Stories
Slash Boxes
Comments

SoylentNews is people

posted by Fnord666 on Thursday January 14 2021, @05:48PM   Printer-friendly

$119+ BeagleV powerful, open-hardware RISC-V Linux SBC targets AI applications

Running Linux on RISC-V hardware is already possible, but you'd have a choice of low-end platforms like Kendryte K210 that's not really practical for anything, or higher-end board like SiFive HiFive Unmatched or PolarBerry for which you'd have to spend several hundred dollars, or even over one thousand dollars to have a complete system.

So an affordable, usable RISC-V Linux SBC is clearly needed. We previously wrote about an upcoming Allwinner RISC-V Linux SBC that will be mostly useful for camera applications without 3D GPU, and a maximum of 256MB RAM. But today, we have excellent news, as the BeagleBoard.org foundation, Seeed Studio, and Chinese fanless silicon vendor Starfive partnered to design and launch the BeagleV SBC (pronounced Beagle Five) powered by StarFive JH7100 dual-core SiFive U74 RISC-V processor with Vision DSP, NVDLA engine, and neural network engine for AI acceleration.

[...] Based on our previous article about SiFive U74 core, performance should be similar to Cortex-A55, so a dual-core U74 RISC processor will not have that much processing power compared to other Arm boards, but the network accelerator should make it competitive against other AI boards like Coral Dev Board mini.

One obvious item missing from the specifications is a GPU, and I was told while the first batch scheduled in March will be GPU less, but the next batch – slated to be manufactured in September – will come with an Imagination Technologies GPU.

BeagleV will be supported by mainline Linux and a Debian-based software image will be provided. I can also see mentions of Fedora and FreeRTOS. The RISC-V Linux SBC will be open-source hardware just like other boards from the BeagleBoard.org foundation meaning hardware design files, firmware, and the software will be made available publicly.

Related: SiFive Announces HiFive Unmatched Mini-ITX Motherboard for RISC-V PCs


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

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: 4, Insightful) by DannyB on Thursday January 14 2021, @06:24PM (15 children)

    by DannyB (5839) Subscriber Badge on Thursday January 14 2021, @06:24PM (#1100149) Journal

    I would like to see RISC V succeed and thrive commercially. Just as Linux has. Despite all the Linux and open source naysayers 20+ years ago.

    It is amusing to see Linux run on multiple processor ISAs already. I'm sure this is an oversimplification, but it seems like once you have the compilers ported to the new architecture, it becomes easier to port the Kernel. (not to underestimate that effort, because this is different hardware!) Once there is a kernel, I suspect many applications can simply be recompiled. Others with processor dependencies can be modified. [any informed clarification on this appreciated] It was amazing to me to see how quickly a vast array of Linux applications were available on Raspberry PI (ARM). And then how many different distributions were ported. Including desktop environments.

    It also seems that if you work at a fairly high level, away from direct hardware, Linux feels the same regardless of underlying processor. For example Python doesn't seem different. Java certainly doesn't seem different. Nor browswers, JavaScript, etc. Even rudimentary "textbook" style C programming doesn't seem different. Lisps and other languages seem to work the same.

    Apple has managed to switch processors before, twice, successfully. I expect them to successfully transition to ARM.

    Microsoft has tried (twice, I think) to offer their software on ARM. But the real value of Windows is the legacy software. (And that's also the real value of Intel ISA, the legacy software.) A lot of Windows legacy software is wedded to the Intel ISA, and/or 32bit, and even 16bit code. Therefore I don't expect Microsoft to have any easier going of moving their software base to RISC V than they had moving it to ARM.

    Modern Windows apps may be easy to move to other processor architectures. But they also may be easy to move to other OSes than Windows. (Witness Microsoft porting their Edge browser to Linux desktop, and earlier porting SQL Server to Linux.) If you're not running legacy Windows apps, there is less need to have a Windows OS.

    If we live in a world where the vast bulk of everyday software can be "transparently" moved to new processors (thanks to not insignificant efforts of others), this will enable new ideas in hardware design to be tried. Researched. Practical experiments done. Etc. We might see hardware improvements accelerate as much as software improvements have. That would be cool. Even research into different ISA concepts.

    I still want to see more and more cores come to everyday processors in common use.

    --
    People today are educated enough to repeat what they are taught but not to question what they are taught.
    • (Score: 3, Interesting) by looorg on Thursday January 14 2021, @07:18PM (14 children)

      by looorg (578) on Thursday January 14 2021, @07:18PM (#1100174)

      I would quite like to see this to and I really do hope that it comes faster then Linux, cause that is more like 30+ years ago now somewhat depending on what you consider to be Linux and such. It's not for not that "year of the linux desktop" became somewhat of a running and longstanding joke phrase -- it's always next year!

      The Rasperry has somewhat shown that it would be or is possible. It should somehow just scale up to consumer levels now, since apparently we all need gigs of CPU power and memory these days to do anything ...

      Microsofts issue then will never be resolved if there is always the need to be completely backwards compatible and run legacy software -- unless the future becomes so good as to emulate it won't pose any issues. One would like to think that these legacy applications eventually go out of style. But then I know from personal experience that things I wrote in the mid 90's are still running and are used in all their DOS shell glory. That said I would prefer it if the machine didn't have to waste system resources on software emulation, but that might just be a personal preference.

      • (Score: 3, Insightful) by Anonymous Coward on Thursday January 14 2021, @07:25PM (11 children)

        by Anonymous Coward on Thursday January 14 2021, @07:25PM (#1100179)

        ...is.

        It's a proprietary SiFive U74 core, properietary nvidia ai core, proprietary DSP with plans to replace it with a *POWERVR SGX GPU*, and I am forgetting what else, but it had a couple other proprietary components given its full specification list.

        This is just trading the x86 bullshit and the arm/rpi bullshit for a third set of bullshit. There is no more freedom to be had in this software platform than in the previous ones, and now that they are all unified on UEFI we don't even have a real bios platform that doesn't assume lockdown from initial boot.

        Think very carefully about what you support from here on out, because we're very close to never having a user controlled piece of hardware again.

        • (Score: 3, Insightful) by DannyB on Thursday January 14 2021, @07:48PM (7 children)

          by DannyB (5839) Subscriber Badge on Thursday January 14 2021, @07:48PM (#1100187) Journal

          The key difference is that there is no impediment to prevent a dozen competitors to SiFive. It isn't only SiFive's core. Or, at least, I hope that is not the end game. I thought the whole point is that others who make silicon could also get into this game and compete on implementation -- while agreeing on interface. (a saying that's popular in the Java world, where multiple implementations implement identical interfaces.)

          If RISC V is available from multiple vendors, then the market will support what the customers want. Or do not want -- like "management engines", or "managing your digital rights", etc. The presence or absence of these features might be verifiable.

          --
          People today are educated enough to repeat what they are taught but not to question what they are taught.
          • (Score: 4, Informative) by requerdanos on Thursday January 14 2021, @09:37PM (1 child)

            by requerdanos (5997) Subscriber Badge on Thursday January 14 2021, @09:37PM (#1100216) Journal

            the market will support what the customers want. Or do not want

            I "do not want" something requiring proprietary drivers, firmware or software. I want something that will support the use of all free software.

            • (Score: 3, Informative) by DannyB on Thursday January 14 2021, @10:00PM

              by DannyB (5839) Subscriber Badge on Thursday January 14 2021, @10:00PM (#1100220) Journal

              I want that too.

              --
              People today are educated enough to repeat what they are taught but not to question what they are taught.
          • (Score: 2) by driverless on Friday January 15 2021, @05:15AM (3 children)

            by driverless (4770) on Friday January 15 2021, @05:15AM (#1100411)

            The key difference is that there is no impediment to prevent a dozen competitors to SiFive.

            But the same could be said of any other ISA. All you need to do is hire a few hundred highly skilled engineers and managers, spend a few hundred million on development, and have your own billion-dollar fab, and you can do anything you want.

            In fact arguably ARM is less proprietary/locked-in at the moment because I can go to dozens if not hundreds of different vendors to pick and choose what I want, while with SiFive there's, well, SiFive. Take it or leave it.

            • (Score: 2) by DannyB on Friday January 15 2021, @02:50PM (2 children)

              by DannyB (5839) Subscriber Badge on Friday January 15 2021, @02:50PM (#1100550) Journal

              You seem to miss the entire point.

              ARM is proprietary. Like Windows.

              You have to license it.

              --
              People today are educated enough to repeat what they are taught but not to question what they are taught.
              • (Score: 2) by driverless on Friday January 15 2021, @03:19PM (1 child)

                by driverless (4770) on Friday January 15 2021, @03:19PM (#1100562)

                That only matters if you have a billion-dollar fab in your basement and can make your own silicon. For everyone else, you go out and buy ARM CPUs by the truckload from any vendor you want, no licensing involved. And they're cheaper, faster, and more capable than SiFive's silicon.

                As far as 99.99999% of the world's population is concerned, the choice is: ARM = cheap, universally available, RISC-V = expensive, single vendor (if you want one you can run Linux on). This is why the entire planet runs ARM and not RISC-V. It'd be nice if there was more RISC-V, but it's not providing anything that the market cares about.

                • (Score: 2) by pe1rxq on Friday January 15 2021, @09:43PM

                  by pe1rxq (844) on Friday January 15 2021, @09:43PM (#1100845) Homepage

                  ARM might be cheap, but it is gone when NVIDIA feels like it.

          • (Score: 0) by Anonymous Coward on Friday January 15 2021, @03:01PM

            by Anonymous Coward on Friday January 15 2021, @03:01PM (#1100554)

            We hoped RISC-V would mean fully open ISA (instruction set architecture) and implementations, and that didn't happen. But people can make open implementations to the RISC-V ISA, which still makes it a more open ISA than ARM or x86/x86_64. I don't think it will happen, but I hope high performance fully open RISC-V implementations eventually exist.

        • (Score: 0) by Anonymous Coward on Thursday January 14 2021, @07:53PM (1 child)

          by Anonymous Coward on Thursday January 14 2021, @07:53PM (#1100189)

          https://www.powerpc-notebook.org/ [powerpc-notebook.org]

        • (Score: 0) by Anonymous Coward on Friday January 15 2021, @11:28PM

          by Anonymous Coward on Friday January 15 2021, @11:28PM (#1100897)

          i agree that i want all open hardware, but what binary blobs does this thing require? is it not a step in the right direction? It would be nice to see a blog post with a critical breakdown of everything from the perspective of an freedom idealist. I'm not well versed in embedded hardware enough to do it.

      • (Score: 2) by DannyB on Thursday January 14 2021, @08:10PM (1 child)

        by DannyB (5839) Subscriber Badge on Thursday January 14 2021, @08:10PM (#1100196) Journal

        Microsofts issue then will never be resolved if there is always the need to be completely backwards compatible and run legacy software -- unless the future becomes so good as to emulate it won't pose any issues.

        Emulation of legacy becomes more feasible with each passing day. The Linux on my wristwatch is a more powerful machine than my first Linux desktop PC in June 1999.

        Look at how successfully MAME made obsolete ancient video games continue to have new life. It became possible to emulate these machines at real-time performance of their originals.

        One would like to think that these legacy applications eventually go out of style. But then I know from personal experience that things I wrote in the mid 90's are still running and are used in all their DOS shell glory.

        I also know this from personal experience. In my employment in the 80's I was developing (specialized) accounting software in Pascal. We were aware of the Y2K problem far away in some distant future. But by then we would have an entirely different system, long since rebuilt into something more closely resembling perfection and ultimate bliss with unicorns and rainbows everywhere.

        Y2K arrived. A few years before we got on fixing it -- in that same codebase I was talking about. Even though we had already finally begun rewriting into a new GUI system. (the web wasn't on our radar yet as a user interface) The trick was that the UCSD p-System stored dates in 2 bytes. 7 bits for the year. Our insight was that technically, the year could go up to 127 in those 7 bits, meaning up to 2027. We revised our text-based green screen interface to record years from 00-27 as 100-127 in those seven bits of a date. Now all date arithmetic and sorting just worked, everywhere. Lucky us.

        I'm sure many programmers have been astonished how long software systems end up living on. Heck, people are still using COBOL on Mainframes. The real value of Mainframes is that it runs gazillions of lines of COBOL code written from the 1960's on.

        --
        People today are educated enough to repeat what they are taught but not to question what they are taught.
        • (Score: 2) by looorg on Friday January 15 2021, @02:10AM

          by looorg (578) on Friday January 15 2021, @02:10AM (#1100312)

          Look at how successfully MAME made obsolete ancient video games continue to have new life. It became possible to emulate these machines at real-time performance of their originals.

          I'm not entirely sure this translates as well. If we are talking classic arcade machines (which is more or less the target for MAME) they are usually very dedicated machine that only really runs one thing (ok you can swap ROM:s and such and make it run other games but besides that). It's mostly emulates one or several 8-bit, sometimes 16-bit, processors usually running in the single digit Mhz range. So it's now well within emulation range with current hardware, and has been for quite some time really. Errors are usually when they triggered or did something fancy with the hardware that is missed in the emulators. That said emulating some 30 year old hardware should be doable if we fancy running old DOS applications or some early Windows etc, no issue I think, unless it requires some very specific hardware perhaps. As things gets newer tho it might be more trouble then it's worth if you have to run newer and newer versions of Windows sucking down resources like the bloated hog that it is.

  • (Score: 2, Informative) by Anonymous Coward on Thursday January 14 2021, @06:35PM

    by Anonymous Coward on Thursday January 14 2021, @06:35PM (#1100154)
  • (Score: 0) by Anonymous Coward on Thursday January 14 2021, @07:41PM (5 children)

    by Anonymous Coward on Thursday January 14 2021, @07:41PM (#1100184)

    Is it only me or does it say 0 comments for the last four or five stories even tho there are comments in them?

    • (Score: 2) by DannyB on Thursday January 14 2021, @07:52PM

      by DannyB (5839) Subscriber Badge on Thursday January 14 2021, @07:52PM (#1100188) Journal

      It's not just you.

      It must be because Silicon Valium has too much power! Censorship! Silicon Valium has more power than the rest of the grid!

      --
      People today are educated enough to repeat what they are taught but not to question what they are taught.
    • (Score: 0) by Anonymous Coward on Thursday January 14 2021, @07:57PM (2 children)

      by Anonymous Coward on Thursday January 14 2021, @07:57PM (#1100190)

      On a site with as many self-identified very smart people as here, I would expect the ones in charge to have fewer problems with writing and running Perl scripts.

      • (Score: 5, Funny) by Anonymous Coward on Thursday January 14 2021, @08:16PM (1 child)

        by Anonymous Coward on Thursday January 14 2021, @08:16PM (#1100199)

        Writing and running Perl scripts is easy, its reading and fixing them that gets hard.

    • (Score: 2) by canopic jug on Friday January 15 2021, @11:17AM

      by canopic jug (3949) Subscriber Badge on Friday January 15 2021, @11:17AM (#1100501) Journal

      I have a possible duplicate in the queue which could be merged or deleted outright: First Affordable RISC-V Board Designed to Run Linux [soylentnews.org].

      --
      Money is not free speech. Elections should not be auctions.
(1)