7 ms·
Interesting article. I wonder what is really happening. Looking at moves by Chinese vendors and others, I'm going to hypothesize that ARM said to itself: "We
by ChuckMcM 6y ago
Interesting article. I wonder what is really happening.
Looking at moves by Chinese vendors and others, I'm going to hypothesize that ARM said to itself:
"We can't compete with 'free' and RISC-V is 'good enough' that its going to wipe out the commodity micro-contoller licensing stream that is over 50% of our revenue. That means we won't have the resources to invest in designing higher end features in the high margin parts and we're going to die. We should sell now, take our cash, and let the buyer eat the future losses."
I don't know of course. I'm an outsider like most people. But I have been part of, and then watching the semiconductor industry since about 1984.I was personally on the "inside" at the development of the SPARC architecture, and watched it grow, and die, as it was mismanaged by Sun. And I've been associated with American high tech companies for more years than I care to remember :-) and have seen a bit about how such companies survive, grow, and fail to grow. A common theme is that there is a "cash cow", a product that has exceptionally high margins relative to its "class" and that revenue funds the R&D that does the "innovation" work. These businesses live or die on finding "the next thing" before their cash cow dies. Failing to find that thing, as SGI did, as Sun did, and as I think Google is in the process of doing, starves the company of the cash to pay the innovators as they focus more and more on keeping the cash cow alive for longer and longer. The company essentially morphs from a mission of doing cool new things, to one that is preserving the 'one thing' at all cost. That thing eventually dies in spite of their efforts and what is left of the company is bought by a healthier company for its assets and maybe a couple of its projects that were cool, but not good enough to replace the cash cow.
If I am correct, this is the right move for ARM, and a really risky buy for anyone who would buy them now. A smart buyer, knowing how this will evolve, will wait for the desperation to set in so that they have maximum leverage in the acquisition.
And another thing that will be true (if I am guessing correctly here) is that RISC-V is going to really take off and have lots of people making them for cheap.
- Google234 6y agoRISC-V still seems overhyped to me. It would be nice to see working examples being used competitively.
- jacquesm 6y agoIt is work in progress on a very impressive trajectory.
- bsder 6y agoIt's effectively MIPS warmed over. And MIPS never went anywhere because it never cracked the cellphone market (which is where the real volume is). When someone puts out a RISC-V phone, you've got a battle on your hands. Until then, you're fighting over table scraps.
- jacquesm 6y agoI actually liked the MIPS architecture and used it for many years as my daily driver. True, it wasn't designed with mobile in mind, and ARM was a very good match there. But I don't see any reason why it could not be adapted to that.
- ithkuil 6y agoCould you summarize the main architectural (not implementation) differences between arm and risc-v that apply to being a good/poor match for mobile in particular?
- jacquesm 6y agoI actually don't think there are any at the lowest level, it's mostly a matter of ARM being able to reduce power consumption by aggressively throttling down the clock rates and various advanced sleep modes. In principle you should be able to do this with any CPU, given a roughly equal transistor count power consumption should be close. Obviously ARM is much further along in the practical implementation of these features.
- imtringued 6y agoARM has big little on every chip except embedded ones. The mix of super scalar and in order CPUs significantly improves power consumption during idling.
- jacquesm 6y agoARM management has historically been anything but clueless, they must have their reasons for putting it on the market and yours is as good as any that I've seen. RISC-V is a serious threat to their business model. In a way this is good, CPUs and the surrounding IP should be a (free) commodity, just like operating systems and utilities.
- klelatti 6y agoI'm not a CPU designer but if I were I'm not sure I'd agree that CPUs and the associated IP should be free. I think we all benefit from the effort of those who push forward CPU design and they deserve to be paid for it. I'm with you if this means that CPUs should be a collaborative effort with shared overheads and common standards but ARM is pretty close to this already.
- jacquesm 6y agoSure, but this stuff is two decades old by now and really should be a commodity. Newer and innovative designs are welcome to some protection. But at this level the main questions are can I have a toolchain that is unencumbered and how much power does a particular design consume. In a way that is where the battlefield is, extracting as much computation as possible from a given energy budget. Which led me to speculate the other day that mobile phones should be programmed in lower level languages for an immediate jump in battery life. Nothing my phone does requires gigahertz processors.
- klelatti 6y agoThe reason why I'm supportive of the ARM business model is that the fees are really not that significant (eg compared to x86) and actually I think it's good to support a firm with a primary focus of getting the most performance per watt from a mobile chip. A few dollars in licensing fees is really a small price to pay.
- Qwertious 6y agoI think there's "X should be free", and then there's "there should be a free implementation of X you have the option of using". The former is up for debate, but I doubt many people would disagree with the latter.
- klelatti 6y agoI understand the attractions of RISC-V and hope it does prosper but I'm not sure that it's 'free' to the extent that seems to be assumed. Some implementations are open source but others are not and someone has to do the work (and in most cases get paid for it). Chinese firms may rush to RISC-V but given the geopolitics (e.g. Huawei) I can easily see barriers being put up in the US and elsewhere to cheap chinese RISC-V chips being used everywhere.
- ChuckMcM 6y agoA couple of things I've noted: Western Digital adopts RISC-V [1] Microchip's new push [2] Analysts covering the microprocessor marking making some pretty big projections [3] [1] https://riscv.org/2017/11/ee-times-article-risc-v-spins-drives-ai/ https://riscv.org/2017/11/ee-times-article-risc-v-spins-driv... [2] http://ww1.microchip.com/downloads/en/DeviceDoc/00002911A.pdf http://ww1.microchip.com/downloads/en/DeviceDoc/00002911A.pd... [3] Semico Research estimated that there will be 62.4 billion RISC-V cores shipping in 2025. It’s not just for microcontrollers, Asanovic said. There is demand for every performance level. -- https://venturebeat.com/2019/12/11/risc-v-grows-globally-as-an-alternative-to-arm-and-its-license-fees/ https://venturebeat.com/2019/12/11/risc-v-grows-globally-as-...
- klelatti 6y agoI agree (and hope) that RISC-V will become very, very significant. That doesn't mean that ARM 'is going to die'. RISC-V is not 'free' in the sense that someone has to design an implementation that works for you - and then support it. The economics work for WD but not necessarily for everyone - not everyone has WD's scale. Plus price isn't the only factor in choosing an ISA / implementation.
- HeadsUpHigh 6y agoIt's entirely possible that due to the modularity of Risc-V some kind of common ground will be found in terms of common extensions( a de facto standard, i.e. a standard that people actually follow). Then you might have companies licensing designs for bits of the SoC so the total cost of developing a full SoC drops. This is happening to an extend already seeing as you can buy gpu or hardware accelerator tech from various companies.
- cpeterso 6y agoSeeing the writing on the wall, perhaps ARM could use their expertise to design and sell proprietary RISC-V instruction set extensions. This is the classic Innovator's Dilemma, a hard pill for ARM management to accept. (See Intel and Xscale.) ARM management would probably need to firewall RISC-V development in a subsidiary company.
- bfuclusion 6y agoThe day a RISCV SOC on par with the Raspi 1, is the day I never buy another ARM board. It'd be nice if the whole SOC is open cored like the CPU, but I'll take what I can get.
- klelatti 6y agoGenuine question. Why is this the case? What would a RISC-V board enable you to do that you can't do with an ARM board today?
- bfuclusion 6y agoMostly have a hope of having a bootloader process that doesn't hinge on binaries. I mean at some level I may have to, but it's really annoying I can't move an image from machine to machine even though the arch is all the same.
- klelatti 6y agoBut that's not an ARM issue is it? PI has strange boot by GPU but issue is SoC implementation I think not CPU architecture.
- bfuclusion 6y agoRight, it's not CPU arch, but there's a much higher chance of having the entire thing open arch if the CPU is. I expect the probability of reaching a de-facto "standard" SoC to be much higher in RISC-V and I'll take the pain of being the early adopter.
- pjmlp 6y agoMost likely that board would be using RISC-V extensions, so no juice.
- klelatti 6y agoI have a lot of sympathy but fear that RISC-V will lead to even more fragmentation than we see with ARM - eg proprietary extensions. PC standardisation is largely due to the level of control that a couple of dominant vendors have not due to the status of the IP in the core.