The Powder Toy

Subframe TC blocks by LBPHacker

LBPHacker
LBPHacker
162 / 3
12th Jun 2020
12th Jun 2020
Tired of all the subframe computers which require external tools to program? Have an algorithm that you'd really like to implement in TPT? Try the Subframe Turing-complete blocks! Heavily inspired by lesio's save, see the link inside.
programming subframe stcb

Comments

  • TheNik
    TheNik     13th Jun 2020
    Not to mention persistent memory across different blocks. Unfortunately I can not spend nearly enough time to truly appreciate your save and thus I don't even have a basic understanding of this system. But kudos for implementing in mere days that, which I though to be impossible at all - and all in subframe.
  • TheNik
    TheNik     13th Jun 2020
    When I first saw the save by lesio and commented that it would need "more than just memory" for turing completeness, I was absolutely certain that any attempt at such a programming language would fail due to conditional jumps to arbitrary adresses (how did you implement that? How are the block adressed?).
  • LBPHacker
    LBPHacker     13th Jun 2020
    Thanks! Cool for you. Sadly, the one thing my degree taught (is teaching) me is that the goal of the academic environment this part of the world isn't to involve you in interesting things but to get you a degree :P
  • Schmolendevice
    Schmolendevice     13th Jun 2020
    @LBPHacker Funny enough I'm also into computer graphics and compilers, but only got to take a core course for the latter. One thing my degree taught me is I should have gotten an FPGA instead of 74HC DIP chips all those years ago. Anyways, great work with this.
  • LBPHacker
    LBPHacker     13th Jun 2020
    Tl;dr relatively modern v95 subframe black magic :P
  • LBPHacker
    LBPHacker     13th Jun 2020
    It deserialises the two addends multiplied by 4 into the temperature of two HSWC particles, then then 0 is deserialised into the temp of two other HSWC. This is followed by exactly one of the HSWC equalising the temperature between all four of them, thus setting the temp of all of them to the sum, then TSNS is used to read this sum. Except all this is done in a 2-adder carry-select configuration because the adders are 16-bit and temp only has a precision of 13 bits, of which I only use 11 or so.
  • RaconTPT
    RaconTPT     13th Jun 2020
    I don't get how your adder works
  • BomberPlayz
    BomberPlayz     13th Jun 2020
    really good +1!
  • East_Europe3339
    East_Europe3339     13th Jun 2020
    this is madness +1
  • docRoboRobert
    docRoboRobert     13th Jun 2020
    Fantastic. 1+