Part 1: 1990 to 1993 —The Dinosaurs Wake
Part 2: 1994 to 1996 — The Spark Catches Fire
Part 3: 1997 to 1999 — The Wild West

(These are the opening three sections in a six-part history of emulation series, and they are still very much a work in progress. Tracking down reliable sources for material that has long since disappeared is not always easy, so the record here will continue to grow and improve over time. I also plan to add more images, screenshots, quotes, and other material as this project develops, so feel free to check back periodically for updates. If you have additions, corrections, comments, or insights to share, please reach out—your contributions are welcome and will be included as this project evolves.)

Game Emulation History Part 2: 1994 to 1996 — The Spark Catches Fire

By the middle of the 1990s the world of home computing had entered a new age of possibility. The Intel 486 had already given way to the early Pentium processors, VGA graphics had been replaced by SVGA cards capable of sharper resolutions, and sound hardware such as the Sound Blaster 16 gave ordinary users audio fidelity that only high-end studios had enjoyed a few years earlier. Optical drives were becoming standard, CD-ROMs were beginning to replace floppy disks, and hard drives were finally affordable enough for families to keep growing software libraries installed permanently on their systems. For most consumers these changes meant faster games, smoother productivity, and the promise of multimedia titles that could blend video, animation, and music into experiences that felt futuristic. But for a small underground community, the leap in processing power meant something else entirely. It meant that, for the first time, personal computers were strong enough to begin convincingly impersonating video game consoles. What had been fragile and proof-of-concept in 1990 was about to become a global underground movement.

(HEY YOU!! We hope you enjoy! We try not to run ads. So basically, this is a very expensive hobby running this site. Please consider joining us for updates, forums, and more. Network w/ us to make some cash or friends while retro gaming, and you can win some free retro games for posting. Okay, carry on 👍)

The earliest experiments had already proven that emulation was possible. Haruhisa Udagawa’s Family Computer Emulator on the FM Towns in 1990, Yuji Naka’s secret Mega Drive Famicom project in 1991, and Nobuaki Andou’s Pasofami in 1993 showed that consoles could indeed be tricked into running on machines they were never designed for. But these efforts were limited either by their secrecy, by their narrow user base, or by crude performance that left them as novelties more than platforms. By the years 1994 to 1996, however, the ground had shifted. CPUs were faster, memory was more abundant, and developers were discovering that code written in assembly or portable C could take advantage of these gains. Suddenly emulation was not a rumor, not a laboratory trick, but something spreading across borders and into the hands of ordinary users.

The distribution networks of the time gave this underground movement its identity. Long before modern websites and torrents, software moved across bulletin board systems (BBSes), where dial-up users traded ZIP files overnight. Communities sprang up in Usenet newsgroups, with alt.binaries.emulators becoming one of the earliest hubs for sharing both emulator executables and ROM images. By the mid-1990s the early web began to catch up, with fan pages hosted on GeoCities, Angelfire, and Tripod, each one curating lists of downloads, FAQs, and compatibility tables. The arrival of dedicated emulation portals such as Zophar’s Domain (founded in 1996) and Archaic Ruins gave users a place to find the latest builds and to read news about what projects were in development (Zophar’s Domain history). These networks did more than just distribute software. They created culture. They introduced a shared vocabulary around terms like ROM dump, save state, and compatibility list, giving the movement its own slang and lore.

The cultural climate of the mid-90s made this even more electric. Nintendo and Sega were locked in the most visible console war of the decade, trading barbs in television ads and magazine spreads. The Super Nintendo Entertainment System and the Sega Genesis were battling for dominance in homes around the world, while arcades still roared with the sounds of Street Fighter II, Mortal Kombat, and Virtua Fighter. To fans, these machines were rivals and icons. But to underground programmers, they were just hardware puzzles waiting to be cracked. The irony was not lost on anyone that while Nintendo and Sega’s marketing departments were insisting their consoles were unique, hackers were proving in real time that the differences could be collapsed into software. To run Super Mario World on a PC, or to make Sonic the Hedgehog appear outside of a Sega console, was heretical and thrilling. It was rebellion with code.

Technically, the leap was finally feasible. A Pentium 90 MHz machine with SVGA graphics could handle not just the 8-bit NES or Game Boy but, with careful programming, could begin to approximate the 16-bit complexity of the SNES and Genesis. While performance was still rough and sound often garbled, the sheer novelty of seeing Final Fantasy IV or Super Mario Kart boot up on a DOS PC was breathtaking. It was not accurate, but it was alive, and for the users who discovered these programs on Usenet or in a ZIP file from a friend, it felt like forbidden magic.

By 1994, the release of VSMC, the first public SNES emulator, made the unthinkable a reality (Emulation General Wiki). In 1995, ZSNES began circulating, written in raw assembly and optimized for speed, quickly becoming a darling of the BBS scene. By 1996, Snes9x appeared, bringing portability and accuracy to the table, while NESticle exploded into mainstream awareness with its irreverent style and unprecedented usability. Alongside these headline projects were others that deserve recognition: Virtual GameBoy (VGB) brought handheld emulation into view in 1995, Callus allowed Capcom’s CPS1 arcade classics like Final Fight and Street Fighter II to be played on DOS PCs, and Genecyst gave Sega fans a reliable Genesis emulator for the first time. Even ambitious multi-system projects like MESS (Multiple Emulator Super System) began to appear in 1996, planting the seeds for the archival mindset that would later define MAME (RetroReversing on early emulators).

The underground nature of emulation also gave it a rebellious identity. Because emulators required ROM images, players had to learn the language of cartridge dumping, hex editing, and file-sharing. Because companies like Nintendo and Sega were fiercely protective of their intellectual property, emulation existed in a legal gray zone, with debates raging on newsgroups about whether owning a ROM without the cartridge was theft or preservation. This gave the scene the energy of a counterculture. To download Super Mario Bros. 3 in 1995 was not just to play a game but to participate in a new digital underground that rejected the limitations imposed by hardware manufacturers.

By the end of 1996, emulation had grown from a rumor into a storm. It was no longer about whether consoles could be emulated—it was about which emulator was fastest, which had the best compatibility, and which site had the newest dumps. Communities were forming not just to trade files but to collaborate on development, submit bug reports, and argue about accuracy versus speed. For the first time, emulation had a sense of continuity, with projects like ZSNES, Snes9x, and NESticle being updated and improved over time. The scene had begun to professionalize, even if it remained underground.

The spark that had been lit in 1990 was by now a flame that could not be extinguished. The years 1994 to 1996 marked the true birth of emulation culture, the point when personal computers were strong enough, when distribution networks were fast enough, and when programmers were bold enough to turn fragile experiments into tools that millions would one day use. It was messy, controversial, and at times dangerous, but it was also exhilarating. These years were the ignition point—the moment when the underground began to roar.

SNES Emulation Emerges (1994–1996)

By the mid-1990s, the Super Nintendo Entertainment System stood as one of the most beloved consoles in the world. Released in 1990 in Japan as the Super Famicom and in 1991 in North America, the machine was a technological leap from the 8-bit era, boasting a Ricoh 5A22 CPU, the famed PPU (Picture Processing Unit) that allowed for rotating and scaling backgrounds through its iconic Mode 7, and audio hardware designed by Sony’s Ken Kutaragi, who would later be remembered as the father of the PlayStation. For millions of players, the SNES represented the pinnacle of console gaming, home to The Legend of Zelda: A Link to the Past, Final Fantasy IV and VI, Street Fighter II, Donkey Kong Country, and countless other classics. For emulator developers, however, the SNES represented an enormous challenge. Unlike the relatively straightforward architecture of the NES, the SNES was a complicated beast with multiple processors, specialized chips, and a sprawling library of games that made heavy use of custom cartridge hardware.

Still, the same improvements that made PCs multimedia machines in the early 1990s—Intel 486 and Pentium CPUs, SVGA graphics cards, and Sound Blaster 16 audio hardware—also meant that for the first time, it was at least conceivable that a personal computer could begin to approximate the SNES. Between 1994 and 1996, this leap went from theory to reality, as a handful of programmers across the world released the first true attempts at SNES emulation.

VSMC (1994): The First Public SNES Emulator

The honor of being the first known SNES emulator goes to VSMC, released in 1994. Its name, Virtual Super Magicom, paid homage to a popular disk copier peripheral that allowed Japanese users to dump Super Famicom cartridges onto floppy disks. The programmer behind VSMC was Chris George, better known online as “The Brain”, who was part of the Damaged Cybernetics group (SnesLab Wiki, Zophar’s Domain archive).

Chris George, better known in the underground as “The Brain,” was a central figure in the mid-1990s emulation scene and a member of the notorious group Damaged Cybernetics, a loose collective of coders and hackers who pushed boundaries both technical and cultural. His most remembered contribution was VSMC (Virtual Super Magicom), the first publicly released SNES emulator, which he coded in 1994 and spread through BBS networks and Usenet, making him one of the earliest to prove that 16-bit emulation was possible on consumer PCs (SnesLab). Within Damaged Cybernetics, George earned a reputation as both a pioneer and a provocateur, embodying the chaotic spirit of the emulation underground where technical breakthroughs mixed freely with controversy, irreverent humor, and defiance of the gaming industry’s guarded walls. (More about Damaged Cybernetics later on in this publication).

VSMC was crude by any measure. Running on a 486 PC, it could display some graphics and handle limited gameplay, but it was agonizingly slow and compatibility was almost nonexistent. Sound was entirely absent, and even basic rendering features such as transparency were broken or missing. Nevertheless, the fact that any SNES software could be coaxed into running on a DOS PC was historic. It was the first step toward proving that a console as complex as the SNES could be emulated, and it spread quickly through BBS systems and Usenet groups where curious users tried to run Super Mario World or F-Zero, only to find their machines grinding nearly to a halt.

For preservationists, VSMC was the “proof of concept,” the moment when theory crossed into practice. It gave the community the confidence that SNES emulation was possible and inspired other programmers to take the concept further.

ESNES and NLKSNES: Sound versus Speed

The mid-1990s saw two competing SNES projects that embodied different philosophies.

ESNES, short for Expert SNES, was developed by Lord ESNES and Ishmair, programmers known in early emulation circles (Zophar.net). ESNES focused on sound emulation, an aspect that most early projects ignored. This made it attractive to users who wanted to hear the SNES’s iconic audio tracks, from the sweeping score of Final Fantasy VI to the jazzy bounce of Super Mario World. Unfortunately, ESNES sacrificed speed for this accuracy, running slowly on most PCs of the time.

Meanwhile, NLKSNES was built with the opposite philosophy. Designed for raw speed, it stripped away sound entirely and focused on producing playable frame rates for action-heavy games. While the developer’s name is less well-documented, NLKSNES gained traction among users who wanted smoother gameplay even at the expense of fidelity.

The existence of these two projects side by side revealed an early tension in the emulation scene: should emulators strive for accuracy, faithfully reproducing every aspect of a console’s behavior, or should they prioritize speed and accessibility, even if it meant cutting corners? This debate would resurface for decades in projects like ZSNES versus Snes9x and later Dolphin versus Cemu, but in the mid-1990s it was embodied by ESNES and NLKSNES.

NLKE: The Attempted Fusion

Recognizing their complementary strengths, the developers of ESNES and NLKSNES attempted to merge their work into a single project, known as NLKE. The goal was simple: combine ESNES’s sound emulation with NLKSNES’s speed. In practice, the result was buggy and unstable, but it represented a spirit of collaboration that was beginning to emerge in the emulation scene. The fact that separate developers would attempt to merge their code at all was a sign that the community was evolving beyond isolated experiments into something more cooperative and iterative.

The reality, however, was less than perfect. NLKE proved buggy and unstable, with compatibility issues that prevented it from achieving wide adoption. Still, the very attempt signaled that the emulation scene was evolving beyond isolated solo efforts toward something more iterative and community-driven. Even though NLKE itself never rose to prominence, its role as an experiment paved the way for more polished projects like Snes96, Snes97, and eventually Snes9x, which would finally strike the balance between speed, accuracy, and portability. According to its Zophar’s Domain listing, NLKE was officially discontinued on January 1, 2000, closing the chapter on one of the era’s more interesting collaborative experiments. It paved the way for more mature projects like Snes96, Snes97, and eventually Snes9x, which would successfully balance speed, accuracy, and portability.

Snes96 and Snes97 (1996): Toward Usability

By 1996, two new SNES emulators arrived that signaled the underground was maturing: Snes96, created by Gary Henderson, and Snes97, written by Jerremy Koot. These were not the first, nor the most accurate, but they were polished enough to attract a growing user base that was hungry for something more stable than the glitch-ridden experiments that had come before. Unlike those rough early builds, Snes96 and Snes97 offered menu-driven interfaces, save states, and noticeably better compatibility with popular titles. They still faltered when faced with the SNES’s trickiest features—transparency effects, sprite layering, or cartridges that used special enhancement chips like the Super FX—but they could run Super Mario World, Final Fantasy IV, and Street Fighter II with enough playability to keep fans glued to their monitors.

Both developers brought different strengths and priorities. Henderson’s Snes96, a DOS-based emulator that also saw ports to Unix-like systems, demonstrated his focus on portability and cross-platform accessibility, even at a time when most emulators were Windows-only. Koot’s Snes97, by contrast, took advantage of Windows 95 and DirectX, making it smoother and more user-friendly for the growing number of PC gamers running Microsoft’s new operating system. Together, these projects highlighted the different directions emulation could take: Henderson’s eye toward reach and maintainability, and Koot’s toward playability and responsiveness.

What makes their collaboration so significant is not just the individual achievements but the way they converged. In July 1997, Henderson and Koot decided to merge their efforts, combining their codebases into what became Snes9x, one of the most important emulators in history (Wikipedia). Written in portable C, it could run not only on DOS and Windows, but also on macOS, Linux, SunOS, and eventually even mobile devices. This emphasis on portability echoed Henderson’s early vision while carrying forward the usability focus Koot had brought into Snes97.

Their work also became wrapped in small pieces of emulation lore. A 1998 MacOS SNES emulator called “Silhouette” was rumored in some corners of the community to have been a stealth release derived from their work, possibly even by Henderson himself, though this remains unconfirmed and is often dismissed as an elaborate hoax (Nintendo Life). Regardless of speculation, what mattered was that Henderson and Koot’s contributions turned SNES emulation from a patchwork of hobbyist demos into a serious, cross-platform project that could be maintained and improved for decades.

Even before their famous merger, Snes96 and Snes97 represented a turning point. Emulators were no longer obscure curiosities traded in zip files on bulletin boards. Thanks to developers like Henderson and Koot, they were beginning to look like real tools—software that ordinary users could download, configure, and enjoy with relative ease. Their partnership did more than bring Super Mario World to a PC screen. It helped redefine emulation itself, transforming it from an underground experiment into a culture of preservation, playability, and shared access.

Super Pasofami (1996): Commercial Emulation Returns

In 1996, Nobuaki Andou, the developer of the infamous Pasofami NES emulator, returned with Super Pasofami, a commercial SNES emulator. Like its predecessor, Super Pasofami was distributed as shareware and marketed as a legitimate product rather than a hobbyist experiment (Emulation Fandom).

Super Pasofami carried over many of the quirks of the original Pasofami, including its controversial copy protection routines, which could reportedly damage system files if tampered with. Yet it also provided a more structured and user-friendly experience than many of its freeware counterparts. By continuing to market emulators commercially, Andou raised difficult questions about the legality and ethics of emulation. Was it preservation? Piracy? Software innovation? Nintendo, already aggressive in defending its IP, undoubtedly saw it as a threat.

Despite the controversy, Super Pasofami gave users in Japan an accessible way to experience SNES games on their PCs, making it an important piece of the era’s emulation landscape.

The Cultural Impact of Early SNES Emulation

If the technical achievements of these early emulators were modest, their cultural impact was enormous. For the first time, users could load Super Mario World or The Legend of Zelda: A Link to the Past on a PC and watch them spring to life without a console in sight. Even in their broken, stuttering forms, these experiences carried a sense of magic and rebellion. Players swapped ROM images on BBSes, debated compatibility in newsgroups, and curated lists of which games worked best on which emulator.

The arrival of SNES emulation also expanded the scope of the underground community. NES emulation had proven nostalgic, but SNES games were current, still being sold on store shelves. To play them on a PC felt like trespassing into forbidden territory, as if users were glimpsing the future of gaming before it was sanctioned. Emulation became not just a technical curiosity but a cultural statement: a rejection of hardware restrictions, a challenge to corporate control, and an embrace of digital freedom.

The SNES emulators of this period were awkward, incomplete, and far from accurate, but they mattered because they showed that emulation was viable beyond 8-bit consoles. VSMC proved it could be done, ESNES and NLKSNES embodied the accuracy-versus-speed debate, NLKE represented collaboration, Snes96 and Snes97 pointed toward user accessibility, and Super Pasofami carried the banner of commercial emulation into the 16-bit era.

Together, they transformed SNES emulation from a dream into a living reality, one that would soon be perfected by ZSNES and Snes9x, and later refined by projects like bsnes/higan in the 2000s. But it all began here, in the messy, glitchy, experimental years of 1994 to 1996, when programmers dared to make the impossible happen.

Genesis / Mega Drive Emulation (1994–1996): Sonic’s First Steps on PC

While the Super Nintendo often takes the spotlight in emulation history, the mid-1990s also saw the birth of emulators for its fiercest rival, the Sega Mega Drive—known as the Genesis in North America. By 1994, Sega’s 16-bit console was still alive and thriving, with Sonic the Hedgehog 3 and Mortal Kombat II commanding attention in homes and arcades alike. For many fans, the Genesis represented speed, attitude, and a direct challenge to Nintendo’s dominance. To emulate it was to capture Sega’s swagger, and in 1994, that dream became reality in the most fragile but historic way.

Megadrive Emulator (1994)

The very first known Genesis emulator appeared in 1994, simply titled “Megadrive Emulator.” It was the work of Steve Riddle, better known by his handle ]TcG[ (The Careless Gamer), with contributions from fellow scene figures Baalzamon, Berzerker, and Stegg (History of Genesis Emulators). The program was primitive even by the standards of the time. Running on DOS PCs, it could load Sonic the Hedgehog, but the results were painfully slow, riddled with glitches, and entirely lacking in sound. Even scrolling—a central mechanic in Sega’s high-speed platformers—was barely functional. Yet, as with Haruhisa Udagawa’s early Famicom experiments four years earlier, the point was not polish but possibility. For the first time, Sega’s 16-bit mascot existed outside the console that defined him.

The story of this first Genesis emulator is as much about loss as it is about invention. According to recollections preserved in fan histories, Steve Riddle misplaced the source code, effectively halting further development before the project could mature. Without backups or collaborative repositories (which were rare in 1994), the emulator became a digital fossil—proof that Genesis emulation had begun, but also a reminder of how fragile these early efforts were. Much like the incomplete NES and SNES emulators of the early ’90s, Megadrive Emulator was a stepping stone, a fragile crack in Sega’s once-impenetrable armor.

GenEm (1996): The Sound of Sega on PC

By 1996 a meaningful leap forward occurred in the world of Genesis emulation with the arrival of GenEm, the second known emulator aimed at Sega’s 16-bit powerhouse. Unlike the fragile Megadrive Emulator of 1994, this DOS based project was the work of Markus Gietzen, a programmer whose name survives today in emulator listings and archives on SegaRetro and Zophar’s Domain (SegaRetro, Zophar’s Domain). His emulator marked a leap in both ambition and user experience, setting a new bar for what was technically possible at the time.

The most striking innovation GenEm introduced was preliminary sound emulation, a milestone that transformed the authenticity of gameplay. For the first time, players could hear the distorted but unmistakable tones of Sega’s YM2612 chip spilling out of their PC speakers. The effect was crude, the notes were off, and timing was imperfect, but it was revolutionary. To load up Sonic the Hedgehog and hear even a faint garbled echo of the Green Hill Zone theme was enough to send chills through those who had grown up with the console’s signature audio. As one user later reflected on the early days: “I remember using NES emulators like Pasofami and using GenEm Genesis emulator in DOS around 1995 or 1996, it didn’t have sound yet” (Reddit). That mix of frustration and joy captured the era perfectly, where progress was slow but every step forward felt monumental.

Of course, GenEm was far from perfect. Performance remained sluggish, compatibility with commercial titles was limited, and many games would crash or refuse to run entirely. Even so, its addition of sound emulation changed the conversation. Emulation was no longer just a question of getting sprites on the screen, it was about recreating the full sensory experience of the original hardware. By doing so, GenEm helped legitimize Genesis emulation as more than a novelty, opening the door for later, more refined projects like Genecyst in 1997.

In retrospect, GenEm’s significance lies less in how well it worked and more in what it proved. It showed that the Genesis’s complex audio hardware, once thought impossible to duplicate, could be imitated in software. That revelation helped shift expectations for what emulation could accomplish and inspired developers to push further. GenEm may not have been fast, and it may not have been reliable, but it made the Genesis sing, however imperfectly, on a machine that was never meant to play it.

Although the most famous Genesis emulator of the decade—Genecyst by Bloodlust Software—wouldn’t arrive until 1997, the years between 1994 and 1996 laid the foundation. They showed that Sega’s 16-bit powerhouse, once thought too complex and proprietary to touch, could be pulled apart, studied, and imitated in code. For the Sega faithful, these first attempts were both frustrating and thrilling. Frustrating, because the games ran poorly, often unrecognizable. Thrilling, because for the first time, the possibility of preserving and playing Genesis games without Sega hardware had become real.

What makes these early Genesis emulators fascinating is how they paralleled the SNES experiments of the same period. Just as VSMC and ESNES stumbled forward with half-functional builds, Megadrive Emulator and GenEm were clumsy, glitch-ridden, and easily forgotten by later users. Yet they mattered. They were the first attempts to translate Sega’s arcade-inspired console experience into the language of PCs, and they set the stage for the emulation boom that was about to arrive.

Virtual GameBoy (1995): Putting Nintendo’s Handheld on the PC

By the mid-1990s, handheld gaming had become one of Nintendo’s greatest strengths. The Game Boy, launched in 1989, was more than just a portable system—it was a cultural icon. Its monochrome screen, endless library of puzzle and adventure titles, and the global phenomenon of Pokémon had turned it into a fixture in backpacks, bedrooms, and playgrounds across the world. But by 1995, a quiet revolution began to unfold in the underground emulation scene. For the first time, Nintendo’s beloved handheld could be played on a computer screen.

The breakthrough came from Marat Fayzullin, a prolific Russian developer who had already made his mark with fMSX, one of the earliest and most respected MSX computer emulators. In 1995 he unveiled Virtual GameBoy (VGB), the first emulator capable of running commercial Game Boy titles on PCs. Unlike many hobbyist experiments of the time, VGB was engineered with a strong emphasis on portability. Written in ANSI C, it was designed to compile across multiple operating systems, including DOS, Windows, Unix, and Mac OS (Fayzullin’s official page). This cross-platform vision set VGB apart and foreshadowed how future emulators would be expected to reach as many users as possible.

Technically, VGB was astonishing for its time. It not only played a wide range of commercial titles but also supported features that felt futuristic in 1995. Save states, joystick support, and even Game Genie/GameShark cheat functionality were built in. It also handled Super Game Boy enhanced titles with surprising competence, and it was one of the first handheld emulators to provide sound emulation that, while imperfect, captured the Game Boy’s distinctive audio character. For those who loaded up Tetris or The Legend of Zelda: Link’s Awakening and heard the familiar bleeps and bloops, the sense of awe was overwhelming (Zophar’s Domain VGB listings).

But VGB was more than just a technical project. It also embodied the evolving ethos of the emulation community. Fayzullin distributed VGB as shareware, allowing users to try it for free but encouraging registration for full functionality. This model, similar to Nobuaki Andou’s commercial experiments with Pasofami, sparked debates about whether emulators should be free tools for preservation or monetized products. Critics argued that charging for emulators blurred the line between preservation and piracy, especially since the legality of distributing and playing ROMs was already a contested issue. Supporters countered that Fayzullin’s structured approach and constant updates helped VGB achieve a polish that many freeware projects lacked.

Community memories from the era underline how significant VGB was. On early forums and BBS boards, users marveled at the ability to play Game Boy classics without hardware. Some saw it as a godsend for preservation—an opportunity to save cartridges that were already prone to battery death. Others embraced it as a novelty, a chance to run handheld games on a big monitor with better sound. And for developers, it was another proof of concept: if the compact, custom-built hardware of the Game Boy could be translated into C and made to run on a PC, then any console might eventually be within reach.

VGB’s influence spread rapidly. It inspired a flood of rival Game Boy emulators in the late 1990s, but few matched its balance of portability, stability, and accuracy. Even today, Fayzullin’s official VGB site remains online, with source code and documentation preserved for posterity (fms.komkon.org). That longevity makes it not just a historical curiosity but a living reminder of emulation’s earliest portable success.

In hindsight, Virtual GameBoy marked the beginning of handheld emulation as a serious pursuit. It was a project rooted in technical rigor, but it also carried with it the same mix of excitement and controversy that defined the broader scene in the 1990s. It gave players the thrill of experiencing Game Boy titles in new ways, but it also forced the community to confront the messy legal and ethical questions of translating proprietary games into open software. Above all, it proved that Nintendo’s smallest console could be just as vulnerable to emulation as its larger siblings—and once the door was opened, it would never close again.

Arcade Emulation (1995–1996): Bringing the Arcade Home

By the mid-1990s, the arcade was still the crown jewel of gaming culture. Even as consoles like the SNES and Genesis battled for living-room dominance, nothing could quite replicate the booming sound, flashing lights, and fierce competition of standing shoulder to shoulder in front of an arcade cabinet. For an entire generation, games like Street Fighter II, Final Fight, and Ghosts ’n Goblins weren’t just pastimes—they were rites of passage. To imagine these games running on a home PC was almost unthinkable. Consoles were one thing, but the arcade? Those machines were mysterious, custom-engineered beasts. They weren’t meant to be copied.

And yet, by 1995, emulation had advanced to the point where the walls of the arcade were beginning to crack. It was not Nintendo or Sega that led the charge this time, but a small underground developer group who called themselves Bloodlust Software. Already known for their irreverent style and hacker ethos, Bloodlust stunned the community with a release that would redefine what was possible: Callus.

Callus (1995): Street Fighter II Comes to DOS

Released in 1995, Callus was the first emulator to convincingly run Capcom’s CPS-1 arcade hardware on consumer PCs (Emulation General Wiki). This was no half-working experiment. Callus ran Street Fighter II, Final Fight, and a handful of other CPS-1 classics with a level of quality that stunned early adopters. For the first time, arcade-perfect gaming wasn’t locked behind a cabinet and a pocketful of quarters—it was available on a DOS machine.

Bloodlust Software, helmed by programmers like Dave and Sardu (Icer Addis), brought their trademark mix of technical brilliance and underground bravado to the project. Their reputation for pushing boundaries was already known from their work on console emulators, but Callus elevated their profile into legend. Suddenly, the arcade—once the untouchable holy ground of gaming—was being replicated in code.

Technically, Callus was a revelation. It provided smooth scrolling, crisp sprite rendering, and even offered joystick support, giving players a way to mimic the feel of arcade controls at home. Its biggest limitation was sound—while functional, it wasn’t always accurate, and certain audio channels dropped in and out—but this did little to diminish the thrill of throwing Hadoukens on a desktop PC.

For fans, Callus was a revolution. On early BBS boards and emulation forums, users traded stories of booting up Street Fighter II for the first time outside of an arcade cabinet. “It’s like magic,” one fan wrote on a 1996 discussion thread preserved by enthusiasts. Others worried openly about what this meant for arcades themselves: if the same games could run for free at home, what was left to lure players into smoky, neon-lit halls?

The release of Callus also intensified debates about emulation’s legality and ethics. With console emulation, companies like Nintendo and Sega had already shown their willingness to defend IP aggressively. But arcade boards felt different—Capcom was still actively selling and profiting from their cabinets in 1995. For many, Callus wasn’t just a neat trick, it was a shot across the bow of the arcade industry.

Despite these controversies, Callus’s impact is undeniable. It demonstrated that arcade hardware—once considered far too exotic for duplication—could be understood, mapped, and re-created on consumer hardware. It opened the door for other arcade emulators, eventually leading to the creation of MAME (Multiple Arcade Machine Emulator) in 1997, which would formalize and expand the preservation effort into a global phenomenon.

Bloodlust Software was never just another underground coding outfit—they were a paradox of irreverence and brilliance, a team that blended crude humor with technical mastery. Icer “Sardu” Addis, later to become infamous as the creator of NESticle, was already reshaping the scene with tools that would soon define emulation culture, from save states to sprite editing and gameplay recording (Overclocked.org, Wikipedia, Hacker News). His partner Dave brought design flair and a knack for shareware distribution, the same partnership that produced cult oddities like Executioners and Timeslaughter, notorious for their violent satire and DIY shock value (VICE).

Their synergy reached a peak in Callus (1995), a near-perfect emulator of Capcom’s CPS-1 arcade hardware that astonished players by running Street Fighter II and Final Fight with unprecedented accuracy. Community lore, later confirmed in an EmuFAQ, even claimed that Sardu had experimented with network play for Street Fighter II, long before “netplay” was a standard feature of emulators (Overclocked.org).

Yet with this brilliance came turbulence. Behind the scenes, Bloodlust’s rise was shaken by breaches of trust and conflicts within the emulation underground. Most infamously, Sardu’s own NESticle project would later be derailed after a damaging intrusion into his development machine—a story tied to the notorious group Damaged Cybernetics that we will cover in detail in the next entry. For now, Callus stands as both a technical marvel and a cultural milestone, proof that even the mighty arcade could be conquered in code, and a reminder that innovation in the emulation scene often came with chaos attached.

In retrospect, Callus was more than just an emulator. It was a cultural moment. It brought the arcade home at a time when arcades were still thriving, shifting the balance of power and igniting conversations that still resonate today. More importantly, it proved that the ethos of emulation—preserve, replicate, and share—was not confined to consoles. The neon glow of the arcade could live forever in code.

MESS (1996): The Birth of Universal Emulation

If Callus represented the thrill of bringing a single arcade board into the home, MESS, released in 1996, embodied something far grander. Where most emulators of the early to mid-1990s focused narrowly on one machine—an NES, a Genesis, a Game Boy—MESS dared to ask a bigger question. What if a single program could emulate everything? Not just consoles, but home computers, calculators, and electronic oddities that might otherwise be lost to time.

MESS stood for Multiple Emulator Super System, and the name captured its ambition perfectly. Built on the same architecture as MAME (Multiple Arcade Machine Emulator), which would debut in 1997, MESS shared much of its DNA with the preservationist ethos of that project. But while MAME’s focus was arcade cabinets, MESS cast its net across the entire landscape of digital machines. By attempting to emulate dozens of platforms, MESS became the first true “universal emulator” (segaretro.org, Emulation General Wiki).

The project originated with Marat Fayzullin’s open-source philosophy and was further carried by Nicola Salmoria and other developers already working on MAME. Early builds of MESS were rough, often incapable of running commercial games smoothly, but that wasn’t really the point. Its significance lay in its scope. Emulating a Sega Genesis was impressive. Emulating a Game Boy was remarkable. Emulating both under the same framework, alongside computers like the Commodore 64, the Apple II, and the ZX Spectrum, was a conceptual leap that signaled a new era.

For users in 1996, downloading MESS was an adventure in itself. The list of supported systems was astonishing—dozens of consoles and home computers, some obscure even then. Most worked only partially. Graphics were garbled, controls inconsistent, audio often missing. Yet the program’s sprawling compatibility list sent a message: no machine was too strange, too old, or too complex to attempt preservation. As one Usenet poster in 1997 remarked, “MESS isn’t about playing games, it’s about making sure we can.”

The preservation angle mattered enormously. Whereas projects like NESticle or Callus often carried a “for fun” reputation, MESS deliberately aligned itself with archiving and digital history. It attempted to document the quirks of each machine, not just hack together code to run Mario or Sonic. This created tension in the community. Some users mocked it as impractical—a jack of all trades, master of none. Others defended it passionately, pointing out that by keeping the framework modular, MESS could evolve as PC hardware improved.

Technically, MESS also demonstrated the power of modular emulation design. Its developers created a plug-in style architecture, where new systems could be slotted into the existing framework without rewriting the entire program. This philosophy not only ensured that MESS itself could expand endlessly, but it also influenced later emulator projects across the board. By designing for extensibility, MESS was thinking like a preservation library, not a toy.

MESS’s path wasn’t free from tension. The slave to accuracy, it often fell short on performance and compatibility, showing the trade-offs preservation entailed. Systems like the Macintosh couldn’t run, not for lack of effort, but because true emulation required precise recreation of every IC and controller—leading to notoriously clunky performance. Some critics called it impractical, while others celebrated its breadth and ambition as unmatched.(E-Maculation)

Despite its rocky early years, MESS gradually grew into a cornerstone of emulation history. When it eventually merged back into MAME in the 2010s, the vision came full circle: one unified emulator covering arcades, consoles, and computers under a single project (MAMEDev.org). That integration traces directly back to the mid-1990s, when a small team of coders decided to build something absurdly ambitious at a time when most people were happy just to play Super Mario World on their PCs.

In retrospect, MESS in 1996 was less about functionality and more about philosophy. It announced that emulation wasn’t only for gamers—it was for historians, archivists, and anyone who cared about digital heritage. It gave the underground scene a new sense of mission, showing that preservation was possible, even necessary, and that nothing in the digital world should be dismissed as impossible to emulate.

Miscellaneous and Multi-System Emulation (1995–1996): Beyond Consoles

While much of the emulation buzz in the mid-1990s centered on the familiar battlegrounds of Nintendo and Sega, the underground was far more diverse than casual observers might think. From personal computers to handheld ports on obscure hardware, these years also produced experiments that broadened the scope of what emulation could mean. They might not have carried the glamour of NESticle or the excitement of Callus, but they quietly reshaped the conversation, proving that emulation wasn’t only about replaying childhood favorites. It was about exploring and preserving all digital machines.

UAE (1995): The Universal Amiga Emulator

In 1995, a German developer named Bernd Schmidt released the first version of the Universal Amiga Emulator (UAE), targeting the Commodore Amiga 500. The Amiga was legendary for its graphics and sound in the late 1980s and early 1990s, but its custom chipset made it notoriously difficult to emulate. Early UAE builds were virtually unusable—sluggish, buggy, and demanding far more power than PCs of the era could provide (Wikipedia).

And yet, the sheer fact that an Amiga could boot at all on a standard PC was astonishing. For Amiga loyalists, UAE offered hope that their beloved machine could survive in software long after Commodore’s bankruptcy in 1994. Over time, UAE would become one of the most advanced and celebrated emulators in history, but even in 1995 its mere existence signaled that no system was beyond reach.

Wzonka-Lad (1996): Game Boy on the Amiga

If UAE symbolized the herculean effort of recreating the Amiga on other platforms, Wzonka-Lad represented the reverse: bringing Nintendo’s Game Boy to the Amiga. Written by Ville Helin and first released in 1996, Wzonka-Lad was coded in highly optimized 68020 assembly, squeezing performance out of the Amiga’s hardware in ways few thought possible (Wikipedia).

Wzonka-Lad supported most commercial Game Boy titles and even offered audio emulation, though full sound required registering the software. This decision sparked small debates in the Amiga community—should emulators be monetized, or should they remain open experiments? Despite the minor controversy, Wzonka-Lad became a cult favorite among Amiga owners, many of whom were thrilled to see Nintendo’s handheld classics running on their aging but beloved machines.

Other Experiments (1995–1996)

Alongside UAE and Wzonka-Lad, a handful of lesser-known but important projects appeared.

• PC-Engine / TurboGrafx-16 emulators began circulating in embryonic forms, though none gained wide traction until later in the decade.

• MSX emulation continued to expand through Marat Fayzullin’s fMSX, building on his earlier work and cementing his reputation as one of the scene’s pioneers.

• Developers experimented with multi-system frameworks, anticipating the grand ambitions of MESS, which would officially arrive in 1996.

These projects may not have dominated headlines, but they hinted at the boundless scope of the emulation idea. If an Amiga could run Nintendo games, if a PC could pretend to be an Amiga, if multi-system packages could even be imagined—then emulation was not about one console or one generation. It was about the universal logic of machines.

Conclusion of Part 2

By the end of 1996, the emulation world had transformed from a handful of glitch-ridden experiments into a sprawling underground culture. Nintendo’s NES and SNES had fallen, Sega’s Genesis had been cracked, the Game Boy had found its digital double, arcades were being pulled out of smoky game halls and onto desktop PCs, and even complex computers like the Amiga were taking their first steps into software form.

What united all of these projects was not perfection—they were slow, buggy, incomplete—but possibility. For every failed boot, every missing sound channel, every broken sprite, there was also a flash of revelation: this works. And once it worked, even a little, there was no going back.

The dinosaurs had woken. By 1996, the spark had caught fire, and emulation was no longer the dream of a few daring coders. It was an unstoppable movement.

Previous: Part 1: 1990 to 1993 —The Dinosaurs Wake
Next: Part 3: 1997 to 1999 — The Wild West