The Graphics Plateau Became A Cliff: The Uneasy State of Games!
Sometimes I feel like I was born a generation too late.
The kinds of games I love and that I loved growing up aren't getting made that much anymore. There was an entire sub-industry dedicated to smaller handheld games, continuing the tradition of pixel art games. Handheld games got killed in favor of the modern mobile games industry which prioritizes dark patterns and a constant dopamine drip. Development times used to be quicker across the board. You could make a full priced game in 2 years or less, now it's 5 years on average if you're lucky (OR MORE if you're not).
TLDR: The games industry is at an inflection point. The graphics arms race driven by the biggest and most expensive games has become untenable. Realism is a trap. We reached the point of diminishing returns in regards to resolution and visual fidelity two console generations ago. Graphics technology has plateaued by all meaningful metrics, and now requires AI upscaling to even move the bar forward. But the bar keeps being raised for a shrinking audience that do not care at the expense of the people who make the games. In order to grow, the games business needs to embrace the fun and whimsy of decades past as inspiration for new and exciting experiences that can be led with by less specialized teams. Projects need to be smaller, more experimental. Instead of blowing the budget on art assets, iterate and find new mechanics or fresh takes on classic genres. Instead of betting the farm with one $500 million project, you could diversify the investment with 20(!) $25 million projects.
DISCLAIMER: Throughout this essay I'm going to be using the royal "we" when referring to the games industry. As someone who has been actively working in game development for over 10 years, I'm painting with a BROAD brush. This is meant to be a deconstruction and examination on my feelings about the current state of the games industry, how we got here, and how I hope we can do better. I'm going to do my best to keep my opinions about outside factors to a minimum - like politics and the global economy. But I would be remiss to not point out how we live in a capitalist hellscape, and that the "number must go up" mentality is the only thing that gets the oligarchs/powers-that-be who steer the ships we're all strapped to, out of bed. As long as the ship reaches its final destination and the captain who steered it gets their treasure, I mean who gives a shit if the ship treads water from neglect and all of the crew drown pointless deaths, right?
Gestures to the current state of things™
...OK, let's get spicy...
I // 4K: The Arms Race Toward The Top The Bottom
I joined the games industry in 2014 at the cusp of games being made with 4K as the target. Early on I could see the friction this was causing in the pipeline of my own discipline, let alone what it was doing to every other discipline under the development umbrella. It's simple math. Every discipline under the umbrella is now hyper specialized, what used to be one person's job is now 3+. Tasks take longer to do because we're now authoring at higher and higher fidelity, we're all spinning more and more plates, and more people now need to be involved in every decision. Why are we chasing 4K resolutions? Because the games industry has been propped up for its entire existence on the mercurial target of realism. Advancements in graphics technology, is the easiest selling point for new games. Box B pushes more pixels out than Box A; it's visibly more than the box that's already sitting under your TV. It markets itself. But the arms race, perpetuated through the AAA space, hit mach speed with the advent of HD TVs. Soon game engines would be incorporating new pipelines for more realistic lighting and more realistic texturing than their competitors. It seemed like there were new processes and software to learn each year. PBR (physically based rendering) would soon be the standard, even if realism wasn't the end goal for your game.
On the media side of things... pixel peeping - the act of zooming in on recorded footage and counting how many pixels are rendered, comparing how specular the highlights are on your gun based on the box (console) or how shiny reflections look with ray-tracing turned on - would become a whole cottage industry (derogatory) and become an overly emphasized portion of the conversation about a game instead of... I dunno... talking about if the game is even fun. How dare one box run a game at 900p instead of 1080p! The horror! (P.S. Digital Foundry's initial praise of DLSS 5 and backtracking of that praise after the negative audience reaction is the most mask off moment I've seen in a while. Treating technical analysis as a form of taste making has been such a shitshow, and I know they're just fans of technology, but the grifters that have popped up in that space is a real problem that we've yet to contend with.)
It's exhausting. This conversation has always been exhausting. Because so much of it takes place in bad faith echo chambers, amongst chuds who are looking for a "gotcha" against people who bought a different box like they're fans of a rival sports team. I'm attempting an honest take based on my own personal experiences.
When authoring things at 4K we're zooming all-the-fuck-way-in to like 400% in Photoshop, and doing what I jokingly describe to friends and family, who occasionally will listen to my soapbox, as "pixel fucking". We're basically pixel peeping ourselves because no one wants to be called out and dogpiled on Reddit or Twitter. We've been brought up in an industry that has been and has rewarded "perfectionism" i.e. realism. This is a part of crunch culture, but maybe I'll get to that another day... I'm trying to stay focused.
I was a child of the 90s. Toy Story was a watershed moment in entertainment, and I grew up reading or hearing some portmanteau of "it looks like a Pixar movie" all the time as a description for new games. There's always been a new landmark game to push the bar forward. One that lights the fire in artists hearts, and presents them with a new benchmark to strive towards in the never ending march towards realism. It used to be how big and detailed the sprites could be in 8/16-bit games. Then with the advent of 3D it became how detailed the textures were, how long the draw distance was, how many tris/polys were rendered on screen, and how dynamic the camera was. GTA 3 showed that an entire 3D city could be a sandbox of creativity, and Gran Turismo showcased how good hard surface rendering was coming along. Crysis melted GPUs that attempted to max out the graphics settings. The Last of Us's anal retentive perfectionism became the template for pretty much the entirety of the first-party PlayStation portfolio. But we've reached the point of diminishing returns. Heck, I'd argue that we reached that point during the Xbox 360 and PlayStation 3 generations when $50m was an eye-watering sum that only the biggest studios could command before 4K was even a twinkle in the eyes of a marketing team. Fuck, we can't even hit those 4K resolutions natively without AI upscaling. So when you're running DLSS or the equivalent, your machine is running the game at lower settings under the hood and using AI to interpolate an approximation what it looks like at the highest settings. None of this shit is real. It's saw dust that's sold as fairy dust, and the industry keeps touting it and ray-tracing as the "next big thing", I think partly because the reality of confronting our utter lack of creativity is too hard to stomach, and partly because ray-tracing and AI upscaling are directly tied to NVIDIA's rise as a cornerstone of the world's economy through its AI push, and thus... it's all connected to the number going up.
All I've been hearing for a majority of my career in games is that the budgets are getting too high. This was when the budgets were encroaching on $100m mark. We're so far past that these days with estimated budgets of AAA games reaching mid 9-figures on average, and the GTA 6 (expected to be the largest entertainment launch in history) is rumored to be well past 1 billion in the hole once marketing is factored in. It's crazy pants, man. Now we're getting AI shoved down our throat (another essay for another day) from the corporate class that led us down this road as a solution to these budgets, but it's a band-aid on top of a festering wound. How can they sell the next game or the next box if it doesn't "look better" than what came previously? This shift in how games get made came at the same time as a shift in the business. MBAs in charge are now betting the farm on "forever games" or "live-service games". Games that can print money in perpetuity. They've uncovered an "infinite money glitch". This push for the prettiest games and the siren call of unlimited money pulling from both ends has made for an industry that is essentially a house of cards built on top of shifting sand. We feel the effects of each disastrous decision as they ripple across the industry, and it can be seen with every mass layoff since the influx of VC and investment money into the space since the COVID boom.
The industry already had a retention problem long before COVID. Now, I think we've got a retention crisis. I call it a crisis, not to be alarmist, but because people are leaving in droves - either through the onslaught of constant layoffs through no fault of their own, or by choice. We were sold a bill of goods with remote work that turned out to be a lie (oops, we actually spend a lot on expensive office space that we need to justify and need you to return to office) and they rightfully will not relocate their families cross country for a job that might disapear in another year. This industry is simply too unstable. In today's landscape you can get laid off after delivering a critically acclaimed hit, or have a studio shuttered for a game that sold well but maybe not "infinite money glitch" well... And well... The number has to keep going up!
II // You Don't Know What You Don't Know
You might have noticed a rising trend of games having performance problems. I firmly believe that part of this can be chalked up to the industry at large having forgotten how to make games performantly and cost-effectively. There are techniques and best practices that have been lost and found over and over again. We've just been stuck reinventing the same wheel. A lot of that knowledge gets lost to the ether as people leave the industry or get pushed out. I can't tell you how many project management pipelines I've been part of at this point in an attempt to wrangle the chaos. It's an inside joke that we try to "pivot to Agile" every few years.
Again, I'm painting with a broad brush, so I'll use a personal anecdote here. There's a lot of stories that aren't mine to tell, but on Borderlands 3 my team consisted of 0 people that had worked on a Borderlands game before. Most of the designers on my team had never shipped a game with Scaleform (middleware software that allows for UI graphics built in Adobe Flash). Hell, half of BL3's development was spent with us learning Unreal Motion Graphics (UMG) after the switch to Unreal Engine 4, before we had to pull the rip chord for reasons, and Scaleform was the pipeline that was already proven on previous games.
Speaking only for myself, BL3 was my first AAA game. I learned a lot about how to make a game and how not to make a game during its development. It was a great way to cut my teeth and develop as a designer, but I spend not an insignificant amount of time thinking about how I would do things differently NOW that would have made that game's UI a hell of a lot better. But we were seriously hurting bad from brain drain during this time, and there's not exactly a guidebook to reference on how to do certain things or a button you can click in Unreal that automagically makes the game run better. So a lot of us were flying by the seat of our pants and making things up as we went along, making the most informed decisions with the knowledge we had at the time. Simply put, you don't know what you don't know. That's game dev bay-bee!
We called it Tribal Knowledge. Things that were already figured out on previous titles, some that we had access to through documentation but some thinge that we were trying to figure out again. (insert O'Reilly "we'll do it live" meme). And then there are the Eldritch learnings of how things operate under the hood that you simply can't know until you're forced to reckon with how things have been built, because no one had the time to document best practices and because middleware can be a black box. For example, there was a point late in development where we had to go back and convert all of the UI animations from Classic Tweens to Motion Tweens in Flash for performance reasons. Classic Tweens were OK in previous titles. I know this because I was using old Flash files to reverse engineer how things were made on older titles. But for a variety of reasons, this was a performance win we could make without rebuilding all of the UI (again). Game Development is a series of whack-a-mole encounters, always putting out little fires. Maybe we should have had a much more strict process of using textures that were either a power of 2, or divisible by 4. Why? Well, computers like those numbers and it's more efficient on memory. It's broad knowledge like that that's not something that's ever spelled out as a newcomer to game development, which fundamentally reshapes how you'd think about making art for games.
III // The Roblox of It All
Games have always been a hit driven business. Trendsetters that can be copied and reproduced for more hits. Pac-Man and Asteroids spawned knock-offs in the arcades. More hits means more money.
Then comes software-as-a-service (SAAS) trends, popularized by Adobe in the early 2010's and adopted by the vast majority of all tech companies in the years that followed has been a net negative for consumers all around the world. It helped normalize enshitiffication for products, and never owning your software again. Paired with free-to-play (F2P) business model of games popularized in Asia, you get games-as-a-service (GAAS). The GAAS model promises predictable recurring revenue, if your game is popular (load bearing part of the statement). And if your game is really popular you now get access to infinite money through in-game purchases, usually either through micro-transactions in the form of cosmetics to alter your in-game appearance (skins) or gambling on the random chance of a new character/item/skin.
I'm not going to go deep into all of the problems surrounding Roblox, but I'm of the opinion that it's a gross product that needs to be regulated into the ground, along with lootboxes/gacha style games. Maybe you're unfamiliar with the game, I mean most people that traditionally play games tend to not be aware of the issues, but it's essentially a service that allows users to make experiences/games inside of the Roblox format that can then be monetized for a revenue split between the creator and Roblox as a company. The libertarian attitude with how anything can be monetized is super fucking heinous and has few guardrails for protecting kids from dark pattern behaviors vying for their attention and their Robux gift cards. This is a problem, because kids are either getting burned by their horrid experience of Roblox taking their money and they leave games as a hobby entirely, or they can get sucked deeper into some increasingly insidious patterns under the guise of "it's just a game".
So... Kids aren't buying games these days. Kids are on Roblox. Studies have shown that kids are more interested in the social experience of playing with their friends in Roblox or Minecraft or Fortnite - or whatever else the industry has come to label "forever games" (games that become the destination for all of a player's time vs players interacting with a variety of releases). 2/3 of these previously mentioned "forever" experiences, visual fidelity is not even in the top 10 selling points. They're not exactly pretty, and kids don't care. What the industry is grappling with is a generational shift in the habits of people who play games. Instead of reckoning with these very obvious trends that threaten the core business of the entire industry; we've seen companies triple and quadruple down on squeezing their current consumers for micro-transaction money in order to...? That's right! Keep the numbers going up!
It's been a not-so-secret secret for a while now in the industry that all of the innovation in games is coming from the indie space. The AAA space has become so calcified and inflexible, with the budgets as big as they are, that decision making from the top down is extremely conservative. Which means when the bets from on high aren't trying to replicate Fortnite money - we see the landscape for what it currently is with the rut of sequels, open-world checklist games, and narrative driven third-person action games with a "cinematic" level of polish that only the biggest studios are able to pull off.
There's a reason "friendslop" as a genre has risen in popularity in a way that I don't think can be replicated at scale by larger teams. There's a homegrown quality to the lofi aesthetic of most friendslop games that would make a C-Suite of a publicly traded company shudder. But it works because it's earnest. It's authentic. It's charming. It's human. You feel the love of the craft. Also, these games are a lot of fun with friends.
IV // Take a Step Back - Emphasize The Fun
For too long this industry has had a chip on its shoulder, often using other media like movies and prestige television as direct influences. For a time it seemed like the more cinematic the game, the more it sucked up all the attention in the space - with lots of attention paid to simulated film grain post processing, how its glossy cutscenes felt like a movie, how realistic the motion capture animation and facial capture were, or how sad the dads are in these games.
Nintendo saw this coming. Back in the early 2000's after a lackluster generation with the GameCube, which had comparable technology under the hood to the Xbox and PlayStation 2, Nintendo realized that the graphics arms race is a losing battle. They adopted the Blue Ocean strategy which targeted new ways of play with the Wii and Nintendo DS, in order to appeal to more gamers on the periphery of the space. They had the right idea, and on paper their hardware was lackluster when compared to their direct competitors in the Xbox 360/PlayStation 3 and PSP respectively. But the games didn't suffer. In fact, they flourished. Sure visually they looked stunted in comparison. Motion controls became a gimmick, certainly due to low-quality bargain bin releases trying to cash in on the meteoric success of the Wii. But the core games were good, and were thoughtfully designed. Mario Galaxy is still considered the best 3D Mario by many. I'd sooner break out my Wiimote to play Wii Sports or Punch-Out than some of the latest live-service trend chasing. The DS is still, 22 year later, a perfect system. The DS Lite with its rounded corners and soft edges is a sublime piece of industrial design, and an absolute joy to play. It was the last bastion of pixel art as a mainstream aesthetic. As a fan, I have a passion about my experiences with the games on those consoles that I simply do not in modern AAA games. This isn't to say that the Wii did everything better than the 360 or PS3. They hadn't figured out online features quite yet. Not every game utilized the hardware well. Heck, 20+ Years after Xbox Live made an online ecosystem that was a novel and engaging experience (for a generation of people who might not have played games on the PC) Nintendo is still figuring out what online play can be. But you know what... that's fine because they allow me to play games in a handheld format while sharing the TV space with my wife. That's an experience that can't be replicated with the other boxes in a seamless way that feels good for all types of games, as much as they may wish streaming would be an answer. PlayStation Portal or streaming from a modern console to a handheld like SteamDeck is still pretty clunky, and introduces a level of latency that makes me avoid games that demand precise inputs.
What I'm getting at though is the Nintendo philosophy of making hardware in support of new ways of play, and how they evolve their core franchises with new mechanics afforded to them by the hardware. Nintendo acts like a toy company. The act of play, and the player experience is at the center of all of their creative decisions. The majority of the games industry operates like the tech industry - treating games only as software, and relying too much on metrics like player retention and monthly active users.
As things stand, I don't think anyone wants new boxes... Which worries me, because there's an underlying feeling like things are about to implode if we start a new generation where the only argument to upgrade is "more graphics". Hell if games weren't my livelihood, and I was just a casual fan I feel like I could have gotten by swimmingly without upgrading to a PlayStation 5 or Xbox Series X.
I'm pretty lost in the sauce when it comes to gaming discourse de jour. And since the very start of the PS5/XSX console generation, the general sentiment I have heard overwhelmingly in enthusiast circles is that folks are still waiting for this generation to start. Part of this is likely the warped sense of time and collective trauma from the COVID lockdowns that we as a society haven't grappled with yet. But at the time of writing, based on previous console timelines we're probably staring down the barrel of new boxes supposedly coming in the next year or two (if the AI bubble doesn't pop and collapse the economy, or if World War 3 doesn't kick off over Taiwan). No one I've talked to in my personal or professional life is excited about a new box, especially when it feels like the current boxes never hit their stride.
We've lost the plot so bad as an industry and it hurts. I wish more people in positions of power in the games industry would make more decisions with the ethos of making something to ship instead of hand wringing about the fidelity of everything. Rough edges are good. Rough edges are character. Rough edges feel human. If you sand away all of the rough edges, you're left with something shiny that also feels hollow. Labeling something AAA feels like a self-report that something went wrong on your project. If you're starting a project with the expectation of a mid 9 figure budget, something is wrong. If you overshoot that budget, something is really wrong. Stop betting the entire farm on the latest trend. Collectively, we need to make games smarter. AI will not save us. Simply ask the question, "OK, but would this make the game more fun?". This is a subjective question that can only be answered depending on the kind of experience you and your team are making, but be brutally honest with yourselves. Is the answer yes? OK, go with God and spin your wheels a little. Sweat over how convincingly realistic the subsurface scattering of your skin is, I guess. Is the answer no? Fucking move the fuck on, players WILL NOT care how much you stressed over the fidelity of your art. Get it to 80%. Get it good enough, and if you have time polish later. Just because the box gives you more room doesn't mean it needs to be used. Be more efficient. The modern consumer responds to authenticity - the most 4K, most ray-traced graphics scream corporate which is the antithesis to what they're looking for. Elden Ring is one of the biggest successes of this generation and from a technical perspective it looks marginally better than Dark Souls, which was never known to be a technical powerhouse. FromSoftware, like Nintendo, emphasizes the fun. Their games are sticky, and players notice. I'm sure if the biggest games of the world devoted their immense resources to something akin to a PS2's rendering fidelity but with modern improvements in animation and gameplay, they'd still do numbers, and continuing to support them with post launch content would be even easier and sustainable.
Sustainability is key here. I have dreams and ambitions for making my own games someday, but I'd also like to be able to retire making games with my friends. It's a big ask, I know. Obviously I'm being cheeky here, but maybe this blogpost can serve sa a timecapsule I can return to after being in the industry another 10-15 years, and future me can compare notes.
If you made it all the way here to the end, thanks for reading. I'm recovering from a bad cold, and maybe got too candid and border on too pessimistic. Maybe it's the DayQuil and low-grade fever. I appreciate your time though, regardless of who or where you are.