lightness of play and the objects behind our screens become heavy again and crash through the monitor. Their weight will be acknowledged as a social and energetic necessity, because play is no longer a tenable framework to describe what we do. At the moment, video games and game engines facilitate computation into larger structures, images, and trends – the contemporary. Even more unexpected, a massive subterranean shift: the technology of the game engine sits now as the base of a superstructure spanning machine learning, robotics, animation, film compositing, construction, architecture, etc. A self-driving car, warehouse bot, or robotic arm makes claims to the GPU. The largesse of play now lays the groundwork for a considerable portion of the real. In the brain of every machine learning agent is a world conceptualized through the framework of the game engine. And through the eyes of our smartphones a world augmented and reprojected in so many ways. The server rooms of the city-state and the massively multiplayer online game. The next decade we will observe our engine becoming the ubiquitous motor of computation.

The “we” is a very inclusive one spanning adjacent creative industries, anyone that considers the GPU an essential apparatus: motion graphics, animation, rendering. And the “them” is the blockchain, web3,  and pay-to-earn idealogues. To get through to the heaviness is through them and what they get right and what we get wrong. One is the renewed understanding of the energetic reality behind video games and two is the need for the economic and social engines of play to disconnect themselves from the video game.

One of the most computationally expensive and expansive tasks, the simulation of video games distributes swathes of high powered silicon onto any available geography. In the United States the energy grid underwrites play with ~35 terawatt-hours (TWh) of energy every year3 and the industry generates revenue in the low tens of billions. We did not have to reckon with this expenditure until GPUs started disappearing off shelves and the disclosure of a new form of valuable computation was made, cryptocurrency. Our distributed system of GPUs had accidentally made another arrangement possible, one in which the gamer unplugs their rig from the game world and instead uses it to manufacture consensus for a decentralized system. The second largest cryptocurrency is within the ballpark of energy consumption of video games (~25 TWh/year) and its capitalization is roughly one year of global game industry revenue (~$300B USD). But this isn’t where we are headed, into some intricate energy calculus between them and us, this is instead about the surprising revelations that happened exactly when we started doing this calculus. 

The most common calculus is a question like so: what is more carbon friendly, traveling to expos and conventions to pitch a game prototype or self-funding by selling virtual assets on the blockchain? Strictly speaking, in terms of economic impact divided by carbon footprint, it’s the latter. This is a contentious point that needs to be expanded as all on-chain values remain speculative, however, the material gains can be realized up until a correction. In light of this calculus, a value judgment has to be made, both in the literal sense (is this a preferred action) and in the figurative sense (do we value this form of value). The answer to that is not as important as understanding that when we examine our actions under an economic and environmental microscope the best decisions are often the least intuitive or socially acceptable. Nonetheless, the irreconcilable differences remain. 

Not only have we been balancing the environmental cost of computation in the blockchain moment, we now have also had to weigh the moral cost of the play-to-earn genre. Most designers, including myself, intuitively know this concept is undesirable, the coercion of players towards some monetary end is insipid. Many games do have a financial end, gambling and poker are both “gaming” after all. The main heresy of play-to-earn is not money, it lies in its pyramid shaped model and instrumentalization of players as mere economic actors. But I would say that is not why we distrust it. The proponents of it are outside the industry and have the same video game history amnesia the public suffers multiple times in any given year. We are in the midst of another relearning of the Steam marketplace, Diablo auction house, and maybe the logical end of play-to-earn: the sweatshops of the World of Warcraft gold mining industry. Video games financialize their players to degrees of success and failure, but we do not want our failures to be reenacted by the du jour gaming venture capital. Forgiving the forgetfulness of video games history, this question should be perennial: how might the institutions and industries of play better be reoriented to better their thriving virtual populations?

Asking the question is difficult, and each developer earnestly answers it in their own way. Much larger things are at stake, the headwinds in the industry are strong as we fight for a vision of a more equitable industry for developers. To re-center and re-imagine the figure of the “gamer” and the planetary playing populations is well trodden ground, but I believe how we as designers of games exist in the world is still through this question.

Should the industry that helped to define the entire category of virtual worlds not be intellectually curious to the development of the last few years? There is a blind utopianism in the current rebranding of games and the web that are happening outside our walls, we seem to be sitting on a good “I told you so” when the house of cards falls. But the uneasy conclusion of energy calculus still remains. What was revealed and quickly buried was that many real world things we take pleasure in doing (travel) would of course be better simulated from a purely energetic perspective—and this is true the other way, many things we simulate should not be simulated. Let’s call out Animal Crossing games for what they are, fake pastoral fantasies that simulate engagements in a community we wish the urban environment provided. Instead we simulate this fantasy rather than making it real because we no longer have the resources and wealth to make it real. 

Jon Rafman – You Are Standing in an Open Field (Mental Traveler) – 2020

In essence, composing and transporting bodies is costly while composing and transporting virtual bodies is not. However, equally costly is an industry that continues to show non-stop double digit growth, with vast computational resources and profit, but no intellectual program other than simulating realities or modeling behavior we might wish to see come along. This mode fits well for games, filling the socioeconomic vacuum of the desires and wishes for worlds we want to see but do not have the power to enact (the pastoral fantasy refrain). Soon there will be an energetic dimension added to this reality, whether through per capita limits on plane travel or the waning of the petrodollar. The transporting and composing of virtual bodies will be the business of games. But this is only a positive future to opt into if video games can address the disconnection between their reality and the local, physical and economic realities of their players. What parts of reality could be made virtual and given their value, their weight, and given to players? And once this happens, how will the role of game designer intervene directly in the world? Could we acknowledge the work and agency of the player and the weight of the virtual? If this is the one thing we can learn from blockchain or play-to-earn then let it be that.

This turn from lightness to heaviness is not prescriptive for all of play, in fact for most video games it is not possible. Play is surprisingly obstinate, we are in a ludic hangover of our own attempts to coerce play into genres like serious games and not-games. The ludic turn was not to be found in the executable file but in the tertiary markets of streaming, eSports, and influencing—assembling and organizing computation into the images and scenes of popular culture. Ultimately these forces have outstripped all attempts to steer play into more productive paths. All the while the game engine has been tunneling out and is about to break the surface. 1,2,3 

This rough sketch of all the moving parts puts us on different trajectories that will cleave the game community. We will still play our games as before, but a second head has grown next to our normal one and it has started to make quite a racket. The question arises and it is an old one: if games and play arbitrate the majority of virtual creation and value, could this computation be put towards a new end. Resoundingly the answer has been no—play should not be used towards an end and its disinterested pursuit is its fundamental enjoyment. But how can this be true anymore, there is no disinterested play, we don’t step out of the game any more than we step out of our home on the way to work. A common example many have experienced—you are playing a first person shooter and are killed, the player tag is a streamer’s, you thought you were playing but instead it turns out your multiplayer lobby was at work. Your medium desktop tower with its RGB lights ablaze and GPU whirring away on its rack. 

Lets ask the perennial question of how play might be reoriented. Either we acknowledge that players might soon need to be compensated for simulating the worlds we create or we continue to invoke video games as a symbolic medium that cannot make direct claims on the real. Maintaining the lusory attitude that we have used to define play means staying with the latter. However, I think it is clear this is no longer an option. Far from this being a failure or disappointment it is instead one of the many answers to the question of how a game designer might exist in the world.