🍵 There have been discussions about monero-pow randomX improvements for newer processor versions. One idea came to my mind: How about doing a check if a tee (trusted execution environment) is available?

Those could provide shortcuts to make CPU’s more efficient compared to that CPU’s that are currently used by ASIC-producers?

AMD:

  • Platform Security Processor (PSP)
  • AMD Secure Encrypted Virtualization and the
  • Secure Nested Paging extension

ARM:

  • TrustZone
  • Realm Management Extension / Confidential Compute Architecture (CCA)

IBM:

  • IBM Secure Service Container, formerly zACI, first introduced in IBM z13 generation machines (including all LinuxONE machines) in driver level 27.
  • IBM Secure Execution, introduced in IBM z15 and - LinuxONE III generation machines on April 14, 2020.

Intel:

  • Trusted Execution Technology
  • SGX Software Guard Extensions
  • “Silent Lake” (available on Atom processors)

RISC-V:

  • MultiZone™ Security Trusted Execution Environment
  • Keystone Customizable TEE Framework
  • Penglai Scalable TEE for RISC-V
    • rideOP
      link
      fedilink
      arrow-up
      1
      arrow-down
      2
      ·
      10 months ago

      Well, but just the proofable fact, that it’s there, could put those chips in a different position. You probably wouldn’t need to use it for anything sensitive.

      • original_ish_name@lemm.ee
        link
        fedilink
        arrow-up
        4
        arrow-down
        1
        ·
        10 months ago

        You probably wouldn’t need to use it for anything sensitive.

        Are you encouraging ASICs? How is my main pc not sensetive

  • itsmect
    link
    fedilink
    arrow-up
    4
    ·
    10 months ago

    The newly announced “ASIC” is neither an ASIC nor more efficient then modern ryzen processors, making the news a nothingburger. If anything it is impressive that even with semi-custom chips they can’t outperform regular CPUs, and a testament that RandomX works as intended.

    • rideOP
      link
      fedilink
      arrow-up
      1
      arrow-down
      2
      ·
      10 months ago

      Thanks for bringing this statement, but it’s not nothing if they find buyers for their old miners. They have been mining two years with it and at the beginning time they did better than usual CPU’s on the market. In my opinion they should be discouraged to do further mining hardware attempts by making the power of their devices noncompetitive.

  • magnetosphere @beehaw.org
    link
    fedilink
    English
    arrow-up
    2
    ·
    10 months ago

    This picture shows up in my feed after several posts from botart (where a bot posts AI generated images).

    At first, I wasn’t sure if this photo was real or not.

    • rideOP
      link
      fedilink
      arrow-up
      1
      arrow-down
      1
      ·
      10 months ago

      That’s an illustration, i didn’t expect bot or ai questions about it.

    • rideOP
      link
      fedilink
      arrow-up
      1
      ·
      10 months ago

      This initial post “TEE & randomX POW” is about a PoW Algo that potentially uses parts/capabilities/characteristics of CPU’s.