Mur
Mur
55 / 2
13th May 2015
13th May 2015
I FINALLY UNDERSTAND HOW CRAY/ARAY LOGIC WORKS *laughs maniacally*
harddrive idea memory electronic electronics ram0 flash cray aray

Comments

  • EgrOnWire
    EgrOnWire
    14th May 2015
    @Schmolendevice we need directional CONV for that, so that tech will be bigger in comparsion to normal electronics
  • pkmarci
    pkmarci
    14th May 2015
    This is incredible! +1
  • ogsterduck
    ogsterduck
    14th May 2015
    ID:1788266 I made a KiloByte RAM with different technology.
  • Schmolendevice
    Schmolendevice
    14th May 2015
    @Mur: Sigh, well 'twas just some past demonstration saves. I do have some working robust 60 Hz adder technology though. Hopefully things will go better once I have a proper high speed demonstration that prints out 60 16-bit binary operands and their respective sums in one second, hehe. But indeed, ISP and exam season's coming up.
  • Mur
    Mur
    14th May 2015
    @Schmolendevice: well, I just took a look at your stuff, and you've got some hardcore subframe stuff in there, and it deserves more attention, I think...
  • Mur
    Mur
    14th May 2015
    @Schmolendevice: lol, you commented right before I finished my second comment. And yes, I am also aware about subframe mechanics. Well, to finish my thoughts, I think it is more fruitful if people aim at making something small, with a reductivist aproach, and from scratch, because Microsoft's code is awfully bloated, and probably so is Apple's. And a computer running @ 60Hz with 1kB of RAM is quite powerful already...
  • Schmolendevice
    Schmolendevice
    14th May 2015
    I guess it wouldn't be so possible to store a whole C compiler on TPT but once we achieve 60 Hz CPUs and high speed graphic devices we might at least achieve a simple text based operating system that doesn't take too long to load. Programs would take seconds to a minute to execute and I'm trying to design a 120 by 100 pixel graphics device with 50 frame update.
  • Mur
    Mur
    14th May 2015
    @groudon: I know it is possible to make coloured FILT memory that stores a rather large amount of data in a small area, but we still have the speed limit (remember that not everyone has a powerful computer to run TPT @ 60FPS+). Maybe trying to replicate a VIC-20, ZX81, ZX80 would give some success, but it would still be awfully slow, I think.
  • Schmolendevice
    Schmolendevice
    14th May 2015
    Ah, wonderful. By the way, have you ever looked into TPT's FILT mechanics and its computational applications. And indeed megabits of data would be a lot for TPT even with FILT. At most I'd try making a 49 kB FILT device assuming that each pixel stores 3 bytes. I'm not sure if you have seen but I have been working on some 1 frame per operation TPT computing technology. So 60 fps would mean a good 60 operations per second.
  • Mur
    Mur
    14th May 2015
    @groudon: Apple or Microsoft? I think it would be much better if people aimed at making a computer that runs MINIX or some other lightweight *NIX of sorts. But even at 60FPS (yes, I am aware of tpt.fpscap) it would take days to boot a useable shell. And 1000bits is 1kilobit, 1megabit = 1 000 000 bits.