• kitering@lemmy.world
    link
    fedilink
    English
    arrow-up
    6
    ·
    23 days ago

    I view it as more of an improvement on F-RAM. The price will determine if it is used more than flash or other NVM technologies. Depending on price, it could be attractive for embedded systems where power loss is likely and/or recovery from power loss is very important.

    • jqubed@lemmy.world
      link
      fedilink
      English
      arrow-up
      5
      ·
      23 days ago

      Like DRAM, FeRAM’s read process is destructive, necessitating a write-after-read architecture.

      So that’s why it’s still called RAM? It can hold the data a long time but the data is lost when it’s read?

      • DaPorkchop_@lemmy.ml
        link
        fedilink
        English
        arrow-up
        5
        ·
        23 days ago

        RAM just means the access latency is more or less the same regardless of which particular bit is being addressed, unlike e.g. a spinning rust drive where you have to wait for the platter to rotate into position under the drive head. EEPROM is also RAM - it’s memory and you can read any particular bit in constant time.

      • UnfortunateShort@lemmy.world
        link
        fedilink
        English
        arrow-up
        4
        ·
        23 days ago

        RAM = accessible anywhere (very simply put) D = Dynamic = loses data if power is gone

        This is NVRAM tech, NV standing for ‘non volatile’. That means it keeps the data if power is lost. You only need to re-write if you read that data. There is probably a hardwired option to do this immediately, perhaps even capacitors to ensure this goes through even if you happen to lose power that very moment.

        Consumes extra energy, may add some latency to reads, but there’s more to being fast and non-volatile at the same time. May be wort it, may be not.