• Please review our updated Terms and Rules here

Replica of a Soviet ZX Spectrum clone

GlueLogic

GlueLogic

Member
Joined
Mar 2, 2025
Messages
21
Location
UK
lgrd2_fs.jpg lgd10_fs.jpglgd18_fs.jpg
I've recently finished designing a replica of the Leningrad-1, a Soviet ZX Spectrum clone from 1988. It's one of the simplest Speccy clone designs out there that doesn't use a programmable logic array, with 44 ICs (45 if you count the regulator) on my version.

It can be constructed entirely out of Western-made components, so you don't need to seek out rare Soviet ICs.

This first version has a few hardware bugs that I hope to iron out in the future (video is currently monochrome only...), but it's currently functional enough to run most software.

More information (including annotated schematics and Gerbers) is available on my website.
 
Do you know if any of the russian spectrums omitted the memory arbitration of the ULA and just let it "snow" ? I always figure the Spectrum could be greatly simplified while maintaining a lot of compatability by losing some of the quality.

Also I'm curious as to whether they had ICs like PALs in the soviet countries?
 
cj7hawk said:
Do you know if any of the russian spectrums omitted the memory arbitration of the ULA and just let it "snow" ? I always figure the Spectrum could be greatly simplified while maintaining a lot of compatability by losing some of the quality.
Click to expand...
On the Russian websites dedicated to Spectrum clones, I haven't seen any mention of ULA snow on clones.

The hardware of most Soviet clones is surprisingly far removed from the original Spectrum; the Leningrad only has one bank of RAM, made up of eight 4164 DRAM ICs (64k x 1). Technically this means there's 64KB of RAM on the board, but the first 16KB is disabled to make space for the ROM.

cj7hawk said:
Also I'm curious as to whether they had ICs like PALs in the soviet countries?
Click to expand...
Not during communist times, at least for consumer-grade stuff. Even finding all the 74-series chips in stock to build something like this back then would've been a challenge.

As Spectrum clones grew in popularity in the early 90's, some chip fabricators started producing dedicated ULA clones such as the Т34ВГ1.
ula-ka1515xm1-216-t34vg1-pro-didaktik-m-a-didaktik-kompakt-98757145.jpeg
 
I don't think any single architecture was cloned more than the ZX Spectrum.

I was always curious as to why it took off so quickly in second world countries, and how it became so popular. It's very difficult to find out any real estimates of how many were made, though I got the feeling that it may have been far more than have historically been recorded.

When I was working in my first job, my friend told me he made his own spectrum from TTL chips, and when I asked about the z80, he said they were impossible to get and he never saw one for a long time and they made their spectrums with local chips. ( He was from Romania ). So for a long time I assume he had made a z80 from TTL chips too, and didn't realized he meant he had other local versions of the Z80, though the rest of the computer would have been regular TTL chips. It's interesting to finally learn more about them.

That T34VG1? it looks like it used wait states for bus arbitration, so would have scaled interestingly. Did they ever make those machines with different video display modes with more color and resolution, or was it always just the original Spectrum screen layout?
 
cj7hawk said:
Did they ever make those machines with different video display modes with more color and resolution, or was it always just the original Spectrum screen layout?
Click to expand...
Later Russian clones from the 90's such as the Pentagon supported the Timex 8x1 attribute mode and occasionally supported x2 horizontal resolution.

Since the 2000's, upgraded versions of the clones from the 90's have been designed that support hi-res and up to 4096 colors.

There's an excellent site here that has tons of history on Spectrum clones, both old and new.
 
hunterjwizzard said:
Is there russian-language software available for it? That'd be a trip to see.
Click to expand...
Most Spectrum software from Russia was written for TR-DOS, as the Leningrad-1 was later replaced by Spectrum clones with 5 1/4" floppy drives. (Some people added a floppy interface to their Leningrad-1, but this required a veritable spaghetti of bodge wires)

Some early Russian software from around 1986-1991 is available in tape format. Here's an example: the first Spectrum port of Tetris was written in the USSR.

Some clone manufacturers added a Cyrillic character set to the Sinclair BASIC ROM that could replace the regular Latin charset with a keyboard shortcut (like on the C64). There's an example of one of these ROMs on my site.
 
GlueLogic said:
I've recently finished designing a replica of the Leningrad-1, a Soviet ZX Spectrum clone from 1988
Click to expand...

Neat project, have fun :)

cj7hawk said:
It's very difficult to find out any real estimates of how many were made, though I got the feeling that it may have been far more than have historically been recorded.
Click to expand...

Lower bound would be number of copies printed for game guidebooks, magazines etc. https://zxpress.ru/books.php has this info.
 
GlueLogic said:
Most Spectrum software from Russia was written for TR-DOS, as the Leningrad-1 was later replaced by Spectrum clones with 5 1/4" floppy drives. (Some people added a floppy interface to their Leningrad-1, but this required a veritable spaghetti of bodge wires)

Some early Russian software from around 1986-1991 is available in tape format. Here's an example: the first Spectrum port of Tetris was written in the USSR.

Some clone manufacturers added a Cyrillic character set to the Sinclair BASIC ROM that could replace the regular Latin charset with a keyboard shortcut (like on the C64). There's an example of one of these ROMs on my site.
Click to expand...

So did people just use these things in English?


That wouldn't be hard to believe. I'm going to tell a long and boring story now that you can ignore, but it gives some useful context. I went to high school in 1999-2003 where I took Russian as a second language. Keep in mind we're only ~10 years after the fall of the soviet union, here. My russian teacher was a very old man who'd been to the soviet union several times before the collapse, and plenty of trips after. My school was one of maybe 16 schools left in the entire state of California to offer Russian, which it did thanks to a program started back in the 1960s. The government asked schools to begin offering Russian as a second language so that high school graduates could be recruited to act as translators for the CIA, military, etc. And by coming out of school with at least a working knowledge, it'd be faster to train them up.

I bring this up because on the other side of the iron curtain, our comrades were doing the same thing with English - but as a fascist regime, they were doing it a lot better. So its not difficult to believe there were enough people walking around Russia in the late 80s with enough knowledge of English to use an early computer.
 
hunterjwizzard said:
So did people just use these things in English?


That wouldn't be hard to believe. I'm going to tell a long and boring story now that you can ignore, but it gives some useful context. I went to high school in 1999-2003 where I took Russian as a second language. Keep in mind we're only ~10 years after the fall of the soviet union, here. My russian teacher was a very old man who'd been to the soviet union several times before the collapse, and plenty of trips after. My school was one of maybe 16 schools left in the entire state of California to offer Russian, which it did thanks to a program started back in the 1960s. The government asked schools to begin offering Russian as a second language so that high school graduates could be recruited to act as translators for the CIA, military, etc. And by coming out of school with at least a working knowledge, it'd be faster to train them up.

I bring this up because on the other side of the iron curtain, our comrades were doing the same thing with English - but as a fascist regime, they were doing it a lot better. So its not difficult to believe there were enough people walking around Russia in the late 80s with enough knowledge of English to use an early computer.
Click to expand...

Mid-80s, I worked in a computer R&D lab for my first job, in Australia, as the only Australian in a department entirely filled by Romanian refugees and political asylum individuals. They all spoke english nearly perfectly and all exhibited an obsessive desire to learn new words when I surprised them with something I said, making me the defacto english dictionary/thesaurus of the office.

My colleage - a guy I still respect - about the same age as I was, made a Spectrum out of components in Romania. He was also a hardware developer and designer, and when he told me my designs were "absolutely crazy like no person would ever conceive of making" - it was high praise indeed. What he meant was that I made pretty complex designs using simple logic, where others would do it differently. And yes, he did respect my designs. We got on extremely well.

I don't think english was that uncommon in technical circles in the 80s in soviet-associated countries. The only person who had a lower command of English ( as in he still spoke it fluently, but sometimes it was noticeable ) was an ex-fighter-pilot. Computers and computer technology development mostly came from the west. Documentation came from the west. If you wanted what they were making, you had to learn english, and to a reasonably high level of ability. As for me? I learned some Romanian words and found they programmed CPUs differently to how I had learnt. But generally, they spoke romanian exclusively unless I was around, and then they always switched to english.

And of course, Western operating systems dominated even if they still programmed the same way they did in Romania....

Though I wonder what a "Windowski" operating system would look like? How they might have tackled the GUI had Microsoft and GEM not already forged the way.
 
In the 80's, Soviet schools had classes for learning BASIC. The computers (school computers were mostly PDP-11 clones) used standard commands; PRINT, FOR, INPUT etc. Students were taught the meaning of these words in English.

Speaking of the PDP-11, one of the USSR's few 70's-80's tech firsts was the Elektronika MK-85, the first pocket sized 16-bit computer. Released in 1986, it was based on the same CPU used in many Soviet PDP-11 clones.
 
cj7hawk said:
Also I'm curious as to whether they had ICs like PALs in the soviet countries?
Click to expand...
Yes! They also had БМК (Basic Matrix Crystal, similar to GA or ULA we would use). The most common БМК I've seen in Soviet Spectrum clones is the 64-pin КА1515ХМ1-216 - you could essentially build a Speccy using one of these with a handful a chips: a Z80, the ROM, the RAM and 2-3 chips to handle cassette/video. The Сириус (Sirius) is one example.

I've now also got an example of a 100-pin БМК called "МЭДЖИК" (I'm assuming a simple transliteration of the word "MAGIC" ... ?) that came in the MAGIC-06 clone.

The MAGIC-06 is post-Soviet Russian and has quite a nice keyboard (for a clone) ... the МС7007 ... that was used on other machines.

I'll definitely check out the design by @GlueLogic ... regarding colour ... if we try the Revision 0 board, when you get colour working do you think it'd be something we could bodge on these original PCBs, or do you feel that maybe a totally new PCB would be needed? I'm quite happy to bodge the Revision 0 board if it's not too difficult.
 

Attachments

  • IMG_9352.jpeg
    IMG_9352.jpeg
    2 MB · Views: 3
  • IMG_9355.jpeg
    IMG_9355.jpeg
    1.4 MB · Views: 3
GlueLogic said:
It can be constructed entirely out of Western-made components, so you don't need to seek out rare Soviet ICs.
Click to expand...
I've actually got a Soviet/post-Soviet hall-effect keyboard with Speccy key caps that would be perfect for this! Of all the Soviet/post-Soviet Speccy clones with atrocious keyboards I've come across, none of them will fit this keyboard due to their case design ... your design might just be perfect for it!
 
clueless_engineer (Brett) said:
Regarding colour ... if we try the Revision 0 board, when you get colour working do you think it'd be something we could bodge on these original PCBs, or do you feel that maybe a totally new PCB would be needed? I'm quite happy to bodge the Revision 0 board if it's not too difficult.
Click to expand...
I'm working on a small PAL colour composite video encoder (based on the NIMF 4 encoder seen on this site) that connects to the combined audio/video/power output connector on the right of the board.

If it works, I'll probably just remove the SCART connector from the Rev. 1 board to save space, since a SCART adapter that connects to the A/V connector can be made with only a few resistors.

Rev. 1 will be a big upgrade over the current version. I've found potential fixes for all of the hardware bugs in the Rev. 0 design - despite this, the component count will be lower due in part to the usage of 64x4kbit DRAM chips (only 2 of which are needed instead of 8 4164s).
 
I'll add that I've only tested the SCART output on one display so far, that being a 90's Samsung TV/VCR. I got a raster (which means the sync signal is probably working), but with a black screen (which suggests nothing else is working, for some reason).

There's a small possibility that it might work on other displays, but I can't guarantee that it will.
 
Leningrad-M_1.png Leningrad-M_2.png Leningrad-M_Keyboard-1.png
The new version is complete. Since it contains substantial differences compared to the original, I've given it a unique name to avoid confusion.
List of some of the changes in the Leningrad-M (the M stands for "mini" or "маленький/malenki", which means "small" in Russian):
  • Fewer components (38 ICs, 34 resistors, 28 capacitors and 15 diodes, among other things).
  • Smaller PCBs (220x141mm instead of 254x151mm).
  • The issue that made the border flicker when programs ran AY music commands has been fixed.
  • 4-bit RAM chips, of which only 2 are needed.
  • Rare ICs (MC1456 and 74LS295) have been replaced with ones that are readily available for a low price on Mouser/Digikey.
  • The Kempston joystick interface that didn't work has been replaced with a Sinclair interface - since it's mapped to the keyboard, less decoding circuitry is needed. It also allows for dual joystick connectors.
  • Since I know what configuration of components work in the original, there are no more "optional" components, which should make construction more straightforward.
  • The A/V connector is now DE-9. This'll make it possible for adapters to plug directly into it, similar to the Amiga A520.
I'll send the Gerbers off to JLCPCB soon. If it works, I'll upload the files to my website.
 
Last edited:
You must log in or register to reply here.
Back
Top