Sony's NEWS workstations (68k, MIPS)

The story of a small team - in Sony, mostly a consumer goods company(*) - creating a workstation and then a line of workstations.

Doi had imagined a continuation of Sony’s efforts in office automation (OA), particularly building on the legacy of the SMC-70, a business oriented computer introduced years before. But the team had other ideas. They wanted to build a UNIX workstation that could replace the DEC VAX systems they relied on daily.

(The Playstation hook is a bit far-fetched, but that’s only the final paragraph!)

The first prototype was ready in just six months. By October 1986, the project was announced, and in January 1987, the first NEWS workstation, the NWS 800 series, officially launched. It ran 4.2BSD UNIX and featured a Motorola 68020 CPU. Its performance rivaled that of traditional super minicomputers, but with a dramatically lower price point

By the end of 1989, Sony introduced the NWS 3860, its first NEWS model powered by a MIPS R3000 processor. The move to MIPS architecture was a major turning point. While Motorola chips had served them well, MIPS offered a simpler and more scalable design philosophy. It also came with a practical benefit: Sony’s own semiconductor division could license and build on MIPS technology in-house.

Sony’s foray into portable UNIX machines also began during this time. In 1990, the company released the NWS 1250, a heavy but movable workstation equipped with a 68030 CPU. Two years later, it followed with BigNEWS, model NWS 3150, which used a MIPS R3000 running at 40 megahertz.

In 1995, Sony released the NWS 7000. It was a technical powerhouse, featuring four MIPS R10000 processors running at 200 megahertz with full 64 bit processing. On paper, it was one of the most capable UNIX workstations ever built by a Japanese company.

From the article:

(*) But see also a previous article: The Birth of Sony Computers [unlocked]

Inside, the Series 35 ran on an 8-bit Z80 CPU with a proprietary Sony operating system. It was a single-tasking machine, with only the printer control running in the background through basic spooling. Every boot required a system disk, with a second floppy for saving documents—there was no hard drive, no non-volatile memory. Yet for its time, the Series 35 was a marvel of compact, reliable design.

One of the most intriguing aspects of the new system [SMC-777, still Z80 based] was its promotion of “Dr. Logo,” an educational programming language developed by MIT’s Dr. Seymour Papert. Sony saw the educational value in “Logo” and worked with Digital Research to port it to the SMC-777. This made this updated model an appealing choice for students and teachers alike, introducing a simple yet engaging way to learn programming by drawing shapes on screen using the “Turtle” cursor. Alongside Dr. Logo, the 777 was bundled with 777-BASIC, 777-Assembler, and 777-MEMO, among other software, further enhancing its appeal to hobbyists and educators.

via Liam Proven on mastodon

7 Likes

I’m a sucker for vertical aspect ratio, and the keyboard is a bit more compact and less lopsided than the usual AT-like layout.

Most of the rest is pretty unremarkable looking, but this one has a pleasant aesthetically pleasing design.

1 Like

I missed this story last year, but have to comment since I was part of the probably very few people in Europe who used these.

In the 1990s the Finnish technical university Helsinki University of Technology (hut.fi) (now part of larger Aalto university) Computer Science Lab bought a classful of these workstations (I understand there were some personal connections to Japan, that’s how and why). Later they also bought more of the MIPS-based servers.

There’s a history of the CS Lab here, all in Finnish, but Google Translate will give you good enough idea. Historia - Niksula - the Sony was the “Niksula II” phase. I actually started already on the “Niksula I”, and graduated before “Niksula III”.

Why are all the monochrome screens so small? For a person with weak eyes. bigger is better.

Screens were just small. In the mid-1990s, 14" diagonal tube size (viewable image was more like 13"?) was normal for home computers, and maybe 17" for workstations. Before that, things were smaller; 12" screens on the 8-bitters of the early to mid 80s were common. By the late 1990s, 17" was not unusual for home computers, and 19+ for workstations.

Why were they small? Accelerating electrons is expensive. Large glass vacuum envelopes are heavy. RAM for framebuffers was scarce, so the pixels became large quite rapidly.

1 Like

The Apple ][ was commonly paired with 9" surveillance monitors… I used that combination back in 1981, and that was perfectly fine (as long as you were happy with 40-column, B/W text :slight_smile: )

For UNIX workstations in the 1990s 17" to 19" was pretty much the range. And since they were CRTs, they required about the same or more depthwise.

We were younger. Eyes were better and most didn’t need reading glasses. Oh to be that way again.

-Gordon

2 Likes

Large displays tended to be blurry and screen geometry was also an issue. Hence, large high resolution displays often came with small deflection angles and were thus massive.

(I recall once coming to a small office space here in Vienna in what must have been around 1990 and finding a well-known local avant-garde artist, Valie Export, behind a Mac with a large high-res B/W or grayscale display, and this display must have been about 3/4 m deep, at least. The weight of this must have been substantial as must have been the price of it.)

That’s my understanding too - a television picture is quite forgiving, but text display needs a good quality raster. I think I recall that oscilloscopes were quite deep in part to have low deflection angles, in order to have low distortion.

I have a couple of 9inch mono screens.

Normal 80 x 24 (mono) displays are fine for me. It is the portrait displays like the Mac that are just a little too small. The same goes for message text input boxes. A adjustable font / window size would be better for composing. is that a , or .