RYAD - The Soviet attempt to clone the IBM S/360

In the late 1960s, the Soviet Union and a number of allied countries embarked on a bold attempt to significantly increase their computer technology and catch up to the west, via an ambitious effort to produce a clone of the popular IBM S/360 line of mainframes.

The result of this program was the RYAD line of Soviet mainframes, a whole family of computers that were a significant leap forward for Soviet computing, and a program whose story is a fascinating one both for the ways in which it succeeded as well as the ways in which it fell short.

The script can also be read over on our Substack account

#soviet #sovietcomputer #ryad #documentary

Update: It has been brought to my attention that I badly mispronounced "peripheral" every single time I said it. I have corrected my pronunciation going forward and future videos will have...well they may have other inadvertently mispronounced words, but "peripheral" will not be one of them ;)

Credits:

[andraskovacs]

I was a mainframe operator on an R-35 in 1984; yes, I was typing in
// ASSGN SYSIN, SYSRES
on a Cyrillic main console keyboard.

[Hectico]

Please don’t stop making cool videos like these!
These are amazingly researched work, thank you for sharing this!

[ivan_pozdeev_u]

I was very surprised seeing "RYAD" as the name. In the USSR, they were exclusively known as "ЕС ЭВМ" (Unified System [of] Electronic Computing Machines).
"Ryad" (lit. "row"; also means "series", "lineup") was but a bureaucratic designation for each stage of the official plan to design and produce different sets of models (Ryad 1, Ryad 2 etc). It was likely largely unknown outside of the circles directly involved in the planning.

[MoultrieGeek]

I love how random the YouTube algorithm is and I'm very glad it is. I'm a nut for mainframe computer history (my dad was a Honeywell systems engineer from 1971 through the early 2000s) so your video is right up my alley. I can't imagine how much research went into this but I'm sure it took ages. Excellent presentation and you've got a new sub.

[gerrymcdonald]

We had an IBM S/360 where I worked in the 1970's.When it came time to upgrade to a much more powerful mainframe the company tried to sell it back to IBM. IBM refused. So the company put it up for sale on the open market. No bites (no pun intended). A company in Manitoba, Canada bought it .. but not to use. Rather that company foresaw the coming likelihood that companies worldwide who depended on S/360's would become desperate for parts. So that company in Manitoba bought all the S/360's they could, for pennies on the dollar, and made a fortune. Their biggest customers were insurance companies. When the demand for S/360 parts dropped to the point that it was no longer profitable to stockpile the parts that company folded and the owners lived happily ever-after.

[wovenvideo]

My father worked on this project. I was born in 1972 and we left Russia in 1979. He told me stories about getting the manuals to the mainframes to start reverse engineering under the guise of the ministry of health. He said he had the manuals as soon as they came out in the west. Do you know what they were doing with that processing horsepower? Crypto for the in flight guidance of their ICBMs. We got out of Russia because of those programs. My father passed away a few years ago.

[philipgrice]

I was in the IBM compatible peripherals group Control Data Corporation bought the RYAD 1040. I was an experienced IBM 360 operator and systems programmer so became involved in the setup and testing prior to and after the purchase. The CDC systems engineer discovered that the RYAD had no FDD in the processor chassis to install firmware. The Selector Channels all were completely compatible with the CDC disk and tape drives but the Byte Multiplexor Channel timing needed adjustment causing the card reader and printer to time out. Unable to install firmware without a FDD the CDC engineer asked the "Finnish" engineers for help. Using the CDC engineer's portable Tektronix oscilloscope (which they coveted highly) a 'Finnish' engineer tweaked the byte-mux timing by adjusting some pots in the back of the mainframe with a screwdriver. It might not have been 'high tech' but it worked perfectly. Then everything worked great.

The system was imported into the USA and setup at the CDC HQ in Bloomington, Minnesota. Once setup and running in the CDC data center Governor Rudy Perpich was invited to attend a demonstration before it was reshipped to Washington to demonstrate just how compatible it was. During the early setup and testing I was able to install OS/DOS and OS/MFT and compile and run several suites of Assembler, Cobol and PL/1 programs successfully.

CDC bought the system to demonstrate the current level of Soviet computing technology. CDC had been blocked by Washington from selling a CDC 3300 first generation computer to the Soviets because of Pentagon objections that it was too powerful and could be used for nuclear physics research. The sale was important as a node in a worldwide weather reporting system that already partially existed around the world using multiple CDC 3300 systems all running the same complex software. The system proposed was one that had returned from a long lease and was effectively a museum piece but such was CDC's justified reputation for high performance computers they never were able to complete the Central Asian node.

As an amusing aside, CDC had an internal product brokerage group. These people found market for products made in the eastern Bloc and elsewhere used to purchased CDC products when the buyers did not have suitable negotiable hard currency, i.e. dollars. I am aware of one sale that was paid with 20, 000 Polish rubber Wellington boots and another from Czechoslovakia paid for with 600 beautiful competition shotguns. BTW, the Soviets bought several IBM 360 series computers from a French leasing company, who resold machines as they came beck from lease, via an Austrian computer broker. They were interesting times in mainframe computing.

[ejomatic]

The quota issues remind me of a factory job I had for a bit; we were assembling small parts for heart and lung machines and I was repeatedly told by my immediate supervisor to put any production over the day's quota into a bin for tomorrow so our quota wouldn't go up. A lack of flexibility from management (no understanding that there are good and bad days and things even out over time) created production bottlenecks because we wanted to keep our jobs.

Turns out that punishing your workers for exceeding expectations creates problems in any system lol.

[mr_noobx]

this is the kind of video i'd expect from a channel with at least a million subs, you having 5k actually surprised me. Hope to see you reaching those numbers soon 💪

[xRbtt]

Thanks for the trip to the past. I used to work part time as a student on Minsk-32 you mention. As I recall, we used IBM 360 manuals for everything. I always thought it was a direct clone. And as you say, the only options you had in languages were machine or assembler, for the most part, with a bit of Fortran thrown in.

[barbaranostrand]

I used to have a co-worker who worked on the IBM-360 clone project. He said that the Soviet Union indirectly bought an IBM-360 via Siemens.

[critical_analysis]

Your stories are amazing and so well researched and find it very objectively made.

Well done! You deserve millions of subscribers for this tremendous work. All the very best.

[IronFist.]

I was on the fence about subscribing to this channel but once I saw all the love for *‘peripheels’* I knew I had hit that sub button with both hands 🙌 ❤

[vladimirrodionov]

The decision to clone the S/360 was controversial back then and oh boy it's controversial in Russia now. In the alternative history the ingenious Soviet BESM-6 system would have been selected as the basis for the future computer lines. It was actually a well designed and performance competitive computer often compared to the CDC- 6600 and it ran a variety of advanced operating systems and had the full range of compilers developed for it. It was actually continued to be manufactured alongside the Ryad.

[unklekal]

I visited Honeywell on a school field trip in the mid seventies, and I was surprised to see they were still building circuit boards by hand for their mainframe computers.

[Moppup]

Thank you for the obscure (actually not boring) story.

[AureliusR]

Say it with me now -- PER-IF-ER-AL ... Peripheral... PER-IF-ER-AL ... four syllables

[LMB]

15:20 the Eastern Bloc not only didn't want to get dependent on the Soviet Union in computing, but also was weary of getting punished for using IP obviously stolen from the West.

That's why our traffic lights (Poland) were controlled by a British ICL mainframe.

(Of course it ran into the 1990's, because sanctions do work, people!)

[lahma]

Awesome video! It is pretty clear that this video took a lot of time and effort to research but ultimately, I think it paid off with the production of an extremely informative video providing lots of little-known information. Thanks for all the hard work expended to provide this video to all of us for free. Definitely earned my subscription!

[Lupinicus]

This is an excellent review of the topic. Congratulations. Very well researched and presented clearly and instructively. Great trove of old films and images too. Subscribed and look forward to more excellent material.