back to article RISC-V needs more than an open architecture to compete

Interviews with chip company CEOs are invariably enlightening. On top of the usual market-related subjects of success and failure, revenues and competition, plans and pitfalls, the highly paid victim knows that there's a large audience of unusually competent critics eager for technical details. That's you. Take The Register's …

  1. werdsmith Silver badge

    Vision V??? I think the Sipeed Nezha RISC V SBC from last year must have a better claim to be first as it appeared last year. That is also expensive and slow.

  2. Paul Crawford Silver badge

    I think you are failing to factor in the geopolitical aspect, there may well be a lot of companies that chose to use RISC-V, or are forced to by their governments, to avoid the risk of future sanctions.

    1. LDS Silver badge

      Yes, and for the same reason there will be countries that will need to hinder the RISC-V business to avoid to give away an advantage to those under sanctions...

    2. Jason Bloomberg

      there may well be a lot of companies that chose to use RISC-V, or are forced to by their governments, to avoid the risk of future sanctions.

      Or, of course, forced into using RISC-V because of sanctions by other governments.

      China, India and others will have seen the impact of sanctions being applied to Russia and will have asked themselves what would happen if they were in the same boat.

      Nato expansionism has properly upset the cart. I am not sure if the inevitable consequences were unforeseen or unintended.

      1. 3arn0wl

        The biggest catalyst for the adoption of RISC-V...

        ... has been Trump's tech trade dispute, which Biden seems to have continued. Alibaba's T-Head had their family of XuanTie processors by 2019 - they've sold in excess of 2.5 billion of them, apparently.

        Perhaps a new iron curtain is being made of silicon, rather than clay. :/

        1. Justthefacts Silver badge

          Re: The biggest catalyst for the adoption of RISC-V...

          Are you under the impression that 2.5billion cores shipped (over a four year period) is a big number?

          See, this is your problem. ARM has shipped 200billion cores. For all the trumpeting, RISCV has shipped *1%* of ARM. ARM ships about 2billion cores per *month*. And ARM are by no means the only player at the low end. A quick Google finds - about 25billion PIC controllers, for example. And most engineers can name another dozen MCUs that *each* shipped 25 billion cores.

          2.5billion RISCV cores, over a multi year period, doesn’t put it in the top ten, I’m afraid. And likely never will.

          1. 3arn0wl

            Re: The biggest catalyst for the adoption of RISC-V...

            2.5 billion cores ain't nothin' either - especially given that some of the XuanTie family can run Linux, and others are server class.

            But Alibaba's T-Head isn't the only company producing RISC-V processors - SiFive has sold well over 2 billion, Andes upwards of 10 billion. And there are any number of other companies too. In fact, since chip manufacturers don't have to disclose that they're using the ISA unless they use the logo, it's impossible to know exactly how many chips have been deployed.

            A November 2019 report from Semico Research Corp. predicted the market for RISC-V CPU cores will reach 62.4 billion by 2025. I believe that number has now been increased to 80 billion. (As of 2019, the 35 year old Arm architecture had shipped 150 billion cores.)

            https://riscv.org/announcements/2019/11/9679/

            1. This post has been deleted by its author

            2. Justthefacts Silver badge
              Stop

              Re: The biggest catalyst for the adoption of RISC-V...

              Andes have *not* shipped ten billion RISCV cores. You need to read more carefully what they say on their own website:

              “ cumulative shipments have surpassed 10 billion” ok……and read more carefully lower down:

              “What’s more remarkable is that while 1% of Andes-embedded SoCs shipped are based on RISC-V, 99% are contributed by Andes processors of the third-generation architecture (V3) over the years”.

              So, they’ve shipped 1% of 10billion = 0.1billion

              http://www.andestech.com/en/2022/03/08/andes-technology-announces-over-10-billion-cumulative-shipments-of-andes-embedded-socs-and-records-all-time-high-annual-and-monthly-revenue-in-2021/

              I’ve tunnelled into the 2billion quoted by sifive too. If you go closer, you’ll discover that isn’t cores *shipped*. In fact, sifive has no means to even know how many cores have been produced. Almost all of these are *licenses* at bulk rates, not royalties. It costs about $300k to $600k to *license* a core from them to ship an infinite number. The only purpose of the “cores shipped” statement is for them to take it to their lenders and investors and say “see, we have a business”.

              This is a company whose *total revenue* is in the low single digits millions

              The amount of silicon they themselves ship is not precisely known, but never in commercial volume. They’ve been unable to supply, probably because they’ve hit yield problems, which is why they’ve shifted over to partner with Intel for production on Horse Creek.

              1. 3arn0wl

                Re: The biggest catalyst for the adoption of RISC-V...

                Interesting. Thanks

      2. oiseau Silver badge

        Nato expansionism has properly upset the cart.

        Ahh ...

        Finally, some fresh air.

        +1

        ... not sure if the inevitable consequences were unforeseen or unintended.

        Well ...

        I've recently read that, since 1776, the US has not been at war for only 21 out of 239 years.

        Bear in mind that the all powerful corporations that manufacture the tools of war are not the only interested/benefited parties.

        War is a very lucrative activity for a great many and there's always a fantastically huge amount of moolah to go around.

        Draw your conclusions.

        O.

        1. TeeCee Gold badge
          Facepalm

          Nice bit of rabid tinfoil-hattery there.

          Shame it overlooks the obvious snag: There would almost certainly be no expansion of NATO were it not for Mad Vlad doing a passable impersonation of Hitler, with the Russian army gleefully adopting the role of the Waffen SS.

          You are a Fancy Bear newzbot and ICMFP!

        2. Anonymous Coward
          Anonymous Coward

          Isn't it baffling, that all these free, democratic countries are freely choosing to join NATO, when they have a bunch of mad, evil Nazis to the east who keep invading their neighbours and threatening nuclear war?

    3. DS999 Silver badge

      Which governments?

      China is putting their muscle behind the homegrown Loongson architecture for internal use, so RISC-V is too little too late to make any big impact there.

      Even Russia has their own homegrown architecture, but probably are suffering such a brain drain from Putin's stupidity lately that effort will never advance any further.

      1. fg_swe

        Anglo Conjectures

        "Even Russia has their own homegrown architecture, but probably are suffering such a brain drain from Putin's stupidity lately that effort will never advance any further."

        If you knew a bit about ELBRUS, you would figure the opposite.

        First, it is used mostly in olive green applications and secondly they are now forced to substitute American CPUs. Olive green gets more funding due to obvious and the second might be even more important to do.

        Even though their elite has made very serious mistakes, they have figured that AI controlled technology is of strategic importance. In addition to the strategically important tech they build since U2s invaded their airspace.

  3. 3arn0wl

    Plenty of reasons to choose RISC-V

    Control

    In an increasingly divided world, regions are seeing the wisdom of digital sovereignty. China, the EU, Russia and smaller nation states, have all been talking about it, legislating for it, building it. RISC-V belongs to no one, and is available to everyone.

    Stability

    The Nvidia debacle - Arm was once considered to be the Switzerland of ISAs. Now its independence seems unsure, and its future... somewhat dodgy. The problem hasn't gone away either - Softbank are still wanting to sell off Arm, and who's to say a competitor company won't try to amass a controlling share?

    Openness

    People are recognizing the imperative for open knowledge, and tech transparency. And companies are seeing the advantage of collaboration. There's also a hungry consumer market for open hardware (and designers of proprietary RISC-V chips need to be mindful of that too).

    Ease of use

    The base of RISC-V is 47 instructions on a single side of paper (with no kinks or workarounds). University students have been using it for a decade. There's open source software available for design and verification, and even funding available too. Moreover, there's no prohibitive license agreement to haggle over and pay for, and no official body monitoring what you can and can't do - in fact, creativity is positively encouraged. In short, there seems to be little impediment for anyone to use RISC-V.

    But none of these things would matter in the least if it turned out that RISC-V wasn't a decent instruction set. Industry luminaries such as Dave Jaggar and Jim Keller have endorsed it. Others have left lucrative, high-powered positions at OEMs to start-up RISC-V design companies. And it has been shown time and again that RISC-V can beat ARM designs on (their metric of) PPA.

    - There's a boom in chip implementation at the moment, and RISC-V can go head-to-head in design of new silicon and, just possibly, produce solutions more quickly.

    - China accounts for 20% of the tech market (and is one of the biggest manufacturers of tech) - it's moving over to RISC-V (and open source software) quickly. If other countries want to retain a piece of that market, then they'll need to produce goods which are US-IP free.

    1. werdsmith Silver badge

      Re: Plenty of reasons to choose RISC-V

      There's also a hungry consumer market for open hardware

      The vast majority of consumers don't know what this is, nor do they give a shit. However, I have no doubt is would be possible to market-influence such a hunger into them. "Intel Inside"

      1. DS999 Silver badge

        Re: Plenty of reasons to choose RISC-V

        Some people live in an alternate reality and actually think consumers care about stuff like that. I can only assume that their only contact with other people is online via self selected forums, and they haven't talked to a normal person outside of their family since they finished school.

      2. martinusher Silver badge

        Re: Plenty of reasons to choose RISC-V

        The 'consumer' may not care but developers do.

        Having parts unencumbered by supply problems, international politics and ITAR regulations is worth a great deal. So what if the parts are not quite as fast as the latest -- for the majority of embedded applications you don't need the fastest, you need 'adequately fast'. You are also likely to make up any performance gap by improved software design (a lot of modern software design is relatively dirty, its designed primarily around reactivity, speed to market, rather than efficiency; it represents a huge opportunity for a potential competitor).

        1. Justthefacts Silver badge

          Re: Plenty of reasons to choose RISC-V

          If your design or use is affected by ITAR…..then your *first* hurdle is the ITAR status of the FPGA hardware. Almost all the FPGA manufacturers are US: Xilinx, Altera, as was. Selecting a particular IP core as being non-US, and slapping it down on a US part is just such nonsense.

          Of course, if you are implementing on that EU BRAVE shit, you are so far lost from real engineering that it is no longer worth discussing.

          1. fg_swe

            Thanks !

            I was not aware of a french FPGA. Thanks for telling me.

            https://www.nanoxplore.com/

            Maybe we get the IT Airbus after all !

    2. thames Silver badge

      Re: Plenty of reasons to choose RISC-V

      India are also very interested in RISC-V as well, for all the reasons you state plus another one as well. The additional reason is that allows Indian companies to participate in the market on an equal footing to those of other countries. This will allow the Indian technology sector to grow without being subsidiary to the global plans or priorities made in other countries.

      1. fg_swe

        Yeah Charity Will Work

        India has been too corrupt to design, build and market their own processor and this will change because some do-gooders donate an ISA to them ?

        That will work as nicely as donating tractors to a nation of illiterates.

  4. Arthur the cat Silver badge
    Headmaster

    the other UK chip industry

    Not chips! CRISPS, dammit!

    1. werdsmith Silver badge

      Re: the other UK chip industry

      Indeed, if you follow the link you see the correct word used in the Statista report.

    2. Qarumba

      Re: the other UK chip industry

      I'm sure he's right, it's the chips industry. The number of fish/black pudding/sausage/haggis suppers added to the frozen chips market must make it nearly as big as the market for chips! No?

  5. Anonymous Coward
    Anonymous Coward

    Dear Rupert

    I generally respect your long pedigree of articles and opinion pieces over the years, but you look in the wrong direction if you just compare the utility of an architecture to that of one capable of running a PC or flagship phone. There are trillions of MCUs running everything else for which Risc-V will be the deal platform. It can be argued that the evolution of this architecture is outstripping the largely add more local memory or make features smaller tinkering "evolution" of mainstream OS capable processors. We are looking forward to more innovative MCUs for use in smarter and innovative out of sight devices and it's going to be Risc-V that fills this environment. GM

  6. herman Silver badge

    Embedded FPGAs

    RISC V chips are good for FPGAs used in complex interface devices in embedded systems.

    1. Justthefacts Silver badge

      Re: Embedded FPGAs

      Why?

      Given that most of the FPGAs of the size that you’d want a CPU core on, ship already with a hard ARM core on, for free. You either use it, or it sits idle. And it runs at 3x clock speed a soft core will.

      Or, you buy a cheap teensy little FPGA that doesn’t have a hard embedded CPU on it, and just use any of the over a dozen free CPU soft IPs that you’ve been able to download for over twenty years now. Microblaze, openMIPS et al.

      I know bad engineers tend to put this shizz onto their designs to pad their CVs out with buzzword bingo, but honestly it’s not doing you any favours.

      1. werdsmith Silver badge

        Re: Embedded FPGAs

        But putting a soft core on an FPGA and making it do something is fun, well it is for me anyway.

  7. Anonymous Coward
    Anonymous Coward

    still not getting it

    The commentards that is.

    Rupert is 100% on the mark in this article.

    It doesn't matter how much pro-RISC-V propaganda various outlets - including the RISC-V Foundation - spit out to the public. They're playing to the gallery that can't tell the difference between an ISA Spec PDF and a chip.

    Extolling the virtues of RISC (as in Reduced Instruction Set) ISA's is nice, and might make one feel good, but it's not a chip that can be seated on a board and make a decent performance server.

    For RISC-V to provide a viable and competitive alternative to x86_64, it needs to prove itself as realized silicon, not just PDF specs on GitHub, or atta boy enthusiastic statements.

    To date, no-one has produced a viable RISC-V alternative to Xeon. Play-toy SiFive development boards don't count.

    I have nothing against RISC-V. It's an updated version of MIPS, or SPARC without register windows or delay slots. Yay!

    RISC-V's problem is that it's playing second fiddle to ARM, and ARM has been playing second fiddle to x86_64 since ... 1999?

    ARM has already sold their ARM64 licenses to just about everyone interested in buying one. And? Where are they today, and where are they going? Nowhere, that's where.

    If this RISC-V alternative to x86_64 ever materializes, it will be in the US, and maybe the EU. Not in China, not in Russia and not in India. The only thing these three countries have proven so far, in terms of silicon engineering, is that they massively suck at it, sanctions or no sanctions.

    The fundamental problem being: no-one - not even ARM themselves - has managed to make a compelling case why ARM64 is worth tossing out hundreds of millions in x86_64 infrastructure investments, and replacing them with ARM64. So, where's the corresponding case for RISC-V?

    1. Anonymous Coward
      Anonymous Coward

      Re: still not getting it

      The fundamental problem being: no-one - not even ARM themselves - has managed to make a compelling case why ARM64 is worth tossing out hundreds of millions in x86_64 infrastructure investments, and replacing them with ARM64.

      I think that the work that Apple is doing proves that Arm can be a viable alternative to X86_64.

      I'm sure that they have only just started. Others in the industry are working day and night just to get where Apple was a year ago.

      IMHO, Intel is in cahoots with Microsoft at the moment to keep Windows/ARM out of the hands of Joe Public. Sooner or later manufacturers who currently use X86 will see the writing on the wall and want to go where the money is and that is not X86_64.

      1. Anonymous Coward
        Anonymous Coward

        Re: still not getting it

        > I think that the work that Apple is doing proves that Arm can be a viable alternative to X86_64.

        Yeah. Where? On mobile devices, i.e. smartphones or laptops?

        Nobody cares. Apple is an outlier in the laptop/desktop market (around 14% market share) and marginal in the smartphone market (around 15% market share). They've been stuck in that percentage for more than a decade:

        https://www.statista.com/statistics/576473/united-states-quarterly-pc-shipment-share-apple/

        https://www.statista.com/statistics/216459/global-market-share-of-apple-iphone/

        Which is why they have to keep raising prices and keep increasing the bullshit decibel level.

        As far as smartphones are concerned, that market already belonged to ARM anyway, with or without Apple.

        How and where, exactly, is ARM competing with x86_64 and Xeon? And where, exactly, does RISC-V fit in this picture?

        1. werdsmith Silver badge

          Re: still not getting it

          Yeah. Where? On mobile devices, i.e. smartphones or laptops?

          Nobody cares. Apple is an outlier in the laptop/desktop market (around 14% market share) and marginal in the smartphone market (around 15% market share). They've been stuck in that percentage for more than a decade:

          Since M1 arrived, Apple have had a jump in sales of Macbooks because people like the idea of longer battery life and cooler running devices with increased performance.

          If you consider that Apple operate in the premium laptop sector only, their market share in that field is much higher. They don't make high selling cheapo units, which boosts the market share of the Windows vendors.

          1. Anonymous Coward
            Mushroom

            Re: still not getting it

            > They don't make high selling cheapo units, which boosts the market share of the Windows vendors.

            I don't think you understand statistics at all.

            1. werdsmith Silver badge

              Re: still not getting it

              I don't think you understand statistics at all.

              I don't think you like these statistics at all.

              1. Anonymous Coward
                Facepalm

                Re: still not getting it

                > I don't think you like these statistics at all.

                Here you go, Genius Math Boy: According to Statista, for every 100 laptops/desktops/PC's sold in the world, Apple sells a bit less than 15. The rest of a bit more than 85 are sold by someone else. I.e. not Apple.

                It doesn't matter if they are high end, low end or whatever end. 15% is Apple's market share, and has been Apple's market share for more than 12 years. 15% is marginal.

                So, quit ARM-gasming over Apple's relevance in this market. It's nil. Apple is relevant to Apple fanbois. In reality, it's about as relevant as Microsoft's Surface.

                According to Statista, Apple ranks third in market share behind Dell and HP:

                https://www.statista.com/chart/26039/most-popular-laptop-brands-us/

                If you ask Gartner:

                https://9to5mac.com/2021/07/13/mac-market-share-q2-2021/

                Apple has only 8.5% of the laptop market. IDC puts it even lower, at 7.4%. IDC is quoted in the same article.

                These aren't numbers to brag about. But I am certain you can make up your own numbers in your head, and go with those.

                Still don't see how this relates to RISC-V in any way.

                How did you do in High School math? Not so good I take it.

                1. Justthefacts Silver badge

                  Re: still not getting it

                  I’m not so familiar with the laptop market. But in smartphones, Apple also have 15% market share. However, they make *50%* of all the profit. At the other end, the “bottom” 50% of the smartphone volume market….make *5%* of total smartphone profit.

                  And worse still, if you look at the bottom 25% of smartphone volume….they are operating at *net loss*. That’s net loss, despite the fact they have zero R&D. It’s not even remotely sustainable for a Western company, and not intended to be. That’s China subsidising it’s local market. Capitalism doesn’t have to work the way you like.

                  If you *want* the EU to be in the same position sponsoring all its manufacturers to make net losses on RISCV, go right ahead.

      2. oiseau Silver badge
        WTF?

        Re: still not getting it

        ... Intel is in cahoots with Microsoft at the moment ...

        At the moment? 8^D !!!

        Not for the past 30+ years?

        Really now ...

        O.

    2. sianag

      Re: still not getting it

      Who cares about x86? I have never not once designed a product with an x86 in it, nor have most EEs.

      The most important chips are right where ARM started. Deeply embedded. No matter what you design, even say a PC peripheral or a gadget of some sort, odds are, it will need a microcontroller. RISC-V offers a price advantage and an extremely competitive performance and power efficiency. Furthermore if you need some sort of special fixed silicon integrated, it's much easier to commission RISC-V embedded to control it as opposed to anything else.

      ARM has fought its way up from the tiniest of devices, from being the heart of Gameboy Advance and Psion Revo, from being just thrown in as control logic into USB WLAN Adapters and the like, USB memory sticks, SD-Cards etc. This is the big field that RISC-V is likely to conquer. Odds are, you have dozens or hundreds of ARM devices in your possession without you having the faintest idea, and this is the sort of market that RISC-V targets aggressively.

      1. Justthefacts Silver badge

        Re: still not getting it

        There’s engineering opportunities all the way up the capability stack, but I agree the majority of the revenue comes from the broad base. If that’s “most important”, so be it.

        But where on Earth are you getting “ RISC-V offers a price advantage and an extremely competitive performance and power efficiency.”? That’s demonstrably and factually opposite to the truth on all three counts. Why are you saying these things? Name *one* RISCV chip that wins against its specific competition at that market point. One.

        “If you need some sort of special fixed silicon integrated”. Like what. *Exactly*? This is *incredibly* rare. At each level of the embedded market, there’s already a solution that does essentially what you need, and very efficiently. If you just want a control logic, there’s MCUs that are cheap, efficient, and the software will wiggle the pins appropriately.

        There just *isn’t* a use-case for that much specialist silicon, where you want to tightly integrate it into the CPU. Because if you have a special application that needs to, say, exponentiate the Wronskian of a 14-dimensional matrix of quaternioms very efficiently, guess what you do? Same as we’ve been doing for twenty years. Code up a silicon accelerator IP block in Verilog, hang it off the memory bus, and for extra credit bodge on an RDMA so it doesn’t tie up the CPU to load the data. Leave the CPU alone. The rationale for integrating it as an *instruction* into the CPU is incredibly weak.

        1. sianag

          Re: still not getting it

          Let's see. You know Gigadevice, GD32F103... the STM32 clone. This same company released a set of pin compatible and equipped with similar peripherals GD32FV103 which is a RISC-V, they're offering it at, not sure now, at one point i looked and it was like half the price, and it is a little faster and substantially more power efficient. However i am well aware that the improvement is IN SPITE of RISC-V rather than due to it.

          Espressif is also going RISC-V in their new low cost offerings, nothing changed much on the power front, and the performance got a very modest nudge up compared to their older Tensilica Xtensa based offering. It just stands to reason that there is no inherent disadvantage for these from adopting RISC-V, both being of the MIPS legacy, but they got a better deal on the core IP.

          The typical advantage of MIPS designs has been specifically that they have a coprocessor interface - that custom DSP instructions and such can be lower-latency since they don't need to traverse the memory bus at all, it's all just register transfers. Depending on core design, they can also take advantage of existing pipelining infrastructure etc. And these are never particularly wild things, maybe you just need to accelerate computing checksums or something of a kind and you want to be able to make do with the smallest and least complex addition to an existing design. Or look at PS1's GTE for a well-documented practical example.

          Chips already exist to satisfy all necessary custom functions? That they kinda do, but they also get refreshed once in a while, for one reason or another, they won't be up to date forever. Say NVMe flash controllers will need to be updated for next PCIe standard and the chip companies will go core shopping again, and at least Samsung seems to have chosen a RISC-V implementation for something like that, for the next round.

          Things that end up winning aren't necessarily the best things, the most sensible things. A lot has been said about inefficiency of x86 as an ISA, and yet on the high end, it's fine, it's not what's holding them back. And ARM? Nobody cared much about them at first, at least until they grew a Thumb; and at that point, they became Poor Man's SuperH, and that's when they started to matter. RISC-V is likely to have a bright future ahead of it as a Poor Man's architecture. It will be taken as an opportunity for new entrants to compete with the established players, like we're likely to see something like Poor Man's PSoC - patents are running out aren't they? Lack of instruction set license costs, a large competitive field of IP suppliers pushing IP costs down, lack of dependence on a single entity for the future of the product lineup, and the cores, they're good, they're fine.

      2. fg_swe

        Japanese Cars Full of PDP11s

        https://de.wikipedia.org/wiki/Renesas_H8

  8. Anonymous Coward
    Anonymous Coward

    am I the only one here who wrote RISC asm in the 1980's?

    I looked at the RISC V instruction set and architecture and its - who cares. This stuff was already obsolete by the late 1980's. The ARM2 of 35 years ago was a much more capable and useful processor. As was the MIPS R2000 I look at the instruction and architecture design decisions in the RISC V and I get the overwhelming sense that no one involved actually knows how to design modern processors. It looks like someones grad school project. From 1986. Which ignores a lot of the lessons from the 1970's. Patterson must have forgotten what he wrote 30 years ago. I see he was involved. I guess Hennessy must have been the one with a clue.

    Maybe for custom SOC embedded products there might be a (small) market. But for everyone else its software, tools, and infrastructure. Who cares if its "Open Source". Just another bloody stupid processor.

    1. Anonymous Coward
      Anonymous Coward

      Re: am I the only one here who wrote RISC asm in the 1980's?

      > I look at the instruction and architecture design decisions in the RISC V and I get the overwhelming sense that no one involved actually knows how to design modern processors.

      I wouldn't go as far as state that no one involved actually knows how to design modern processors -- RISC-V is David A Patterson's brainchild:

      https://en.wikipedia.org/wiki/David_Patterson_(computer_scientist)

      It started out as Berkeley RISC.

      But yes, RISC-V is definitely a child of the early-to-mid-80's RISC designs, with vector instructions added on top, and not much more. Yawn.

      I wonder if RISC-V would have gotten this much attention if it was called MIPS-V instead. I'm betting on no.

      In fact, the RISC-V designers are still arguing whether to add direct bit manipulation instructions to the RISC-V ISA, instead of relying on the old-school RISC shift-and-mask sequence, which is always expensive. In my view, that's a no-brainer, but the current state-of-the-art in RISC-V is shift-and-mask. Hence my earlier comparison with MIPS or SPARC.

      Anyone who thinks MIPS or SPARC are a modern, state-of-the-art RISC ISA, please raise their hand.

      1. fg_swe

        Japan SPARC

        ...was good enough to compete with anybody else.

        Seems they still ARE: https://spec.org/cpu2017/results/cpu2017.html

        They now switched to ARM ISA for their high performance computing, but I guess this is more a matter of fashion.

        [No, I never worked for Fujitsu and have no economic or other relation, except a weak spot for japanese engineering]

      2. Anonymous Coward
        Anonymous Coward

        Re: am I the only one here who wrote RISC asm in the 1980's?

        I did mention Patterson. Read the book when it came out in the early 1990's and though, finally, someone who had a clue. But based on RISC-V, it must have been Hennessy who wrote the good parts of the book

        Here is the thing about RISC. It failed. All the huge claims made in the 1980's for RISC proved unrealisable. What happened was some of the more practical bits of RISC were folded into CISC's and those instruction sets have sailed on successfully ever since. The psychotic mess that is x86 was made super-scaler by the great AMD RISC'ish hack. PPC was about as RISC as the 68060. And so on..

        MIPS was actually quite a nice architecture. At least for us lowly asm writers. SPARC was a mess. Register windows? Were nt they an idea that failed on a Burroughs mainframe back in the 1960's? Why are you trying to solve a problem with the instruction set that can be solved by the compiler. If the compiler can do instruction scheduling it can sure as hell do local variable liveness analysis. Actually they all do that as part of the optimization pass.

        But as I look at the RISC V instruction set I just shake my head. So the 6502 and Z-80 I was writing for in 1978 had richer and more capable instruction sets. With less than 10K transistors..

        There again it took x86 almost 30 years to get PC relative addressing. Something almost everyone else had since the 1970's. And the 1960's on the Big Iron.

        And so it goes. Another decade, same old stupidities from decades past.

  9. Anonymous Coward
    Anonymous Coward

    Just like linux will never compete

    Just like for linux to be successful it had to generate cash like MS Windows. Never mind that the vast majority of servers and containers running, well, anything these days are linux, it just can't compete. Will RISC-V be lucky enough to fail at the same scale?

    1. 3arn0wl

      Re: Just like linux will never compete

      Happily, I think you've put your finger on the nub of the issue - what the other "commentards", including the author of the piece, haven't managed to grasp - or more realistically, are choosing to ignore: it's not a competition. It's not all about hyperbole of the biggest, or the fastest. RISC-V was an academic paper outlining what, in their - Turing-Award-Winning - expert opinion, is the best way to do processing. (And it matters not one jot if their opinion is not much altered since the '80s) . Moreover, anyone is free to add functionality they feel the need of - and to share that with others, or not.

      RISC-V will be available to use freely, for as long as anybody sees the benefits of it. The RISC-V foundation grew by 130% in 2021. It would seem that companies, including Intel, can see the benefit in it.

  10. quadibloc2

    Bad Advice

    This article is giving the RISC-V very bad advice, despite the fact that what it is saying is very true.

    Yes, RISC-V needs to find a niche where it is "first" instead of a runner-up.

    But the way to do that is not to decide on a niche, and then pursue it, and fail because you picked the wrong one.

    The niche where RISC-V will matter... needs to reveal itself. One possibility, for example, might be scientific workstations, because the RISC-V standard includes some nice vector instructions, so some RISC-V processor might come along that out-competes the NEC SX-Aurora TSUBASA for that market. But that will happen if someone makes the investment.

    1. amanfromMars 1 Silver badge

      Great Advice

      For RISC-V to really prosper, it needs its own sector. AI/ML could be one;

      Spooky virtualised paramilitary/augmented remote mercenary force is another one, which actually might also be a not too dissimilar one.

      And Uncle Sam [via the Pentagon] are ACTively openly seeking market help for their own admitted deficits/exploitable vulnerabilities in fields which are both strategically and tactically vital to the maintenance and retention of national security and social advantage/status quo leadership, mentoring and monitoring.

      Here be two pieces of very recent evidence illustrating those facts ...... https://www.nationaldefensemagazine.org/articles/2022/5/16/special-operators-must-learn-to-exist-without-tethers ....... https://www.nationaldefensemagazine.org/articles/2022/5/16/modernizing-sof-acquisition-quietly ........ with one of them prompting this surreal alien but nevertheless sincere response with a very specific RISC instruction set to initiate a novel guaranteed strangely successful strategy ..... and please, let's not be having any of that MRDA lark,... He would say that, wouldn't he. It is what it is.

      amanfromMars [2205161017] ....... shares on https://www.nationaldefensemagazine.org/articles/2022/5/16/modernizing-sof-acquisition-quietly

      [Thank you. Your comment will be displayed soon after reviewing.]

      James, Hi,

      The "quiet professionals" SOF seek and need for future great and Greater IntelAIgent Games use patiently await and welcome any and all leading requests from them for sensitive proprietary other party information they may hold but which does require their initiating engagement for subsequent JOINT Program Developments.

      The thing to bear in mind always though is, where the West are fearful to tread, go the East to lead ..... and vice versa of course. Such is only fair/fit and proper and to not now realise that great fundamental changes are more than just afoot and highly ACTive has one registered as a mere remote hapless spectator rather than live contributory actor in the Greater IntelAIgent Games being played out for realisation of events via media presentation. .......... which coincidentally is a great program and project for the likes of a RISC-V International global organisation too.

  11. lundril

    It's about Software/Ecosystem

    The reason why ARM never made it for Desktops is Software.

    I actually have a perfectly fine ARM based Laptop; but of course it's running Linux.

    How many people actually do that on a Desktop? 5%?

    People already complain that they can't run Photoshop or Quark Express or whatever on x86_64 Linux, not to forget Microsoft Office.

    Now imagine if you not only switch the OS but also the underlying CPU Architecture...

    The same is true why x86 never made it into the mobile market. You need Android there; Intel got the OS working on x86; but then there are probably millions of Apps which contain NATIVE ARM code and these can only be ported if the App developer recompiles... or if you emulate ARM CPUs which usually results in really bad performance (especially performance per watt).

    Apple in contrast controls Hardware AND Software. So they can and did switch between x86 and ARM.

    They could also switch to RISC-V if there is a good reason for it (I can't see one right now; the license fees for the M1 are already payed I guess).

    The only market I see for RISC-V is embedded devices and maybe Servers (like the Amazon Graviton); in both cases you either run your completely customized Software or you use open source Software, meaning recompilation is not an issue.

    For everything else the dependency on already existing Software, which you cannot recompile, is too high to break into a market.

    Of course if China is FORCED to abandon x86 AND ARM, then China might have to adopt an alternative...

  12. Anonymous Coward
    Anonymous Coward

    It’s fine…

    The point is to have an Open ISA. So…. probably at some point ARM will be forced to open their ISA.

    Even if RISC-V doesn’t win…. Open processors will.

  13. NeilPost Silver badge

    Intel Schadenfreude

    The bitter irony left unsaid in the article is that Intel were the market leader in ARM processors with StrongARM and XScale which they flogged for a pittance to Marvell in 2006 to double down on ultimately fruitless x86 ‘low power’ aka Atom.

    XScale running many a Dell Axim, Compaq IPaq, HP PocketPC Smartphones and other mobile devices in the very months before Apple bonded MP3, PDA and Phone together and the true Smartphone was born.

  14. msobkow Silver badge

    I think the author of the article is a tad "old school" to appreciate the shifting undercurrents of an open architecture. Much like Microsoft 30 years ago who pooh-poohed open source and insisted no one would ever trust it, Intel and Arm are in for some serious surprises.

    You don't start off with a multi-billion locked in consumer market like China's without making a major splash on the industry shares, and China has very vocally committed to RISC-V.

  15. BlokeInTejas

    At last, some RISC V Sanity

    I think the article is a useful breath of fresh air.

    RISC V has exactly one unfair competitive advantage - the hype surrounding it. Technically, it's yet another '80's RISC, perhaps with a bit more forethought given to instruction encodings, and without a delay slot. Ho-hum. Nothing wrong, but no unfair technical advantage whatsoever.

    Recall that when it started, it was said to have been brought into existence to simplify the lives of computer architecture researchers. Prior to RISC V, such researchers had to cast about for an architecture which they could modify to explore and then show the fruit of their research. There really weren't many available - MIPS was an obvious choice because it was 'open', at least for a while.

    But the problem for the researchers was that before they could get their brainchild to do useful work, they had to create an OS for it, and a compiler, and a library. And these had to be as good - more or less - as the existing ones for existing architectures. This meant that interesting research could spend only a small fraction of the research effort on the neat new stuff - they also had to do all this adaptation/redevelopment work on the software infrastructure.

    The big thing that RISC V promised was that - if everybody based their research on RISC-V - then there were no architecture licensing issues - it was free and open as an architecture; and that the infrastructure work of compilers and OS and library etc would be done once, and everybody could share. Thus - higher quality research. And for that use, being "just a MIPS without a delay slot" was an advantage - adding neat new architecture to a vanilla old machine was just the playground one would like to show how much better your architecture improvements made things.

    But now, folk are reckoning to get rich selling IP, or maybe SoCs, or maybe both based on designs with RISC-V cores inside them.

    These products have no inherent advantage over (say) equivalent ARM-based products. Anything RISC-V in silicon will have exactly the same supply chain problems as an equivalent ARM-based SoC. Or a MIPS-based SoC. It's fab capacity that's in short supply. Not .pdfs

    And there's nothing in the RISC V architecture which makes it better in silicon than an equivalent ARM. There's no magic.

    So, yes; RISC V is a fine thing for academia. It's unencumbered. You can modify it if you insist. There's an OS port or two. llvm works. You can experiment using it in systems which choose to do things differently - where the fact that it's a RISC V is largely irrelevant, but having a free and unencumbered processor is highly convenient.

    But for products? For profitable business?

    As the article says, entirely unclear how RISC V leads to business success for RISCV IP and silicon vendors. Doesn't mean it won't or can't happen; just that its completely unclear.

    1. fg_swe

      Well

      Maybe it is of little use to compete in the traditional applications. But now imagine a very low power processor that will be powered by a small solar cell on the same chip. A device that costs 3cent/piece and can be attached to each tree in a forest, monitoring its vital parameters. It would run at very low speed and consume very little power.

      Maybe this is a contrived example, but you get the idea.

      There might be unheard-of niches were even tiny license costs are too much. In this space RISC V might be the right solution.

      1. 3arn0wl

        Re: Well

        The RV32EMC ONiO.zero - being tested atm, and coming out next year - harvests enough power from the EM Spectrum not to need a power source! What if a simple circuit using an ONiO.zero was incorporated into consumer devices, removing the need for standby power... a green dividend in energy saved worldwide.

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon

Biting the hand that feeds IT © 1998–2022