Announcement

Collapse
No announcement yet.

Apple Announces Its New M2 Processor

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

  • #31
    Apple’s SoCs are completely destroying AMD64 mobile chips. Qualcomm needs to hurry and release their Nuvia based SoCs. Unless you need a gaming laptop, require CUDA or absolutely hate Apple, I don’t see a purpose of buying an AMD64 based laptop.

    Comment


    • #32
      Originally posted by benjiro View Post

      https://www.notebookcheck.net/AMD-Ry....623763.0.html

      People still believe that stories or what...
      Asus ZenBook S 13 Ryzen 7 6800U 28W 25.5W 10468 374 / 411
      Apple MacBook Pro 14 M1 Pro (8 Cores) 25W 21W 9581 383 / 456
      Its barely a difference of a few percentage when comparing the power vs performance on MT load. Alder Lake is a power drainer, not AMD.

      Apple their benefits are that they hide a lot of their performance under their ( oa media ) encoding engine for a lot of tasks. When the CPU is put pure on CPU tasks, that famous power efficiency scales very close. Notice how a 6nm vs 5nm are pulling very close the same power in MT tasks.

      Apple gains are mostly in the ST tasks where its reporting better performance on single core tasks. We see 4W * 4 + Efficiency cores (21W) result in 9581 in MT. But why can AMD deliver 10468 on a 25W power budget because X86 when not turbo boosted to hell, is actually very efficient.
      Asus ZenBook S 13 Ryzen 7 6800U 28W 25.5W 10468 374 / 411
      Apple MacBook Pro 14 M1 Pro (8 Cores) 25W 21W 9581 383 / 456
      Asus Zenbook S 13 Flüstermodus Ryzen 7 6800U 12W 10W 6725 560 / 672
      Take a look at the 10W result in MT for AMD resulting in 6725. How is it possible for AMD to use only 10W and still deliver 70% of the performance that took Apple M1 21W?? Its funny these results are they not. And that is on 6nm, a process that is not supposed to deliver a increase in power efficiency, compared to 7nm. Unlike 5nm that gives Apple a 20% gain.

      Its been clear for YEARS that AMD and Intel have been turbo boosting their CPU way too much for ST tasks. So why is AMD so efficient in MT tasks? Because CPU are not designed for laptop first, or desktop first. Its server first. Where you want great MT performance at the best possible power usage ( most server CPUs that use the exact same cores are sold with very conservative clock speeds for that reason ).

      Then those CPUs get filtered down to desktop, where they need to show great benchmark/gaming results, so there goes the clock speed up because that is the most easy way to reuse the same design. That CPU then needs to conform to laptops and well, your just trying to shoehorn server / desktop designed CPUs into laptops. And that becomes harder and harder but when you really analyse the result on a more equal playing field, ARM is not that special.

      We already see how Smartphone are becoming hotboxes because of that same drive for more performance at any cost, despite it also being ARM technology.
      Interesting. I have never ever believed the ARM hype for general-purpose computing in desktop and laptop PCs, and this validates my views perfectly.

      Comment


      • #33
        Originally posted by sinepgib View Post
        That's far from proving most modern software does. Further, that's far from proving most software in a system is actually modern.
        most stuff people do today is in fact "Browser" ans as soon as you use more than one tab at the same time you are in the multicore world already. in fact in a modern world webassembly+webgpu is all you need for nearly all the apps you develop.

        Originally posted by sinepgib View Post
        A lot of stuff written in Python will definitely run a single core. Many of the core utils tend to be single threaded.
        Multithreading is something that scares away a lot of developers, so many tend to write single threaded code instead and cross their fingers that it's enough.
        that python is single core only is a big lie as a python developer you already use many libs and functions that is automatically multicore. your language do not need to be multithreaded as soon as you use functions to call in a lib this lib can be C/C++/Rust and can be multicore. and it is not only the libs if your python app calls a kernel function this function can be multicore by the kernel to.

        and to quote you: "in reality, it doesn't matter because you'll still have multiple processes running which actually scale better on multicore than a single process with multiple threads because they don't need to sync)"

        so this means even if you use a single core python app your system will use many more cores than if you just have 1 core.
        this effect was shown in the time of instead of single core cpus the world see dualcore cpus or singlecore cpu with hyperthreading
        the apps speed up even if only a single thread i used.

        Originally posted by sinepgib View Post
        Not the claim under discussion.
        i just with there would be a overpowered singlecore cpu without hyperthreading on the market then you could compare your theory to my one.

        nearly no one would buy such a "overpowered singlecore cpu" because modern OS spread so much threats that any dualcore cpu

        Originally posted by sinepgib View Post
        But just to give an example of software I use that may be slow (if it has many plugins, mostly) and runs on a single thread, Vim.
        I'm not claiming single core performance is more important BTW (in reality, it doesn't matter because you'll still have multiple processes running which actually scale better on multicore than a single process with multiple threads because they don't need to sync), just that saying most applications do use them individually, well, I'd rather have numbers for such a claim.
        in history we already have the numbers "I'd rather have numbers for such a claim" from a time the industry was going from single core cpus to dualcore cpus...

        it is a reason why the mainstream "playstation5" is on 8core cpus ... even 6core gaming PC become a rarity

        but it is also a fact that these mainstream system do not have 64core cpus ore 32core cpus or even 16 core cpus because most people can not utilize it with today software stack.

        today situation is this: 1-6cores is to little 8-12 cores is the golden and 16/32/64 cores is to much.

        thats the reason the apple M2 has 8 cores.

        you can see this in the 5800X3D it has only 8 cores but it beats the 5900X and 5950X in most tasks

        AM5 with zen4 runs at very high clock speed 5,5ghz and more so the 6800X cpu will be very good.

        and even if you just buy the 6core versions you will get big penalty already for the lag of cores.

        Phantom circuit Sequence Reducer Dyslexia

        Comment


        • #34
          Originally posted by WannaBeOCer View Post
          Apple’s SoCs are completely destroying AMD64 mobile chips. Qualcomm needs to hurry and release their Nuvia based SoCs. Unless you need a gaming laptop, require CUDA or absolutely hate Apple, I don’t see a purpose of buying an AMD64 based laptop.
          AMD has similar multicore performance per watt. apple is only better in single core performance per watt.

          i hate apple but if they would start to sell the M1/M2 with linux i would buy it. but right now you better buy AMD based laptop if you want to use linux.

          Asahi Linux
          is not ready yet
          Phantom circuit Sequence Reducer Dyslexia

          Comment


          • #35
            Originally posted by qarium View Post

            right. the point that apple m1 is faster in singlecore tasks per watt becomes irrelevant to the fact that nearly all modern software can uilize 8 cores or more.

            intel did the same and over displayed single core performance in their advertisement.

            apple m1 is still a good product the only one who is overall really the worst product is intel.

            AMD AM5 +zen4 and apple M2 of course destroy everything intel has.

            and this in all cadegories in single core performance in single core performance per watt and in multicore performance and in multicore performance per watt...
            AMD only had the lead for one year and that was due to TSMC. AMD was facing Intel's architecture from 2016(Skylake) until 11th gen was released. Now that they started to focus on CPUs again they've just been hammering out IPC improvements. First with Sunny Cove, 18% IPC over Skylake, followed by Golden Cove with a 19% IPC increase over Sunny Cove. 13th Gen will use Raptor Cove cores which is another 15% IPC increase which will release about the same time as Zen 4. Then Intel will be releasing their next generation hardware with Meteor Lake, and by the name I can guess what it will be doing to AMD.

            Comment


            • #36
              Originally posted by WannaBeOCer View Post
              AMD only had the lead for one year and that was due to TSMC. AMD was facing Intel's architecture from 2016(Skylake) until 11th gen was released. Now that they started to focus on CPUs again they've just been hammering out IPC improvements. First with Sunny Cove, 18% IPC over Skylake, followed by Golden Cove with a 19% IPC increase over Sunny Cove. 13th Gen will use Raptor Cove cores which is another 15% IPC increase which will release about the same time as Zen 4. Then Intel will be releasing their next generation hardware with Meteor Lake, and by the name I can guess what it will be doing to AMD.
              as soon as you count in the price and power consumsion AMD is nearly always the better option :

              https://cpu.userbenchmark.com/Compar...7839vsm1821519

              even the intel loving userbenchmark only claims the 12900KS is 17% faster...

              in germany the 5800X3D is 499€ and the 12900KS is 749€

              https://geizhals.de/intel-core-i9-12...-a2702287.html
              https://geizhals.de/amd-ryzen-7-5800...html#offerlist

              the AMD TDP is 141watt the intel TDP is 273watt

              so to cover your claims intel is using the socket TDP loophole AM4 is 141 watt intels sockel is 273watt
              without this loophole intel would lose.

              AM5 mainboards fix that they also have 170+watt TDP

              also the price of your system a mainboard for the 12900KS is very expensive compared to this AM4 mainboards are cheap.

              your so called "IPC increase" claims is only because of the higher TDP..

              AM5 systems also have higher TDP so intel has a problem again.
              Phantom circuit Sequence Reducer Dyslexia

              Comment


              • #37
                Originally posted by qarium View Post

                as soon as you count in the price and power consumsion AMD is nearly always the better option :

                https://cpu.userbenchmark.com/Compar...7839vsm1821519

                even the intel loving userbenchmark only claims the 12900KS is 17% faster...

                in germany the 5800X3D is 499€ and the 12900KS is 749€

                https://geizhals.de/intel-core-i9-12...-a2702287.html
                https://geizhals.de/amd-ryzen-7-5800...html#offerlist

                the AMD TDP is 141watt the intel TDP is 273watt

                so to cover your claims intel is using the socket TDP loophole AM4 is 141 watt intels sockel is 273watt
                without this loophole intel would lose.

                AM5 mainboards fix that they also have 170+watt TDP

                also the price of your system a mainboard for the 12900KS is very expensive compared to this AM4 mainboards are cheap.

                your so called "IPC increase" claims is only because of the higher TDP..

                AM5 systems also have higher TDP so intel has a problem again.
                You clearly don't know the definition of IPC. You should read some actual reviews. As you can see from AnandTech's review in real world workloads we still see the performance gains of the IPC and Intel's chips don't tend to go that high regarding power usage.

                https://www.anandtech.com/show/17047...-complexity/16

                Power is an interesting topic, and although our peak power numbers when all cores were loaded were above the 241W Turbo power on the box, in real world workloads it didn’t tend to go that high. The P-cores alone on the chip matched the power consumption of Intel’s 11th Generation in AVX2 workloads, but adding in the E-cores does put it over the previous generation. I’m not entirely sure what that says about Intel’s 7 manufacturing process compared to the 10SF used before. A lot of the performance gains here appear to come from IPC and DDR5, and that doesn’t seem to have come with performance per watt gains on the P-cores. It means that Intel is still losing on power efficiency at load compared to the competition.

                Comment


                • #38
                  Originally posted by qarium View Post
                  most stuff people do today is in fact "Browser" ans as soon as you use more than one tab at the same time you are in the multicore world already. in fact in a modern world webassembly+webgpu is all you need for nearly all the apps you develop.
                  Fair enough.

                  Originally posted by qarium View Post
                  that python is single core only is a big lie as a python developer you already use many libs and functions that is automatically multicore. your language do not need to be multithreaded as soon as you use functions to call in a lib this lib can be C/C++/Rust and can be multicore. and it is not only the libs if your python app calls a kernel function this function can be multicore by the kernel to.
                  Pure Python is single threaded. I didn't check the source code to see if any of the C modules in the standard library does release the GIL, but my guess is also no. Stuff like NumPy and what not is of course parallel because those make a point of releasing the GIL.
                  Kernel calls are not multicore from the POV of Python: regardless of if it blocks or is non-blocking, it's a different thread that will take its turn to run, but when it does it will take the GIL. Essentially, pure Python can only parallelize IO.

                  Originally posted by qarium View Post
                  and to quote you: "in reality, it doesn't matter because you'll still have multiple processes running which actually scale better on multicore than a single process with multiple threads because they don't need to sync)"

                  so this means even if you use a single core python app your system will use many more cores than if you just have 1 core.
                  this effect was shown in the time of instead of single core cpus the world see dualcore cpus or singlecore cpu with hyperthreading
                  the apps speed up even if only a single thread i used.
                  Indeed. But the system is not an application, is the whole system. I'm discussing whether most single applications do use multicore properly, which was what you said.

                  Originally posted by qarium View Post
                  i just with there would be a overpowered singlecore cpu without hyperthreading on the market then you could compare your theory to my one.

                  nearly no one would buy such a "overpowered singlecore cpu" because modern OS spread so much threats that any dualcore cpu
                  Sorry, but what do you suppose my theory is, exactly? My only "theory" is that to claim most apps make proper use of all cores is something that needs a lot of data to backup. I never contradicted, and in fact I asserted, that multicore performance is by today's standards absolutely more important than single core performance. Single core performance would only matter if your "system" was actually running a single process at a time, a-la DOS.
                  In fact, if any criticism to the world is to be emitted from what I said, is mostly that I don't think everyone does a good job of parallelization, certainly not that we should go back to optimize for the single core and single process past.

                  I'm not quoting the rest of the post because you're clearly not reading what I'm saying anyway. The numbers I ask for are not performance numbers, but statistics about multicore applications and how they scale. It was never an argument against whether or not it was better to optimize hardware for single or multiple concurrent tasks, as I already stated it's obvious the second case is what's desirable in multi-tasking systems (i.e. everything but the smallest embedded).

                  Comment


                  • #39
                    Originally posted by schmidtbag View Post
                    Apple gains a lot of performance-per-watt specifically because they lock down their platform. Everything is finely tuned and tightly integrated, which makes their design more efficient. They don't allow a bunch of BS 3rd party bloatware. Their compiler takes advantage of all the instructions of the CPU. They don't have to compensate for a bunch of 3rd party APIs which may add overhead. On a RISC architecture, these differences add up quite significantly.

                    What Intel has done with Clear Linux is a good example of how much more performance you can squeeze out of a CPU without disproportionately cranking up the wattage (whereas overclocking or adding more cores will).
                    This.

                    Comment


                    • #40
                      Originally posted by qarium View Post

                      AMD has similar multicore performance per watt. apple is only better in single core performance per watt.

                      i hate apple but if they would start to sell the M1/M2 with linux i would buy it. but right now you better buy AMD based laptop if you want to use linux.

                      Asahi Linux
                      is not ready yet
                      Yeah this is the bottom line. The M series is nice in theory, and only great in practice if you can handle being stuck on OSX.

                      Sometimes thats fine, but sometimes its not.

                      Maybe it will work on linux (and by extension Windows through WINE) some day, but it does not work now, it will not work tomorrow, and it probably wont be full speed for a long time.

                      Comment

                      Working...
                      X