Gaming and extinction: notes on emissions, degrowth & the hardware upgrade cycle

On Friday I gave a keynote as part of the day-long online symposium “Imagining Extinction in Video Games: An International Symposium” organised by researchers at the Universidade Do Porto. It was lovely to be invited, and I found our discussion afterwards lively and engaging, with fantastic responses to my talk by Ed Chang and Patrick Whitmarsh, and great comments in the chat.

My slides from the talk are below. I’ve done a little precis below them, but the body of this post is about some of the things that came up in the responses and discussion afterwards.

In the talk, I cover two of the major dimensions along which games interface with (and exacerbate) extinctions. The first is through the lens of greenhouse gas emissions, and I included my best guesses about where those emissions occur and in what sorts of amounts. I talked about what that suggests for the industry, through the lens of my report on the climate footprint of producing Die Gute Fabrik’s Saltsea Chronicles. This is all happening in the global context of increasing demands for climate action, policy responses from governments ratcheting up, and greater expectations from consumers, employees, and even investors. Much of this I have spoken and written about elsewhere (including here), so devoted GTG readers may just want to gloss over the first half for a refresher.

The second is about the hardware upgrade cycle and a question that I have only recently begun to articulate through writing the talk. It occurred to me as I did so, that we really have precious little clarity and absolutely no certainty about what, precisely, is driving the manufacturing, marketing, and sale of new game hardware from consoles to CPUs to graphics cards. There are several received narratives about it, but once you start considering what the actual evidence base is for any one explanation… things get quite a bit murkier.

The first explanation is that “it happens because companies need to sell products” – Sony, Microsoft and Nintendo have to make a profit, and new devices are a great way to do that especially once you’ve sold the existing one to all your current customers. But that gets complicated by the fact that many consoles are sold at a loss – at least initially – and after that, usually only at very thin margins. So the next explanation kicks in, which is that it is because it enables profit elsewhere, through selling more software, platform lock-in effects, and so on. This is more persuasive, but it suggests a political economy of hardware, software and service sales that is, quite frankly, terrifyingly opaque. If our goal as sustainability advocates is to understand the hardware ecosystem to intervene in it then how do we even know where to begin or focus our effort?

Another received narrative says that the hardware upgrade cycle happens because of genuine technological innovation – and there’s no denying this is a real component of it. The jumps in performance across console generations (to take the most obvious examples) from 8 to 16-bit, to the 3D era, and more recently going from 1080p to 4K, even I must admit, are all genuine advancements. However the differences between each generation have grown smaller and smaller, and the marketing differentiation has had to resort to ever more granular and artificial restrictions.

Look at these leaked details about the PS5 Pro, as reported by The Verge:

Sources familiar with Sony’s plans tell The Verge that Sony is asking developers to create a new PS5 Pro-exclusive graphics mode in games that combines Sony’s new PlayStation Spectral Super Resolution (PSSR) upscaling to 4K resolution with a 60fps frame rate and ray-tracing effects….

Developers could also choose to enable ray-tracing effects and get the PS5 Pro Enhanced label without improving resolution or frame rates. If a developer wants to target 60fps instead of 30fps with the same resolution, this may also qualify as a PS5 Pro Enhanced game.

To be sure there is a small chorus of serious enthusiasts who care greatly about things like raytracing, high refresh rates, and so on. But overwhelmingly I think a consensus is emerging that – at least on graphical upgrades – we’re achieving diminishing returns. There's even some actual evidence for this now, beyond our 'gut feeling', which I'll outline below.

But this leads to what I think is the final received narrative around games and the endless hardware upgrade cycle, which is that it’s just what we do now. In other words, that sheer inertia is behind much of it now. The audience has been trained to expect it, the companies have built business models around it, and so no one has any real sense of how to depress the clutch that unmeshes these two powerful gears from one another. But it’s going to be crucial that we find a way to do so if we are going to achieve the sustainability transition.

There are two pieces of research that I think are going to be super useful for anyone wanting to work through the challenges in this area (and I hope others do!). The first is Graeme Kirkpatrick’s (2012) study of the formation of the gamer identity through gaming magazines in the 80s and early 90s. The key insight of Kirkpatrick’s is the pinpointing of the invention of the term “gameplay” in the mid-80s, emerging in the discourse as a way to separate the field from general-purpose computing and allow it to establish itself as a distinctive cultural activity. Computing for games vs computing for spreadsheets. The effect of this was to compartmentalise game characteristics into elements like “gameplay”, “graphics”, “sound” and “story”. We can probably guess how essential having the aesthetic criteria of "graphics" was to sellers of new devices with more powerful circuitry.

While discussing this topic with Brendan Keogh in advance of the talk, he sent me half a dozen examples of early ads for game consoles that played off this "graphics upgrade" discourse – which you can see examples of in the slides. We talked about how playground discussion (for those of us who were children in the 90s) often revolved around console wars debates riffing on these numerical comparisons between devices. The discourse persists today but at a far lower level, becoming the preserve of the truly enthusiastic. Splitting hairs about the L2 cache speeds between the PS4 Jaguar APU and the Xbox One Scorpio Engine has lost the school-yard approachability of "The new Nintendo has sixty four!"

The second piece that seems crucial to this question, and which I turned up just while looking for work on this topic, is Yinyi Luo and Mark Richard Johnson’s "How do players understand video game hardware: Tactility or tech-speak?” which looks at whether or not players own understanding of their game hardware adopts the marketing discourses used to sell them. Without having read beyond the abstract (yet), I don’t want to get too far ahead, but their findings seem to support the notion that it matters much less, or in a quite different way. They argue that:

Gamers rarely understand their gaming hardware through these marketing and advertising discourses despite their high visibility, instead framing the technology they engage with in intimately tactile, and more broadly contextual, terms.

This is a doubly astounding finding given that at least one of the researchers (first author Yinyi Luo) worked for Paradox Interactive at the time of publication. The interests of hardware manufacturers and game (software) developers seem to be diverging further and further.

So is the fly-wheel of the hardware upgrade cycle losing momentum? What does a more user-centred understanding of game hardware look like, and can it align closer with a sustainability centred understanding as well? My hunch is that it might look like older, more repairable devices in use for longer, and perhaps ideally, modular or upgradeable devices. Though these are often not the most 'efficient' in terms of end-user energy consumption. (Cf: a typical PC vs consoles power use) Tricky questions.

So I think it's absolutely critical to ask who this system of endless new devices and upgrades is actually serving. I really hope other researchers will start to take up different parts of this question – it's far too big for one person and I am getting busier and busier with other work. (At the very least, I hope Brendan Keogh writes that paper idea he’s been threatening to on this topic! I can think of few better placed to understand and describe the dynamics of the modern game production processes.

In the responses to my talk, Ed and Patrick brought up several compelling themes and debates that they saw as threading their way through my talk, and I was deeply humbled by the gracious and thoughtful engagement I received from both. One theme that was raised by both, as well as by Chloé Germaine in the chat (who had to leave before we had a chance to discuss into further) was the spectre of growth vs degrowth.

If you’re not tuned into the eco-socialist debates that have been on a slow boil for the past couple of years online (and I can’t blame you if you’re not!) they have recently reached a new level of intensity following the publication of Kohei Saito’s ‘Slow Down: Degrowth Manifesto’. The book received the following uncompromising review by Matthew Huber and Leigh Phillips in Jacobin. I’m hesitant to offer too much of my own commentary because I am still far from certain of my own conclusions, but my interpretation of the disagreement is that it boils down to whether or not growth can continue and how to respond to it for those of us on the left-socialist-communist spectrum.

For Saito, it is clear that capitalism is already operating outside of what is sustainable for the long-term health of the planet, echoing the warnings of the world’s preeminent scientists and experts in climate. For Huber and Philips, who do not deny that any of this is happening, they emphasise, however, the impossibility of any approach that relies on “using less” at precisely the time the working class (at least in the United States) is being increasingly immiserated by the forces of capitalist exploitation. They therefore argue that the degrowth agenda is a losing proposition, and in any case, that the “planetary limits” the world facing are more of a product of capitalist relations and the need to make a profit than they are "objective" barriers beyond human capacity to influence.

While it’s easy enough to agree that by removing corporate control of food production, for example, we would unlock the ability to feed the world. We produce more than enough as it is, it’s just inequitably distributed after all, and a more democratic control over production would certainly be a benefit. It's less clear to me, however, that the pro-growth camp can overcome all aspects of the planetary crisis of carrying capacity, of the ability to maintain lifestyles at their current level (not to mention, to spread those lifestyles across the entire world). So while there is merit to several aspects of their analysis, even if we might not agree with it entirely, I find it hard to stomach their conclusions about approach. But these are by no means a settled issue and will play out for years to come.

Which brings me back around to the question of extinctions. Unlike the political-economic problem of the distribution of calories in a world with diminishing (yet still quite sufficient) agricultural capacity, the process of a species going extinct lies outside the realm of the control of the forces of production. Once the white rhinoceros is extinct, that’s it. End of story. We might hope to realise our fanciful sci-fi visions of Jurassic Park-ing some animals back from the brink, but a species is not the same as a single animal – it is an entire gene pool of diversity, an unbroken process of the reproduction of life stretching back for thousands and thousands of years, and that is far harder to replace.

There are irreversible processes underway all across the planet and its biosphere, right now. No amount of shackling of the immense productivity of the state is sufficient to undo hundreds of years of damage from the fossil economy. A terrifying recent publication in Nature presented new findings on the amount of economic damage we can expect – already “locked in” by present concentrations of greenhouse gases in the atmosphere. The climate physicists behind this research claim we can expect:

an income reduction of 19% within the next 26 years independent of future emission choices (relative to a baseline without climate impacts, likely range of 11–29% accounting for physical climate and empirical uncertainty).

The only conclusion I can see is that degrowth is coming, whether we want it to or not. Whether the games industry survives the next thirty years may well depend on whether we can come up with a credible plan for how to adapt.