• circuscritic
    link
    fedilink
    English
    arrow-up
    3
    ·
    edit-2
    8 months ago

    Let’s try a little thought experiment, or practical if you have the time and money:

    First step:

    Go buy a Commodore 64 and an RTX-4800

    Next step:

    Make them work together

    Last step:

    Find a Soviet 70s/80s era computer (probably a mainframe style) and network it with your Commodore 64 that has an RTX-4800.

    You’re suggestion boils down to:

    Science says it’s possible, why don’t they just redesign the entire missle?

    But, I am just spitballing here. If you are that convinced that this is doable, or even has been done, go do some research.

    • Aceticon@lemmy.world
      link
      fedilink
      English
      arrow-up
      1
      arrow-down
      3
      ·
      8 months ago

      Your example in highly artificial and uses constraints that are not applicable here: an RTX-4800 interfaces with a computer via a complex protocol called PCI-Express (now in version 4) which is very complex because it had to be designed to pass a lot of data really fast (Gigabits per second) as that was required for the main use case of that board: rendering millions of vertices worth of textured 3D models into a 2D output at least 25 times per seconds (though many people nowadays expect 100 times per second or more).

      Something that provides a direction in a 3D space, on the other hand, has to send all of 12 bytes a couple of times per second (if you’re using a double-precision floating point value for each coordinate), so the design of the interface hasn’t been constrained by the need for ultra-high throughtput and can be very simple.

      As it so happens, the second part of your example (getting a Soviet 70s/80s era computer to talk with a Commodor 64) is quite likely possible using the Serial protocol.

      In fact a more correct version of that example would be to get a Soviet 70s/80s computer to talk to a modern microcontroller - because for any system designed now to interface with older systems one of the sides would always be modern - and that absolutelly is possible as long as that computer has any kind of comms protocol, as even if it doesn’t have Serial (which would be strange, but possible) you can always program a modern microcontroller to bit-bang any protocol that’s not high throughput. (In fact there are tons of projects out there were people interface modern microcontrollers with ancient hardware).

      Also we’re talking about something that already receives GSM data from a GSM module, and my experience of working with that kind of electronics (civilian) is that it’s simple digital data via something like Serial.

      I agree that are for sure be constraints I am entirelly unware of here (as I have no experience with designing military hardware), but your idea of were they are is, from my own experience desiging electronic systems, not at all in the right area.

      • circuscritic
        link
        fedilink
        English
        arrow-up
        3
        arrow-down
        1
        ·
        edit-2
        8 months ago

        Are you just using ChatGPT, or another LLM, to string words and concepts together?

        Kinda sounds like you are, at least for significant sections of your responses.

        Some parts are just confusing as to why even include them, unless you actually have no idea what they mean, or why they’re not contextually relevant, for example:

        GSM? Nope, not a factor…

        PCIE - why specify v4?

        But, whatever, I’m not a Raytheon engineer. Doesn’t matter if you convice me (you haven’t, but doesn’t matter).

        Edit: also, FWIW, Those constraints aren’t artificial. They’re each specific to what you’re suggesting.

        Old American tech, retrofitted and repurposed with brand new tech, and then integrated with, and fired from, old Soviet tech.

        There is a reason that NATO was only able to integrate fire control to the HARM for the SU’s, and not all of its cockpit control features and capabilities. Which is why they have to preprogrammed for each sortie…

        • Aceticon@lemmy.world
          link
          fedilink
          English
          arrow-up
          2
          arrow-down
          3
          ·
          8 months ago

          It makes sense that somebody who thinks radar “radiation” is different from radiowaves can’t follow point-by-point electronics engineering explanations and design engineering considerations so thinks they must be the product of ChatGPT.

          Should’ve dropped this conversation after the comment where you explained radar “radiation” as special.

          Thanks for the lesson.

          • circuscritic
            link
            fedilink
            English
            arrow-up
            3
            arrow-down
            1
            ·
            edit-2
            8 months ago

            Not special, just a lot more power being fed through it vs other battlefield RF.

            I followed, but it was just dumb. Way too many words to explain, or justify, a functionally unsound idea: converting HARM into a new missle, with a different purpose (Counter-EW vs SEAD).

            But okay, you’re cleary an ubermensch, go to Raytheon, or start a new company. I’m sure you’ll get boatloads of investors.