Classic Computers Emulated Part 3: The March of the Clones

May 25, 2021

The Ball Was Take From IBM And The Others Run With It.

Image courtesy of IBM

Since the IBM PC was an open standard, it was only a matter of time before other companies would start making compatibles that were direct clones of the system. At first, IBM wasn't worried because they knew that in order to clone the PC, the manufacturers would have to duplicate the only proprietary piece in the system, the BIOS. The BIOS, or Basic Input/Out System, is a chip located on the motherboard. It handles the settings like the disk drives, keyboard, display, cards plugged in the expansion slots, etc. Earlier clone manufactures just directly cloned the BIOS chip. The BIOS contained code that was IBM's intellectual property. Many early clone makers, like Corona Data Systems and Eagle Computer, were sued by IBM's legal department. They didn't get far.

Enter Compaq

Their was a story about how one company's fall from grace brought the rise of another one. Texas Instruments, or TI, were a major semiconductor manufacturer. They made processors and chips that powered calculators and many of the early computers. With the rise of micro-computing, TI entered into the industry with the TI-99 Home Computer released at the tail-end of 1979. The computer was plagued with design issues. It was originally planned to be a computer.

Later on, with the success of the Atari 2600, it was remolded into a gaming console. With the rising popularity of the microcomputer, the TI-99 was remolded back into a computer. The computer suffered from slow performance from the constant design changes, despite the fact that the machine was powered by a 16-bit TI TMS9900 processor running at 3MHz, which was fast for '79. The company had solid control of the software market as well. Software programs had to be published through TI. Because the system was geared towards schools,

TI only released a few games on their system. They didn't change their course until Jack Tramiel of Commodore started to hand TI their rear-end to them in the marketplace. From their, TI got into a intense price war with Commodore and their C64. Jack Tramiel had a slogan: "Business is War", and the TI-99 crashed and burned in the marketplace from its intense war with Commodore, with the C64 dominating the home market.

They were 3 senior managers at TI who were very disgruntled with how the company ran things. Those three managers: Rod Canion, Jim Harris, and Bill Murto, lost confidence in the company. They decided to create their own startup. They found venture capitalist to fund the new company. They decided on the name Compaq.

It later meant Compatibility and Quality, even though it was later mentioned that this was an afterthought. The first computer would be a portable system. Portable computers means something different today than they meant at the time. In the early 80's, portability meant that the computer had a handle, built in screen, and could (barely) fit underneath a airline seat. Also called luggable computers, they had no battery and were powered by a AC cable.

These machines included the Osborne 1 and the Kaypro. The Compaq Portable would be in the mold of the Kaypro, even though their were fundamental differences between the two computers. The Kaypro was a 8-bit computer with a Z80 processor that ran CP/M. The Compaq Portable was going to be a full-fledged IBM Compatible computer with a 16-bit Intel 8088 processor and running MS-DOS. It would be able to run the same software that the IBM PC could run. The computer had a green-screen, 9 inch monitor and shipped with 128KB of RAM, expandable to 640KB. The machine came with two, 5-1/4 inch floppy disk drives that could use 360KB floppy disk.

The issue encountered was how to reverse engineer the BIOS, which determined how compatible the PC is. The reverse-engineering method had to be done in a way that Compaq wouldn't face IBM's legal department. Companies like Corona Data Systems and Eagle Computers just cloned the chip and shipped it in their computers. Because the BIOS contained IBM's intellectual property, both companies got nuked by IBM (sucks to be them).

The Compaq engineers decided to reverse-engineer the BIOS by using the clean-room method. This is a legal method of reverse-engineering a product.For the BIOS, this method works by having one team analyzing the BIOS without disassembling it, and then writing a description of how the BIOS was able to accomplish the task. The notes would be given to team two who would write the routines imitating the description without disassembling the BIOS or looking at IBM's code. This took awhile, but the Compaq engineers wrote a BIOS that was free of IBM code, but yet almost perfectly compatible with the PC.

The Compaq Portable was announced on November 1982, and started shipping on March 1983. It was universally praised for both its design and high level of compatibility with the PC. Not surprisingly, IBM sued Compaq. Big Blue didn't have a case because a code audit was performed that showed that the Compaq BIOS didn't have any of their code in it. Free from IBM's legal department, Compaq experienced rapid growth in it's early years.

By the second year of operation, it had $100 million in sales, and $329 million by 1986. As mentioned in the Wikipedia article, it was the youngest firm to join the Fortune 500 club. By 1987, the company hit the one billion revenue mark. In 1993, Compaq surpassed both Apple and IBM in sales.

The Compaq Portable is emulated by both PCEM and 86box. For The Compaq machines, I've used 86Box though.

Olivetti and the AT&T PC 6300

Olivetti is a company not that well known here in the U.S. They have a much bigger presence in Europe. They are, for all intensive purposes, the IBM of Italy. Known for their business computers, the company decided to make IBM Compatible machines. They made a very technically competent machine called the M24. Olivetti wanted to sell the computer in North America. Their was one big issue though. They weren't a household name over here. As a result, the M24 would fail to the likes of the Compaq Portable and the IBM PC/XT, even though the M24 had better hardware than those machines (I will get into the technical specs later). They hatched a plan. Partner with a American company who would sell the computer for them. They found that household name that they would partner with.

AT&T was looking to get its foot in the computer business. Previously, they couldn't compete in the computer market because they were a government sanctioned monopoly, and as a result, free enterprise saw that as unfair competition. That was the charge IBM made against the company from the 50's to the early 80's. AT&T could take part in computer research, but they couldn't profit from their work, and as such, went into the public domain. Bell Labs, the research division of the company, developed the transistor in 1948. The transistor became the basis for early computer components and popular electronics such as radios and televisions from the 1950's to the 70's. It became public domain so that the monopoly couldn't profit from their work. The UNIX operating system, developed at Bell Labs as well, was placed in the public domain. Anyone who wanted the source code for UNIX could get it at request. Microsoft was one of the parties that requested the source-code, which became the basis for Xenix, which was the their brand on UNIX. IBM requested this code as well, and their port of UNIX became AIX.

The monopoly came to an end in 1982. AT&T was deregulated and split into multiple telecom companies. They included:

Ameritech (Bit the dust after being acquired by Southwestern Bell)
Bell Atlantic (Which would acquire the independent telephone operator GTE to become Verizon)
Bell South (Bit the dust after being acquired by Southwestern Bell)
NYNEX (Acquired by Bell Atlantic)
Pacific Telesis (Also bit the dust after being acquired by Southwestern Bell)
Southwestern Bell (Who would later acquire AT&T, and then rename themselves as AT&T Inc.)
US West (Became Quest)

And of course their was AT&T. On one hand, their survival meant that they had to compete in the marketplace now. At the same time, that also meant that they could compete in fields now that they were previously banned from, like the personal computer market. Entrance into the PC market also meant that they had the opportunity to taking on their arch nemesis, IBM. They needed a computer, and someone to partner with. This partnership wasn't totally unexpected though. AT&T owned around 25% of Olivette. The company partnered with AT&T and would provide the hardware for the telecom giant to sell. Early on, the M24, re-branded as the AT&T PC 6300, was universally praised for it's superior hardware for its day. This computer was not a 8/16-bit hybrid like the IBM PC/XT. This was a true 16-bit system, that was powered by an Intel 8086. Because the 6300 used a 16-bit board, the 8086 could operate without a performance penalty like the 8088 did on the PC/XT. At the same time, Olivetti went with a faster variant of the 8086, with a clock speed of 7.16MHz, faster that the 4.77MHz 8088 used on the PC/XT. On performance benchmarks, these machines were typically 2-1/2 times faster than the PC/XT. The M24/6300 included a high-resolution, CGA mode special to that computer that made a ideal for desktop publishing. The computer had a 640 by 400, monochrome screen with a bit-mapped display. This computer became a very popular with office applications, competing directly with the PC/XT. The extra screen mode also made this machine popular with engineers running CAD software, which put the high-resolution mode to powerful use. This was one of the first computers that made IBM look silly. And I haven't even gotten to the PCjr yet!

PCEM emulates this particular machine very accurately. The only issues I had was with the mouse. The M24/6300 had it's own mouse protocol, which I had issues with. As a result, I opted to use the emulated Microsoft serial mouse. Beyond that, the software that was specifically designed for this particular model ran without issues. The emulator shows PCEM version 8 and 9 in action, even though I have updated my configuration with the latest binary without issue.

Tandy and the Technical Incompetence of the IBM PCjr.

Tandy had a big role to play in this industry. Part of the original "Trinity" of 1977 with the Tandy TRS-80 Home Computer, along with the Commodore PET and the Apple II. Tandy had a very successful run with their TRS series of computers, along with the Tandy Color Computer, or Coco. The company, like Apple, made attempted entry's into the corporate market, without much success. Their first attempt was with the TRS-80 Model 2 in 1979.

While this computer was technically competent, it didn't have much success in the marketplace because corporate customers did not take Tandy Radio Shack serious. Yes, they make the awesome audio speakers and receivers that allows me to blast Foreigner, Styx, and Boston on the loudest volume. They also make those cute little computers with their little games, but they don't know our corporate needs because they have no presence in the corporate market, like IBM does. If IBM was the Vatican of office computing, then Tandy was that abandoned church where all the stoners go to get high while jamming to the Beetles and Pink Floyd.

Tandy made a second attempt in 1983 with the Tandy 2000 business computer. This computer was technically superior to the IBM PC. It was a true 16-bit system which used the relatively unknown Intel 186 processor, which was a stopgap between the 8086 and the 286. This was a very odd computer because while it had a 16-bit Intel processor and ran MS-DOS, it wasn't a true compatible.

This workstation was more like Tandy's take on the PC with their own proprietary design. It would be tempting to call it a "Not-So-Quite IBM Compatible" that were actually typical for 1982-1983. Companies made their own "interpretations" of the IBM design: partly because they felt they could design a better computer than Big Blue, but also because they felt that directly cloning the system could have IBM nuking them with their legal department. The Tandy 2000 crashed and burned in the marketplace, and again for the same reasons; Fortune 500 companies didn't take Tandy seriously like they did with IBM.

By this time, IBM wanted it all. The PC/XT was conquering the corporate market, and in the process, destroyed Apple's chances of succeeding there. IBM didn't have much presence though in the educational market (dominated by Apple), or the home market (dominated by Commodore). The home market was dominated by 8-bit home computers like the Commodore C64.

The low price-point of the system made it very attractive for home users, who wouldn't be investing into those much more expensive IBM and/or compatible PC's. Also, the C64 built itself up as a very capable computer at a time when the video game crash was in full effect. Console was a dirty word during this time. The C64 that had powerful graphics capabilities also had the ability to run applications and educational software. Jack Tramiel, the CEO of the Commodore until he was ousted in late 1983, personally hated video game though. Ironic as most people brought the computer just to run the games though. Big Blue decided to enter that market with their own computer specifically geared towards home users called the PCjr, which would be an entry level PC.

Nicked-named Peanut during its development stage, it was released on early 1984. It included enhanced CGA graphics that allowed higher resolutions and geared toward games. This was the only IBM compatible machine that incorporated game cartridges as a media. The PCjr also had a sound chip developed by TI that was used on the failed TI-99 home computer. Believe or not, many industry pundits were actually much more excited about the PCjr than they were about the Macintosh, which was also released around the same time. Apple was still establishing themselves, but IBM was conquering a whole market. When the computer was announced on November 1, 1983, Time magazine called it's announcement "D-Day for the Home Computer".

However, once the computer was released, things quickly fell apart for Big Blue. The computer had many design flaws. It had a battery powered, wireless, infrared keyboard that wasn't very responsive, and it drained batteries quickly. Also, the text of the DOS prompt appeared small to person sitting from a distance on the 9-inch display using the wireless keyboard. This made typing DOS commands from a distance awkward.

The keyboard that shipped with the PCjr had a Chiclet keys that made the keyboard awkward to use. The keyboard looked like one of those you found on "toy" computers marketed for children. Many would think that since the PCjr is still an IBM computer, you can still plug in a standard keyboard into the PC, right? Wrong, the wired connection used a different connection than the standard types used on the PC/XT. Thus one would have to buy an adapter to plug in the standard IBM keyboard into the case. The computer was expanded by plugging in external modules that were mounted on the side of the case, thus making the computer grow in size and taking up more space when expanding its capabilities.

The PCjr lacked DMA (direct memory access) on the motherboard. DMA was needed if the computer needed to accommodate hard-disk. Hard-Drive kits were eventually made for the PCjr that got around this restriction by providing DMA on the kit board. Compatibility was a hit-or-miss of this computer as well, which turned off many buyers that believed that this computer was still a full-fledged compatible system. The lack of DMA broke compatibility with a lot of software, including Lotus 123. A version of Lotus 123 was released specifically for the PCjr. The lack of DMA also had a performance hit on the computer.

On the PC, when the floppy disk operated, it used DMA instead of the CPU, which didn't hurt performance when peripheral devices were used. On the PCjr though, which lacked DMA, floppy or peripheral devices used the CPU at the same time the program did, which resulted in slower performance. The machine struggled in the marketplace. Due to it's price, between $600-$1600 (depending on options), this computer was very expensive by the standards of home users, which was the intended target audience for this computer.

Apple took advantage of the situation by selling Apple IIe's to people who were originally interested in the PCjr. Discounts were offered to the educational market. Many remarked the the Macintosh was only a few hundred dollars more, but yet offered a lot more than what the PCjr could. Interest in the Mac increased as a result. The PCjr became a disaster in the marketplace.

Tandy was in a bind. The company was going to release their own clone of the PCjr called the Tandy 1000. They were working on the advertisements and literature of the machine. When it became apparent that the PCjr was becoming a disaster for Big Blue, Tandy rapidly changed course. They removed references of the PCjr name from all their literature, and promoted the computer as a standard PC clone with enhanced capabilities.

The company learned their lessons from the Tandy 2000 in that people would prefer compatibility over features, and as a result, placed emphasis on how compatible their machine was to the PC/XT. Tandy saw what IBM did with the PCjr, and they decided to do the opposite. Crappy keyboard shipping with the PCjr. Ship a professional, business-grade keyboard with the Tandy 1000. Unreliable, wireless, infrared keyboard hungry for batteries shipping with the PCjr. Include a standard, wired keyboard that is proven technology with the Tandy 1000. The PCjr has limited compatibility with IBM software. Remind customers that your machine can run the IBM versions of Lotus 123 and the Microsoft Flight Simulator, the two test used to determine IBM compatibility.

The Tandy 1000 was still an expensive computer, but even with the mid-range, feature-full model selling for $900, it was within reach of many home buyers. It was estimated that about half of those buying the Tandy 1000 were buying them for the home setting. The machine became a very massive success in the marketplace, succeeding where IBM failed. This computer was establishing IBM Compatibles in the home setting.

Due to the success of the 1000, Tandy now had a established presence in the corporate and educational markets as well. They really took the ball from IBM and ran with it. Due to the success of the computer, many game developers coded for the enhanced CGA graphics standards that was originally developed for the PCjr. The enhanced CGA standards were the PC's equivalent to the graphics capabilities of the C64 home computer.

The funny irony was that IBM created the graphics standard, but yet, the success of the Tandy 1000 caused people to refer to enhanced CGA as "Tandy Graphics". Many games that were popular on the C64 were ported to the PC with the Tandy-specific capabilities. What made the computer popular with home buyers was the fact that this machine could play all those fun games, but could also run professional applications like Lotus 123 and WordPerfect.

This appealed to home buyers who need to run the same applications that they used in the workplace, yet provided a family computer that the kids could play games on. Eventually, a GUI environment was specifically released for the 1000 called Deskmate. The success of the Tandy 1000 allowed the company to have a worthy follow up to the TRS-80 model of home computers, which Tandy was retiring at this point.

PCEM has really good emulation of the Tandy 1000. It can run Deskmate, and supports the HX, which has ROM-DOS support (MS-DOS in ROM). Tandy-specific games can run without issues. The emulation is solid.