Even at this time, Roberts wasn’t a novice in the realm of computation. In his teens, he’d been a hobbyist, and he’d built machines using digital logic implemented in relays. He’d even started and run two electronics businesses while in the Air Force: Reliance Engineering, and Reliable Radio. But neither the Air Force nor business were Ed’s goals. His first goal was money and that meant business, but that money was sought only to enable him to own a plane, live on a farm, and complete medical school.

A fellow officer, Forrest M. Mims III, was interested in model rocketry, and he wanted to make and sell small flashing light systems for night launches for rockets. Roberts, Mims, Stan Cagle (civilian electrical engineer whom Roberts knew from college), and Robert Zaller (another officer) met at Roberts’ house to discuss a line of products for rocket telemetry. By the end of that meeting, Roberts was elected president of whatever the company turned out to be. After a series of successive meetings, the name MIT Systems was decided upon due to its similarity to the school, and after that, Cagle suggested Micro Instrumentation and Telemetry Systems. As Reliance Engineering still existed, had a credit rating, and had legally existed, it was also decided that MITS would be a subsidiary of Reliance Engineering, but this didn’t happen. In late 1969, the company was incorporated in Albuquerque, New Mexico as Micro Instrumentation and Telemetry Systems, commonly referred to as MITS. Each of the four were given 950 shares and each contributed $100. Another 200 shares were given to the company’s attorney. The first product was the TLF-1 light flasher. The second was “The Booklet of Model Rocketry Telemetry” written by Mims. Transmitters and telemetry modules followed. Despite advertising in the magazine Model Rocketry, the company wasn’t moving a ton of product.

The group had realized that the market for model rocketry, being constituted primarily of professors and hobbyists, could not properly support a company. As the group had taken a rather severe pay cut to found, run, and work at MITS this wasn’t great. So much so, Zaller had to leave for a bit and get proper work to support his household. Mims and Roberts then had the bright idea to build electronics. The first of these was an infrared voice communicator called the Opticom. The first transmitter and receiver pair was completed that summer, financing to build around a thousand kits was obtained from a friend, and Mims managed to write an article about the construction, operation, and modification of the devices for Popular Electronics receiving a check for $400 (this was large sum, adjusted for inflation to May 2026 ~ $3433). This article was also a good bit of advertising for the young company’s product. Orders began coming in.

Roberts, it would seem, could not get past that desktop calculator. He began working on a kit calculator immediately after having helped finish the Opticom. Meanwhile, Stan Cagle began working on an infrared intrusion alarm kit and a solid state laser kit. These diverging paths created quite a bit of tension, and while Cagle and Mims were contemplating buying out Roberts’ stock, things materialized the opposite way. On the 10th of November in 1970 at a meeting at the Sunport in Albuquerque, Roberts offered to buy the others out for a combination of $300 cash, another $300 the following March, and $350 in equipment. Cagle and Mims accepted. After all, they’d never even drawn a salary from the fledgling company. Before the two had even signed the agreement, Roberts had recruited engineer Bill Yates (second lieutenant from the Laser Division at the weapons lab), and a fresh investment from another officer. The company moved from a former sandwich shop to Roberts’ garage, and then to a rented house on 2016 San Mateo.

For his part, Mims became a freelancer around August of 1971, sold some solid state laser articles to Popular Electronics, and wrote the manuals for the laser kit at MITS as a freelancer. The manuals were good, and Roberts asked Mims to write the manual for the calculator project in exchange for one of the calculators. The calculator was completed in 1971 and appeared in Popular Electronics in November that year, just a month after the laser article had debuted.

The 816 was offered as a kit for $179, assembled for $275, with parts and boards available for purchase separately if needed. Circuit diagrams and other technical instructions could be purchased for just $2. The 816 was a major success for the company, and they finally turned a profit. Despite this success, Roberts wasn’t done. Before the machine was completed and began shipping, he had a 32-step programming unit that would make the 816 a programmable calculator (this was made into a product that could connect to the 816 or 1440). He also began thinking about building a microcomputer, but he found the 4004 and 8008 to be insufficient.

The house the company was working out of was scheduled for demolition to widen the street, so MITS moved to 5404 Coal Avenue, and they then later moved to 6328 Linn Avenue. With more space, the company hired technical writers and a receptionist, but more importantly for history, this space allowed the company to more quickly and efficiently put machines together. More calculators followed from those with more digits, to handheld calculators, to scientific calculators with trigonometric functions. They were doing well, but success is a signal to the market of money to be had. Competitors moved in, and by 1974, MITS was carrying around $200k in debt (about $1.3 million in 2026 dollars).

MITS had evolved into a legitimate consumer product manufacturer by this point, and they knew how to build, market, sell, and deliver products. Roberts decided that it was time to make a far more powerful, more ambitious product, and he knew that his company was capable. The Intel 8080 launched in April of 1974, and this provided Roberts’ one of things he’d need for a more powerful product, a microcomputer. Meanwhile the editor over at Popular Electronics asked if the company could have a microcomputer project ready by January of 1975. Roberts and Yates began working on a prototype with interface logic for the 8080, 256 bytes of RAM, a 2MHz clock, front panel logic for 25 input switches and 36 LEDs. The most important piece was decided by Roberts quite early in the project’s life; the machine would include an open bus allowing for additional memory and peripheral cards after the initial assembly. The cabinet could hold 16 cards, and this meant that a rather strong power supply would be required. They fitted what they thought would be sufficient, but they underestimated. No one had really tried something quite like this before. The completed prototype was shipped to Popular Electronics, and in August, Roberts flew to New York to demonstrate the machine to the editorial staff. Sadly, that prototype was lost or stolen at Kennedy Airport, so Roberts was left to just spread out the circuit diagrams and explain the machine. This wasn’t too well received, and Roberts didn’t feel great on the way home. He had no guarantee that Popular Electronics would publish the article, and he’d need to borrow money to get the machine built. Still, it was try or allow the company to go bankrupt since calculators were no longer profitable. Roberts secured another $65,000, and he sent a non-functional case to the magazine for photography. A second prototype followed in September, it arrived safely, and it worked.

Product names at MITS were normally quite straightforward, but this product was different. The VP of marketing, David Bunnell, suggested Little Brother. Roberts eventually called it the PE-8. Les Solomon over at Popular Electronics is credited by Mims with having said that PE-8 was a dull name. He discussed the name issue with two members of his team, Alexander Burawa and John McVeigh, and they came up with Altair. A separate story credits Solomon’s daughter with having said they should name the computer Altair because the USS Enterprise was heading there that night (a rerun of Star Trek, Amok Time). In any case, it would seem that someone Les Solomon knew was responsible for the naming of the Altair 8800. The article ran in the January 1975 issue of Popular Electronics (this meant December of 1974 for those too young to know how print magazines used to work).

The response to the article was huge. By February, the company had received just over 1000 orders. While the initial shipping time was set to 60 days, the company had not anticipated the number of orders they were going to receive. While the company did have delays on orders, Roberts strategy was to focus on the computer itself first. This is… fine, except that the Altair 8800 was nearly useless without expansions. By the end of May, 2500 Altair 8800s had been delivered. At launch, an Altair 8800 was available for $397 as a kit, or $498 assembled.

Paul Allen and his friend, Bill Gates, had run a company called Traf-O-Data while attending Lakeside School, a private highschool, in Seattle. That highschool had a club, the Lakeside Programmers Group, which got free time on some large computers in exchange for writing software. Gates and Allen saw a business opportunity. In the 1970s, if a municipality wanted to know traffic flows on roads, they’d put sensors out that tracked the number of axles and the timing of each axle, and that data would be recorded by punching paper tape with a 16bit pattern. Paper tape readers existed, the Intel 8008 existed, and the two thought that they could process the data more quickly and more cheaply than the companies already engaged in that business. Both of them lacked the hardware and they didn’t know how to build it, but they knew software quite well. After his graduation, Allen went to Washington State University, and he was able to get time on a PDP-10. Allen wrote an 8008 emulator, and Gates, who’d gone to Harvard, wrote the other pieces. They then hired Paul Gilbert to handle hardware. Over 1972, Gilbert built a working microcomputer around the Intel 8008, and Allen’s emulator was good enough that the software worked when run on the real chip. While they’d hoped to make and sell the machines with the software preloaded, this never worked out. They did manage to make some money processing the tapes, but this ended when the State of Washington began offering free traffic processing to cities and municipalities.

Among the many people who read that Altair article was Paul Allen who recommended to Gates that the two should make a BASIC interpreter for the Altair. Having written an emulator for the 8008, Allen began modifying his old software to emulate the 8080 and specifically the Altair using only the manuals for reference, Gates began writing the interpreter, Monte Davidoff worked on floating point, and the three worked from shortly after the article’s release until March of 1975. The interpreter, I/O system, and line editor, fit in just 4K RAM. Gates called MITS claiming that they had a working BASIC, and he setup a meeting. They then punched the software onto paper tape, and Allen got a plane to Albuquerque, New Mexico. Of course, they’d gotten ahead of themselves a bit. They had BASIC, but they did not have a bootstrap loader. No worries though, because Allen managed to write one in 8080 assembly while on the flight. The loader, the interpreter, and the I/O routines all worked flawlessly when demonstrated. Microsoft was formally established on the 4th of April in 1975 with gates as CEO, and the headquarters were physically in the same building (but around the corner) from MITS. This first version of Altair BASIC was later referred to as 4K BASIC as it could run on a 4K RAM machine with 790 bytes available to program code.

By December of 1975, prices had increased. Of course, the base system now included BASIC, enough memory to accommodate BASIC, and the Altair’s bus had become the standard for home microcomputers as the S-100; a standard for which more than 300 peripheral adapters were now available. At this time, an Altair kit cost $439, assembled $621. For memory costs, a 1K RAM kit was $97, 2K was $145, 4K was $195, serial interfaces were $119 or $124 depending upon the type, a parallel interface was $92, and a cassette adapter was $128. Every listing from the era shows the kit price more prominently, and this follows information I can find. Most of these machines were sold as kits. Partially, this would be due to economic conditions of the time, and partially, this would be due to most home computer users being hobbyists and tinkerers. By this time, there were now multiple BASIC varieties for the Altair. 4K BASIC was available for $150, 8K BASIC was $200, Extended BASIC was $350 and required the user to have at least 12K RAM. A text editor, assembler, and system monitor was available for $175. The Altair Disk was an eight inch floppy disk drive, and it cost $1480 as a kit.

The Altair DOS that made this more easily usable was another $500, but it wasn’t available until 1977. Also, saying more easily usable is a little ridiculous. The user would still have to toggle front panel switches to load the OS, and these switches entered binary code.

The late release of Altair DOS was problematic. Gary Kildall had been working on CP/M and had done a sort of limited release of version 1.3 of his operating system in 1977 at the West Coast Computer Faire. In 1975, Gary Kildall began working with IMSAI who released the IMSAI 8080 in December of 1975, and they released CP/M for this machine in 1977. The IMSAI was an S-100 bus machine with an Intel 8080. Version 1.4 was broadly available to the market in 1978 and by this time, S-100 bus systems build around the Intel 8080 running CP/M were a sort of standard for the industry. With this came standardized software packages such as WordStar, SuperCalc (clone of VisiCalc originating on the Osborne), and dBASE.

In March of 1976, David Bunnell organized the World Altair Computer Convention which took place in Albuquerque, New Mexico. This was, as far as I can find, the world’s first personal microcomputer conference, and it was a success given the relatively small market size. In total, 7 countries were represented, 46 US states were represented, and total attendance was 700 people.

In October of 1975, MITS announced the Altair 680 built around the Motorola 6800 at 500kHz. The machine saw some delays, and it didn’t ship until May of 1976. The computer was physically smaller than its predecessor and this likely lent to its popularity as a machine controller. With the single 100pin expansion slot, risers were common and cards would lay horizontally inside the chassis. These cards had to be physically smaller than those for the 8800, and they were not compatible with the S-100 standard. With a CPU that was incompatible with the 8800, largely incompatible expansion cards, and less overall expandability, the 680 was never as popular as its older sibling. This machine was incredibly important to history none-the-less. Microsoft was once again asked to create a BASIC for it. Ric Weiland handled the translation from 8080 to 6800, and the resulting language interpreter was licensed to MITS for $31,200. Later, Weiland and Marc McDonald used the 6800 port as the basis for the 6502 port that was licensed to Commodore and Apple.

While iterations on the design of the 8800 were quick in coming, so were competitors. Roberts had largely accomplished his original goals by the latter weeks of 1976, and he hadn’t forgotten how quickly the calculator market had changed. The company had grown to 230 employees, had achieved sales of $6 million, and Roberts was thinking of an exit. He thought Pertec would be a good candidate for a buyer as MITS had used Pertec’s floppy drives, so the two companies already had a good relationship. Pertec also understood the products, the market, and the customers. Things came together on the 3rd of December and the sale was completed in May of 1977 for $6 million (around $33 million in 2026 dollars). Roberts received $2 million of that sum with the rest split among the other shareholders. Pertec itself was sold to Triumph-Adler for $120 million in August of 1979. The Altair location in Albuquerque was closed in December of 1980.

As for Roberts, he moved to Georgia, the home state of his grandparents where he’d spent much of his youth. He became a farmer and a doctor, as had been his goals, and he practiced medicine for the rest of his life.

While MITS was not a long-lived company, the S-100 bus was not a long-lived standard, the Intel 8080 was soon eclipsed by the 8086, the method of operation of the 8800 was tedious, and the physical design of the computer was more similar to a minicomputer than later micros, nearly everything about the PC and workstations of the next 30 years began with the Altair 8800. The expectation of socketed CPUs, upgradable RAM, removable storage, expansion slots, display adapters, communications equipment, operating systems, ready-made availability or DIY. MITS had also been pivotal in early computer shows, users’ groups, newsletters, the software industry and more. MITS traditions continued with S-100 machines remaining somewhat popular until the launch of the IBM PC. But it is arguable that without MITS and the Altair 8800, there might not have been a market for the PC, the Tandy TRS-80, the Commodore PET, or the Apple ][. After all, the Altair was not the first personal microcomputer, it was simply the first successful one, and that success molded what succeeded it.

My dear readers, many of you worked at, ran, or even founded the companies I cover here on ARF, and some of you were present at those companies for the time periods I cover. A few of you have been mentioned by name. All corrections to the record are sincerely welcome, and I would love any additional insights, corrections, or feedback. Please feel free to leave a comment.

Katsushige Mita had become the fourth president of Hitachi in 1981 as the former president Hirokichi Yoshiyama moved into the position of chairman. A company that had originally focused on power generation and motors had become an international conglomerate with a heavy presence in computing, industrial automation, robotics, fiberoptics, nuclear power, locomotives, and consumer electronics. Hitachi had been wildly successful. The company had helped to rebuild Japan after the Second World War, and they’d then helped Japan become an economic and technological leader.

When Mita came in as president, he’d stated:

We cannot live with tradition alone. I have to make Hitachi a more modern company.

He meant his words in a way that many others do not. Hitachi began a mission of automation (as seen with the 1983 B-16), cost cutting, and innovation. The R&D spend of Hitachi, by 1990, accounted for a bit more than six percent of all R&D expenditures in Japan, and that kind of spending coupled with the company’s pool of intellectual firepower meant that Hitachi held more patents than any other company in Japan. Those patents turned into a stunning array of products from bullet trains to drawing tablets for microcomputers. Within the realm of computing, it was entirely possible for someone to work in a lab and access a Hitachi supercomputer while using a Hitachi workstation that utilized a Hitachi CPU, Hitachi HDD, Hitachi FDD, and Hitachi display and communicated through Hitachi-made networking equipment.

Hitachi’s success somewhat mirrored the transformation of Japan’s overall perception in the West. For those of you too young to remember this, Japan was once seen as making low quality, inferior, cheap, and simple products. Throughout the 1980s, this perception quickly changed. Hitachi was making extremely high quality semiconductor products and consumer appliances, and there was nothing more fashionable than Sony. Thus, what had been unthinkable just twenty years prior, came to pass with little fanfare on the 1st of May in 1989; Hitachi bought out National Semiconductor’s holdings in National Advanced Systems. This gave Hitachi 84 percent of the company, and National Advanced Systems became Hitachi Data Systems on the 23rd of October in 1989. Electronic Data Systems held 16%. HDS systems were around twenty percent faster than the competition while being roughly the same price (or cheaper). Despite ambitions in various computing segments, Hitachi also formed a partnership with IBM (following IBM having reported a 21% revenue drop in Japan for 1990, where Japan represented around 19% of IBM’s total revenues) in 1991 allowing Hitachi to resell IBM laptops in Japan under the Hitachi brand name. Another partnership was created with HP that allowed Hitachi to use PA-RISC CPUs, but also to make their own derivatives of PA-RISC. Even in research, Hitachi wasn’t shy from partnering, and the company formed the Hitachi Cambridge Laboratory in 1989 with the University of Cambridge. This lab would study quantum phenomena.

When talking about Hitachi’s automation, it’s important to mention what controller Hitachi preferred to use. As of 1988, that had been the Hitachi H8 microcontroller. The H8 family began as a series of 8bit microcontrollers with a few 16bit registers, and an ALU that had some 16bit operations. These were decidedly CISC and largely implemented the DEC PDP-11 architecture. The H8/300 and H8/500 were primarily 8bit. The H8/300H and H8S were primarily 16bit, and the H8SX was primarily 32bit. It is nearly impossible to provide a complete history of these controllers due to the number of variants. Given their use cases, these could be ordered from Hitachi Semiconductor with an extremely wide variety of peripheral controllers, ROM capacities, and RAM capacities. Hitachi used these in industrial controls, robotics, digital cameras, printers, smart cards, chess computers, synthesizers, and sold them to anyone who wished to do the same. Some of the more famous uses were the LEGO Mindstorms RCS, Nokia 2110, Namco System 12, Yamaha FS1R, and Roland SC-55. Otherwise, as these were microcontrollers, their use is meant to be invisible.

Immediately prior to the IBM laptop agreement, Hitachi released wonderful consumer-segment laptops. The 1989 HL400C and HL500C were Intel 80x86 laptops with a six and a third inches color LCD built using thin-film transistors. This was a big deal. The screen was active matrix with 384k pixels and each pixel was controlled by its own transistor. The HL400C used an 80286 and the HL500C used an 80386SX. Both shipped with 1MB of RAM as standard (though the 500C could be upgraded), a 1.44MB FDD, 20MB or 40MB HDD, an 8bit ISA slot, and a 16bit ISA slot. They weighed in at fifteen pounds. While these were the most noteworthy in the series, there were non-color versions, and even an 8088 model starting at half the price of the lowest spec 80286.

On the 28th of March in 1990, a long legal battle between Motorola and Hitachi concluded in Austin, Texas. U.S. District Judge Lucius Bunton found that both parties had violated their patent-sharing agreement and had to pay each other some penalties. For Motorola, this made the 68030 effectively illegal to sell, and for Hitachi, it made their H8 microcontroller effectively illegal to sell. Honestly, both Motorola and Hitachi had the resources to develop other products. The true losers in this scenario were companies heavily invested in the 68k such as Apple, Atari, and Commodore.

On the 8th of April in 1991, Katsushige Mita became chairman and Tsutomu Kanai became president. This came as Hitachi was stagnant. While the company was operating 38 research centers across the globe, breaking records with bullet trains, and generally being rather awesome, competition in their primary markets of computing, appliances, and consumer electronics was thinning profit margins, and hindering growth.

Hitachi’s workstations continued in the 1990s with the 3050 series that launched in April of 1991. These systems made use of two different CPUs, the Motorola 68040 and the HP PA-RISC 50MHz. All in this series were treated to HI-UX with X Windows and Motif.

The M68k models came in two distinct physical forms: the LT laptop, and the plain 3050 desktop. The LT was built around a Motorola 68040 at 25MHz and 16MB to 48MB of RAM. The desktop utilized the same CPU, but offered RAM capacities from 8MB to 72MB. The 3050/R was a desk-side machine built around a PA-RISC CPU at 50MHz and 16MB to 112MB of RAM. The highest of high-end machines in this line was the 3050Rsv which use the same PA-RISC CPU at 50MHz but increased the RAM capacity to 128MB. All of these machines had a range of disk options, could be configured with either 16Mbps token ring or 10BASE-T LAN. The 3050 and 3050LT achieved 3.5MFLOPS, and the 3050/R achieved 57MIPS or 15MFLOPS. I’ve never before seen such a clear presentation of the CISC/RISC war of the era. Naturally, the PA-RISC workstations were more expensive.

Announced in 1992 and seeing release over the course of 1993 and 1994, some of Hitachi’s workstations were upgraded with the PA/50L and PA/50M CPUs designed and manufactured by Hitachi themselves. These were in the PA-RISC 1.1 family, and the L model was the low-cost option delivering about 55MIPS while the M model offered 100MIPS. These were 32bit chips with an integrated and pipelined FPU, seven 32bit shadow registers allowing fast interrupts, data prefetching, sleep states, support for SDRAM, an 8K L1 instruction cache, 4K L2 data cache, a 32/64-entry translation lookaside buffer (2-way set, 4K-page, each +2 additional block entries), and a 256K to 32MB block translation lookaside buffer. These chips were built of 1,280,000 transistors on a 600nm, three metal layer CMOS, and placed into a 160pin plastic QFP. Machines not upgraded with Hitachi’s PA-RISC chips would gain the PA-7100 or PA-7300. All of these new machines would feature ECC RAM, advanced fault recovery systems, and HI-UX/WE2.

In the pizza-box form factor were the 200, 220, 230, 235, and 255. The 200 utilized the PA/50M, the 255 utilized the PA-7300LC, and the rest utilized the PA-7100. Otherwise the differentiation for these were cache and RAM. The 200 had the 8K/4K cache and 144MB of RAM, the 255 had 64K/64K L1 cache, 1MB L2 cache, and 256MB of RAM.

The desktop models included the 310S, 320S, 320, 330, and 340. The 310S and 320S had 64K/64K cache, 16MB to 272MB of RAM, and 566MB to 1GB HDD. The 320, 330, and 340 had 256K/256K cache, 32MB to 416MB of RAM and 566MB, 1GB, or 2GB HDD. All of these utilized the PA-7100. There were alternate models the 320G, 330G, and 330T which were graphics-oriented workstations.

The desk-side models with seven expansion slots in the full-size tower were the 430 and 440. The 430 utilized a PA-7100 clocked at 80MHz with 256K/256K cache, and up to 768MB of RAM. The 440 was largely the same but clocked at 100MHz. The 440 could hit around 131MIPS or 40MFLOPS.

The 3500 series were branded specifically as servers. These had the model designations of 310, 410, 510, 520, 530, 540, 630, and 640. The 310 was the lone desktop, the 410 through 540 were desk-side, and the 630 and 640 were cabinets. The noteworthy differences here were far higher RAM capacities, from 16MB to 1GB, and higher HDD capacities from 21GB to 208GB. The peak performance of the 3500 640 roughly matched the 3050RX 440.

The 3050RX 100C was built around the PA/50L at 33MHz, 16MB to 80MB of RAM, a 500MB SCSI disk, 21MB floptical, and a 10.4 inch TFT capable of 1024 by 768 pixels at 512 colors. Being a laptop form factor, this UNIX workstation could be expanded with PC Card, but was otherwise limited. External ports included SCSI, VGA, ethernet, serial and parallel. Top performance for this machine was 42MIPS.

The Hitachi SuperH, or SH-1, began development in 1990 as the replacement for the H8 series of microcontrollers, and it was targeted at roughly the same market; that is to say, the low cost, low power, embedded market where new products were being made such as PDAs and mobile phones. The SuperH was important for Hitachi not only due to reliance on others’ chip designs had proven risky and litigious, but also because the memory market had been a race to the bottom. This humble CPU was the first 32bit, single-chip, RISC microcontroller, and it was designed to be the best choice in terms of instructions per second per watt. The folks responsible for the first SuperH were Tsugio Makimoto, Shumpei Kawasaki, Keiichi Kurakazu, Yasushi Akao, Shiro Baba, Toshimasa Kihara, Shinichi Yoshioka, Ikuya Kawasaki, Hideo Inayoshi, Jim Slager, Ehsan Racid, Takaki Noguchi, Kunio Uchiyama, Masahiro Kainaga, Nobuyoshi Domen, and Hideo Maejima. These project members were spread across multiple research groups in Japan and the USA. In general, SuperH was a 32bit, 56 instruction, RISC CPU with a five-stage pipeline where both the address length and data length are 32bits. Unlike other designs, the instruction length is fixed at 16bits to optimize for memory use and power consumption. SuperH being RISCy had simpler decoders than contemporary x86 chips, but it achieved comparable code density thanks to its use of 16bit instructions each of which could perform multiple functions.

Tsugio Makimoto became the general manager of Hitachi’s semiconductor division in 1992, and his primary goal and focus was the successful launch of SuperH. Mass production of the chip began that September with a ceremony held at Hitachi’s Musashi factory to celebrate the first batch. SH-1 was publicly unveiled at Hitachi’s Microprocessor Technical Seminar in November of 1992, and it was built of 600,000 transistors on an 800nm CMOS process with a variety of packaging options. Depending upon power, thermals, and the specific workload the SH-1 could achieve 16 to 26MIPS reliably while consuming less than a watt (I’ve seen published data showing power consumption between 0.1w and 0.8w).

Why did Hitachi target microcontrollers first? Well, partially they did have PA-RISC chips for high-end, but I imagine it’s also because more microcontrollers are sold than large CPUs. Then, there’s the competition, as Makimoto stated:

SH-1 was the world's first 32-bit single-chip RISC microcontroller that was superior to the conventional chips in terms of MIPS per watt of power. We have deliberately chosen a market segment that differs from the Intel-dominated market. While Intel has been pursuing the development of high-end products for high-speed processing, we have focused on the development of low-power, low-cost chips that can achieve high performance. We have been anticipating the growth of the embedded business in the information appliance field.

Yet, the name of Hitachi didn’t magically make sales appear. Hard work by sales professionals did that. There was also one potential customer in Sega. Hitachi’s sales team had been encouraging Sega to consider the Hitachi PA-10 (a low power, low cost Hitachi PA-RISC chip offering around 10MIPS). Sadly, Sega had no interest, and it appeared as though the games company was going to choose a RISC chip from NEC (while I didn’t cover it, there were options). Still, if it was the cost of Hitachi’s PA-RISC chips then Shunpei Kawasaki thought they may have interest in an even lower cost chip.

In the summer of 1992, Kawasaki, Shirou Baba (direct supervisor), Yasuhiro Ueda (boss’s boss), and Junichi Tatesaki (engineer) went to Sega’s corporate home. They got there only to find out that no conference rooms were available. This was a presentation for which Kawasaki had diligently prepared and rehearsed, and he must have felt frustrated or even insulted. For this group, the chance was just too important. They pushed aside their emotions, and followed the person who’d received them to the cafeteria. It wasn’t as though the cafeteria were empty either. Sega employees were clustered around tables chatting, drinking tea, and having some snacks. Finally, they were met by Kazuhiko Hamada (manager from hardware R&D), and they made their pitch. For his part, Hamada betrayed no interest, but he did say that SuperH was “interesting” and the team got a second date.

As discussions continued, it became apparent that Sega would need some things changed. Multiplier performance needed to be increased, the chip needed to support synchronous DRAM, and it needed to be generally more capable. At Hitachi, this moved SuperH from being a simple microcontroller to a series of chips that were upward compatible and offered a range of performance characteristics. While many people from the SH-1 team were bewildered to hear of SH-2 being worked on so soon, Takaki Noguchi and a coworker went to work anyway. Noguchi had worked on the 16bit multiplier in SH-1, and delivered a 32bit version along with the DRAM interface in just two months (May to July of 1993). Another team added multiprocessor support to the SH-2 at the request of Nobuyoshi Doumen at the Hitachi System Development Laboratory. From an ISA standpoint, the largest change from SH-1 to SH-2 was the addition of 64bit multiplication support. The SH-1 had 56 instructions while the SH-2 had 62 instructions.

SH-2 was demonstrated to Sega in the summer of 1993, and they weren’t impressed. They felt that the 25MIPS on offer from SH-2 was simply insufficient especially as they had it on good authority that Nintendo was working on a 64bit system with Silicon Graphics. Leadership of the two companies convened in Hakone in September of 1993. As it turned out, the multiprocessor support requested and apparently desired only by the System Development Lab, was the answer. The Sega Saturn would utilize two SH-2 CPUs. By March of 1997, more than 7.5 million Saturns were in the world (and just short of a million 32Xs which also used the SH-2), and that meant more than 15 million SH-2 CPUs. Of course, while the SH-2 was used as the CPU in the Saturn, the SH-1 controlled the CD-ROM drive and handled the console’s infamous copy protection. Notably, SH-2 saw improvements over its life and die shrinks were one. This size reduction led to variations, one of which was placing two SH-2s on a single chip.

Part of the deal struck between Hitachi and Sega allowed Hitachi to release their own Hitachi-branded Saturn console. This unit was no different from the Sega-branded model except that included the Saturn Movie Card also known as the Video CD Card. As the name would imply, this card would add VCD playback capabilities to the Saturn. There were multiple versions of the card from Hitachi: Video CD Card, Video CD Card 2, Movie Card. JVC also released such a card, and Victor Video released three models. Within games like Lunar, a VCD card offered high quality FMV playback during cut scenes.

As a microcontroller, Hitachi SuperH did well. SuperH was present in musical instruments, VCRs, CD-ROM drives, navigation systems, household appliances, and more. In general, the SH-1 had a 20MHz clock and offered about 20MIPS. The SH-2, as already noted, was at the center of both the Sega Saturn and Sega 32X, but it also found use in networking hardware, and more notably in engine control units from Mazda, Mitsubishi, and Subaru among others. The SH-2A that merged two SH-2s on one chip was used in multimedia but found many wins in the automotive sector in various roles. The SH-DSP was technically a variant of the SH-2 and was used primary in mobile phones and in cameras. The SH-2 generally had a clock of 28MHz and offered 78MIPS. The SH-2A can be found with clocks as high as 200MHz in later models. By 1996, more than 35 million SuperH devices had been shipped, and ARM had licensed the SuperH patent portfolio to create the Thumb instruction set. These kinds of sales meant that Hitachi was the global leader in RISC with a 35% market share (MIPS Technologies was at 29%) and over a thousand design wins.

The success of SH-1 and SH-2 caught the attention of Casio who were working on a handheld PC that would run Microsoft Windows CE. The SH-3 brought another six instructions, for a total of 68, and it got an MMU. These changes made the SH-3 a rather great fit for the early Casio Cassiopeia handhelds (later models switched to NEC-built MIPS chips). The SH-3 DSP was used in many PocketPC devices from a variety of vendors, and it was used in digital cameras, networking equipment, printers, and phones. Models of the SH-3 DSP were later upgraded with USB supporting a transfer speed of up to 12Mbps. The most famous usage of the SH-3 was in the Japan Aerospace Exploration Agency’s Hayabusa and MINERVA.

The SH-4 was introduced in 1998. This was a superscalar model with vector processing capabilities. It was capable of 1.4GFLOPs, and it was the first CPU with 4D vector instructions. This CPU became the heart of the Sega Dreamcast.

In May of 1991, Hitachi had introduced the FLORA series of personal computers. The first models in this series shipped with a 20MHz 80386SX with the high models using either a 386 or 486 at 33MHz, 2MB to 4MB of RAM, 80MB to 200MB HDD, one or two 3.5 inch FDDs, and Windows 3. On the 11th of November in 1998, Hitachi announced three new models in this line called the FLORA Prius 330J set for release on the 5th of December. These machines were built around the Pentium II at 400MHz, 64MB of SDRAM, a 6.4GB HDD, 24X CD-ROM, and a 100BASE-TX ethernet card. The cheapest SKU provided just the computer, the middle included a 14.1 inch color TFT, and the highest added the TFT and a WinTV Video capture card. These machines shipped with both Windows 98 and BeOS R4.0J. By default, they booted into Win98. Accessing BeOS required either using BeOS boot floppy or installing either the BeOS boot manager or the BeOS launcher from the BeOS Backup CD-ROM.

Etsuhiko Shoyama had been the company’s president since 1999, and he became the CEO of Hitachi in 2003. Hitachi had just suffered through the Asian financial crisis of 1997 and the Dot Com Bust. The company needed to reorganize a bit, trim less profitable business units, and focus more on their core strengths. The company posted a loss of nearly $4 billion for 2001, income of $372 million for 2002, and just $150 million for 2003. Shoyama successfully returned the company to profitability and success. He was followed by Takashi Kawamura who became CEO in 2009.

In 1999, Hitachi had bought the remaining 16 percent of Hitachi Data Systems held by EDS and began transitioning to become a storage solutions business headquartered in Santa Clara, California. At this time the wholly owned subsidiary employed around 2300 people, was active in thirty countries, and had annual revenues of around $1.6 billion. At this point, HDS was still servicing enterprise customers’ needs for CRM, ERP, data warehousing, data mining, CJIS, and eCommerce systems on S/390, Windows NT, and UNIX. However, all of this was focused on what Hitachi referred to as the “No Limits” concept where the company aimed to provide customers access to their information from any computer, anywhere in the world, at any time. For this, they offered clustered storage systems in the Freedom Storage family.

The transition to being a storage provider was completed on the 6th of January in 2003 with Hitachi’s acquisition of IBM’s hard disk business. The disk business of HDS and the disk business of IBM combined to become Hitachi Global Storage Technologies, or HGST. HGST manufactured and sold: Ultrastar enterprise drives, Deskstar consumer desktop drives, Travelstar mobile drives, Endurastar rugged drives mostly used in the automotive industry, and CinemaStar drives which were quiet and meant for audio-video use cases. HGST sold to Western Digital (sans the European manufacturing unit responsible for 3.5 inch drives there, which went to Toshiba) on the 8th of March in 2012 for $3.9 billion and 25 million shares of stock. While Hitachi would no longer manufacture their own storage systems, they did now own about 10% of Western Digital.

On the 1st of April in 2003, Mitsubishi and Hitachi merged their CPU and semiconductor manufacturing operations to create Renesas.

Kawamura was followed by Hiroaki Nakanishi on the 1st of April in 2014.

Hitachi was moving back toward energy, heavy industry, and electric motors. The remaining computing systems were mainframes, but these ceased in 2017. After that, Hitachi sold IBM z systems and the only major difference is that Hitachi-sold IBMs run VOS3. Otherwise, the company has a joint venture with LG that manufactures ODDs, they still make some equipment for telcos, and they manufacture ATMs. The remainder of HDS was merged with Pentaho on the 19th of September in 2017 to form Hitachi Vantara, a wholly-owned subsidiary focused on enterprise cloud.

Masakazu Tokura became CEO of Hitachi on the 1st of June in 2021. While Hitachi still has some computing business, the corporation is more present in nuclear energy, rail, electric motors, appliances, building systems, microscopy, test and measurement equipment, particle therapy, cell culturing equipment. The company has many subsidiaries and joint ventures. Two large former subsidiaries of which Hitachi is no longer owner are Hitachi Astemo which is an automative supplier, and Hitachi Construction Machinery.

As for SuperH, the CPU that led to me researching Hitachi, the patents expired. The SH2 ISA was then implemented in BSD-licensed VHDL as J2.

My dear readers, many of you worked at, ran, or even founded the companies I cover here on ARF, and some of you were present at those companies for the time periods I cover. A few of you have been mentioned by name. All corrections to the record are sincerely welcome, and I would love any additional insights, corrections, or feedback. Please feel free to leave a comment.

Namihei Odaira was born on the 15th of January in 1874 in Ienaka, Shimotsuga, Tochigi Prefecture, Empire of Japan to Sōhachi and Chiyo Odaira. This was the Meiji era, and at the time of Odaira’s birth, the Japanese Industrial Revolution was well underway and Japan’s first hydroelectric generator was completed in Kyoto. In keeping with the times, Odaira’s parents insisted that their children be well educated, and that, as a matter of business practicality, their children learn English. Around the age of sixteen, Odaira’s father fell ill and passed away. Odaira’s older brother then left school to support the family and to ensure that Odaira could continue his studies.

As he was completing his secondary education, Odaira felt anxious about his future. What should he study? At the age of twenty two, Odaira enrolled in Tokyo Imperial University’s electrical engineering department. He would have graduated after concluding his third year, but being young, he became absorbed by hobbies (photography in particular), and he had to repeat a year. He graduated at the age of twenty six in 1900.

The rise in electrification around the world meant a rise in copper mining, and those copper mines needed electricity themselves. Odaira was hired by Fujita & Co and assigned to the Kosaka Mine in Akita under the direction of mine chief Fusanosuke Kuhara. His first job was to serve as the engineering section chief, but shortly after he started in that position, Kuhara set him to work on a power station for the mine. He completed this task in just two years, and he was now convinced that electricity was the future of Japan. He left the Fujita zaibatsu and went to work for Tokyo Dento. This job had him working on power transmission equipment.

Kuhara contacted Odaira sometime in late 1903 or early 1904, and informed him Odaira of his intention to acquire the Akazawa copper mine in Hitachi City in Ibaraki. Kuhara wanted to avail himself of Odaira’s expertise and capabilities. Odaira accepted the job as chief of engineering, completed the Nakazato power station shortly after joining, and then went to work overhauling of the Akazawa mine’s machinery and facilities. In 1905, the Akazawa mine was renamed the Hitachi mine, and sometime shortly after Kuhara started the Kuhara zaibatsu and moved his various enterprises into it.

Largely due to the work of Odaira and his engineering team, production at the mine increased, and this brought about the need for even more power. The result was the design and construction of the 4000 kW Ishioka power station.

Odaira was likely a little obsessive. While working incredible hours for his employer, he and his team were also working on designs for electrical equipment. The first product of their research and design efforts was a 5hp/3.7kW electric motor completed in 1910. He took this to Kuhara, there was a brief debate, and Odaira was permitted to pursue the idea. Offices were finished that November, and a team of about four hundred was assembled. He chose to use the name Hitachi 日立 for this venture, combining the kanji for sun and rise (fitting for a company in the land of the rising sun), and it was a natural fit given the name of the city and mine.

Initial production was difficult, defect rates were high, and this spurred Hitachi to start an in-house training school. This helped, but after setting up a large generator for a customer resulted in another failure, Odaira established a formal equipment testing department, and a formal research and development department.

The company moved to Tokyo in 1917 only to burn down two years later. Despite the loss, this allowed Hitachi to become fully independent of Kuhara zaibatsu (which would later become Nissan) in 1920 as Hitachi, Ltd with Odaira as the managing director. On the 1st of September in 1923, an earthquake damaged Tokyo’s power grid and generation systems. Hitachi’s work in turning this around gained them social credibility and name recognition. The following year, Hitachi completed the ED15 electric locomotive, which combined with their recently received reputation, spurred orders from the Japanese government. Electric fans followed in 1926 and became the company’s first export. Those exports, however, were short lived with the arrival of the Great Depression in 1929. Odaira became the president of the company that year. In 1932, the company began producing elevators, and refrigerators. In 1939, the company built water turbines. As with most companies in countries involved in WWII, Hitachi was required to build for the war effort. This made their factories prime targets during the war. Odaira left the company in 1947 by direction of the US occupation. After the war, the company rebuilt yet again. In 1949, the company began producing construction machinery greatly adding to the country’s reconstruction efforts.

While Hitachi operated in many fields (electrical generation, consumer electronics, locomotives, nuclear reactors, auto parts), the rest of this article will focus on (mostly) on the company’s computing efforts.

Hitachi began research into computing in 1951. The earliest work was in analog computers, and digital computing research began in 1956 with parametron logic. Why would a company that dealt in electrical power generation, construction, appliances, and mining be researching computers in the 1950s? Well, for Hitachi, knowing the exact sag and tension of power lines was rather important. For these calculations, the company completed the HIPAC MK-1 in 1957. The HIPAC MK-1 (HItachi Parametron Automatic Computer) was loosely based upon the EDSAC (and the ILLIAC to a lesser extent), but it was quite obviously different and original as it made use of parametrons. The MK-1 used magnetic drum memory, paper tape input (also used for system bootstrap), and could output to paper tape or a printer. The MK-1 performed its duties well and saved Hitachi quite a bit of time.

The MK-1 was good. Hitachi then enhanced it, provided standardized input, output, and housing for it, and displayed it at a computer exhibition in Paris in June of 1959. This was the HIPAC 101. Shipments of the machine began in July of 1960. An improved version, the 103, shipped in December of 1961. The updated HIPAC utilized core memory, 48bit words, with the capability to address 8192 words. It was exceedingly CISCy with 110 different types of instructions. It was capable of hardware floating point, and it shipped with FORTRAN. This was the final parametron computer from Hitachi. The most impressive bit, to me, reading about the HIPAC is that the mean time to failure was high. Even with the MK-1, it would seem that it ran for a solid seven hours for a single calculation. This is the kind of figure expected of fully transistorized machines, and even more particularly of machines that employed integrated components.

In the mid-1950s, a team at the Japanese National Railways’ Railway Technical Research Institute led by Mamoru Hosaka and Yutaka Ohno began designing logic for a computer that could handle seat reservations on a locomotive. This design was completed in 1958 and then given to Hitachi to implement. The machine, MARS-1 (magnetic-electronic automatic reservation system), was completed in July of 1959 at Hitachi’s Totsuka Plant by a team led by Yasuhiko Tani.

The MARS-1 was unique. The logic was implemented with 4500 diodes, 1000 transistors, and seventy vacuum tubes. Memory was provided by a magnetic drum that was just shy of twelve inches in diameter and another twelve inches in length spinning at 3000 RPM with 200 tracks. The drum provided 400,000 bits of main memory, and 1800 bits of register memory. CPU registers were provided by using the surface of the drum as a series of delay lines.

MARS-1 was used to figure out how the automation of the rail system would work, and MARS 101 took over the job in February of 1964. MARS 101 handled eight days of reservations, 127000 sets on 238 trains, their schedules, fares, and other important data. The MARS 101s across Japan were upgraded over time, and replacements came in the 102, 103, 104, 105, 201, 202, 301, 305, and 501 (this last entered service in 2002). The 103, however, was significantly different from its predecessors.

Hitachi had been working on transistor technology for some time, and the company’s first fully transistorized computer was completed in April of 1959 and delivered to the Japan Electronic Industry Development Association. This was the HITAC 301 and it was similar to the ETL Mark IV (junction type transistors, synchronous clock at 180KHz, magnetic drum of 1000 words) but it had a larger drum of 1960 words that rotated at 12000 RPM. The HITAC 301 achieved 3.4ms per addition, 4.8ms per multiplication, and 6.4ms per division. Importantly, where the HIPAC was intended for scientific and engineering purposes, the HITAC was intended for business processing work.

In 1949, Japan’s wartime asset freeze, which had prevented monetary flight throughout the Second World War, was lifted. At the same time, the Foreign Exchange Law came into effect with the goal of balancing international payments and stabilizing the country’s currency. A side effect of this was that IBM of Japan couldn’t make remittances to its corporate parent or to the IBM WTC. Some attempts were made both by IBM of Japan and by IBM WTC to ameliorate the situation but these efforts came to no avail. The solution finally came in Q4 of 1960 when the Japanese Ministry of International Trade and Industry accepted a proposal by IBM to license IBM’s patents to Japanese computer manufacturers and in return allow remittances to be made by IBM’s subsidiaries. Hitachi was among the very first Japanese companies to take advantage of this new arrangement along NEC, Fuji Tsushinki Seizo (Fujitsu), Oki Electric, Tokyo Shibaura Denki (Toshiba), Mitsubishi Electric, Matsushita Electric Industrial (Panasonic), and Hokushin Electric Works. Later licensees included Yokogawa Electric, Shimadzu, Shiba Denki, Sharp, Yamura Shinko, and Tokyo Denki Onkyo.

Despite Hitachi having been eager to get such an agreement, they didn’t immediately make any use of it. The next several computers used Hitachi’s own proprietary technologies, methods, and often enough materials (lest we forget that Hitachi was originally a mining technology company, then a construction equipment company, electrical generation, and everything else too).

Of course, the 301 was really only affordable for large enterprises, and this left a market niche unaddressed. That niche was filled by the HITAC 201 which was completed in March of 1961. The 201’s development differed from most other computers at the time by having the goals set largely by the potential customers. Hitachi asked what was most important, and Hitachi’s customers responded with inexpensive magnetic tape drives, a kana-character line printer, at least 4000 word memory, and an instruction set suited for accounting systems. These goals were achieved. The HITAC 201 would ship with Hitachi’s second tape drive system (the first having been developed for the KDC-1 at Kyoto University), magnetic drum memory of 4000 words, paper tape reader, tape-punch, and a line printer capable of 120 lines/minute. The 201 could actually support up to four tape decks with each capable of 1000 digits per second with 300,000 digits per reel. As for computation, this was a fixed-point machine that used base ten, had thirty seven instructions, could perform additions in 4ms and mult/div in 30ms.

The Hitachi HITAC 5020 had begun development in April of 1960 by Kenro Murata and Kisaburo Nakazawa, and this would be a largely diode-based machine operating at about 18MHz. The first prototype was completed in May of 1963, and the first delivery was completed in 1965 to Kyoto University. It used core memory, delay line registers, and swappable cards made up logic elements (not too conceptually dissimilar from the cards made at DEC around the same time).

This was the first computer made by Hitachi to feature an operating system. This system, while primitive, was certainly a welcome addition. It featured a system monitor handling I/O and providing simple hardware abstractions, and a job monitor that would load and run each job automatically. It also included the MCP, or main control program, that acted as the scheduler and interrupt handler. The software package offered FORTRAN IV (optimized for the 5020), Autocode (enhanced with file controls, error handling, I/O buffering, scheduling, and macros), and a set of software libraries for common maintenance, operations, programming, debugging, and more. I cannot find as much information about this system as I could for SHARE, and if any reader knows more, please let me know.

In late April of 1965, Hitachi announced the HITAC 8000 series of computers. These were based upon the Spectra 70 series machines from RCA, and were therefore ISA compatible with the IBM System/360. Like the RCA machines, the Hitachi computers came with a range of capabilities but were upward compatible such that software written for smaller machines would run perfectly well on a larger machine. The 8200 was the smallest, the 8300 and 8400 made up the mid range, and the 8500 was the largest. These were IC-based machines with standardized connectors, standardized I/O devices, and utilized an 8bit data format. This last part was vital for the Japanese market where kana was needed. This was also where the primary incompatibility with both RCA and IBM was found. While the English-language software could certainly be run on a HITAC and Japanese-language software could be run on a Spectra, it would require that all of the software libraries, printing capabilities, and so on also be present. These machines could run POS (Primary OS), TOS (Tape OS), and TDOS (Tape Disk OS) from RCA at the time of their introduction. Based upon these, however, Hitachi introduced DOS, EDOS, and EDOS-MSO. The latter of which was interesting for supporting batch, remote batch, and online processing at rather large scale and offering storage management and automatic file allocation for all three modes.

Hitachi was active in the minicomputer market starting with the HITAC 10 announced in February of 1969. This machine was mostly 16bit, and implemented entirely in ICs (except for memory). In its most humble configuration it featured 4096 words of core memory, but it could be configured with up to 32,768 words of memory. The HITAC 10 featured hardware multiply and divide, and featured up to sixty four I/O channels. The computer could be ordered as a rack mount unit or desktop, and it could make use of many other pre-existing Hitachi peripherals such as tape readers, memory expansions, card or paper tape punch, and so on. The software side of things was quite interesting because it reveals Hitachi’s thoughts about the computers use cases. The HITAC 10 shipped with a macro assembler, FORTRAN, a floating point arithmetic package, a set of mathematic subroutines, a debugger, and a premade calculator program. This minicomputer was clearly targeted toward scientific and mathematical environments, and not so much toward business.

This machine’s successor shipped in 1973 with the HITAC 10II. Physically, it was half the height of the original, cycle times were reduced, and the machine gained support for BASIC. The target market had clearly and quickly changed. Two more updates were shipped, the 10II/A and the 10II/L which added IC memory and floppy disk support respectively. In 1975, Hitachi introduced the HITAC 20. This model’s CPU had more instructions, allowed for up to 256 peripherals to be attached, supported newer CRT consoles, and made communications controllers a standard feature. This made the machine popular as a network controller, and enabled remote connection to mainframes across Japan. This minicomputer was frequently the machine allowing terminals at bank branches to communicate to an off-site mainframe system, thereby greatly lowering the cost of computing in business settings. In the 1980s, the 10II was replaced by the E-600 series of minis, and the 20 was replaced by the E-800 series of minis. These were LSI implementations of higher performance with E-600 being 16bit with full backward compatibility and the E-800 being 32bit with M series compatibility. Both series were of the standard nineteen inch rack mount form factor.

In 1973, Hitachi adopted a large monkeypod tree in the Moanalua Gardens as a corporate symbol. Hitachi promised annual payments of $20,000 to Damon Estate for exclusive rights to the tree’s image. The funds helped to maintain the gardens. The tree is protected by law by the county of Honolulu to this day. This particular monkeypod tree is known as the Hitachi Tree. More recently, John Philip Damon, president of Kaimana Ventures, purchased the gardens from the Estate of Samual Mills Damon for $5.05 million. A new contract was signed between Kaimana Ventures and Hitachi on the 28th of December in 2006. The new fee was $400,000 per annum and would last ten years. This was renewed in 2016. Hitachi seems to really love trees, and various Hitachi subsidiaries have dedicated themselves to protecting rainforests and planting trees around the world.

In 1970, RCA sold its computing division to Sperry Rand, and this forced Hitachi to once again chart its own route in computing. Funny enough, this wasn’t a problem. Three years prior, the Japanese government had funded a project to develop Japanese-designed and Japanese-manufactured high performance computers. So, in 1970, the company announced the HITAC 8700, and the following year, the 8800. These were available on the market in August of 1972, and they featured 32bit virtual addressing, multiple processors, pipelining, and buffered I/O with a dedicated I/O controller. The 8800 was one of the fastest computers available at the time, and offered up to 8MB of core memory. Both the 8700 and the 8800 were used for a wide variety of workloads including time-sharing systems.

To make these advanced computers easier to use, Hitachi developed OS7. This allowed for expansion via virtual memory, and swapping to a high-speed magnetic drum. Like the earlier 8000 series machines, the 8700 and 8800 could operate in batch, remote batch, and on-line interactive modes, but these machines could do so for multiple users and multiple programs simultaneously thanks to their advanced hardware coupled with OS7. The operating system was also aware of the multiple CPUs in the computer and could make accommodations for hardware failures so long as one CPU was still fully-functional. OS7 shipped with FORTRAN, PL/I, COBOL, BASIC, and an assembler. The standard utility suite for OS7 offered a linker, sort/merge, file and volume managers, debuggers, test data generator, batch job manager, terminal handler, symbolic and program library manager, math library, statistic library, graphic routines, and plotter routines.

Development of the HITAC M series of machines began in the autumn of 1971, and these machines took advantage of the new technologies developed for the 8700/8800 including virtual memory, buffers, and the use of a dedicated I/O processor. These were descendants of the IBM S/370 compatible machines, but more advanced and proprietary to Hitachi. The high-end M-180 was announced in November of 1974. It offered two CPUs, up to 16MB of memory, up to 64K of buffer memory for 16 channels, an advanced ALU, and a dedicated array processor for vector calculations. The M-170 could be ordered with only a single CPU, up to 8MB of memory, and up to 32K of buffer memory for eight channels. The M-160 reduced the main memory to 256K to 4MB, zero buffer memory, and just five channels. The M-150 reduced main memory to just 1MB, and channels were reduced to four.

A quick note for those of you who’ve never dealt with mainframes of this era, a channel is comprised of a small processor (usually called a control unit) on its own bus to the CPU. It operates largely independently of the main system processor and can run a channel program (list of 8‑byte channel command words [CCWs]) that can read memory, write memory, address the I/O device on that channel, and then interrupt the CPU when necessary. This design frees the main CPU from having to handle every single event on a peripheral device whether that’s tape drives, hard disks, floppy disks, card readers, teletypes, terminals, or anything else and the CPU can just be made aware of whatever information is coming from the device or going to the device. So, the channel could fetch its own program from memory, interpret the channel commands, issue I/O requests, transfer data over its bus, and signal the CPU when it either completes a command or needs something from the CPU. As the channel handles the attached device, from the CPU perspective it is just sending things to a channel address rather than the device directly. This prevents the extremely expensive computer from sitting idle while handling some I/O event or waiting for data to return.

The HITAC M series were treated to VOS, or Virtual Operating System. Larger machines used VOS2 and smaller machines used VOS1. VOS2 supported virtual memory of up to 16MB, could processes up to fifteen jobs simultaneously, had more complete abstractions of the underlying hardware, supported backgrounding/foregrounding of processes, had a basic concept of permissions allowing for shared resources when specified, and had emulators for DOS, EDOS, and EDOS-MSO. In addition to those emulators, VOS2 shipped with an assembler translator from the 8000 series to the M series, and a COBOL converter for the same.

Smaller machines released by Hitachi in the early 1970s (and in particular the 8250) had used NDOS. By smaller machines, we mean those whose memory was somewhere in the 32K to 64K range. This enhanced DOS offered three main processing modes: single, continuous, and spool. It also expanded general capabilities in day to day use supporting console displays, interactive job control, and better disk space allocation controls. It continued the support of COBOL, FORTRAN, and PL/I but it also added RPG. The real-time functions of the system were enhanced, and better remote batch facilities were added (where the NDOS machine was sending a job to a larger machine such as the 8800).

The NDOS advancements were hybridized with features of VOS2 and this became VOS1. We see batch, on-line, remote batch, and remote batch terminal (the NDOS system) modes all present. Like NDOS, VOS1 supported COBOL, FORTRAN, RPG, and PL/I. VOS1 was more fault tolerant than NDOS with ability to reconfigure devices and flag bad memory/storage for avoidance. VOS1 was largely compatible with NDOS, but not as compatible with VOS2.

In 1976 in the USA, Itel Corporation began selling what were essentially clones of IBM mainframes 370/148, 370/158-1, and 370/158-3. The basic idea being to deliver systems that were equivalent in performance but at lower cost. This attracted quite a bit of attention. Hitachi and National Semiconductor formed a joint venture, National Advanced Systems, in Mountain View and took over Itel’s IBM clone business in October of 1979. This included the production, service, support contracts, inventory, and other assets and operations of Itel’s IBM-compatible business. They then released four new lines of systems named AS/3000, AS/5000, AS/7000, and AS/9000. All of these systems were compatible with System/360, System/370, 4300, and 303X software, and for the System/370, its features were replicated in hardware.

A key breakthrough at Hitachi was their Twin-Well Hi-CMOS process in 1978. Comparing the Hitachi HM6147 to the Intel 2147 on Intel’s HMOS, the Hitachi unit met the Intel’s performance at 87% less power consumption. The team that achieved this was led by Toshiaki Masuhara. Masuhara stated:

Mr. Yasui and I decided to develop the 4K CMOS SRAM. At that time, Intel was the top manufacturer of the high speed SRAM. And the part name was Intel 2147. We decided to develop CMOS SRAM, which was as fast as the Intel nMOS SRAM 2147 and has much less power and much less die area. Since we were using nMOS type polysilicon-load cell for the SRAM cell, the area was very small. They were contained in a single well. And we were using CMOS peripheral, so the power was much lower. I remember it was the year of 1977, we designed the circuit and we manufactured the device at Musashi works by using three-micron process. And we submitted the paper to ISSCC and that was accepted and we had the first paper in 1978. That was made into a project, actually a product later named 6147.

The work by this team proved to be among the most valuable to Hitachi’s computing efforts for the next ten years, for while most of the industry was using NMOS, they were using CMOS.

In September of 1978, Hitachi released the first home microcomputer of Japan, the MB-6880. This little machine was initially named the Page Master MB-6880, but was renamed the BASIC Master Level 1 to differentiate it from other models in the line. This computer was not actually developed by Hitachi’s computer division, but rather by the TV division as this wasn’t considered a serious computer. If anyone has information on precisely how this computer made it to market, I’d love to hear about it. Until we know for sure, I prefer to imagine that it was a drama fit for film.

The 6880 was built around a Hitachi-made Motorola 6800 at 750kHz, 4K RAM, and 8K ROM (featuring an integer BASIC). The display output was either 32 by 24 characters or 64 by 48 pixels. For mass storage, the computer could make use of a 300bps cassette interface. The very next year, the 6880L2 was released doubling the RAM and ROM and adding floating point.

In 1980, Hitachi released the 6881, and here RAM is expanded to 16K and the price is lowered.

The final 6800 8bit from Hitachi was the MB-6885 or Basic Master Jr of 1981. This was a 16K RAM and 16K ROM unit with floating point capable BASIC. It also had dedicated VRAM allowing a resolution of 256 by 192 pixels, the PSU was integrated into the main computer housing, and had a cassette speed of 1200bps. Other immediately noticeable changes are the printer port and expansion port on the rear of the machine.

The sudden name switch to Basic Master Junior is due to the release of the 6809 based Basic Master Level 3 or MB-6890 in 1980. These machines shipped with Microsoft BASIC, and they had far more “professional” peripherals including floppy drives, color displays, printers, and so on. The 6890 was built around the Motorola 6809 at 1MHz, 32K RAM, 24K ROM, and could support a top resolution of 640 by 200 or 320 by 400 in eight colors. This machine also included six expansion slots. This was followed by the 6891 which increased the ROM size allowing for full kana support. The 6892 followed and increased RAM to 64K.

Of course, the world of microcomputers moves quickly. By 1982, an 8bit CPU was a low-end device. Hitachi then released the MB-16001 with an 8088, 320K RAM (expandable), and MS-DOS. This machine also supported color graphics, full kana, and kanji. As this was coming from Hitachi’s computer division who’d by this time realized that microcomputers were, indeed, real computers, the 16001 had FORTRAN and COBOL available.

This, of course, is Hitachi. While they made computers, they also made pretty much everything else somewhere… so very similar machines were created for OA (office automation) and FA (factory automation). These were 8088 machines with RAM ranging from 256K to 512K with 5.25 inch floppy disk drives and 10MB hard disks. They were made available in 1983 as the B-16 range. The main feature to set these apart from anything else available was the software selection. I haven’t read of another computer sold with robotics routines and robot monitoring as standard.

With many different machines capable of taking the high-end market, the Hitachi HD68B09E (Motorola 68B09) at 2MHz was used to create the MB-S1 series of computers in 1984. The S1/10 shipped with 48K RAM, 48K VRAM, and Microsoft BASIC. The form factor was now a desktop machine rather than the wedge, and the formerly expensive and high-end peripherals were quotidian. The S1/20 was essentially the same machine but added a ROM card containing kanji. The S1/30 and S1/40 were the next SKUs in this line, and these integrated the floppy disk drives into the computer’s housing, and these were higher capacity 1MB floppies. In 1985, the S1/10 was released as an S1/10AV adding six voice audio, and Atari-compatible joystick ports. The S1/15, and S1/45 took the 10 and 40 and added a communications ROM and an RS232C port.

While Hitachi had been releasing small machines, the company hadn’t lost sight of the truly large machines either. In August of 1982, the company announced the second Japanese super computer, the HITAC S-810. This computer implemented the IBM System/370 ISA on the scalar Hitachi HITAC M-280H CPU at 35.71MHz with a 256K cache. The vector processor registers were 256 elements wide where each element was 64 bits, and the vector processor ran at 71.43MHz. Main memory was implemented in CMOS ICs, and could be up to 256MB. Expansion memory could be up to 3GB. Total performance peaked at 630 MFLOPS. Just five years later, the company would follow this with the S-820 offering performance up to 3 GFLOPS, 1GB of RAM, and 24GB of extended memory. The S-820 was the first supercomputer to have an integrated video output system as a standard feature.

Both the mainframe and minicomputer HITAC brands continued with the L series. The L-470 and L490 were decidedly more mainframe-like while the L-450 could be order either in a mainframe-like form factor or “Desk-Type.” These machines were implemented in VLSI with 20k gates per chip in the logic circuits. Main memory for these machines could be up to 12MB and maximum disk capacities up to 5.4GB. The most amazing aspect of these machines was that they didn’t require any special power or air-conditioning considerations.

In September of 1985, Hitachi entered the UNIX workstation market with the Creative Workstation 2050 running their in-house HI-UX OS based on 4.2BSD. I’ve never used HI-UX, but it sounds like a great addition to the UNIX ecosystem. The OS added Japanese language support, defaulted to csh, had cross compilers for the 80186/80286 and Motorola 68K CPUs, and shipped with TCP/IP and ethernet support. There’s mention here of a “multi-window screen,” but I do not have much information on which graphics server was in use. In images of the 2050 series, a tiled window manager is clearly seen, but before the end of Hitachi’s workstations, HI-UX was using X11 with CDE. This UNIX lineage still exists at the time of this writing with HI-UX/MPP which makes use of the Mach microkernel. The IBM-compatible mainframe HI-UX releases are of a different lineage entirely coming from Interactive Systems corporation. For the hardware, the 2050 was built around the 68010 with 2MB to 4MB of RAM, 40MB to 65MB HDD, a 3.5 inch floppy disk drive, and either a 15 inch or 20 inch color flicker less, anti-glare treated, adjustable, CRT display.

Simultaneous to the release of the company’s UNIX workstation, Hitachi released a DOS-based workstation, the 2020, based around an Intel 80286, 1MB of RAM, one or two 5.25 inch floppy disk drives, a hard disk ranging from 10MB to 40MB, and a commodity CRT display of either twelve or fifteen inches with the buyer able to select either monochrome or color. For these, Hitachi didn’t ship commodity MS-DOS, but rather they shipped a compatible DOS that offered multi-tasking. I initially thought that this must have been Concurrent DOS. This would explain the presence of cross compilers on the 2050’s HI-UX port as well as DRI provided compatibility libraries for UNIX. This does not appear to have been the case. I have no reason to suspect that Hitachi ever had dealings with DRI, and Hitachi had extensive OS development experience. From everything I’ve been able to find, Hitachi made their own M68K port of UNIX System V release 2 with 4.2BSD, made their own multitasking DOS, and made their own compatibility libraries, cross compilers, and runtimes. If anyone has more information, let me know.

These lines were updated in May of 1988. The 2050/32 brought the 68020 at 20MHz with the 68881 coprocessor and 64K cache, up to 16MB of RAM, and an 88MB HDD. The 2020/32 brought the 80386 with an 80387 and 64K cache, up to 16MB of RAM, and an 80MB HDD. The 2050/32 was released in 1989 with a 68030 at 25MHz, up to 52MB of RAM, and a 168MB HDD. The 2020/32E was released in 1990, and was cost reduced (and physically smaller) with a 386SX at 16MHz, a 387SX, 4MB of RAM, and a 40MB HDD.

With the likes of SGI taking UNIX workstations into the world of graphics, Hitachi chose to compete. This brought about the 2050G in 1988, and the 2050G/ET and 2050G/EX in 1989. The 2050G was a machine built around the 68020 at 20MHz with a 68881 and 64K cache. The machine could be fitted with up to 16MB of RAM, a single 3.5 inch FDD, up to a 2.9GB HDD, and either a fifteen, twenty, or twenty six inch color display capable of 16.7 million colors and a resolution of 1280 by 1024 pixels. The rather high display specifications were enabled by a bespoke LSI chip for two dimensional graphics. The 3D performance was enabled via the same LSI, the 68882, and software. The G/EX used a 68030 at 33MHz with a 68882, increased the maximum RAM to 64MB, and added a RISC coprocessor that functioned as both a 2D and 3D accelerator. The ET was roughly the same as the EX but offered a larger 3.3GB HDD.

The Hitachi Proset 30 was built in 1989 and Hitachi called it a “personal work tool.” It was built around an 8MHz 80286, standard RAM was quite low at just 256K but it could be upgraded to 4.5MB. While the machine wasn’t generous with RAM, it did ship with a 40MB HDD, two 1.2MB FDDs, a 640 by 400 color CRT, SCSI, RS232C, modem, telephone, keyboard, and a mouse. As one might guess by these specifications, the machine was oriented around word processing, drawing, email, phone, and fax. Its OS did allow for window switching, and it came with software for the capture and display of still images.

By 1990, Hitachi had been around for eighty years, they’d entered many industries, and their computing division had developed everything from supercomputers to portables, and their TV division had made some home micros. Notably, Hitachi fabricated their own chips, and they had some good history with fabrication. Their CMOS process was as good as any competitor, and their vertical integration assured them supply of almost all of the components they’d require for any endeavor. The company was well known and well positioned for the competition of the 1990s.

My dear readers, many of you worked at, ran, or even founded the companies I cover here on ARF, and some of you were present at those companies for the time periods I cover. A few of you have been mentioned by name. All corrections to the record are sincerely welcome, and I would love any additional insights, corrections, or feedback. Please feel free to leave a comment.