home.social

#transputer — Public Fediverse posts

Live and recent posts from across the Fediverse tagged #transputer, aggregated by home.social.

  1. Slides eines Talks über Transputer aus 2013. Zwei Seiten Quellen und Links, die Titel im Verzeichnis sind vielversprechend. Locker die Hälfte der Links sind 2026 kaputt, Dokumente, Seiten oder ganze Sites verschwunden. Ja, Wayback-Machine existiert, ist aber teilweise mühsam und kennt auch nicht mehr alles. Ich mache mir mittlerweile von jedem Schnipsel, den ich später mal vielleicht noch brauchen könnte, ein PDF für das lokale Archiv. #transputer #vintagecomputing

  2. I've spent days trying to work out why Inmos TDS3 compresses all its output onto one line on RISC OS. Turns out it's a TDS bug!
    If TDS hadn't been open-sourced I'd never have found it.

    If the GOTO X/Y definition in the ITERM (termcap) file ends with a X or Y coordinate substitution, its offset parameter will be sent as a literal byte instead of the coordinate!

    The workaround is to end the definition with at least one literal character, e.g. a null.

    #transputer #transputers #retrocomputing

  3. @Kroc @psf I’m intending my #transputer emulator project to do this. Code running on the emulated CPU is connected over a link to another system that has the screen attached. A simple protocol over that link allows the code to access host resources, including terminal (and later, graphics). When I have an OS, & a display, that same protocol can be implemented by on the transputer, making it all self-contained.

  4. Have been contemplating displays for forthcoming #Transputer emulator-driven devices. I looked at resolutions...

  5. @timixretroplays Just found this on AliExpress - thanks! - and each board is SEVENTY FIVE UK PENCE WITH FREE SHIPPING. I’m building my #transputer emulation system with Pi Picos, and if I can get PIO working off something this small, I’ll be able to build a cluster for next to nothing! Incredible!

    aliexpress.com/item/1005008513

  6. So I have the USB CDC link to debug, then integrate the slow GPIO link, and get the separate link adapter build working.

    Then it’s time to get it all building on all platforms - with some extra IServer code for Windows ‘cos serial I/O is a completely different beast on the Fischer-Price OS (of course it is, blessss…)

    And I’m maxed out with other stuff involving different hats… so I’m not setting deadlines, other than “the future” :)

    #Transputer

  7. Hmmm. I hooked Link0 of a Transtech TTM15F-85-F module up to the host card, and it's flagging Error and failing to communicate. I can see the peek command on LinkIn but no reply on LinkOut. Very odd. #transputer #Transputers

  8. Turns out if you want a Replicator, all you need to do is learn Occam. #transputer

  9. In which I post an update to my blog (taking Joan Westenberg’s recent advice to blog more), and try the RSS Parrot… @birb - Parachute Post-winter-solstice update: mattgumbley.wordpress.com/ #Transputer

  10. Got my first transputer program running from my #PC6502

    Very simple code that spits out 0xAA000000 if it’s a 16 bit transputer and 0xAAAA0000 if it’s 32 bit.

    Output is -
    << READ =$AA
    << READ =$AA
    << READ =$00
    << READ =$00

    Next write something simple to confirm the tool chain I have works.
    Then a server program to run on the 6502, so I can then run some Mandelbrot code on the transputer.

    Long way to go, but fun

    #retrocomputing #transputer #6502

  11. So the question; what should I try and get running on my #Transputer?
    Obviously this is going to be one of my first to try - geekdot.com/1st-code-example/
    But I'd like something that does a memory test maybe, or ideally something that scans a network of Transputers (Yes I have more transputers and yes I am thinking of hooking them all up just because).
    links to any suitable tools (sadly all of my current Transputers are integer only) would be gratefully received :)

  12. Here it is, and oh boy does it get toasty!
    Next I best figure out how to get it doing something, like say..... a Mandlebrot!

    #Transputer #PC6502

  13. FINALLY!!!
    The #PC6502 is talking to a #transputer !!
    I can now write and read to the memory on the TRAM !!!

    Colour me very happy!!!

    #Retrocomputing

  14. Trying to build a project that can be built for desktop OS and Pi Pico, using three different stdio implementations (C++ iostream, Pi Pico stdio and direct over USB CDC) depending on which part is being built, for which platform… is “tricky”.

    Samuel L. Jackson may have a more appropriate term.

    #Transputer emulator release is currently blocked by such shenanigans. A restructuring too far that I’ll have to stare at for a few more hours or revert.

  15. The Pico bit-banged #Transputer link works! At a staggeringly-fast 96 bytes/second! If I remove the oversampling receiver it could go 16x that, and I could run the timer interrupt 10x faster - but that would max the CPU out. This is good enough for a proof of concept release. PIO later. Still got to do another link implementation over a USB-CDC interface, and route logging over a separate one, then wire all this into the emulator. Looking tight to release by the solstice :(

  16. Looking at it a bit more closely - if I ignore the top clock signal which is a bit wild, the rest of the traffic looks perfectly cromulent. Second #Transputer Link is acking as expected, and reflecting the received byte back.

    I’ve slowed the click down 10x, but it still has gaps. Can’t explain that yet, nor the link overrun error I see in the test harness.

  17. Bother. I did not expect to have a wobbly clock. First trace from the #Transputer link interface (bit-banged). The top trace rising edge is the start of the timer interrupt; falling edge is end of the ISR. Rising edges should be regularly-spaced. The transmitting link reported a framing error. Perhaps I need to clock it slower - it's dead slow anyway without PIO, but if the clock pulses are that close together, there'll be no time to run the emulator.

  18. Pretty close with the #Transputer link implementation - but I’m a renowned expert at estimation so who knows 😂 - fixed a few bugs, including one that strict adherence to TDD could have caught, but I just wanted to ship something! Will self-flagellate later… (how do non-TDDers do it? How are they so brave?)

    Submitted one PR, got a couple of improvements to make tomorrow.

    Will try to take part in the 2m AFS RSGB #Hamradio contest as well. Got to research contesting from home, not tried before.

  19. Going to be a bit busy with #Commercialmas, but hoping I can get the release of the #Transputer emulator out by the winter solstice (21st). One slow-speed Pico link plus one USB CDC link to the host for now, single core. Fingers crossed.

  20. Wooot today’s haul of chips from someone at work having a clear out of his loft.
    12 Inmos T414 #transputer chips and 7 TMS32020 DSP chips
    I’m looking forward to connecting these to my #PC6502 stack when the PCBs arrive (couple of weeks time)

  21. I tooted a while ago about wanting a cross-platform top-level build tool that handled a software project's lifecycle, and release process - in a manner similar to #Maven but not so heavyweight. I'm quite busy trying to get the #Transputer emulator released by the end of the year, and didn't want to go on a long side-quest... so I prompted Claude.AI and it generated a plausible python script to do this. I've reviewed it, there are some small bugs, and plenty of improvements... 1/n

  22. "My CPU is a T800 Transputer, a parallel processing computer."
    #retrocomputing #transputer

  23. Continuing with bit-banging the #Transputer emulator link on the Pico for now, although it won’t be fast enough to work against real chips (that’s not my goal). 5ns is probably the fastest timer interrupt I could achieve. It is fully TDDed though; a huge advantage IMHO. And it will enable me to get the API to the rest of the emulator correct. To reach 10 Mbps/s and beyond (which is my goal) requires mastering PIO, which is going to be hard to TDD. There are PIO simulators; research required.

  24. Been staring at #transputer datasheets some more. Seems like the Scott connectors Inmos specified for the Transputer TRAM boards are very much no longer available -- but Harwin do the H3192-05 which looks like a very close match. Except it'll need a plastic shroud to set the correct height. Hmm. Not impossible.

  25. The last thing I expected when looking up a piece of chip art[1] at lunchtime was to find that it's Fat Freddy's Cat, from the Fabulous Furry Freak Brothers.

    I guess the Inmos engineers needed a break after finishing the chip layout.

    1: people.cs.bris.ac.uk/~dave/S42 top right, credit David May, people.cs.bris.ac.uk/~dave/tra
    #transputer #inmos #transputers

  26. If you're the person who spunked £500 down and properly outbid me on the CSI / CSA Transputer Education Kit on Ebay UK a few weeks ago, I have just one question...

    ... can I please have a copy of the software disks, and some photos of the ISA card?
    edit: or scans of the C manual?

    If you can't do mastodon, my email address is on philpem.me.uk/contact .

    Boosts appreciated, feel free to share this post with non-Fedi friends!

    #digipres #transputer #inmos #electronics #retrocomputing

  27. Putting a Teensy to Task as a Transputer Link - One downside of working with the old Inmos Transputer devices is the rarity and co... - hackaday.com/2025/10/20/puttin #retrocomputing #transputer #teensy #inmos #c011

  28. Well it looks like I might have somehow killed all 4 of my #Transputer TRAMs :(

    I've no idea what has happened, but I'm pausing this for a few weeks whilst I get my head straight.

    #gutted

  29. Well it looks like I might have somehow killed all 4 of my #Transputer TRAMs :(

    I've no idea what has happened, but I'm pausing this for a few weeks whilst I get my head straight.

    #gutted

  30. Well it looks like I might have somehow killed all 4 of my #Transputer TRAMs :(

    I've no idea what has happened, but I'm pausing this for a few weeks whilst I get my head straight.

    #gutted

  31. Well it looks like I might have somehow killed all 4 of my #Transputer TRAMs :(

    I've no idea what has happened, but I'm pausing this for a few weeks whilst I get my head straight.

    #gutted

  32. Well it looks like I might have somehow killed all 4 of my #Transputer TRAMs :(

    I've no idea what has happened, but I'm pausing this for a few weeks whilst I get my head straight.

    #gutted

  33. Having failed to get my first version of the C011 #PC104 #transputer adapter to work, I've gone for a simpler approach.
    This has just the Inmos C011, 5MHz oscillator, decoding logic (CPLD/74*245) and of course some LEDs.
    I've added more this time which I hope will give me of an idea when something is working or not.
    I've also brought in all 20 bits of the ISA address bus to the CPLD (not that I think I'll need it but just in case).
    I've also used DIP chips in case I need change the speed.

  34. Wohooo!!!
    My #Transputer #pc104 adapter lives!!!
    Turns out the CUPL code has two pins inverted (active low) that should not be inverted.

  35. Same sequence of events on my pc104 PC controller by QBasic, I don’t get a response from the TRAM for first read :(
    So something isn’t right with the PC timing or the CPLD.
    #transputer #pc104

  36. #PC104 #Transputer woes...
    It seems I can write a value to the TRAM (a T425 based one) but when I try to read it back I get rubbish :(

    This is my writing 0xAA55AA55 to address 0x8000FF00 (this is a valid address)

    but when I read it back, I just get 0x10101010

  37. And finally the dual #Transputer TRAM #PC104 board is done

    Going to let them sit for a couple of days in my head, and then order PCBs.