Stories
Slash Boxes
Comments

SoylentNews is people

posted by janrinok on Friday August 17 2018, @09:10PM   Printer-friendly
from the doing-more-with-less dept.

Arm Unveils Client CPU Performance Roadmap Through 2020 - Taking Intel Head On

Today's roadmap now publicly discloses the codenames of the next two generations of CPU cores following the A76 – Deimos and Hercules. Both future cores are based on the new A76 micro-architecture and will introduce respective evolutionary refinements and incremental updates for the Austin cores.

The A76 being a 2018 product – and we should be hearing more on the first commercial devices on 7nm towards the end of the year and coming months, Deimos is its 2019 successor aiming at more wide-spread 7nm adoption. Hercules is said to be the next iteration of the microarchitecture for 2020 products and the first 5nm implementations. This is as far as Arm is willing to project in the future for today's disclosure, as the Sophia team is working on the next big microarchitecture push, which I suspect will be the successor to Hercules in 2021.

Part of today's announcement is Arm's reiteration of the performance and power goals of the A76 against competing platforms from Intel. The measurement metric today was the performance of a SPECint2006 Speed run under Linux while complied under GCC7. The power metrics represent the whole SoC "TDP", meaning CPU, interconnect and memory controllers – essentially the active platform power much in a similar way we've been representing smartphone mobile power in recent mobile deep-dive articles.

Here a Cortex A76 based system running at up to 3GHz is said to match the single-thread performance of an Intel Core i5-7300U running at its maximum 3.5GHz turbo operating speed, all while doing it within a TDP of less than 5W, versus "15W" for the Intel system. I'm not too happy with the power presentation done here by Arm as we kind of have an apples-and-oranges comparison; the Arm estimates here are meant to represent actual power consumption under the single-threaded SPEC workload while the Intel figures are the official TDP figures of the SKU – which obviously don't directly apply to this scenario.

Also at TechCrunch.

See also: Arm Maps Out Attack on Intel Core i5
ARM's First Client PC Roadmap Makes Bold Claims, Doesn't Back Them Up
ARM says its next processors will outperform Intel laptop chips

Related: ARM Based Laptop DIY Kit Ready to Hit the Shops
First ARM Snapdragon-Based Windows 10 S Systems Announced
Laptop and Phone Convergence at CES
Snapdragon 1000 ARM SoC Could Compete With Low-Power Intel Chips in Laptops


Original Submission

 
This discussion has been archived. No new comments can be posted.
Display Options Threshold/Breakthrough Mark All as Read Mark All as Unread
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
(1)
  • (Score: 2) by requerdanos on Friday August 17 2018, @10:16PM (8 children)

    by requerdanos (5997) Subscriber Badge on Friday August 17 2018, @10:16PM (#722966) Journal

    ARM Aims to Match Intel 15-Watt Laptop CPU Performance
    posted by janrinok on Friday August 17, @05:10PM
    from the doing-less-with-less dept.

    I have made a minor correction to the above quoted text.

    Affordable ARM chips/systems are incredibly slow (but very efficient) compared to Intel or AMD offerings.

    Fast ARM chips/systems are merely slow compared to Intel or AMD offerings.

    If you say something like "Let's set an arbitrary 15-Watt limit" then the ARM chips look good in the resulting benchmark tables, but we don't say things like that at my house. We say "hold on tight, and watch out 'cause the lights kind of dim when we fire this thing up" before we do things on our computers.

    On the plus side, ARM chips are not known to have built-in rootkits controlled by not-you, unlike AMD and Intel chips, and that kind of anti-feature can be a deal-breaker, whereas having to say "pretty fast considering how slow they are" is kind of pathetic but not a deal-breaker for more use cases.

    • (Score: 0) by Anonymous Coward on Friday August 17 2018, @10:21PM (2 children)

      by Anonymous Coward on Friday August 17 2018, @10:21PM (#722968)

      Do they not have a management engine equivalent?

      • (Score: 4, Interesting) by requerdanos on Friday August 17 2018, @10:32PM (1 child)

        by requerdanos (5997) Subscriber Badge on Friday August 17 2018, @10:32PM (#722976) Journal

        Do they not have a management engine equivalent?

        While a vendor could produce a total system based on an ARM design that had such a thing, nothing like that is part of the design. You would have to graft one on at the hardware design phase for your specific use case if you wanted one, which I submit is the appropriate level of effort for having a permanent rootkit built into your CPU/SOC.

        There's some areas of grey, also. The Raspberry Pi, for example, requires being fed a nonfree blob in order to so much as boot. That blob probably is not, but might be, a supervisory spy engine and rootkit. I have read of an effort to make a free software replacement for that blob, but don't know how that's going because such a concept for me was an obvious nonstarter and I chose not to play.

        • (Score: 1, Informative) by Anonymous Coward on Friday August 17 2018, @11:24PM

          by Anonymous Coward on Friday August 17 2018, @11:24PM (#722993)

          I have read of an effort to make a free software replacement for that blob, but don't know how that's going

          Seems it isn't [github.com] and Broadcom chips should be avoided. Is TrustZone the arm version of the TPM?

    • (Score: 2) by takyon on Friday August 17 2018, @10:29PM (4 children)

      by takyon (881) <takyonNO@SPAMsoylentnews.org> on Friday August 17 2018, @10:29PM (#722972) Journal

      If you say something like "Let's set an arbitrary 15-Watt limit"

      It's not arbitrary. The "U-series" 15W TDP laptop CPUs are a particular segment of the market. Pretty low power consumption, but better performance than fanless ~4-8 Watt designs.

      ARM wants more powerful ARM chips in laptops. There are already fanless ARM Chromebooks and the like. The 15W i5-XXXXU chips are a bit higher on the ladder. Then above that you have your 28W-35W laptop CPUs, maybe 45W, and so on until you start putting desktop Ryzen or Xeon chips into "mobile workstations".

      As for what a (still relatively low performance) ARM laptop can bring to the table, you can find some info about that in the Related stories. I doubt you'll be impressed (4G LTE connectivity can sub in for your missing backdoors).

      --
      [SIG] 10/28/2017: Soylent Upgrade v14 [soylentnews.org]
      • (Score: 2) by requerdanos on Friday August 17 2018, @10:44PM (3 children)

        by requerdanos (5997) Subscriber Badge on Friday August 17 2018, @10:44PM (#722980) Journal

        It's not arbitrary. [citations of chips using 4, 8, 15, 28-35, 45, and up watts].

        If I am going to gather a bunch of chips and say "let's see how fast these are and how fast per watt they are", that's reasonable.
        If I further say "and to mix it up we are going to set a totally not arbitrary limit where no chip may be fed more than 15 watts", then that's less reasonable.
        It's the difference between "Best (something) in class" versus "Best (something)".
        I realize that ARM is setting the particular target of matching a certain class of chip, but benching well against a severely limited selection of chips != benching well.

        As for what a (still relatively low performance) ARM laptop can bring to the table...I doubt you'll be impressed

        I suspect that your doubt is misplaced. Their freedom-friendliness is more important for many use cases than the raw speed of the spy-rootkit chips from Intel and AMD. I have three ARM-based systems running at my desktop (slow ones), and I am evaluating the purchase of a Pine64 RockPro64 for primary desktop use. I find the 4G convenient moreso than a spy feature, but I agree that it could serve the purpose.

        I have an old Thinkpad for my laptop (not old enough to be secure) but it will probably be replaced by some ARM laptop when it finally gives up the ghost.

        • (Score: 0) by Anonymous Coward on Friday August 17 2018, @11:00PM (1 child)

          by Anonymous Coward on Friday August 17 2018, @11:00PM (#722988)

          The rk3399-based boards are definitely attractive, but I'm going to wait until I see what odroid is coming out with before I make a decision to purchase one or not (they were originally planning a rk3399 board for release this month but it was canceled due to issues sourcing memory, but have said they will announce something new in September IIRC).

          • (Score: 0) by Anonymous Coward on Saturday August 18 2018, @08:10PM

            by Anonymous Coward on Saturday August 18 2018, @08:10PM (#723169)

            hmm, a dual Gb eth ports with atheros wifi chip capable of AP mode would be cool. add a dedicated sata for nas use? even better.

        • (Score: 2) by shortscreen on Saturday August 18 2018, @08:43AM

          by shortscreen (2252) on Saturday August 18 2018, @08:43AM (#723064) Journal

          TFS says the 5W ARM chip matches the single thread performance of a 3.5GHz Intel. The fastest Intel chips, which have 10x or more power consumption, are up to what now? 4.8GHz? Not even 1.5x as fast. The speed difference is negligable compared to the difference in heat dissipation, battery life, and reliability for any portable system.

          If ARM's performance really is that close to Intel's then it's silly to call them slow. Back in the old days, before 130nm chips, who would even bother to upgrade for a piddly 1.5x speed up?

  • (Score: 1) by anubi on Saturday August 18 2018, @07:27AM (2 children)

    by anubi (2828) on Saturday August 18 2018, @07:27AM (#723059) Journal

    After studying the intent behind the Parallax Propeller, I am surprised the big boys by now aren't using machines with at least a thousand cores.

    Each one running a single process. No multitasking. Context switching eats up a lot of time. Spawn off another process? NEW another core. Instantiate and delete processes very similar to memory management in C++.

    All this task switching consumes time. The optimization is in getting as many cores as possible running in parallel.

    I think Chip Gracey of Parallax has thunk up one helluva architecture in his Propeller chip. I find it to be a great programmable I/O processors... I can program them to whatever protocol I want...serial, I2C, SPI, ModBus, DMX, TCP, whatever! Even serial VGA ready to send to a monitor. Then change it later if I need to.

    I think I see the potential in his architecture, even though I do not understand the nitty-gritty of it yet.

    --
    "Prove all things; hold fast that which is good." [KJV: I Thessalonians 5:21]
    • (Score: 1, Insightful) by Anonymous Coward on Saturday August 18 2018, @09:39AM (1 child)

      by Anonymous Coward on Saturday August 18 2018, @09:39AM (#723075)

      All this task switching consumes time. The optimization is in getting as many cores as possible running in parallel.

      Memory/IO become quickly the bottleneck, no?

      • (Score: 1) by anubi on Saturday August 18 2018, @10:58AM

        by anubi (2828) on Saturday August 18 2018, @10:58AM (#723090) Journal

        Give each core some local memory... and some multiport shared. The multiport is for I/O, while the program and local variables are in the local. A harvard-like architecture.

        ( I am obviously not a professional chip designer, but when I saw Chip Gracey's design of the Propeller chip, I was pretty impressed. I'd love to know more about that chip, but its not lack of information, actually I have a rather nice book on the chip published by McGraw-Hill from Parallax, but its sheer lack of time for me to sit down with a few chips and code up a few thingies. I would really like is to understand how to change the VGA driver to take I2C instead of serial TX/RX, so I can put it on the my I2C line along with all my other interfaces. I'll just pick an unused address and use that. And write to it in a similar manner as I presently write to LCD displays. Then use the other cogs to emulate yet more UARTS so to make talking with other things that insist on RXD/TXD assigned to yet more I2C address. I envision one emulating 4 UARTS, with one pre-assigned for VGA duty, answering to four consecutive I2C addresses. )

        --
        "Prove all things; hold fast that which is good." [KJV: I Thessalonians 5:21]
(1)