RISC-V is making moves, but it has work to do if it wants to hit the mainstream RISC-V has been talked up as a challenger to Arm and x86, offering an open royalty-free architecture that promises flexibility and innovation without licensing costs. But for all the noise, you're more likely to find it buried inside IoT gadgets and obscure embedded systems than powering anything that'll typically grab a …
When it comes to the infrastructure - and I'd argue that the ISA falls under that - you should never bet against Free.
Free has an insidious appeal that helps it spread remarkably well. Especially if there are software shims available - which there will be in this case.
If you'd told me in 2000 that Linux would be installed on billions of devices and powering much of the modern world,, I'd have looked at you like you'd drunk a bit too much. Whether what you'd drunk was koolaid or beer would be my next concern. Remember, in 2000 IBM was touting running Linux on your mainframe, and it was most successful as a webserver - where it was out of sight and out of mind.
And yet here we are.
ARM has only just started to look at putting a proper Plug & Play infrastructure for desktops in place. I suspect that RISC-V will follow within a couple of years, as it seems to be on an accelerated track that replicates what ARM has done in the past.
So would I bet on RISC-V laptops being shipped by a big name like Dell within the next five years? Hmmm.. probably not.
But would I bet against that happening? Definitely not.
Because I think it's more likely to happen than not happen, and the awkward variable for the bets is simply the timescale involved in the bets.
Never bet against free is certainly good advice, but then it can be pretty difficult to distinguish between "free" and "very cheap", especially as the "free" part of Risc-V gets lost on the fact that one still has to pay for the silicon regardless of the cost of the ISA.
ARM has been uncannily good at pricing its products so that it's other functions - i.e. controllers of what the ISA is - become important. If you write software for a particular flavour of ARM, you know it's going to work on any chip that claims to implement that flavour. It's a bit more complex on Risc-V, but that seems to be enough to keep ARM firmly in play.
Plug and Play; there's ARMs aplenty that support PCIe, and I presume that firmware and OSes that support that on ARM are perfectly capable of detecting what devices are actually plugged in. It's been a mixed bag in terms of non-PCIe peripherals on the chip, but that's seems to have been getting sorted out with a lot of work in the Linux kernel project aimed at simplifying how ARM peripherals are described to a kernel.
People compare it to open source software like Linux, but it is nothing like that. With software I really am getting something for free, assuming it pretty much works out of the box. I can download Fedora, install it on my PC and it costs $0 versus costing whatever to install Windows.
The royalties for SoCs are tiny compared to the cost to design and fabricate them - costs that apply just the same to RISC-V. You don't have any costs like that associated with software, even in the days when you might need to have software delivered on a CD the "fabrication" costs were like a buck. A leading edge SoC will cost perhaps $50 just to have TSMC fabricate for you. Even if you're paying a dollar to ARM in core royalties (pretty sure it is less than that in reality) that's a tiny tiny potential savings.
Qualcomm is selling their latest SoCs for $200+. That's not because they're ARM that's because they had to pay TSMC for fabrication, they had to amortize their design work, they are including a 5G modem, they're charging you to license a bunch of their patents, and they are taking a generous cut of profit on top of that. If they switched to RISC-V, they'd charge the same for it but pocket that extra dollar or whatever it is they're paying ARM themselves.
Where RISC-V can make a difference is in the really low cost embedded market. If you have some little chip made with older technology that costs only a buck to fabricate saving a dime or two in ARM licensing makes a big difference in your profit margin. People need to quit dreaming that RISC-V will come to smartphones (other than maybe the sub $50 market) let alone PCs. It will never happen, because the economic incentives aren't there unless ARM gets super greedy and increases their licensing costs by 600%.
RISC-V isn't about saving a few cents on a chip, it's about 100 companies having freedom to innovate and compete, not just Intel. AMD, Arm, Qualcomm, and Apple.
> If you have some little chip made with older technology that costs only a buck to fabricate saving a dime or two in ARM licensing makes a big difference in your profit margin
There are very popular RISC-V chips that retail for $0.10 in qty 50 (i.e. you have to invest $5 to get that price). Search for CH32V003. There is a retro-PC kitset using that chip -- with more capability than a ZX81, twice the RAM, 100x the speed -- with PS/2 keyboard and VGA connectors that sells for €1 for the whole computer, not just the chip.
https://www.olimex.com/Products/Retro-Computers/RVPC/open-source-hardware
https://www.youtube.com/watch?v=dfXWs4CJuY0
> People need to quit dreaming that RISC-V will come to smartphones (other than maybe the sub $50 market) let alone PCs. It will never happen
Never is a very long time. Both of those are going to happen by 2030.
You can already get multiple laptops using RISC-V processors, including a main board for the high quality Framework Laptop 13. At present they are slow, a similar speed to a late Pentium III or a very early Core 2 (e.g. original MacBook Air). They'll be hitting mid-life Core 2 Quad speeds sometime this year, and early Core i7 (maybe Sandy Bridge-ish) next year, maybe Zen2 / Apple M1 in 2027.
>RISC-V isn't about saving a few cents on a chip, it's about 100 companies having freedom to innovate and compete, not just Intel. AMD, Arm, Qualcomm, and Apple.
That can be looked at in a different way. ARM providing a chip manufacturer a ready-to-go core (which implicitly includes all the software dev tools for that core - complete, finished, supported, documented, etc) allows the chip manufacturer to innovate in ways that matter more to their customers; namely, what sort of peripherals, what memory there is, etc on the device.
Ultimately a core is just a core, a way of running software, and a ready-made core that's extremely well understood is a valuable thing for a chip manufacturer. They can more cheaply participate in the market.
Sure, Risc-V allows chips manufacturers for no license fee to start tinkering with the ISA. However, that's a very complex undertaking. To differentiate themselves in any meaningful way they've got to take on an awful lot of extra expertise and a whole lot of extra work. Whether that extra effort amounts to a world-beating difference is very unlikely.
>You can already get multiple laptops using RISC-V processors, including a main board for the high quality Framework Laptop 13. At present they are slow, a similar speed to a late Pentium III or a very early Core 2 (e.g. original MacBook Air). They'll be hitting mid-life Core 2 Quad speeds sometime this year, and early Core i7 (maybe Sandy Bridge-ish) next year, maybe Zen2 / Apple M1 in 2027.
Well, to make Risc-V performance competitive with today's best is going to mean booking a large production run on TSMC's line, with a Risc-V design dotted with pipelines, caches and memory controllers sufficient to exploit the performance of TSMC's finest transistors. That's not going to happen unless some major player decides to abandon the ARM ecosystem (which has served them very well) and go it alone. Thing is, they'd also have to undertake to persuade peripheral vendors to re-write all the device drivers for the peripherals they want to glue into systems. That might be non-trivial, if the peripherals are not their own and are not blessed with OSS drivers (see a lot of WiFi, graphics, touchscreen devices). And then they'd have to persuade software / application vendors to support it too...
That's kinda ARM's strength. They've made it easy for everyone to use ARM, and there's a lot of OSS and proprietary software inertia (especially in the mobile space). They don't bight the hand that feeds them - in fact, they barely nibble - so there's not a lot of motivation to divert that inertia.
And when you say innovate….you mean add non-standard instructions? The RISCV argument *doesn’t make any logical sense*. Either you want it the same, or you want it different. If you want it different…..then it’s balkanised and proprietary, the exact opposite of what you claim to believe in. If you want it the same, then it’s just an argument that you want to get somebody else’s design work, but not pay for it.
The ISA doesn’t really cost anything. If you like, I will sell you an ISA, and the full rights to it, in perpetuity, for 50 quid. The intellectual input in an ISA, for somebody who has done this for a living, is less than the work involved in designing a pretty pattern for a T-shirt. The work is in clever implementations, and that is very much not free, whoever does it.
I love the way that you proudly boast that a RISCV chip can be had for $0.10, and you think that is cheap. But there are dozens of non-RISCV CPUs, considerably more capable and powerful, with proprietary ISAs that cost *5* cents. There’s several at 3 cents. *More than half the price of the cheapest RISCV CPU* is the implied cost of complying to some RISCV philosophy that’s completely unnecessary. The idea that the ISA must scale across all price points is just foolishness.
Even the M cortex series from ARM has excellent ecosystem from Keil.
The day ARM makes it very cheap or even free, RISC-V as an application processor will face tough times.
Conversely, if MS had kept the price of Visual Studio low, Linux and Python would have had more challenges in establishing itself.
Cortex M1 and M3 have been free for a while for use on FPGA for pre-production and education etc, since 2018 I think.
But I think for anyone thinking of using these cores in a commercial product, the license fee isn't the major cost. Somebody has to implement the cores into a device, then they have to be fabricated and packaged. None of those things are free.
> But I think for anyone thinking of using these cores in a commercial product, the license fee isn't the major cost.
Indeed. The processor cost is X, and some part of it is silicon and some is IP.
At least, that's true as long as the license fee remains flat for a given part. If ARM starts demanding a percentage of the sale price of the *final product*, as has been mooted, then suddenly the equation changes.
Anyone can buy a $0.80 RP2350A chip, or a $5 Pico 2 board, from Raspberry Pi, install the free SDK and tools, be up and running on its RISC-V cores within minutes. Or they can stick to using the ARM M33 cores and the same application source code.
In fact it's the perfect way to discover platform and architecture doesn't much matter in terms of software, apps or development.
Silicon is where it's at. RISC-V may be playing catch-up due to being late to the party but that's neither surprising nor insurmountable.
That "Google retreat" on Android is, for me, over rated as a setback, it is just an inconsequential part of the journey. It doesn't mean Google have entirely abandoned RISC-V.
Everybody's looking at AI, servers, PC's and smartphones, but RISC-V is already ARM's lunch in the MCU space. This is bound to have an impact on its bottom line in the near future.
Higher end computing devices with RISC-V will come in the future, probably driven by the Chinese who are desperate to free themselves of the government controlled ARM and X86 ISA's.
Cheap silicon is great and there are definitely enough people (and companies) interested in using kit like various fruit & vegetable pi's and whatnot for there to be a market.
What's lacking IME are various sets of standards around the architecture and platform, such that systems designers and integrators and, ultimately, sellers and resellers, can put together a package which is attractive enough to end-consumers of many sizes and shapes.
One of the issues with ARM and wider adoption is every new board is essentially a new bring-up project for systems folks; both software and hardware. There's some homogeneity in the Raspberry Pi family, but even so-called mundane things like cases and power supplies aren't necessarily compatible across the product lines. RockPi et al might leverage RaspberryPi peripherals occasionally, but there are no promises around e.g. I/O ports cut-outs, board stand-offs, etc.
To be sure, x86 has issues in these areas sometimes too, but e.g. it's a fair bet if you've got an ITX motherboard that it will fit in most any ITX case you're likely to buy.
Similar things can be said for operating system and software support. Again, there's often an arms race of sorts for Linux (and BSD) kernels to implement support for Intel & AMD's latest processors and other chipsets. But ARM seems to be another kind of different in this area, which is why it's not uncommon to see different OS images which are specific to RPI, ROCKPI, and others.
I could be wrong, but I suspect some of this applies to both desktop and server markets. The companies deploying ARM at scale may be focusing on a specific server platform (Ampere et al?) and likely have enough in-house systems expertise to deal with whatever OS and software integration and support are needed. Whereas x86_64 platform solutions are more likely to be off-the-shelf today.
I think the ARM ecosystem (lacking a better term) may eventually get there, and it'd be neat if RISCV avoids some of the "every board is a new snowflake" symptoms of predecessors.
The one overriding advantage that RISC-V has is that its unencumbered by political decision making in the US. Our government has looked into restricting the use of the RISC-V ISA by the Chinese but all they would have discovered is that abstractions like ISAs don't respect borders, sanctions, entity lists or tariffs. (We should have learned our lesson from encryption where our monopoly over encryption through the widely used DES algorithm rapidly eroded when we tried to control who could use the algorithm and how they could use it.)
I think our popular model of 'computing' being a laptop running Windows is doomed. Its why I don't see Dell or anyone like that bringing out RISC-V PCs. They key to the future is in devices like Chromebooks, devices that won't necessarily run Google's Chrome (encumbered by both government and the dead weight of adware) but will serve the vast majority of everyday needs. Put simply, the bottom will just fall out of the market.
I think we are in for a sudden tipping point in both software and hardware, like the polarity of the earth's magnetic field every so often. Linux will be under threat by modern systems - perhaps built on "safe" languages - like Redox and possibly Genode. This will in turn open the door for refreshing the rest of the "stack" and include new architectures. So much cruft has built up over the years that I believe a clean sheet for much of computing is desirable.
The problem with Redox is that the author is taking too much hay on his fork. He's trying to make a general purpose desktop operating system that competes with Windows (which has a 40 year head-start) and Linux (which has a 30 year head start and is finally usable enough for the novice desktop user).
If I were to make an operating system I'd focus on server or embedded, maybe networking routers or signage displays even. That makes it somewhat palatable for a single developer or small group to see it to completion in a workable time frame.
The world of computers is much larger than presented here:
1.) Apple ARM Mx being a highly innovative, high performance, low energy CPUs for notebooks.
2.) The world of automotive/train/aerospace/medical MCUs, which have totally different priorities than "IT" CPUs. Think of highest reliability requirements. They use PowerPC, ARM, Aurix, DEC PDP ISAs, ARM, Atmel, and probably 25 other ISAs.
3.) Fujitsu SPARC and ARM servers, a viable alternative to x86.
4.) Ampere ARM servers, working like a breeze at Hetzner, Amazon and so on.
5.) IBM S/390 still running most of banking, insurance, large government agencies and so on.
6.) Special CPUs like Elbrus and Loongson, which are very real things in the world of olive green. Maybe the Indians have their own ISA in the shadows...
