Announcement

Collapse
No announcement yet.

ARM Announces Cortex-A76 Processor, Mali-G76 & Mali-V76

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • ARM Announces Cortex-A76 Processor, Mali-G76 & Mali-V76

    Phoronix: ARM Announces Cortex-A76 Processor, Mali-G76 & Mali-V76

    ARM has announced their next-generation wares...

    Phoronix, Linux Hardware Reviews, Linux hardware benchmarks, Linux server benchmarks, Linux benchmarking, Desktop Linux, Linux performance, Open Source graphics, Linux How To, Ubuntu benchmarks, Ubuntu hardware, Phoronix Test Suite

  • #2
    Yet another core that's vulnerable to Spectre?

    Comment


    • #3
      Now I would like to see some SoC using this. I have been waiting for years to replace my power hungry x86-x64 equipment with a low power consumption arm board... I have been looking for low power x64 cpus (because of steam) but non convinces me because they perform really poor...

      Comment


      • #4
        Originally posted by numacross View Post
        Yet another core that's vulnerable to Spectre?
        Well, of course. There vulnerabilities to come, thus don't expect to find a high-performance silicon which is secure from this.

        BUT, they did some some of issues: https://developer.arm.com/support/ar...-vulnerability

        Variant 2 and 3 are solved. Variant 1 and Variant 4 (just announced) are still present.

        Comment


        • #5
          So we'll probably see this in budget smartphones in about 3-5 years right? Looking forward to it.

          Comment


          • #6
            Strange that they do not even mention any Spectre mitigation attempts. One would think that every customer wants to know if a new core is still susceptible to these dire security flaws.

            Comment


            • #7
              Originally posted by numacross View Post
              Yet another core that's vulnerable to Spectre?
              From the press release:
              • Decoupled branch prediction and instruction fetch: Built to hide latency at high bandwidth, the in-order Cortex-A76 front-end is able to fetch 4 to 8 instructions per cycle, using multi-level branch target caches and hybrid indirect predictor to sustain the maximum throughput.

              Originally posted by TheOne View Post
              Now I would like to see some SoC using this. I have been waiting for years to replace my power hungry x86-x64 equipment with a low power consumption arm board... I have been looking for low power x64 cpus (because of steam) but non convinces me because they perform really poor...
              This isn't gaming. This is office productivity. And at the low, casual users and students end. Future wise, any foray into the gaming desktop / workstation performance space will go through a console manufacturer like Sony, Nintendo or Microsoft. Once the major engines are ported, more open PC-like designs might be an option. But that's 2-4 years from now...

              Comment


              • #8
                "7nm"

                Comment


                • #9
                  Originally posted by TheOne View Post
                  Now I would like to see some SoC using this. I have been waiting for years to replace my power hungry x86-x64 equipment with a low power consumption arm board... I have been looking for low power x64 cpus (because of steam) but non convinces me because they perform really poor...
                  Yeah, I've been searching for a Linux compatible ARM-based laptop for years. But, most of them are poor quality, have useless GPU drivers, or lack too many features (that I know other ARM devices support). I really want the Asus NovaGo, but it isn't clear if Linux can run on it, let alone well.

                  Originally posted by quaz0r View Post
                  "7nm"
                  Yeah, it might not actually be 7nm (I hear some companies exaggerate the size based on how the shape and layout of the transistors) but I believe it would be close enough, and achievable. Unlike x86, ARM chips are expected to have a low TDP and low frequency. I get the impression one of the main limitations of modern x86 CPUs getting die shrinks is because they can't maintain their high clock speeds, probably because such speeds require higher voltage, which promotes quantum tunneling. Even if <=10nm nodes are unstable for the demands of x86, that could be largely irrelevant for ARM.

                  Comment


                  • #10
                    Hard to get excited about new Mali graphics when ARM refuses to support open drivers.

                    Comment

                    Working...
                    X