Why ASIC technology could start the next Amiga hardware revolution

For decades, the Amiga community has kept its machines alive through patience, repair and ingenuity. Capacitors have been replaced, battery damage has been cleaned, cracked traces have been patched, and dead motherboards have been brought back from the edge by people who understand that these computers are not disposable electronics. They are creative instruments, cultural artifacts and, for many users, the most characterful personal computers ever made. Yet preservation by repair has a limit. Every year, original machines become older, donor boards become rarer, and the supply of working custom chips becomes smaller. The Amiga community has done remarkable work keeping the platform alive, but it is becoming harder to ignore the central weakness of the current model. The scene can remake cases, keyboards, power supplies and motherboards, but the custom silicon remains finite. That is why the debate around ASIC-based retro hardware matters. The arrival of modern systems such as the Neo Geo AES+ has shown that retro hardware does not have to choose only between original machines, software emulation and FPGA recreation. There is another path: newly manufactured, purpose-built silicon designed to reproduce classic hardware as a stable product rather than a flexible approximation. For the Amiga, that path is not merely attractive. It may be necessary.

The preservation argument now points beyond donor hardware

Replacement PCB projects have become essential to the modern Amiga scene. They rescue machines damaged by battery leakage, corrosion, cracked traces, broken sockets and decades of electrical stress. They allow original chips to be transferred into fresh boards, keeping machines operational long after their original motherboards have failed. This work is valuable and should continue. But it also reveals the problem it cannot solve. A replacement board still depends on original custom chips, and those chips usually come from donor machines. One Amiga is saved by sacrificing another. The total number of viable machines does not grow. It merely gets rearranged. That is not a future. It is a controlled decline.

The strongest case for ASIC is that it changes this arithmetic. New silicon would allow the community to stop treating original Amigas as parts reservoirs. Instead of moving the last surviving custom chips from one old board to another, the scene could begin producing replacements at scale. That would turn preservation from a salvage operation into a manufacturing strategy. A motherboard can be reproduced. A case can be recreated. A keyboard can be rebuilt. The real preservation frontier is the custom chipset, and ASIC is the only route that directly addresses that frontier in a permanent way.

FPGA has been essential, but it should not be mistaken for the destination

Any serious pro-ASIC argument must give FPGA the credit it deserves. FPGA systems have transformed retro computing. They have allowed old machines to be studied, recreated, tested and improved in public. They provide flexibility, and that flexibility has made them indispensable for development and preservation. But flexibility is not always the final virtue. At some point, a platform that wants to survive as hardware needs stability, availability and repeatability. That is where ASIC becomes more compelling.

FPGA is excellent for research. It is excellent for testing. It is excellent for learning the strange details of old hardware. But in a mature product, the same flexibility can become unnecessary cost and complexity. An ASIC, once properly validated, is cleaner. It is built for one job. It does not carry the overhead of a general-purpose reconfigurable device. It can be manufactured as a dedicated component, tested as a dedicated component and sold as a dedicated component. That matters for the Amiga because the goal should not be to keep building clever substitutes forever. The goal should be to restore the possibility of real, reproducible Amiga-compatible hardware that does not depend on 35-year-old donor chips or expensive programmable logic. FPGA should be the proving ground. ASIC should be the preservation-grade result.

Why ASIC is better suited to a normal Amiga hardware market

If new Amiga motherboards are ever to become normal rather than boutique curiosities, the platform needs parts that can be produced consistently and priced sensibly. ASIC is the more natural fit for that future. An FPGA-based solution often remains tied to the cost of the FPGA itself, the surrounding board, level shifting, configuration memory, firmware support and small-batch assembly. That can be perfectly acceptable for enthusiasts, developers and specialist hardware. But it is not always the best route to affordable, widely available replacement parts.

ASIC changes the cost structure. The upfront development cost is higher, but the per-unit cost can fall dramatically once production volume rises. That is precisely why ASIC is used in consumer electronics. It is not chosen because it is easy. It is chosen because, when the design is stable and the market exists, it is the right way to manufacture dedicated hardware. For the Amiga, this distinction is crucial. The community does not need endlessly reconfigurable chips inside every finished machine. It needs reliable replacements for known hardware. Paula should behave like Paula. Denise should behave like Denise. Agnus and Alice should respect the timing and memory behaviour expected by old software and expansions. Once those behaviours are sufficiently understood, there is a strong argument for fixing them in silicon. That is not a step away from authenticity. It is a step toward a more durable form of it.

The economics of ASIC are difficult, but they point in the right direction

The main objection to ASIC is cost, and that objection is real. ASIC development requires non-recurring engineering investment before a product can be sold. Design, verification, layout, prototyping, packaging, testing and possible revision spins all have to be paid for upfront. A failed revision can be expensive. A small production run can make the economics look brutal. But this is also why the Amiga community needs to think beyond one-off projects and small batches. ASIC only makes sense when treated as infrastructure.

If development and validation cost €250,000, the project needs to recover €50 per unit at 5,000 units, €25 per unit at 10,000 units, or €10 per unit at 25,000 units before ordinary production costs are included. If development rises to €1 million, those numbers become €200, €100 and €40 per unit respectively. Those figures are challenging, but not absurd if the product is positioned as the foundation for multiple boards, repairs and complete systems over several years. The key is not to build a single luxury machine. The key is to create a family of reusable silicon components that can serve restorers, motherboard designers and complete-system manufacturers. A Paula replacement could serve repairs. A Denise or Lisa replacement could serve both repairs and new boards. A larger integrated chipset could eventually power a complete Amiga-compatible system.

That is how ASIC becomes practical. Not as one expensive collector product, but as the base layer for a new hardware ecosystem. A realistic ASIC-based Amiga-compatible board could plausibly sell in the €399 to €599 range if production is disciplined and the design avoids unnecessary luxury features. A complete boxed machine with case, keyboard, storage, modern video output, warranty and support might sit closer to €699 to €899. Those prices would not make it a mass-market computer, but they could make it a serious enthusiast platform rather than a museum object. For comparison, many Amiga users already spend heavily on accelerators, scan doublers, recapped boards, replacement cases, keyboards, memory expansions and storage solutions. The money is already in the ecosystem. The question is whether it continues to flow into scarcity, or begins to fund renewal.

ASIC offers something FPGA cannot fully provide: confidence in permanence

The permanence of ASIC is often described as a weakness. Once manufactured, it cannot be changed easily or just not. That is true, and it is why validation matters. But permanence is also exactly what gives ASIC its preservation value. A mature ASIC creates a fixed target. Developers know what they are writing for. Repairers know what they are installing. Manufacturers know what part they are designing around. Users know that one machine should behave like another. That repeatability is important if Amiga hardware is to become normal again.

FPGA cores can evolve, and that is useful during development. But constant evolution can also create uncertainty. Which core version is correct? Which behaviour changed? Which timing issue was fixed? Which bug was introduced? For hobbyist experimentation, that is acceptable. For a preservation-grade hardware platform, the eventual goal should be a stable reference. An ASIC can provide that reference. The Amiga community already understands the value of fixed hardware. Much of the platform’s magic came from developers knowing the machine intimately. They pushed it because it had boundaries. They knew its timings, strengths, flaws and tricks. A properly executed ASIC Amiga would restore that kind of target. It would say: this is the machine, this is the behaviour, and this is the hardware contract.

Accuracy must be proven before silicon is frozen

A pro-ASIC position does not mean rushing blindly into fabrication. Quite the opposite. Because ASIC is permanent, the verification process must be more serious than anything normally expected from a hobby product. A credible ASIC Amiga project would need public compatibility goals, test suites, reference comparisons and broad community involvement. It would need to survive games, demos, productivity software, trackers, copy-protected disks, floppy edge cases, accelerator interactions, display oddities, genlock behaviour and expansion hardware. The demo scene in particular should be treated as an ally, not an inconvenience, because it has spent decades finding behaviours that ordinary software never touches.

The point is not to claim perfection on day one. The point is to create a process that earns trust before production. FPGA prototypes, real-machine comparisons and staged chip replacements would all be part of that path. By the time an ASIC is manufactured, it should not be a guess. It should be the result of years of accumulated evidence. This is where ASIC becomes the mature conclusion of FPGA work, not its enemy. FPGA gives the community a way to discover the truth. ASIC gives the community a way to preserve that truth in a manufactured part.

New motherboards should become normal, but only if new chips follow

There is no technical reason replacement Amiga motherboards should remain rare forever. The scene already has the skills to reproduce and improve boards. Modern PCB manufacturing is accessible, and the repair community has shown extraordinary discipline in documenting faults, layouts and revisions. The obstacle is not the board. It is the chipset.

A future in which new Amiga motherboards become normal requires newly manufactured chips. Otherwise, every new board increases pressure on the same finite pool of original silicon. That is why ASIC is not a luxury add-on to the replacement-board movement. It is the missing piece that would allow that movement to become sustainable. The most promising future is layered. Original machines remain preserved as historical artifacts. Replacement PCBs continue to rescue damaged computers. FPGA remains the development and validation environment. ASIC provides the long-term supply of stable replacement parts. Complete new systems emerge only after the chip-level foundation is strong enough. That future is more convincing than endless dependence on donor hardware. It is also more respectful. It allows original machines to remain intact instead of being harvested.

What an ASIC-based Amiga should include

A serious ASIC-based Amiga should not be a toy wrapped in nostalgia. It should be a practical machine designed around compatibility, repairability and long-term availability. The first priority should be chipset behaviour. The machine must run the software people actually care about, including the difficult software that exposes timing flaws. It must respect original display modes, audio characteristics, floppy behaviour and expansion expectations as far as reasonably possible. Modern conveniences should be included only where they do not compromise the core machine.

That still leaves plenty of room for improvement. HDMI output can sit alongside original video options. SD or compact flash storage can coexist with floppy support. USB can be available without replacing the feel of original input devices. Ethernet, diagnostics, safer power delivery and modern memory can make the system easier to own. The principle should be simple: modern features should support the Amiga experience, not redefine it. The most valuable product may not be a single all-in-one machine. It may be a set of ASIC components that allow multiple products to exist: drop-in repair chips, replacement motherboards, compact new systems and developer boards. That would prevent the future of Amiga hardware from depending on one company, one product or one interpretation of the platform.

A modern ASIC based Amiga should not be trapped at 1987-1994 speeds

A serious ASIC-based Amiga should not only reproduce the original machines. It should also leave room for an enhanced tier that reflects how the Amiga community has actually used the platform for decades. Many users no longer run stock machines alone. They use accelerators, extra memory, RTG-style graphics, faster storage, network cards and modern display solutions. A new Amiga-compatible system should respect the original hardware contract, but it should not pretend that the platform’s history ended with a 7 MHz 68000 and a few megabytes of RAM. This is where ideas associated with SAGA, faster CPU modes, expanded memory and modern video become important. The goal should not be to replace OCS, ECS or AGA compatibility with a fantasy chipset that breaks the software library. The goal should be to provide a carefully separated enhanced mode: original behaviour when original behaviour is required, and a more capable Amiga environment when users want to run modern software, new games, productivity tools or higher-resolution Workbench setups.

An ASIC-based approach could make this cleaner than many bolt-on expansions. A new system could offer a faithful compatibility mode for old software and an enhanced mode with more MHz, more RAM, faster chip and fast memory access, improved graphics bandwidth, higher resolutions, chunky pixel modes, better audio options and modern storage. Done properly, this would not weaken the Amiga identity. It would acknowledge what Amiga users have always done: expand the machine beyond its factory limits. The key is discipline. Enhanced features must not contaminate compatibility. A demo that expects old chipset timing should get old chipset timing. A game that relies on original display behaviour should see original display behaviour. But new software should not be forced to live inside the smallest historical box. An ASIC Amiga can preserve the past and still give developers a larger canvas.

The enhanced Amiga tier

A future ASIC Amiga could offer two clearly defined personalities. The first would be a compatibility mode designed for original software, demos, games, floppy behaviour, display timing and classic expansion expectations. The second would be an enhanced Amiga mode designed for new software and serious daily use. That enhanced tier could include SAGA-style graphics ideas, higher CPU clock speeds, more RAM, faster memory access, modern storage, cleaner digital video, improved audio paths and optional networking. The point would not be to turn the machine into a generic PC. The point would be to create the Amiga that many users have been building through expansions for years, but with a cleaner, more integrated and more affordable hardware base. In this model, ASIC becomes more than a preservation tool. It becomes the foundation for a new Amiga standard: one machine that can behave like the past when needed, and grow beyond it when invited.

The authenticity argument should favour ASIC, not oppose it

Some purists will argue that a machine with newly manufactured chips cannot be a real Amiga. As a historical statement, that is understandable. A new ASIC-based machine is not the same physical object as a Commodore-built computer. It does not have the same factory history, the same board revision or the same material continuity. But authenticity in computing is not only about original atoms. It is also about behaviour. A computer is an active object. Its identity is expressed when software runs, when signals move, when timing matters and when peripherals interact. If a new chip reproduces that behaviour more reliably than a failing original part, then the authenticity argument becomes more complicated.

An original chip that is dying is authentic as an artifact, but unreliable as a working component. A newly manufactured ASIC that behaves correctly may be less authentic as history, but more authentic as a living machine. For a platform like the Amiga, which was defined by use, creativity and hardware intimacy, that distinction matters. The goal should not be to replace historical machines in collections. The goal should be to make sure the Amiga can still exist as hardware when those historical machines are too rare or fragile for everyday use.

The real choice is between renewal and managed decline

The Amiga community has already proved that it can preserve the past. The harder question is whether it can manufacture a future. If the answer remains dependent on original custom chips, then the platform is locked into scarcity. Prices will rise. Repairs will become harder. Donor machines will continue to disappear. Working systems will become more precious and less usable. That may satisfy collectors, but it does not serve a living hardware culture.

ASIC offers a different direction. It would not replace original machines, FPGA systems or emulators. It would give them a stronger foundation by ensuring that Amiga-compatible hardware can still be built when donor silicon is no longer practical. That is why the pro-ASIC case is ultimately the pro-preservation case. It is not about rejecting FPGA. It is not about dismissing original hardware. It is not about selling nostalgia with a modern port on the back. It is about accepting that a machine defined by custom silicon needs new custom silicon if it is to survive as hardware.

A pledge for new Amiga silicon

The Amiga community should make a clear commitment to ASIC as the long-term goal for hardware preservation. FPGA should remain the research platform, original machines should remain the historical reference, and replacement PCBs should continue to rescue damaged computers. But the destination should be newly manufactured, validated, affordable custom silicon. That pledge should come with standards. No ASIC should be rushed to market without serious testing. No product should hide behind branding while ignoring compatibility. No modern feature should be allowed to compromise original behaviour. No project should pretend that donor chips are an infinite resource.

The Amiga needs a path that moves from reverse engineering to FPGA validation, from FPGA validation to chip-level replacement, and from chip-level replacement to complete new systems. That path is difficult, but it is also the only one that treats the Amiga as more than a memory. If every original part is eventually replaced, some will say the machine is no longer the same Amiga. They may be right in a museum sense. But if the machine preserves the timings, the quirks, the sound, the display behaviour, the expansion logic and the intimate conversation between code and hardware, then it keeps something more important than untouched material. It keeps the Amiga alive. And if the choice is between a shrinking world of donor chips and a new generation of purpose-built silicon, the preservation argument should be clear. The future of real Amiga hardware should be ASIC.