One downside of working with the old Inmos Transputer devices is the rarity and cost of the original silicon. Obviously, you can’t sidestep the acquisition of the processor—unless you emulate—but what about replacing the IMS C011/C012 link chip? You need this (expensive) part to interface the transputer to the programming host, but as [Erturk Kocalar] discovered, it’s perfectly possible to coax a Teensy to do that job for you just as well.

Transputers work by utilizing an array of bit serial interfaces to connect a network of devices, allowing for cooperative computation on tasks too large to fit on a single device. This protocol is, at its link level, a simple asynchronous bit serial affair, with 11-bit data messages, and a raw two-bit frame for the acknowledge. The C011 device at its heart is just a specialized UART—it takes 8-bit parallel data from the host, dealing with handshaking, and pushes it out to the first transputer in the chain at 5, 10 or 20 Mbps, but inverted and with two start bits and a single stop bit. In parallel, it performs the same task in the reverse direction.

[Erturk] realized that the Teensy UART has an inverted mode and, crucially, a 9-bit data mode. This allows the second start bit to be generated as bit 0 of the word, with the remaining eight bits forming the payload. Simple stuff. Additionally, the Teensy UART is capable of the maximum transputer bitrate of 20 Mbps, without breaking a sweat.

There is a slight issue, however, in that there is no way to send or receive the two-cycle acknowledgement frame directly. Since the protocol stop bit is a low, it is possible to implement this by simply sending a dummy data word with all 9 data bits low (since the acknowledge is a ‘1’, ‘0’ pattern). In one specific corner case, that of a direct memory PEEK operation, the command is clocked into the transputer, which sends back a two-cycle ACK—almost immediately followed by the 11-cycle data packet with the result. But, since the Teensy UART is still busy ‘fake decoding’ the full 11-bit dummy ACK message, it will miss the data packet entirely.

It turns out that the easiest way to get around this is to speed up the link and run at the maximum 20 Mbps rate. That way, the Teensy will have fully received the overly-long ACK long before the transputer has completed the PEEK command and started to send over the result. Why you would voluntarily run the link slower escapes us, once you’d got the design dialled in and reliability was a given, anyway.

We like transputers, a cool technology that died too soon. Here’s a quick guide to these innovative devices. Some people are really into transputer hardware, like this person. Finally, with the genuine hardware finicky to work with, expensive and hard to find, you could play along with your trusty web browser, and tick it off your nerdy bucket list.

Has anybody heard of the ATW800 transputer workstation? The one that used a modified Atari ST motherboard as a glorified I/O controller for a T-series transputer?  No, we hadn’t either, but transputer superfan [Axel Muhr] has created the ATW800/2, an Atari Transputer card, the way it was meant to be.

The transputer was a neat idea when it was conceived in the 1980s. It was designed specifically for parallel and scientific computing and featured an innovative architecture and dedicated high-speed serial chip-to-chip networking. However, the development of more modern buses and general-purpose CPUs quickly made it a footnote in history. During the same period, a neat transputer-based parallel processing computer was created, which leveraged the Atari ST purely for its I/O. This was the curious ATW800 transputer workstation. That flopped as well, but [Axel] was enough of a fan to take that concept and run with it. This time, rather than using the Atari as a dumb I/O controller, the card is explicitly designed for the Mega-ST expansion bus. A second variant of the ATW800/2 is designed for the Atari VME bus used by the STe and TT models—yes, VME on an Atari—it was a thing.

The card hosts an FPGA module, specifically the Tang 20k, that handles the graphics, giving the Atari access to higher resolutions, HDMI output, and GPU-like acceleration with the right code. The FPGA also contains a ‘synthetic’ transputer core, compatible with the Inmos T425, with 6Mb of RAM to play with. Additionally, the board contains an original Inmos C011 link adapter chip and a pair of size-1 TRAM slots to install two physical transputer cards. This allows a total of two transputers, each with its dedicated RAM, to be installed and networked with the synthetic transputer and the host system. The FPGA is configured to allow the host CPU and any of the transputers direct access to the video RAM, so with proper coding, the same display can mix 68K and parallel computing applications simultaneously. The original ATW800 couldn’t do that!

In addition to the transputer support and boosted graphics, the card also provides a ROM big enough to switch between multiple Atari TOS versions, USB loop-through ports to hook up to a lightning-ST board, and a MicroSD slot for extra local storage. What a project!

If you don’t know what the transputer is (or was), read our quick guide. Of course, forty-year-old silicon is rare and expensive nowadays, so if you fancy playing with some hardware, might we suggest using a Pi Pico instead?

Thanks to [krupkaj] for the tip!

We remember when the transputer first appeared. Everyone “knew” that it was going to take over everything. Of course, it didn’t. But [Oscar Toledo G.] gives us a taste of what life could have been like with a JavaScript emulator for the transputer, you can try in your browser.

If you don’t recall, the transputer was a groundbreaking CPU architecture made for parallel processing. Instead of giant, powerful CPUs, the transputer had many simple CPUs and a way to chain them all together. Sounds great, but didn’t quite make it. However, you can see the transputer’s influence on CPUs even today.

Made to work with occam, the transputer was built from the ground up for concurrent programming. Context switching was cheap, along with simple message passing and hardware scheduling.

The ersatz computer has a lot of messages in Spanish, but you can probably muddle through if you don’t hablar español. We did get the ray tracing example to work, but it was fairly slow.

Want to know more about the CPU? We got you. Of course, these days, you can emulate a transputer with nearly anything and probably outperform the original. What we really want to see is a GPU emulation.

You can’t fake that feeling when a $4 microcontroller dev board can stand in as cutting-edge 1980s technology. Such is the case with the working transputer that [Amen] has built using a Raspberry Pi Pico.

For a thorough overview of the transputer you should check out [Jenny List’s] longer article on the topic but boiled down we’re talking about a chip architecture mostly forgotten in time. Targetting parallel computing, each transputer chip has four serial communication links for connecting to other transputers. [Amen] has wanted to play with the architecture since its inception. It was expensive back then and today, finding multiple transputers is both difficult and costly. However, the RP2040 chip found on the Raspberry Pi Pico struck him as the perfect way to emulate the transputer design.

The RP2040 chip on the Pico board has two programmable input/output blocks (PIOs), each with four state machines in them. That matches up perfectly with the four transputer links (each is bi-directional so you need eight state machines). Furthermore, the link speed is spec’d at 10 MHz which is well within the Pico’s capabilities, and since the RP2040 runs at 133 MHz, it’s conceivable that an emulated core can get close to the 20 MHz top speed of the original transputers.

Bringing up the hardware has been a success. To see what’s actually going on, [Amen] sourced some link adapter chips (IMSC011), interfacing them through an Arduino Mega to a computer to use the keyboard and display. The transputer architecture allows code to be loaded via a ROM, or through the links. The latter is what’s running now. Future plans are to figure out a better system to compile code, as right now the only way is by running the original INMOS compiler on DOS in a VM.

Listen to [Amen] explain the project in the first of a (so far) six video series. You can find the links to the rest of those videos on his YouTube channel.

Back in 2016, Hackaday published a review of The National Museum of Computing, at Bletchley Park. It mentions among the fascinating array of computer artifacts on display a single box that could be found in the corner of a room alongside their Cray-1 supercomputer. This was a Transputer development system, and though its architecture is almost forgotten today there was a time when this British-developed microprocessor family had a real prospect of representing the future of computing. So what on earth was the Transputer, why was it special, and why don’t we have one on every desk in 2019?

This microprocessor family addressed the speed bottlenecks inherent to conventional processors of the day by being built from the ground up to be massively multiprocessor.  A network of Transputer processors would share a web of serial interconnects arranged in a crosspoint formation, allowing multiple of them to connect with each other independently and without collisions. It was the first to feature such an architecture, and at the time was seen as the Next Big Thing. All computers were going to use Transputers by the end of the 1990s, so electronic engineering students were taught all about them and encountered them in their group projects. I remember my year of third-year EE class would split into groups, each of tasked with a part of a greater project that would communicate through the crosspoint switch at the heart of one of the Transputer systems, though my recollection is that none of the groups went so far as to get anything to work. Still how this machine was designed is fun to look back on in modern times. Let’s dig in!

Not Quite RISC

We were told as EE undergraduates that the architecture was RISC, but reading up on it nearly 30 years later I learn that, while it had a relatively simple instruction set, it achieved its one instruction per cycle not by RISC techniques but by clever use of a ROM microcode. Whether this was teaching by lies-for-the-children or the effect of Inmos’ marketing for the processor is unclear, but it’s certainly true that they were making a lot of noise about what the Transputer could do. I remember seeing the video below with its two-screen butterfly demo, ray tracing, and Mandelbrot set, and being bowled over by something that took my Commodore Amiga hours being performed in almost real-time by the Transputers. Yes, relatively low-definition ray-tracing of silver balls was a big deal back in the early 1990s.

The Transputer range was developed from an initial 16-bit offering in the early 1980s to a 32-bit version, then versions for SoCs including an ill-fated collaboration with Sinclair Research, and versions with inbuilt floating-point capabilities. The advantages of the Transputer architecture were eventually whittled away by advances in conventional processor performance, and by the early 1990s with SGS-Thomson in ownership of the company the Transputer development was halted. It had found its way into a range of niche products, but had somehow failed to break into the mass-market dominated by more conventional  microprocessors from rivals such as Intel and Motorola.

Today both Inmos and the Transputer are footnotes in the history of computing. Oddly enough we are now surrounded by mass-market computers with multiprocessor architectures of British origin, but they feature the true-RISC ARM cores whose ancestors were in development at Acorn in Cambridge while the Transputer was grabbing the limelight. The Inmos semiconductor plant in South Wales is today owned by International Rectifier and is still in production, though its days producing Transputers are far behind it.

That Transputer development system at Bletchley is to be part of a restoration project giving the museum an exhibition of Inmos history. Meanwhile the Transputer itself may be dead, but it does have a descendant that is very much still in production. Xmos are another Bristol-based semiconductor company that specialise in CPUs for demanding audio applications, and since their founders include former key Inmos employees their cores are heavily influenced by the Transputer. I may never have encountered a Transputer after leaving university, but a quarter century later as part of a contract working on a high-end audio product, I came as close as it’s possible to get to one because it had an Xmos CPU.

The Transputer then, a bold vision of a semiconductor future that eventually happened, but not quite in the way that its creators hoped. If you find one, hang on to it, it’s a real piece of history!

[Main image source: Inmos IMST425 by Konstantin Lanzet CC-BY 3.0]