Slashdot Mirror
Startup Offers A Chip Based On The Open Source RISC-V Architecture (computerworld.com.au)
angry tapir shared this news from Computerworld:
An open-source chip project is out to break the dominance of proprietary chips offered by Intel, AMD, and ARM... A startup called SiFive is the first to make a business out of the [open source] RISC-V architecture. The company is also the first to convert the RISC-V instruction set architecture into actual silicon. The company on Thursday announced it has created two new chip designs that can be licensed... but the company will not charge royalties. That makes it attractive alternative compared to chip designs from ARM and Imagination Technologies, which charge licensing fees and royalties.
One of RISC-V's inventors co-founded the company, and he says that support is growing -- pointing out that there's already a fork of Linux for RISC-V.
One of RISC-V's inventors co-founded the company, and he says that support is growing -- pointing out that there's already a fork of Linux for RISC-V.
But you have to pay a shitton of money to get the license. Well a shitton to a regular person anyway. If you can afford to manufacture one of these chips the license cost is probably a drop in the bucket.
The component part of the licensing, that is. I imagine if you are mass manufacturing a specific device and need mostly *some* CPU functionality for performance and battery life you can avoid paying for the parts you don't need, as opposed to buying "bulk" CPU functionality. So this might be a way to pack more processing power in the device for the same cost. The only question is how the mostly theoretical RISC-V design will hold against the well baked Intel and Arm architectures that have had so many real life special use cases baked into them over the decades.
I see you've been courted by ARM's marketing department. They've been coming to us as well with exciting charts about how affordable ARM licensing is versus the evil expensive RISC-V. But look closely, it's bullocks.
If all you want to do is mark Cortex-M0 chips, then ARM's DesignStart license is cheaper than hiring an engineer to put together a RISC-V. ($40K iirc). Of course ARM makes that up if you manage to ship a lot of M0 units. But if you need a wide range of ARM products, the licenses quickly get more expensive. And sadly ARM's marketing department doesn't tell you their top tier license is 20x more than their bottom end.
E51 Coreplex IP
The E51 Coreplex comes with Native TileLink interface bus. We offer bridges to connect the ports to various legacy bus protocols.
System Port
Supports Burst transactions. Used for accessing main memory and high speed peripherals.
TileLink to AXI4
+$25,000
TileLink to AHB-Lite
+$25,000
Peripheral Port
Supports Atomic Memory Access. Used for accessing peripheral devices and shared memory locations.
TileLink to APB
+$25,000
Front Port
Allows other masters to write into DTIM, ITIM, System Port, and Peripheral Port address space.
AXI4 to TileLink
+$25,000
AHB-Lite to TileLink
+$25,000
Heavy on the "rah rah", light on the details. None of the things they are saying will matter if the chip they produce isn't good. The current chip makers out there make chips that are VERY good for their given purposes, and they have a lot of R&D going in to that. It isn't as though designing a CPU that is fast, efficient, highly capable, etc, etc is some easy feat.
Now maybe these guys did that... but then let's see some info. What are the specs on the chip(s) and what are they designed to compete with? Let's see how it all stacks up. Talking about being "open" and low cost really doesn't matter that much, there are few markets where those would be the primary concerns. Even in things like lower end mobile devices a company will pay more to get a chip that is physically smaller and lower power consumption. So it is going to have to offer something competitive in any market it tries to go in to. What you need varies by market, the primary demands in a mobile chip are very different from a HPC chip, but regardless they have to show how they can at least compete with the existing players.
Any time I see lots of fluff and little numbers... well that makes me think perhaps the numbers aren't all that good.
A regular person won't have a silicon fab in their basement.
If only there was this way to program a gate array in the field.
So, this is on silicon already? What do I have to do to get my hands on a few samples?
It's everywhere and has been around for decades. Anything else that's been around as much and as long, has become a standard and free of any patents. We need more companies putting out Intel-compatible CPUs to break up their monopoly. And discrete graphics boards, as well.
A Pi-like SoC offering low end desktop equivalent memory+i/o capabilities and/or a Desktop compatible chip capable of using standardized bus technologies to interface with expansion cards, peripherals, and memory busses to allow it to compete, even 'unfairly' from a performance/utility point of view, against the modern Wintel desktop PC design, perhaps opening the floodgates for other cpus/systems utilizing standardized busses and expansion cards on a variety of alternative architectures and operating systems. Homogenous architectures on the level of the entire world are an increasingly bad thing, and unless we start returning the power of choice to the people they will only continue to get worse.
And no, none of the current alternatives ARE alternatives because they all make the same bad (for the customer/owner) design decisions which result in personal insecurity in exchange for institutional security theater. Much like transportation authorities and law enforcement have become in the US and abroad.
I have been following the whole RISC V thing for a while now and all I can say is finally!
This is good for security. Finally a CPU that isn't backdoored out of the box. Right now there isn't really a good alternative although there have been many attempts at producing free (as in freedom) hardware.
https://www.crowdsupply.com/raptor-computing-systems/talos-secure-workstation
https://puri.sm/ (not completely free)
Sounds a little RISC-E. Yep, I went there.
This may be useful for IOT sensors. But beyond that, you need 64 bit. You can't make a modern smartphone without a 64 bit CPU. You can't even make an Amazon Echo-like device without 64 bit.
I will wait for the 64 bit version which is arriving "soon".
That MUST be a good solution. You know that spending a lot of money on an FPGA makes the best products.
There's a reason CISC won out in the big arena. These folk are on acid.
This is about security. Cost is second. If I can implement an open source RISC-V processor, don't you get that I could audit every instruction executed, as well as the means to execute those instructions? Most people blindly support a black box known as Intel or AMD to execute their instructions. There could be a dozen things undocumented hiding in that box that you never knew about, whether planted there unintentionally as bugs, or intentionally by governments and large companies.
It's not the only question - there is also the question of "what can I run my existing stuff on that is better than a now very old RISC based system?".
Exactly. With the RISC-V you just have to fab a more or less orphan design yourself and then invent the entire support infrastructure from top to bottom as you do it. As opposed to buying a universally-supported device for $1 or so (at the performance level of the RISC-V), in units of millions if you really need that many, from any random vendor or supplier you care to name. If you're Microchip (PIC32) or Atmel (AVR32) you can afford to do your own custom architecture (and even those are somewhat niche-market things), but for anything else you need to stick to ARM.
Look, I get that RISC-V is geek cool. It's not, and never will be, a commercially viable product, because it's competing with an entire, vast ecosystem, not just one chip sitting in isolation. Let's not pretend that it's anything other than a cool geek toy, and we can at least have an honest discussion on the topic.
The point of RISC-V is not to come out with a standalone chip. The biggest market is replacement of existing cores in SoC designs. If you're already making a SoC, switching out the core is not a huge complicated ordeal.
A perfect application would be something like the ESP8266 WiFi module https://en.wikipedia.org/wiki/...
These modules sell for less than $2 a piece on AliExpress. I'm sure the manufacturer does not want to pay $1 in royalties for the CPU core. There are many more of those kinds of IoT devices that need reasonable processing power at minimal cost. The RISC-V would be a perfect application for those.
They're not paying $1 per IP core, that's why the whole module can sell for under $2. They'll be paying some insignificant percentage that's lost in the noise. It's an irrelevant amount. What they're getting in return is a complete support ecosystem that lets them build a module that sells for under $2. This can never happen with RISC-V because the licensing cost is an infinitesimal fraction of the total cost. Sure, if you state the licensing as being $200K then that sounds a lot, until you spread it across 2 billion units shipped (that's not the ESP8266, that's the Tensilica Xtensa CPU which is the bit with the ARM license). That's noise, you don't even see it in the BOM.
Now compare that with the cost of fabbing your own RISC-V (or at least paying to have it fabbed), or paying some other boutique place who's getting it fabbed. This can never, ever compete with existing devices because it's not competing with the hundredth-of-a-cent licensing costs per ARM core shipped, it's competing with tens of billions invested in making ARM devices as cheaply as possible.
... CPU cores typically are publically described in minute detail. After all people need to directly write software for those...
Today the problem lies in proprietary hardware. Hardware for which you cannot write a decent driver as there is no public documentation available. That's the problem with modern SoCs, and that's why the mobile operation system scene is so dead right now.
Cost is second. If I can implement an open source RISC-V processor, don't you get that I could audit every instruction executed, as well as the means to execute those instructions?
If cost is second, you can also get an source license for an ARM core, and audit every instruction.
the Tensilica Xtensa CPU which is the bit with the ARM license
The Tensilica CPU is not an ARM. It is presumably cheaper, but not free, and which burdens them with the cost of learning a relatively unknown CPU architecture. If you're going to take that cost, you might as well drop in a RISC-V core, and pay nothing, plus you get to benefit from the growing open source infrastructure around it.
Now compare that with the cost of fabbing your own RISC-V
The company that makes the ESP8266 is already fabbing the SoC, so there is no extra cost for the RISC-V.
it's not competing with the hundredth-of-a-cent licensing costs per ARM core shipped,
ARM charges 1.2% of the chip price for a Cortex. So, for a $1 chip, that's 1.2 cents. May not seem like a huge deal to you, but apparently it mattered enough not to get an ARM.
Ah, yeah, sorry, shouldn't post at 3am :-). Two more comments and then I'd better get some sleep, unless they can magic the masks and other components out of nothing, the cost of creating RISC-V stuff is going to be considerable, and the 1.2% figure is the starting point for negotiations with ARM, not a hard limit. No-one knows, or at least no-one will ever say, what prices are being charged in practice but it's well under 1.2% in many cases. Also, the reason why Tensilica didn't go ARM is because it was founded by ... what's-his-name, ex-MIPS, so they'd be unlikely to become an ARM shop. Cost may have been some factor, but the main reason would have been to do a MIPS-ng.
unless they can magic the masks and other components out of nothing, the cost of creating RISC-V stuff is going to be considerable
Not really. They are already making ASICs with their own stuff. And those ASICs already have a core of some sort. Taking out the HDL from the core, and inserting other HDL for another core, doesn't really change anything in their process. A CPU core is relatively simple piece to synthesise, all straight digital CMOS with standard library components.
If and you're starting with a new ASIC, it's even easier to pick a free core from the beginning. And when you're doing a second ASIC based on the same core, it's almost no work at all.
> There is a benefit. Compatibility. Not even Intel who tried the impossible to kill
> the x86 with HP failed when AMD made 64 bit standard. Funny pentium IV's
> mysteriously started being 64 bit compabile. Hmm my hunch is Intel disabled
> it to make Itanium look better and with a simple patch enabled the other bits.
Yes, there was a conspiracy, but you've got it wrong.
* The original IBM PC ran on an 8088. This was an 8086, 16-bit real mode CPU, with an 8-bit bus. 16-bit peripherals were scarce back then.
* In the early 1980's, IBM was *THE BIG NAME* in computers, and Intel was much smaller than it is today. IBM, being IBM, demanded, and got, a "second source" written into the contracts for 8086/8088 CPUs. The biggest second source player happened to be AMD. There was also the NEC V20 CPU and Cyrix, and possibly other bit players.
* A few years later, Intel stepped up to the 80286 and AMD followed.
* Intel sued, claiming that the original "second source" clause did not cover the 80286. AMD fought it in court, claiming the 80286 was basically a derivative of the 8086/8088.
* Eventually, AMD got their way. The settlement gave them the right to manufacture 80286 and other derivatives descended from the 8086/8088. Part of the settlement involved cross-licencing IP in the 8088 and derivatives.
* Intel came out with 32-bit 80386. The "SX" version had a 16-bit bus, and the "DX" version had a 32-bit bus. AMD followed.
* Intel came out with the 80486 and 80586 (Pentium). Again, AMD followed. Intel was pissed.
* Now for the real conspiracy. Intel deliberately designed the 64-bit Itanium to be so totally different from the 8088-descended CPUs, that the cross-licencing agreement would not cover it. They pushed Itanium very hard, because they badly wanted a market-leading 64-bit CPU that AMD wouldn't have a cross-licence to clone.
* In response, AMD added 64-bit extensions to their Pentium4 clone. This gave them a backwards-compatible 64-bit CPU. Meanwhile, Intel's Itanium flopped badly, being nicknamed "Itanic" in the industry. Part of the problem was that Itanium could only run 32-bit Pentium-class software via painfully slow software emulation.
* Remember the cross-licencing agreement I've mentioned? It was a 2-way agreement. When the Itanium was obviously dead, Intel used the cross-licencing agreement to legally clone AMD's 64-bit extensions. That's how they were able to ramp up 64-bit x86_64 chips so quickly.
I'm not repeating myself
I'm an X window user; I'm an ex-Windows user
...there's already a fork of Linux for RISC-V.
Wrong approach.
The right approach is to get involved with the upstream kernel community and get the changes they need into the kernel. Forking just means it'll always be on the sideline.
Availability of a low cost SoC – a la RaspPi3, Pine64, or ExpressoBin – would be good too.
more like Sigh... yet another "silicon valley startup" ... is this going to be like the $400 bag squeezer?
"unless they can magic the masks and other components out of nothing, the cost of creating RISC-V stuff is going to be considerable"
You have to generate masks for your silicon whatever processor you use.
Admirable effort, but worthless in the end. Unlike the analagous traditional Unix or other enterprise operating systems, there's no significant cost barrier to accessing good microprocessors. Why would anyone need an alternative to AMD, MIPS or Intel ?
SPARC was opened up long time ago.
https://en.wikipedia.org/wiki/OpenSPARC
RISC architecture is going to change everything.