Stories
Slash Boxes
Comments

SoylentNews is people

posted by martyb on Tuesday May 26 2020, @09:23PM   Printer-friendly
from the cortex-designs-to-think-about dept.

Arm's New Cortex-A78 and Cortex-X1 Microarchitectures: An Efficiency and Performance Divergence

Today for Arm's 2020 TechDay announcements, the company is not just releasing a single new CPU microarchitecture, but two. The long-expected Cortex-A78 is indeed finally making an appearance, but Arm is also introducing its new Cortex-X1 CPU as the company's new flagship performance design. The move is not only surprising, but marks an extremely important divergence in Arm's business model and design methodology, finally addressing some of the company's years-long product line compromises.

[...] The new Cortex-A78 pretty much continues Arm's traditional design philosophy, that being that it's built with a stringent focus on a balance between performance, power, and area (PPA). PPA is the name of the game for the wider industry, and here Arm is pretty much the leading player on the scene, having been able to provide extremely competitive performance at with low power consumption and small die areas. These design targets are the bread & butter of Arm as the company has an incredible range of customers who aim for very different product use-cases – some favoring performance while some other have cost as their top priority.

All in all (we'll get into the details later), the Cortex-A78 promises a 20% improvement in sustained performance under an identical power envelope. This figure is meant to be a product performance projection, combining the microarchitecture's improvements as well as the upcoming 5nm node advancements. The IP should represent a pretty straightforward successor to the already big jump that were the A76 and A77.

[...] The Cortex-X1 was designed within the frame of a new program at Arm, which the company calls the "Cortex-X Custom Program". The program is an evolution of what the company had previously already done with the "Built on Arm Cortex Technology" program released a few years ago. As a reminder, that license allowed customers to collaborate early in the design phase of a new microarchitecture, and request customizations to the configurations, such as a larger re-order buffer (ROB), differently tuned prefetchers, or interface customizations for better integrations into the SoC designs. Qualcomm was the predominant benefactor of this license, fully taking advantage of the core re-branding options.

[...] At the end of the day, what we're getting are two different microarchitectures – both designed by the same team, and both sharing the same fundamental design blocks – but with the A78 focusing on maximizing the PPA metric and having a big focus on efficiency, while the new Cortex-X1 is able to maximize performance, even if that means compromising on higher power usage or a larger die area.

While Cortex-A78 will only improve performance by around 7% from microarchitectural changes alone, Cortex-X1 will improve performance by up to 30% due to a wider design, doubling of most cache sizes, and other changes. Cortex-X1 cores are also expected to reach 3 GHz on a "5nm" node, delivering even more performance. The Cortex-X1 cores could use up to 50-100% more power than Cortex-A77/A78. Cores could be arranged in a 1+3+4 or 2+2+4 setup of Cortex-X1, Cortex-A78, and Cortex-A55 cores.

See also: Arm Announces The Mali-G78: Evolution to 24 Cores


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.
  • (Score: 2) by bzipitidoo on Wednesday May 27 2020, @06:10AM (5 children)

    by bzipitidoo (4388) on Wednesday May 27 2020, @06:10AM (#999532) Journal

    Cool. So will the X1 be able to power a desktop with a 4K display? Or a dual monitor setup? One of the main things I've noticed in the past about ARM is that they depend on low resolution displays to perform acceptably. The BeagleBone of a decade ago strained to run just a 1080p. Was okay just keeping up with a GUI, but throw a video at it and make it decode as well as display, and it dragged.

    Starting Score:    1  point
    Karma-Bonus Modifier   +1  

    Total Score:   2  
  • (Score: 2) by RamiK on Wednesday May 27 2020, @07:43AM (1 child)

    by RamiK (1813) on Wednesday May 27 2020, @07:43AM (#999539)

    I've personally seen a few $80-100 TV boxes tiling and dragging 4k video without dropping frames or clipping audio.

    The problem with desktop software is that the GUI widget toolkits force you into a single main loop thread so everything needs complete rewrites which isn't worth the time and money unless targeting mobile first. The equation is only now just starting to shift as Moore's law is dying down. But overall, software development continues to lag behind the hardware and not the other way around.

    --
    compiling...
    • (Score: 0) by Anonymous Coward on Wednesday May 27 2020, @01:01PM

      by Anonymous Coward on Wednesday May 27 2020, @01:01PM (#999591)

      I'm no expert, but it seems to match my experience. One thing I stumbled over a few years ago was that the high definition audio decoder software on earlier versions of Android had serious performance issues. I don't know how much of the problem was ARM, and how much was the software. I ripped our movies to MKV files, and set up Kodi on all of the Android gadgets and computers in the house. Everything played fine on PCs, but on Android devices the ripped Blu Rays wouldn't play or would play for a few minutes and then hang.

      I started reencoding the ripped Blu Ray audio from the default format to AAC, the audio format in MP4 files. After that everything played fine everywhere. But I don't think the audio reencode step is necessary any more, newer versions of Android can play Dolby Digital/AC3 and Dolby TrueHD audio without problems.

  • (Score: 3, Insightful) by takyon on Wednesday May 27 2020, @02:49PM (2 children)

    by takyon (881) <reversethis-{gro ... s} {ta} {noykat}> on Wednesday May 27 2020, @02:49PM (#999671) Journal

    Newer flagship ARM SoCs are probably somewhere between 20-50 times faster than a single-core 720 MHz Cortex-A8 BeagleBone.

    Even a Raspberry Pi 4, which is by no means state of the art performance, can drive two 4K displays (both at 60 Hz if overclocked). It probably can't decode a 4K YouTube video because it lacks VP9 hardware decode for some strange reason. But it can do H.265.

    The Cortex-X1 is a staggering IPC/clock uplift intended to improve single-threaded performance. It will be paired with the already good Cortex-A78 and Cortex-A55.

    https://wccftech.com/qualcomm-snapdragon-875-cortex-x1-super-core-configuration/ [wccftech.com]

    --
    [SIG] 10/28/2017: Soylent Upgrade v14 [soylentnews.org]
    • (Score: 2) by bzipitidoo on Thursday May 28 2020, @12:22AM (1 child)

      by bzipitidoo (4388) on Thursday May 28 2020, @12:22AM (#999986) Journal

      > probably can't decode a 4K YouTube video because it lacks VP9 hardware decode

      That's the rub with video playback on pretty much all low power computing. Hardware decode, or forget it. I'm eager to see AV1 and AV2 displace MPEG4 and H.265 and H.266, but the lag between the release of such a standard with a reference implementation in software, then its implementation in hardware, and finally the release of the hardware to the consumer market, appears to be several years. Throw in an additional year for price drops. I hope that AV1's openness is shortening that time.

      Another bit of hardware I have is a old Toshiba tablet, circa 2012, highest version of Android it can handle is 4.4. Because it has MPEG4 decode in hardware, it can play mp4 video. Throw any other video format at it, and it chokes. On a VP9 file, it will display one of the first frames, and drop all the rest as it plays only the audio.

      • (Score: 2) by takyon on Friday May 29 2020, @05:54AM

        by takyon (881) <reversethis-{gro ... s} {ta} {noykat}> on Friday May 29 2020, @05:54AM (#1000430) Journal

        I just realized that the Chromium browser on Raspbian comes with h264ify [google.com] preinstalled. It's an addon that apparently forces H.264 versions of videos to load on YouTube.

        I did test a 1080p60 version of Big Buck Bunny on YouTube earlier (on a 720p display lol). It plays, but it doesn't look 100% smooth. This may be entirely fixable with OS updates.

        The specs for RPi4 list H.264 1080p60 decode support. 4Kp60 requires H.265. There is no option for me to select 4K on 4K YouTube videos.

        Anyway, I really hope AV1/AV2 are supported on future boards (and all upcoming SoCs/GPUs). And I'm hopeful that the gap between AV1 and AV2 adoption will be shorter due to substantial similarities (let's not forget Google's old plans [soylentnews.org] to release a new VP10, VP11, etc. codec every 18 months).

        --
        [SIG] 10/28/2017: Soylent Upgrade v14 [soylentnews.org]