What If Commodore allowed Amiga clones? The alternate history of a lost computer empire

If Commodore had let other companies build licensed Amiga clones early, the 1990s might have had a third great computer platform. It might also have forced Commodore to confront the hardest question in hardware: would it rather own a smaller kingdom, or tax a larger one? Imagine a computer shop in Europe in 1991. It could be in Düsseldorf, Manchester, Antwerp, Milan, or some side street near a university where the window is crowded with joysticks, disk boxes, printer ribbons, game posters, and handwritten price cards. Inside, the monitors glow under fluorescent lights. A salesman behind the counter knows which hard drives are reliable, which games need extra memory, and which machines parents should not buy unless they are prepared to lose their children for entire weekends. On one side sit the IBM-compatible PCs. Serious machines. Sensible machines. Beige machines. They are getting faster, cheaper, and harder to ignore. Nearby sits the Macintosh, more expensive and more elegant, still carrying the air of publishing houses, design studios, and people who already know what font they want. And then there is the Amiga.

The shop window that never was

In our timeline, that usually meant a Commodore Amiga. An A500 for the bedroom. An A2000 or A3000 for the ambitious. Later, an A1200 or A4000 for those still willing to believe the next leap was just around the corner. But in another multiverse version of the 1990s, the Amiga corner is not one shelf. It is a wall. There is a Philips Amiga for the living room, with a CD-ROM drive. There is a German-built Amiga tower for video studios, heavy and expandable. There is a cheaper Taiwanese Amiga-compatible machine for students. There is a stylish Japanese model aimed at musicians and visual artists. There is a British education bundle with art software, a printer, two joysticks, and a monitor. There is a professional video system sold as a complete production package. Each machine is different, but each carries the same promise: Amiga Compatible.

It never happened that way. Commodore held the Amiga close, and when Commodore collapsed in 1994, the platform was dragged into a long afterlife of new owners, hobbyist revivals, accelerators, emulators, boutique systems, and arguments over legitimacy. Retrospective view on Commodore’s bankruptcy is that the Amiga CD32 arrived too late to save the company, which went bankrupt on April 29, 1994; Escom later acquired the Amiga rights, only to go bankrupt itself in 1996.  The counterfactual still hurts because the Amiga was not short of love. It was short of scale, capital, discipline, and time.

The machine people did not merely use

The Amiga was not loved like a useful appliance. It was loved like a secret. For a teenager in the late 1980s, an Amiga could feel like a future smuggled into the bedroom. It played sampled sound while other computers beeped. It moved color around the screen with a confidence that made office PCs look embarrassed. It could multitask. It could paint, animate, scroll, sing, and play. It could make music in bedrooms, graphics in local television stations, and impossible visual tricks at demo parties where programmers competed to make the machine do things its designers had not expected. That emotional fact matters because platforms are not built from hardware alone. They are built from belief. The Amiga had belief. It had magazines, user groups, bedroom coders, musicians, video people, game studios, and children who could not explain why the machine felt alive but knew that it did.

Commodore owned all of that, yet too often treated it as a product line rather than an economy. That was understandable. Commodore came from a hard world of manufacturing costs, dealer channels, inventory cycles, and price wars. It knew how to sell boxes. But the Amiga needed more than boxes. It needed a wider industrial base. It needed partners. It needed developers to believe the market would grow. It needed a future that was not tied to the cash position of one troubled company. It needed to become bigger than Commodore.

The business problem hiding inside the dream

The phrase “Amiga clones” sounds simple. Let other companies build Amigas. Grow the market. Sell more software. Make the platform too large to ignore. But cloning is not charity. It is a bargain over who gets the profit. If Commodore sold a complete Amiga, it captured the revenue from the whole machine: the motherboard, chips, ROMs, case, keyboard, power supply, packaging, distribution, and brand. If another company sold the machine, Commodore might receive only a royalty unless it also supplied key parts. That difference is not cosmetic. It is the difference between being a manufacturer and being a landlord.

The harsh lesson came later from Apple. Apple licensed Macintosh clones in the mid-1990s, but the economics turned sour. In 1997, Apple moved to buy Power Computing’s Macintosh-based business for $100 million in stock; because Power Computing’s sales had reached nearly $400 million and about 12 percent of the Macintosh market, while Steve Jobs complained that Apple collected only about $50 per clone and still carried the cost of Mac software development, marketing, and developer relationships. Later on, Apple was effectively stopping the licensing of new technology to other clone makers.  That was the trap Commodore would have faced. A clone can be good for the platform and bad for the company that owns it. If a licensed Amiga clone creates a customer who would never have bought a Commodore machine, Commodore wins. If it simply replaces a Commodore sale with a cheaper clone, Commodore may lose far more in hardware margin than it gains in royalties. That is why the economics cannot be an afterthought. They decide whether the clone program is a rescue plan or a slow leak.

Commodore’s numbers tell a brutal story

By fiscal 1993, Commodore was not simply having a bad year. It was cornered. The company’s net sales fell 35 percent to $591 million, down from $911 million in fiscal 1992 and $1.047 billion in fiscal 1991. Amiga unit sales slipped to just over 800,000 from 1 million the previous year, and Amiga revenue fell 40 percent. The Amiga line still accounted for 59 percent of Commodore’s net sales, which meant the company was painfully dependent on a platform whose sales were already weakening.

The geographic picture was just as revealing. Europe accounted for 84 percent of Commodore’s net sales in fiscal 1993, which shows how heavily the company depended on European demand, currencies, retailers, and consumer confidence. A clone strategy might have helped there, because regional manufacturers could have served strong local markets better than Commodore alone. But it also meant that any European downturn or price war hit Commodore especially hard. The margin picture was worse. Commodore’s gross margin was a $132 million loss in fiscal 1993, after positive gross margins of 27 percent in fiscal 1992 and 32 percent in fiscal 1991. The company blamed falling sales, inventory write-downs, asset write-downs, and severe pricing pressure. Research and development fell to $19 million, down from $26 million in 1992 and $31 million in 1991. Selling and marketing expenses were also cut sharply.

This is the financial backdrop to the clone question. Commodore did not merely need more Amiga fans. It needed a business model that could produce cash, protect margins, reduce inventory risk, and fund future chip and operating-system development. A clone program could have helped with all of that. It also could have made all of it worse.

The arithmetic of a clone

Suppose Commodore sells a complete Amiga for a few hundred dollars. In a healthy year, it earns a gross margin on that sale. Not pure profit, of course. Out of that margin must come engineering, marketing, administration, warranty, channel support, and the next generation of products. But at least the full sale belongs to Commodore. Now suppose a clone maker sells an Amiga-compatible machine and pays Commodore a modest royalty. If that machine reaches a new buyer, the royalty is found money. If it replaces a Commodore-built Amiga, the royalty is not found money. It is a smaller substitute for a larger lost sale. That is why pure operating-system licensing probably would not have been enough. Commodore could not afford to become a company collecting a small fee while others captured the hardware business. The Apple example makes this painfully clear: Jobs argued that $50 per Mac clone did not compensate Apple for the costs it alone carried. Commodore, with weaker finances and a more price-sensitive market, would have been in an even more dangerous position.

The stronger model would have been layered. Every clone would need to generate several streams of revenue for Commodore: Kickstart ROMs, AmigaOS licensing, custom chipset sales, certification fees, branding fees, developer tools, reference board designs, and perhaps professional support contracts. Commodore would have to make money from the heart of the machine, not merely from the sticker on the case. In plain terms, Commodore should not have asked, “How much can we charge for the Amiga name?” It should have asked, “How much value can we capture from every Amiga-compatible machine, even when we do not build it ourselves?” That is a very different company.

The factory question

Every romantic counterfactual eventually reaches the loading dock. The Amiga was not a generic PC. Its identity depended on custom chips, ROMs, timing behavior, video modes, audio behavior, memory architecture, and expansion standards. A clone maker could not simply assemble off-the-shelf parts and call the result an Amiga. Commodore would either have to supply the essential chips or approve second-source manufacturers capable of producing them correctly. This could have been powerful. If Commodore supplied chips to licensed manufacturers, every clone became a component sale. Higher chip volumes could spread development costs over more machines. Outside manufacturers could take on cases, keyboards, storage, monitors, and regional distribution. Commodore could reduce its exposure to finished-goods inventory, one of the most dangerous parts of the computer business when prices are falling.

But chip supply would also become a point of fear. Clone makers would ask hard questions. Can Commodore deliver enough chips on time? Will the pricing be stable? Is there a roadmap? Will Commodore favor its own machines when supply is tight? Will it obsolete our product with no warning? Can we trust a company whose own finances are deteriorating? Those questions would not be hostile. They would be normal. A clone maker investing in tooling, marketing, distribution, and support would need confidence that Commodore could behave like a dependable platform supplier. That was not guaranteed.

Inventory, the silent killer

One of the strongest economic arguments for clones is inventory risk. When Commodore built a complete computer, it had to forecast demand, buy parts, manufacture machines, ship them into distribution, and hope the market still wanted them at the planned price. If demand softened or PC prices fell, finished goods could become a burden. Commodore’s 1993 financial discussion describes inventory and asset write-downs, pricing allowances, and competitive pricing pressure as major reasons gross margin collapsed into a loss. A clone ecosystem could have shifted some of that risk outward. If a German tower maker misjudged the professional market, that company would carry much of the finished-goods pain. If a British education bundle underperformed, the distributor would absorb part of the damage. Commodore would still have earned money from chips, ROMs, licenses, and certification.

This is why cloning might have appealed to a healthier Commodore. It could have transformed the company from the sole gambler on Amiga hardware demand into the toll collector for a wider market. But there is a cruel mirror image. If clone makers overproduced and dumped machines cheaply, they could drag down the price of all Amigas. Commodore’s own inventory would suddenly look overpriced. Dealers would demand rebates. Users would wait for discounts. The platform would grow, but the price umbrella would collapse. The economics of cloning are never just about units. They are about who carries the risk when the units do not sell.

Why the IBM PC is the wrong model, and still an important one

The IBM PC shows the power of an open standard, but also the danger to the company that starts it. IBM’s own history says the 5150 team used off-the-shelf parts to meet a one-year deadline and a $1,500 target, published technical reference materials, and embraced open architecture; within a year, more than 750 software packages were available, and companies were reverse-engineering the system boot code to create IBM-compatible machines.  That openness helped create the PC revolution, but IBM did not remain the great winner of the clone economy. IBM’s account notes that its PC market share fell from roughly 80 percent in 1982–83 to 20 percent a decade later, and that the PC eventually became a commodity business IBM exited by selling its PC division to Lenovo.

For Commodore, the lesson was not “open everything.” That would have been suicidal. The lesson was that standards can become bigger than their creators. The company that wins is not always the company that builds the first machine. It is the company whose business model fits the standard once it spreads. Microsoft and Intel fit the PC clone world. IBM did not, at least not in personal computers. If Commodore wanted Amiga clones, it needed to decide where it would sit in the value chain. Would it become the Microsoft of AmigaOS? The Intel of Amiga chipsets? The Nintendo-like certifier of compatible machines? The Apple-like maker of premium reference systems? The answer had to be some combination of all four. That is what made the problem so hard.

The home clone, and the danger of success

The most obvious clone was the home Amiga compatible: a machine that behaved like an Amiga 500, and later perhaps an A1200-class system. It would run the same games, the same paint programs, the same music tools, the same demos, the same educational disks. It would use official ROMs and approved chips. It would pass a strict compatibility test before earning the badge. For users, this would have been wonderful. More choice. Lower prices. Different keyboard designs. Better bundles. Stronger local distribution. A British package could emphasize games and education. A German package could include a better monitor and expansion options. A Taiwanese model could drive the entry price down.

For developers, the attraction would be a bigger installed base without a moving target. More machines that behave like the same machine is exactly what game studios want. For Commodore, this was the most dangerous model because it would attack the heart of its own business. A cheaper Amiga-compatible home machine might not expand the market enough to compensate for lost Commodore sales. The company could find itself funding the platform while partners captured the growth. This kind of clone would need tight controls. Commodore might allow it only in markets where it lacked distribution, or only after an official model had reached maturity, or only with minimum component purchases that protected Commodore’s economics. The clone must not simply be a cheaper Commodore Amiga with someone else’s badge.

The professional tower, where the money looked better

The professional clone was the more promising option. Instead of competing directly with the A500, licensed partners could build machines Commodore was too slow or too constrained to deliver: expandable towers with faster processors, more RAM, SCSI storage, video slots, genlock support, better cooling, graphics expansion, and serious support. These would serve video editors, local television stations, musicians, animators, schools, art departments, and small studios. This mattered because professional systems could carry higher prices and better margins. They could also expand the platform upward rather than cannibalize the home line. A family buying a low-cost A500 was not necessarily the same customer as a production house buying a tower full of storage and video hardware.

A professional clone program could have made the Amiga look more serious without forcing Commodore to design every niche machine itself. It could have let German engineering firms, video specialists, and regional integrators build systems for markets they understood. Commodore would still earn money from the operating system, ROMs, chips, and certification, while partners handled support-heavy configurations. This is probably where Commodore should have started. Not with cheap consumer clones, but with high-value machines that made the Amiga more credible and less dependent on home-computer margins.

The CD-ROM Amiga, and the cost of being early but weak

The living-room Amiga is the most seductive branch of the alternate timeline. The Amiga already understood color, sound, movement, and games. It did not need to bolt on personality through drivers, sound cards, and CD-ROM upgrade kits. It had personality built in. Commodore’s CDTV and CD32 showed that the company understood, at least in outline, that the Amiga could move toward the television. A clone-friendly Commodore could have approached this differently. Instead of making one CD machine and hoping it saved the company, it could have defined an Amiga CD standard and licensed consumer-electronics partners to build around it. A Philips-style Amiga CD system might have targeted families. A Sony-like model might have emphasized design, audio, and video. A keyboard-ready model could have bridged console and computer. A school model could have bundled reference titles, art tools, and authoring software. Economically, this had appeal because consumer-electronics companies knew how to build for the living room, manage retail channels, and design products ordinary families could understand. Commodore could supply the technology and collect from the platform.

But it also carried danger. Console economics are not computer economics. Game publishers expect volume. Retailers want clarity. Families hate compatibility confusion. A CD Amiga that was too computer-like might fail as a console. A CD Amiga that was too console-like might split the Amiga software base. A machine that was not compatible enough would anger Amiga owners. A machine that was too compatible might be held back by old hardware assumptions. The CD-ROM Amiga was not a bad idea. It was an idea that needed money, partners, timing, and ruthless clarity.

The “Amiga Powered” escape valve

Not every Amiga-derived machine needed to be fully compatible. Some devices could have used AmigaOS, Amiga chips, or Amiga development tools in kiosks, editing appliances, educational terminals, music workstations, broadcast graphics systems, or set-top boxes. These machines might have been commercially useful without running a teenager’s game collection. But they should not have been called Amiga Compatible. This is where branding becomes economics. A confused buyer is a future lost sale. A developer who cannot trust the label spends less. A dealer who has to explain too many exceptions pushes another machine instead.

A separate badge, something like Amiga Powered, would have let Commodore monetize the technology without damaging the promise of compatibility. It would tell buyers: this machine uses Amiga technology, but it is not necessarily a full Amiga home computer. That sounds like marketing language. In truth, it is a way of protecting the value of the brand.

The dealer at the counter

The dealer matters. In the early 1990s, computer dealers did more than process transactions. They demonstrated software, installed memory, recommended monitors, repaired machines, handled returns, and told nervous parents what to buy. If Commodore introduced clones badly, dealers could turn from allies into enemies. Imagine a dealer sitting on official Amiga 2000 inventory when a licensed clone maker announces a cheaper, faster tower. Does the dealer celebrate a new product or curse the old stock in the back room? Does Commodore offer rebates? Does the clone maker demand equal shelf space? Does the customer wait for prices to fall?

Channel conflict is not glamorous, but it kills margins. Commodore would need to make dealers believe that clones meant a bigger Amiga business, not just cheaper machines and unsellable inventory. That might mean protected launch windows for official Commodore machines, different categories for professional clones, strict pricing rules, or dealer incentives tied to certified systems. The platform would not only be engineered. It would be negotiated.

The software market and the price of uncertainty

The economic case for clones ultimately depends on software. A larger installed base attracts developers. More developers attract users. More users attract retailers. Retailers attract more hardware makers. That is the virtuous circle every platform dreams about. But developers do not respond to unit counts alone. They respond to confidence. They need to know what hardware to target, what operating-system version matters, what graphics modes are safe, how much memory users have, and whether the market will still exist by the time the product ships. If Amiga clones created a larger but fragmented market, the benefit would shrink. A game studio might decide to target only the lowest common denominator. A video-software developer might support only expensive professional machines. An American publisher might look at the confusion and stay with DOS or Windows.

That is why a clone strategy needed strict layers. The base Amiga had to remain predictable. Professional systems could extend the platform, but not redefine the home target. Amiga Powered devices had to be honest about what they were. The operating-system roadmap had to be public enough for developers to plan. A platform grows when people believe the ground will stay still long enough to build on it.

The good future

In the best version of history, Commodore opens the Amiga in 1988 from a position of strength. It keeps building reference machines but licenses selected partners. It begins with professional towers and regional systems, not bargain-basement home clones. It sells official chipsets, ROMs, AmigaOS licenses, and certification. It creates three clear categories: Amiga Compatible for machines that run the consumer software base, Amiga Professional for expandable workstations, and Amiga Powered for devices using the technology without full compatibility. The economics are carefully designed. Commodore earns something meaningful from every machine. Clone makers get enough margin to care. Dealers get protected categories. Developers get stable targets. Users get more choice without losing trust. Revenue from licensing and chip sales funds new graphics hardware, better development tools, and operating-system work.

By 1991, the Amiga world is broader. There are still Commodore Amigas, but there are also German towers, regional school systems, video workstations, and better CD-ROM experiments. The PC still gains the office. The Macintosh still holds prestige in publishing and design. But the Amiga owns affordable creativity: video, animation, games, music, school media labs, small studios, and the strange cultural space between console and workstation. This is the dream. Not world domination, but durable relevance.

The bad future

In the troubled version, Commodore opens the platform too late or prices the licenses badly. The first clones are exciting. Magazines love them. Users debate specifications. Dealers cautiously order stock. But the economics begin to rot. Clone makers undercut Commodore’s own machines. Commodore collects fees but loses hardware margin. It still pays for operating-system development, developer support, marketing, and future chip design. Partners want lower fees. Dealers want protection. Users wait for discounts. The platform appears larger, but Commodore’s ability to fund the future weakens.

This is the danger Apple ran into later. Power Computing’s clone sales grew, but Apple concluded the success came too much at Apple’s expense, with licensing revenue too small to cover the platform costs Apple alone carried. Commodore could have met the same problem with less cash and fewer options. The bad future is painful because it contains good machines. The clone makers are not villains. They are doing what the contract allows. The problem is that the contract does not make the platform owner whole.

The ugly future

The ugly future begins in desperation. Commodore waits until 1993, when the company is already struggling with falling sales, weak liquidity, pricing pressure, and reduced R&D. Its own filing language raised substantial doubt about continuing as a going concern, with a $356 million fiscal 1993 loss, a $107 million working-capital deficit, defaults on credit agreements, and only about $10 million in cash at the end of June 1993.  A company in that condition is not well placed to build a disciplined clone ecosystem. It is tempted to sign deals for cash. Certification becomes negotiable. Chip supply becomes uncertain. Partners improvise. Machines ship with asterisks.

One clone runs Workbench but breaks games. Another is excellent for video but not really compatible. Another has advanced graphics modes no one supports. Another is cheap and unreliable. The shopkeeper no longer says, “Yes, this is an Amiga.” He says, “It depends what you want to run.” That sentence kills platforms. Scarcity frustrates users. Confusion frightens developers. The ugly clone program does not save the Amiga. It turns the word Amiga into an argument.

What Commodore would have had to become

The deeper question is not whether cloning was a good idea. It is whether Commodore could have become the kind of company cloning required. A successful Amiga clone strategy demanded patience, partner management, documentation, supply discipline, clear pricing, hard certification, and long-term roadmaps. Commodore would have needed to say no to weak manufacturers, no to half-compatible systems, no to short-term licensing cash that damaged the standard, and no to internal pressure to protect every Commodore-branded machine from competition. It would have had to accept that the most valuable Amiga sale might not always be a Commodore-built Amiga sale. Sometimes the better business would be to sell the chips, ROMs, operating system, and certification into someone else’s machine and use that machine to expand the market.

That is easy to say in hindsight. Inside the company, it would have been brutal. Factories need volume. Dealers need stock. Engineers need budgets. Executives need cash. The finished hardware business produces immediate revenue, while platform strategy asks people to believe in a larger future. Commodore needed to become less possessive and more exacting at the same time. That is a rare combination.

The future that might have survived Commodore

The cruelest possibility is that Amiga clones might have saved the Amiga without saving Commodore. That would not be a small thing. A platform can outlive its parent if enough manufacturers, developers, users, and institutions depend on it. Had Commodore built a clone ecosystem early, its own hardware business might still have struggled under the pressure of commodity PCs. It might still have restructured. It might even have failed. But the Amiga would not have been left so dependent on one collapsing company. There might have been licensed manufacturers ready to continue. There might have been professional markets large enough to justify further development. There might have been a cleaner legal and technical path from classic Amiga hardware to next-generation systems. There might have been enough shared economic interest to keep AmigaOS and Amiga-compatible hardware moving forward with less chaos.

Instead, the real Amiga after Commodore became brilliant but fragmented. It survived through devotion rather than structure. Enthusiasts kept machines alive, built expansions, wrote software, argued over successors, and refused to let the story end. That devotion is beautiful. It is also evidence of what was lost. People kept trying to rebuild the ecosystem Commodore never fully allowed to form.

The verdict

Should Commodore have allowed Amiga clones? Yes, but only under conditions that may have been beyond Commodore’s character. It should not have thrown the gates open to anyone willing to pay. It should not have treated the Amiga name as quick cash. It should not have allowed half-compatible systems to pollute the market. It should not have relied on royalties alone. And it should not have waited until collapse made every deal look tempting. The right strategy would have been early, selective, and economically layered. Professional towers first. Regional partners where Commodore was weak. Official chipsets and ROMs. A serious compatibility lab. Clear badges. A protected meaning for Amiga Compatible. A separate category for Amiga Powered devices. A roadmap that gave developers and manufacturers confidence. Clone revenue reinvested into new chips, better tools, and AmigaOS development.

That strategy might not have made Commodore rich. It might not have stopped the PC from dominating the office. It might not even have saved Commodore in the end. But it could have given the Amiga more reach, more time, more software, more allies, and a better chance of surviving the company that owned it.

The machine Commodore held too tightly

The Amiga was a machine of abundance trapped inside a company running out of room. It had more color than the office PC, more mischief than the Macintosh, and more creative intimacy than the workstation. It invited people to make things before they had permission: games, songs, animations, title sequences, school projects, video experiments, strange demos assembled after midnight by people who knew the hardware almost as if it were alive. But a platform is more than a beloved machine. It is a set of promises. A promise to buyers that software will run. A promise to developers that the market will be there. A promise to dealers that the product line will make sense. A promise to partners that the rules will be stable. A promise to the future that today’s machine is not a dead end. Commodore never made those promises strongly enough.

Had it opened the Amiga early and wisely, the 1990s might have had a third road in personal computing. Not the gray corporate road of the PC, and not the polished garden path of the Macintosh, but something stranger, louder, more playful, more European, more televisual, more musical. A road lined with machines built by different companies but animated by the same unruly spirit. Perhaps Commodore was too protective. Perhaps it was too chaotic. Perhaps the economics were always crueler than the dream. But the irony remains hard to escape. To save the Amiga, Commodore may have had to stop holding it so tightly.