Stories
Slash Boxes
Comments

SoylentNews is people

posted by martyb on Saturday October 23 2021, @04:45AM   Printer-friendly

We're closing the gap with Arm and x86, claims SiFive: New RISC-V CPU core for PCs, servers, mobile incoming

SiFive reckons its fastest RISC-V processor core yet is closing the gap on being a mainstream computing alternative to x86 and Arm.

The yet-unnamed high-performance design is within reach of Intel's Rocket Lake family, introduced in March, and Arm's Cortex-A78 design, announced last year, in terms of single-core performance, James Prior, senior director of product marketing and communications at SiFive, told The Register.

San Francisco-based SiFive didn't provide specific comparative benchmarks, so you'll have to take their word for it, if you so choose.

[...] SiFive's latest design, which is set to be teased today, will be christened with a formal name at the RISC-V Summit in December.

The CPU core is said to be about 50 per cent faster than its predecessor, the P550, which was introduced in June. We note that the L3 cache memory capacity has been quadrupled, from the 4MB in the P550 to 16MB in the new design. Up to 16 of these new cores can be clustered versus the maximum of four for the P550. The latest design can also run up to 3.5GHz compared to 2.4GHz for the P550.

Intel's Attempt to Acquire SiFive for $2 Billion Fell Apart, Report Claims

While Intel was interested to acquire RISC-V processor developer SiFive and SiFive is considering its strategic options, the companies could not agree neither on financial terms nor on how SiFive technologies could be used at Intel reports Bloomberg. The latter company is still considering both an initial public offering (IPO) as well as a takeover by a larger player.

Previously: SiFive Announces HiFive Unmatched Mini-ITX Motherboard for RISC-V PCs
Intel May Attempt to Acquire SiFive for $2 Billion
Intel Will License SiFive's New P550 RISC-V Core


Original Submission

Related Stories

SiFive Announces HiFive Unmatched Mini-ITX Motherboard for RISC-V PCs 12 comments

SiFive has announced a mini-ITX motherboard with its SiFive Freedom U740:

At the heart of the SiFive board is a SiFive FU740 processor coupled with 8 GB DDR4 memory and 32 MB SPI Flash. It comes with a 4x USB 3.2 ports and a 16x PCIe expansion slot. The mini-ITX standard form factor makes it easy to build a RISC-V PC.

[...] SiFive Unmatched board will be available by Q4'20 for USD 665, and you can already register your interest. You will get a mini-ITX board, 32 GB MicroSD, and 3-meter CAT5e ethernet cable. SiFive did not speak on the commercial aspect of the product but are very confident about future development. Android and Chrome support is something we can see in the future. The product looks promising and we are excited to see future development in the RISC-V PC ecosystem.

Performance will probably be comparable to a Raspberry Pi 3. Alternatively:

PolarBerry is a Compact, Linux-capable RISC-V FPGA SBC and module (Crowdfunding)

Powered by Microchip PolarFire RISC-V SoC FPGA, PolarBerry is both a single board computer with Gigabit Ethernet and 40-pin GPIO header, as well as a system-on-module thanks to three Samtec board-to-board connectors.

[...] PolarBerry is not available just yet, but LinuxGizmos reports the SBC/SoM will be soon launched on Crowd Supply for $995 and shipments are expected to start in January 2021. Besides the aforementioned crowdfunding page, additional details may be found on the product page.

See also: SiFive Is Launching The Most Compelling RISC-V Development Board Yet

Previously: SiFive to Debut a RISC-V PC for Developers in October


Original Submission

Intel May Attempt to Acquire SiFive for $2 Billion 8 comments

Intel (INTC) Reportedly Offers Over $2 Billion To Acquire the Fabless Semiconductor SiFive as the Consolidation Trend in the Industry Is Nowhere Close to Slowing Down

[According] to Bloomberg, Intel has reportedly offered over $2 billion to acquire the fabless semiconductor SiFive, a provider of commercial RISC-V processor IP and silicon solutions based on the RISC-V instruction set architecture.

Should this deal become a reality, it would mark the climax of growing bonhomie between Intel and SiFive. For instance, back in 2018, Intel was one of the participants in the Series C funding round of SiFive. Thereafter, in March 2021, SiFive announced a collaboration with the Intel Foundry Business (IFB) to develop innovative new RISC-V computing platforms.

Of course, unlike legacy Instruction Set Architectures (ISAs), RISC-V's proponents believe that it addresses the skyrocketing cost of designing and manufacturing increasingly complex new chip architectures, given that that the ISA is layered, extensible, and flexible. It is hardly surprising, therefore, that some believe RISC-V to be the future.

Bear in mind that SiFive was last valued at $500 million, as per the data available at PitchBook. This means that Intel would be paying a premium of over 300 percent relative to SiFive's 2020 valuation.

Previously: SiFive HiFive Unleashed Not as Open as Previously Thought
Qualcomm Invests in RISC-V Startup SiFive
SiFive Announces a RISC-V Core With an Out-of-Order Microarchitecture
GlobalFoundries and SiFive Partner on High Bandwidth Memory (HBM2E)
SiFive to Debut a RISC-V PC for Developers in October
SiFive Announces HiFive Unmatched Mini-ITX Motherboard for RISC-V PCs


Original Submission

Intel Will License SiFive's New P550 RISC-V Core 18 comments

Intel to make a custom SiFive-based RISC-V CPU, will be fabricated on a 7 nm node in a first step towards competing directly with Arm-based chips

The partnership will see Intel license SiFive's IP to create its own SiFive P550-based 64-bit SoC that it will fabricate on its new 7 nm node. It will form the basis of a new development platform Intel is calling Horse Creek, and will be made available to customers interested in exploring its potential in various applications involving embedded SoC tech. This could mean smartphones, but also cars, IoT products and the like. If Intel gets enough interest, it could take the relationship further. Intel hasn't yet revealed the technical specifications of the SoC, so we don't know whether it will be a single-core or multi-core platform, although the latter is likely. It's GPU tech is also unknown at this time, but Xe-based graphics are likely.

While the first Horse Creek SoCs will be ready next year, it isn't likely we will see any Intel RISC-V-based chips in commercially available products until 2023 at least.

SiFive recently announced two new high-performance 64-bit RISC-V cores, the Performance P550 and Performance P270:

SiFive compares the Performance P550 core to Arm's Cortex-A75 with higher performance in SPECint2006 and SPECfp2006 integer/floating-point benchmark, all [in] a much smaller area which would enable a quad-core P550 cluster on about the same footprint as a single Cortex-A75 core.

See also: Ubuntu 20.04/21.04 64-bit RISC-V released for QEMU, HiFive boards

Previously: Intel May Attempt to Acquire SiFive for $2 Billion


Original Submission

SiFive's P650 RISC-V Core Detailed, Claimed to be Faster Than ARM Cortex-A77 17 comments

SiFive Details New Performance P650 RISC-V Core

SiFive's Performance P650 licenseable processor IP core will debut to lead partners in Q1'2022 while the general availability is expected in "summer" 2022. Whether the Performance P650 will make its way into any public SiFive developer boards or the like remain unknown, but hopefully they will come out next year with some performant successor to the HiFive Unmatched.

This successor to their Performance P550 is expected to be the fastest RISC-V processor IP core on the market. Over the P550 should be around a 40% performance increase per-clock cycle. Overall there should be around a 50% performance gain over the P550. SiFive is reporting the Performance P650 will be faster than the Arm Cortex-A77.

SiFive Performance P650 RISC-V core to outperform Arm Cortex-A77 performance per mm2

Building upon the Performance P550 design, the SiFive Performance P650 is scalable to sixteen cores using a coherent multicore complex, and delivers a 40% performance increase per clock cycle based on SiFive engineering estimated performance in SPECInt2006/GHz, thanks to an expansion of the processor's instruction-issue width. The company compares P650 to the Arm family by saying it "maintains a significant performance-per-area advantage compared to the Arm Cortex-A77".

Other architecture enhancements over the previous generation include a higher maximum clock frequency (Liliputing says up to 3.5 GHz), platform-level memory management, interrupt control units, and support for the new RISC-V hypervisor extension for virtualization.

ARM Cortex-A77.

Previously: Intel Will License SiFive's New P550 RISC-V Core
SiFive Teases Fast New RISC-V Processor Core; Intel Acquisition Attempt Failed


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: 1, Insightful) by Anonymous Coward on Saturday October 23 2021, @05:20AM (2 children)

    by Anonymous Coward on Saturday October 23 2021, @05:20AM (#1189829)

    Instruction set is just one part of computer architecture.
    I don't think it's the main factor determining power usage or performance anymore.

    • (Score: 4, Insightful) by driverless on Saturday October 23 2021, @07:32AM (1 child)

      by driverless (4770) on Saturday October 23 2021, @07:32AM (#1189841)

      The yet-unnamed high-performance design is within reach of Intel's Rocket Lake family, introduced in March, and Arm's Cortex-A78 design, announced last year, in terms of single-core performance, James Prior, senior director of product marketing and communications at SiFive, told The Register.

      And once they figure out how to get that performance level without GaN chips cooled with liquid helium, they'll be onto a real winner!

      Before that gets modded flamebait, I actually rather like the whole RISC-V effort, I just wish they'd be a bit more real-world realistic with their claims.

      • (Score: 2) by sgleysti on Saturday October 23 2021, @05:25PM

        by sgleysti (56) Subscriber Badge on Saturday October 23 2021, @05:25PM (#1189907)

        And once they figure out how to get that performance level without GaN chips cooled with liquid helium, they'll be onto a real winner!

        I don't think anyone has a GaN foundry that can reach the process nodes and density required to build these chips. I've only seen GaN used in N-Channel power MOSFETs anyway. I think the higher speed non-Si stuff is usually GaAs; if memory serves, that was used in cellular baseband modems at one point. IBM also has a SiGe graded heterojunction process that is really high speed, but I don't think they've made a CPU out of it. Heterojunction sounds like BJTs anyway, so that might be more targeted at RF and other high speed amplifiers.

        These SiFive CPUs were most likely made at one of the Silicon foundries that does contract work, like TSMC or whoever else is in that market. The most they would have done for cooling is use a large, high power cold plate, but I doubt they're pushing the TDP that far.

        It'll be good to see better data on computational performance and power dissipation.

  • (Score: 2, Offtopic) by Frigatebird on Saturday October 23 2021, @06:59AM (4 children)

    by Frigatebird (15573) on Saturday October 23 2021, @06:59AM (#1189837)

    I owned a Dodge, a 1940's Dodge Truck. Thing about it, or two things about it, I has a 6-volt positive ground electrical system, such as it was. And, it has lug bolts, rather than studs and lug nuts, but that was not the weird thing. The lug bolts on the left/driver's side, were all left hand thread. Taking it in to a modern shop to have the tires replaced was a trip. But the reason my Dodge almost Power wagon had left hand thread lug bolts goes back to wagon wheels. You see, old wagons would have wheel bearings of a sort, usually steel bands running on steel band, in the hub and the spindle. And all that was held together by an axle nut. The nut fastened on the spindle, so on the right side, the motion of the wheel would tend to tighten the wheel nut, since it was right-handed thread. On the other side, the wheel rotated the other way, relative to the spindle, so that wheel nut was given a left-handed thread, so that the forward motion of that wheel would also have the effect of tightening rather than loosening the wheel-nut. So much for Wagons.

    So, my 1940's Dodge retained the tradition of left hand thread on the Sinister side, even though the thread direction of the lug-bolts had no bearing on loosing or tightening of the bolts, unlike with the spindle nut. And now for the relevance of our horse buggy whip- railroad guage- left hand wheel-nut automotive analogy: x86 architecture is exactly like the left handed lug-bolts on my '40's Dodge. A left over from the past that no one understand, but one that is getting in the way when we metaphorically take our PowerWagons Books into the shop to have new tires put on.

    It is time to universalize hardware, to virtualize software, and to make sure that anything can run on anything, but not like Javascript does.

    • (Score: 2) by driverless on Saturday October 23 2021, @08:10AM

      by driverless (4770) on Saturday October 23 2021, @08:10AM (#1189846)

      I has a 6-volt positive ground electrical system, such as it was.

      Wait, a 1940s Dodge? Joseph Lucas supposedly died (meaning returned to the underworld) in 1902, how did he do electrics for Dodge in the 1940s?

    • (Score: 2) by ls671 on Saturday October 23 2021, @08:40AM

      by ls671 (891) Subscriber Badge on Saturday October 23 2021, @08:40AM (#1189851) Homepage

      On the other side, the wheel rotated the other way, relative to the spindle, so that wheel nut was given a left-handed thread, so that the forward motion of that wheel would also have the effect of tightening rather than loosening the wheel-nut. So much for Wagons.

      It reminds me of the oil drilling rigs and a drill string made of pipes. The rotary table is what transfer power to the drill string in order for it to rotate clockwise and cause the drilling bit to dig. In the drill string, everything above the rotary table is left handed for the same reason. I remember seeing this elsewhere, always for the same reason.

      --
      Everything I write is lies, including this sentence.
    • (Score: 1, Informative) by Anonymous Coward on Saturday October 23 2021, @02:22PM

      by Anonymous Coward on Saturday October 23 2021, @02:22PM (#1189887)

      If the wheel-hub pilot wasn't a tight fit, then I believe that your Dodge left hand thread bolts were the correct choice. Imagine the wheel moving slightly relative to the hub, under load....

      For yet another example of left hand threads on the left side, see bicycle pedals. In this case the overhung load (foot not in line with crank arm) will quickly unscrew a pedal that is threaded the wrong way. It's a subtle thing, there is an animation on this page -- https://physics.stackexchange.com/questions/6488/why-are-bicycle-pedal-threads-handedness-left-on-the-left-and-right-on-the-righ [stackexchange.com]

    • (Score: 3, Informative) by sgleysti on Sunday October 24 2021, @03:56AM

      by sgleysti (56) Subscriber Badge on Sunday October 24 2021, @03:56AM (#1190016)

      I wish there was some middle level between the mechanics and actually working on the car in this analogy... since the vast majority of programming targeting x86 today is done in compiled or interpreted languages, it is only people writing compilers, higher precision math routines, or low level drivers / OS stuff who will actually touch the architecture.

      Granted it still affects all of us at some level.

  • (Score: 0) by Anonymous Coward on Saturday October 23 2021, @07:05AM (1 child)

    by Anonymous Coward on Saturday October 23 2021, @07:05AM (#1189839)

    i hope it's not just a "java chip" like most all ARM chips are ...

    • (Score: 2) by sgleysti on Sunday October 24 2021, @04:48AM

      by sgleysti (56) Subscriber Badge on Sunday October 24 2021, @04:48AM (#1190024)

      By "java chip" do you mean only used in Android phones? Because there have been attempts to implement the Java bytecode interpreter in hardware...

  • (Score: 0) by Anonymous Coward on Sunday October 24 2021, @12:38PM (1 child)

    by Anonymous Coward on Sunday October 24 2021, @12:38PM (#1190063)

    I'm really glad to hear Intel didn't get their grubby paws on RISC-V. Hopefully RISC-V will remain independent; and perhaps one day, we'll have processors without processors in them, minding what our processors do. We'll have some freedom in our hardware to go with the free software running on it.

    • (Score: 0) by Anonymous Coward on Sunday October 24 2021, @01:16PM

      by Anonymous Coward on Sunday October 24 2021, @01:16PM (#1190074)

      1. Buying SiFive doesn't deprive anyone of the RISC-V ISA.
      2. RISC-V is easily extensible and will be used to make proprietary and evil products, including backdoors.
      3. RISC-V will end up a poor experience for Linux, just like ARM.

(1)