ZX Spectrum prototype ROM is now available for download courtesy of boffins at the UK's Centre for Computing History Got some unexpected time on your hands and a yearning for simpler times? May we present an original prototype ROM of the Sinclair ZX Spectrum, courtesy of The Centre for Computing History, for your tinkering pleasure. It has been just over a year since Kate and John Grant let the team at CCH get hold of the prototype, replete …
Yeah that sentence struck me as really inaccurate: It wasn't just playgrounds, or the 80s. And nobody who counts ever loved a speccy - all the sensible people were Commodore users ;)
On a more serious note, this is really cool news. I never really used a spectrum (just once or twice back in the day) but I'm all for any preservation efforts and particularly for making the rom available.
His comment about preservation is quite true. If I go to a museum to see the one and only prototype of the Spectrum, I'm not too bothered whether it works - its value as a artefact is the key thing. I'd want to see the original prototype, as it looked, with all its original components. Whereas if I want one for myself, I probably wouldn't care too much about the actual model so long as I can switch it on, use it and tinker with it (although I doubt I'd want a 16K one).
It's a bit like going to a museum to see an amphora they've dug up from ancient Rome. I'm quite happy to see a replica next to it, but leave the original as the bits and pieces you dug out of the supermarket car park, please?
Jupiter Ace had 1 K RAM and could run Pacman written in Forth on that. I had one and even got a 2nd for work to control some test gear.
Allegedly it was a sort of cost reduced Spectrum, according to Wikipedia, because I've forgotten most of what I knew. It had 8 K ROM. I did remember that the designers had worked on the ZX81 and Spectrum. Nowadays in the USA, they'd likely have to work at something else for years due to Serf contracts.
I gave it away maybe less than a year after buying it in 1983. I was using an ACT Sirius 1 (Victor 9000) in work (replaced later by an Apricot) which made the IBM look stupid when I first used it in 1982. I'd had a Spectrum as a test card generator in 1982 to align and check Thorn TX10 CTVs customised as AV/Computer monitor/TVs for Apple & BBC Micro and VHS.
Surprising what you can cram into limited space. I remember writing ping-pong and various other games for the ZX81. Compared to my programmable Casio calculator it had a vast amount of memory and unlimited programming potential. Disliked the keyboard though, some of the keys wore out quickly and became somewhat hit and miss.
That is at least encouraging. A whole lot better than 'magic smoke' and a cry of "fuck!" from the R&D lab.
Next thing to do would be to hook up a logic analyser, plug in a 'ROMulator', start writing and downloading code to exercise signal lines and see what peripherals work and don't. Just like we had to do back in the era.
That sounds like the first stage of what normally happens when you power on a Spectrum. The black screen lasts for maybe a second or two (is it checking the RAM or something, somebody presumably knows/remembers?) and then it clears down to the well-remembered copyright message...
It sounds as though it's nearly getting there, but maybe something in the self-test (if that is what it is) is failing. But, as others have said, even if someone can identify the problematic part, best just to leave it in peace.
The black screen lasts for maybe a second or two (is it checking the RAM or something, somebody presumably knows/remembers?)
Here you go:
I have The Complete Spectrum ROM Disassembly too. I also have The Spectrum Operating System by Kramer which taught me how to use Z80 assembler to interface to the Spectrum ROM, the 8K Interface ROM to talk to Microdrives, the RS232 interface and the ZX Printer. We used to joke, that's no LDA that's my DEFW 1F. Oh we could pull any girl in a nightclub back then.....:-)
Now that brings memories flooding back!
We used to have a special piggy-back clip with a socket on the top which was lovingly referred to as "the plonker"
IIRC I'd go along each ram chip and look for a clean waveform on pin 2 using the oscilloscope until I found one giving of a load of noise. Stick the plonker on - prove that the speccy now stars up correctly and then set about desoldering the chip and adding a socket before replacing with a new one.
This was one of the first repairs I ever learnt to do - aged around 9 or 10 years old! I was immensely privileged that the man who ran the local repair shop (and had already fixed my spectrum+2 a few times) let me come and help him on Saturdays and school holidays... I learnt a huge amount over the 6 years I was there and it set me up for life.
I do miss the days of tinkering with those machines of the eighties. I know we've got Raspberry Pis and Arduinos these days but it's still not quite the same.
In the world of embedded system development, having an extra engineer, let alone a supervisory cat as well, working on a project is more often than you might think an unimaginable luxury. In the 21 years I've been in the industry so far, the firmware development team for almost all of the projects I've worked on has consisted of me. And on around half of them, I've also been the hardware designer too...
Right now I'm in the unbelievable position of working on a project alongside TWO other fulltime FW/HW engineers and, right right now thanks to the current fad for working from home, two supervisory cats as well. Though in reality, I think their focus is more on supervising the contents of their food bowls than anything else. The cats that is, not the other engineers. Probably.
Interesting to read the reminiscences. I skipped the Sinclair stuff, and waited till I could afford something with disks and real OS.....which turned out to be an Osborne 01......CP/M (and all its utility software), Wordstar, Supercalc, dBASE-II, BDS C, two flavours of BASIC. Amazing to think that each application came on a 192K floppy, and the OS and the application software ran in 64K of RAM.
*
What is really surprising is that I'm still using a dBASE clone today (Harbour), and still writing (a little) C today.....so my slow learning curve nearly forty years ago wasn't altogether wasted. Some of the learning curve has been thrown away long since (notably BASIC and 8080 assembler), but then predicting the future has always been a lottery!
The most common problem is in the sockets. Production ICs are soldered in but, prototypes with wirewrap pins are socketed. Most often they are tin or nickel plated, (sometimes gold plated to prevent this very issue) and the IC pins are usually a different alloy. Look at the IC pins - are they bright silver or sort of a dull grey with darkish spots? Atmospheric oxidation and corrosion due to galvanic activity can cause a connection to become a semiconductor.
Usually all that is required is gently lifting each IC out of its socket and re-seating it.
"hints of an early incarnation of the Microdrive software, which was later removed in production."
No, the code wasn't removed. The production ROM that went to customers was masked before the Microdrive code had been put in. This ROM is a continuation of the ROM development, putting code *in*, in the expectation that what Sinclair had been sending out was a temporary measure and would be replaced with the "real" finished version.
Surely they have just re-invented the wheel?
In the 1980's I had a Spectrum and bought a paperback entitled "Over the spectrum". This contained a complete disassembly of the ROM code together with the entry and exit parameters of all the many subroutines in the ROM.
Nowadays there's a well-commented online disassembly of the Spectrum ROM here
http://primrosebank.net/computers/zxspectrum/docs/CompleteSpectrumROMDisassemblyThe.pdf
I wrote some assembler code on the Spectrum to provide a perpetual almanac and sight reduction calculations, which was used for astro-navigation on a small yacht for a while. Eventually replaced by an HP-41C programable calculator which did the same thing without the need for a power-gobbling portable TV while also saving all parameters to use for the next day's ded. reckoning, set & drift and course to steer calculations, which the Spectrum did not do (though I could have programmed it to save to tape I suppose).
More recently I translated the Z80 code into ARM assember, doing so on an almost a line-by-line basis (but re-writing the I/O routines to be compatible with my ARM hardware board). It worked great - though obviously way faster than the Spectrum so software timing loops had to be tweaked. As it is not a Z80 emulator it will not run any Spectrum binary code programs, but works OK with spectrum Basic programs.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
