Announcement

Collapse
No announcement yet.

Intel Rolls Out 10nm Pentium/Celeron CPUs, Previews Rocket Lake

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

  • #41
    Originally posted by coder View Post
    They could still split them in half and execute them as 2x 128-bit parts. Didn't AMD do that in Zen1? I know Pentium did that with SSE, in the Pentium 4.

    In terms of area, AVX bloats the register file, but I wonder if that's even enough to bother about.

    Anyway, let's not forget this is a 10-way core with 6-wide decode! So, we're not exactly talking about a microcontroller or IoT core. And its ancestors had SSE/SSE2 going at least as far back as the 22 nm days, so you'd expect at least basic AVX/AVX2-support wouldn't be completely off the table.

    In case you missed it: https://en.wikichip.org/wiki/intel/m...t#Architecture
    Yeah it looks like a great Atom, not yet available but hopefully it should be arriving soon.

    Comment


    • #42
      Originally posted by vladpetric View Post

      I'm curious what out-of-order decode actually means. Do you know?

      I kinda' doubt that they're doing out-of-order fetch.

      Decoding instructions is for the most part stateless (first instruction doesn't affect the decoding of the second instruction, unless you do some fusion at decode), so it doesn't matter too much the order in which you decode them, as long as you rename them in the correct order.
      It's probably just two-cluster decode ... (decoding two clusters of up to 3 instructions each, in parallel)

      Comment


      • #43
        Originally posted by vladpetric View Post
        I'm curious what out-of-order decode actually means. Do you know?
        You can look at Anandtech's writeup: https://www.anandtech.com/show/15009...e-tremont-core

        Originally posted by vladpetric View Post
        Decoding instructions is for the most part stateless (first instruction doesn't affect the decoding of the second instruction,
        It seems to me that you'd need to decode enough of the first instruction to know how long it is, so you know where to start decoding the one after it, and so on.

        They don't directly address the out-of-order part, but maybe it refers to the fact that one of the engines can concurrently decode the other path of a predicted branch.

        Comment


        • #44
          I'm very interested in the 10nm Pentiums and Celerons.

          Hopefully the Chinese will buy up large numbers of them to put in low-cost laptops like the three Chinese laptops I currently own. I may be wrong, but it seems like the small Chinese OEMs love churning out cheap laptops with Atoms and Celerons, which happens to be just up my alley.

          My existing Chinese laptops with X7-E3950 and N4100 processors have been able to handle what I need them to do so far, and I would expect no less of the Pentium N6000 and Celeron N5100.

          Comment


          • #45
            Originally posted by coder View Post
            You can look at Anandtech's writeup: https://www.anandtech.com/show/15009...e-tremont-core


            It seems to me that you'd need to decode enough of the first instruction to know how long it is, so you know where to start decoding the one after it, and so on.

            They don't directly address the out-of-order part, but maybe it refers to the fact that one of the engines can concurrently decode the other path of a predicted branch.
            Right, decoding two basic blocks makes sense here. And the branch predictor predicts addresses anyway.

            They don't use a trace cache, so this makes sense.

            Comment


            • #46
              Originally posted by vladpetric View Post
              Yeah it looks like a great Atom,
              I just wonder what the discussion was like that lead to Intel's decision to be so ambitious. It's really quite a lot wider than its predacessor. Until this point, their small cores have evolved very gradually and incrementally.

              Was it the threat of ARM creeping into more of their markets, like Qualcomm's laptop play? Was it the opportunity they saw in 5G base stations?

              Comment


              • #47
                Not sure if you're joking? ... If not, how fast does STEM normally go, and what is "78% faster" equal to? And "78% faster" than what? How long does a similar AMD Ryzen take to perform STEM?

                Comment


                • #48
                  Originally posted by Sonadow View Post
                  I'm very interested in the 10nm Pentiums and Celerons.

                  Hopefully the Chinese will buy up large numbers of them to put in low-cost laptops like the three Chinese laptops I currently own. I may be wrong, but it seems like the small Chinese OEMs love churning out cheap laptops with Atoms and Celerons, which happens to be just up my alley.

                  My existing Chinese laptops with X7-E3950 and N4100 processors have been able to handle what I need them to do so far, and I would expect no less of the Pentium N6000 and Celeron N5100.
                  You buy computing equipment from Chinese OEM's? I'm not quite that brave, my data is valuable to me. Anti-virus won't help when the back doors are baked into the motherboard.

                  Comment


                  • #49
                    Originally posted by torsionbar28 View Post
                    You buy computing equipment from Chinese OEM's? I'm not quite that brave, my data is valuable to me. Anti-virus won't help when the back doors are baked into the motherboard.
                    My main concern is IoT and networking gear. I bought my current wifi router (Netgear) specifically because it's a US brand and uses a Qualcomm SoC. I know, shame on me for not using opensource firmware, but I didn't need yet one more thing to fiddle with.

                    Still, the point is there's a lot of inexpensive Chinese networking gear that gets decent reviews which I simply wouldn't touch.

                    Comment


                    • #50
                      Originally posted by Templar82 View Post

                      I know what STEM stands for... I want to know what exactly is 78% faster.
                      Originally posted by torsionbar28 View Post
                      Not sure if you're joking? ... If not, how fast does STEM normally go, and what is "78% faster" equal to? And "78% faster" than what? How long does a similar AMD Ryzen take to perform STEM?
                      The "78%" number comes from https://www.intel.com/content/www/us...on-brief.html:

                      • Students can experience up to 24% faster collaboratively working on STEM projects3
                      • Teachers and students can experience up to 43% faster web browsing4
                      • Students can collaborate and load coding projects up to 36% faster5
                      • Students can play Rocket League Education with up to 22% more FPS6
                      • Teachers and students can experience up to 78% better graphics performance7
                      • Gigabit Wi-Fi enables faster downloads in highly dense network area like a classroom
                      • Seamless sharing and collaborating among students and classroom’s monitor
                      where footnote 7 means:

                      As measured by 3DMark Fire StrikeE graphics score.
                      This wasn't that hard to find. I am not sure why you weren't able to lookup the information yourselves.

                      Clearly, I do not approve of linking "Teachers and students" to "3DMark Fire Strike graphics score".
                      Last edited by atomsymbol; 12 January 2021, 12:18 AM. Reason: Remove colon from link

                      Comment

                      Working...
                      X