(Image credit: GOG)
Looking for old games on a modern console can make the problem seem smaller than it is. Super Mario Bros. keeps returning, Pac-Man runs on almost everything, and familiar RPGs receive remasters, collections and complete reissues. The biggest names remain visible.
Video game history, however, is not limited to the handful of titles publishers can sell again every generation.
The Video Game History Foundation and the Software Preservation Network have published a new study today on the commercial availability of classic video games. Its conclusion is considerably worse than even many collectors may have expected: only thirteen percent of the games studied remain in official circulation. In other words, around 87 percent can no longer be bought normally.
This does not mean every one of those games has been physically destroyed. Cartridges remain in cupboards, CD-ROMs sit in attics and disk images survive on hard drives. Some titles are well documented and actively protected by enthusiasts. Yet anyone trying to buy a legal copy today that can be played without exceptional effort will run into a wall with almost nine out of ten classic games.
The disappearance of old games is not a new discovery. Anyone who has tried to recover a title from their youth knows the pattern. The game is absent from modern stores. A second-hand copy suddenly costs tens or hundreds. The required console has a weak laser, leaking capacitors or a video signal modern televisions cannot understand.
Until now, that experience remained largely anecdotal. Players knew many games were missing, but there was no broad study showing how large the problem really was. This made it easy for publishers and industry organisations to point towards the growing market for remasters, retro collections and digital reissues.
The new study attempts to fill that gap. Researcher Phil Salvador used a random sample of 1,500 games released in the United States before 2010. For each title, the team checked whether it remained reasonably commercially available on 15 April 2023.
Publishers were given considerable benefit of the doubt. An announced reissue already counted. An enhanced edition could count as well, provided it was essentially still the same game. Only complete remakes, substantially different versions, unaffordable collector's editions and releases tied to defunct stores or platforms were excluded from normal availability.
Alongside the broad sample, the team examined three ecosystems in more detail: the Commodore 64, the complete Game Boy family and the PlayStation 2. More than four thousand games were checked in total.
The conclusion remained broadly the same everywhere. Whether a system is beloved, forgotten or still regularly represented by reissues makes surprisingly little difference. Most of its games have left the official market.
The individual figures make the overall result even more uncomfortable. Around 4.5 percent of the Commodore 64 games examined remained available. The Game Boy family reached 5.87 percent. Even PlayStation 2, one of the most successful consoles ever made and a platform whose games still regularly return, stopped at twelve percent.
No five-year period examined by the study exceeded twenty percent availability. For games released before 1985, fewer than three percent remained in commercial distribution.
Historically, that last figure may be the most serious. Those early years were when the rules of the medium were being formed. Developers discovered how controls, scoring, space, speed and interaction might work. Not every early game was good, but failures and forgotten experiments also explain how a medium developed.
A history book containing only the winners is not a complete history. Yet that is roughly what the current market offers. Recognisable brands return because they can make money. Games with complicated rights, lost source code, a small audience or little brand value disappear.
That is not an accusation against every individual publisher. No company can reasonably maintain every old product forever. It does show why commercial availability and cultural preservation are not the same thing.
The timing of the study is not accidental. On 27 March, Nintendo closed the Nintendo 3DS and Wii U shops to new purchases. Previously purchased games can still be downloaded for the foreseeable future, but new customers can no longer buy the affected software, DLC or demos.
The closures removed more than modern download games from sale. The 3DS and Wii U were also official gateways to older Nintendo titles. The Game Boy library was hit particularly hard. According to the study, another 6.5 percent of Game Boy games had been available through these shops before they closed. More than half of the Game Boy games that were still commercially available therefore disappeared in one movement.
This is the weak point of digital reissues. They can make old games available quickly, cheaply and without manufacturing new cartridges. But that availability exists only while the store, licence and technical infrastructure remain active.
A game being reissued once does not mean it has been saved. It can disappear for a second time.
A physical release leaves at least a dispersed number of copies behind. They age, become expensive and grow fragile, but they can change owners. A digital game tied to one closed shop often leaves nothing a new player can legally obtain.
Libraries, museums and archives might seem like the natural answer to this problem. To some extent they are. Specialist institutions collect hardware, create digital copies and document games before media or systems fail.
American rules for video games are far more restrictive than those for books, film and audio recordings, however. Under certain conditions, archives can circumvent protection to keep an unavailable game playable. Digital access must, in practice, remain on the institution's physical premises.
For a researcher, this creates a strange situation. An archive may have safely stored a rare game and made it technically functional, while somebody on the other side of the country cannot study it remotely. Their remaining options are travelling to the building, collecting vintage hardware at home or finding an unofficial copy.
With other media, we have long accepted that a library does more than lock an object inside a room. Video game law still appears to assume that every accessible copy is automatically a threat to the market.
The study was designed partly to test that claim. During earlier American copyright discussions, the games industry argued that an active market for old games already existed and that broader preservation access could interfere with it.
A market offering only thirteen percent of its history can hardly replace an archive.
An uncomfortable truth sits underneath almost every discussion about game preservation: much of the medium's history remains playable because people made copies without permission.
ROM sites, abandonware archives, torrents and private collections perform functions official stores and public institutions cannot fully provide. That does not make every form of distribution automatically right or legal. It shows that demand for access continues after the official route disappears.
For many games, the practical options are now limited to an expensive second-hand copy, fragile original hardware, a visit to a specialised archive or a download from a legally grey source. None is a reasonable default for education, research or a player who simply wants to understand how a particular game worked.
Piracy is therefore not the cause of the preservation problem. In many cases, it is the unofficial response to a market and legal framework that provide no usable alternative.
The irony is that unauthorised copies are often the most widely distributed and therefore the hardest to lose completely. A legally purchased download can disappear with a single store. A ROM copied thousands of times continues to exist somewhere.
The study also makes an important distinction between availability and familiarity. An old name returning to the shop does not necessarily mean the original game has been preserved.
A remaster may remain close enough to count as a reissue. A full remake is a new work, even when the story, characters and locations are largely the same. The original 2005 version of Yakuza, for example, is not the same historical object as Yakuza Kiwami.
For players, a remake is often more attractive. It runs on modern hardware, looks better and avoids old technical limitations. For historians, the differences are the point. Controls, aspect ratio, interface, loading times, voice work and even bugs reveal something about the moment in which a game was made.
Preserving only the improved version also preserves a later interpretation of the past.
The same applies to collections where music is missing, features have been altered or emulation behaves differently from the original hardware. A reissue is valuable, but it does not always replace the original.
Perhaps thirteen percent is enough for somebody who only wants to play the best-known classics. Major Mario games, a selection of Sega titles and a few famous RPGs will probably keep returning. Preservation, however, is not only about saving the best or most popular games.
Mediocre titles, regional releases, educational software, licensed games and small experiments belong to history too. They show which techniques were normal, how publishers thought, which audiences they tried to reach and which ideas ultimately led nowhere.
That material is often the first to disappear commercially. An obscure game has little chance of receiving a remaster and too few customers to justify an expensive rights investigation. When its original platform closes, nobody has an immediate business reason to release it again.
An archive has a different purpose. It does not need to predict which game will be profitable next year. It needs to prevent future researchers from seeing only what happened to remain saleable in our time.
The Video Game History Foundation does not use the report to present a complete new law or distribution service. Its most important step is simpler: demonstrating that the commercial market is not solving the problem by itself.
That evidence now exists. Around thirteen percent of classic games released in the United States remain commercially available, with low availability in every period and platform examined. Digital shops can expand the selection rapidly, but one closure can undo that work in a day.
A new American proceeding on exemptions to the DMCA is scheduled for 2024. Libraries and archives will once again seek more room not only to preserve games technically, but to provide useful research access.
Whether they will succeed is unknown today. What is clear is that the discussion can no longer be dismissed with the claim that publishers already keep their history adequately available.
Almost nine out of ten games say otherwise.
A medium that has grown larger than film and music should not need to wait until its early history is as old as a silent film before discovering that preservation mattered.
Sources: Video Game History Foundation: 87% Missing, Video Game History Foundation: game availability study explained, Phil Salvador: Survey of the Video Game Reissue Market in the United States, Nintendo: Wii U and Nintendo 3DS eShop discontinuation.
