is RISCV mature enough for desktop use? Are there chips based on RISCV that would at least be as good as a AMD/Intel or ARM chips?
If you like working in slow motion, yes, sure.
Source : I have a Banana-Pi SBC https://www.banana-pi.org/en/banana-pi-sbcs/175.html and… it works, running Linux proper, with a desktop environment, which is in itself pretty cool IMHO but damn, you have to be patient. That being said “just” already being at that stage on economically affordable hardware is amazing. We are probably not far, say few years at most, with usable RISC-V chips for mundane tasks, e.g. text authoring, coding, Web browsing, but don’t expect compilation of a browser, Blender, or gaming on this for few more years. IMHO it will go fast because it’s catching up so the path is rather well laid down, which is much harder than innovating and pushing the envelope.
I wonder if the US is trying to slow down the development of RISCV in order to mantain egemony over chip production. I think RISCV poses a big “security” flaw for them being totally open source.
I guess it depends what you mean by “chip production”.
AFAICT mostly via Chip War (2022) and reading a bit on the topic there are few bottlenecks, e.g chip design IP like ARM (UK) or lithography machines like ASML (NL) or high efficiency chip production like TSMC (Taiwan) but overall the grip from the US is mostly on democratization and scale with AMD, NVIDIA, Broadcom or even Intel, namely making a LOT of chips, not necessarily high end (some are) or mobile (also some), for a relatively low price. What I mean is that China is already claiming that they are producing about on-par IPS with e.g. Loongson.
So yes there are for sure incumbents based in the US that do not want RISCV and overall open architectures to make significant progress but is it fair to call them “the US” I’m not sure. Are they heavily leaning on US lawmakers to get their positions strengthened? Maybe. Maybe they do not yet do so simply because they don’t believe it’s a threat yet, nor it might be ever be.
I believe that in chip production you can lock production via innovation but also, like in other sectors, solely with the supply chain. ASML is powerful because they basically own their markets but also because who would contract with newcomers versus a very well established company that can provide all the insurances imaginable that they will indeed deliver on time a specific amount? Why risk it when you are already contracting with the leader?
Sure there is a potential innovator dilemma but what could prevent e.g. NVIDIA or Intel to switch to RISC-V if somehow they can dominate there too thanks to both their existing expertise but also supply chain stronghold?
In what way does it pose a big security flaw? And what are you basing the thought that the USA are slowing down development?
Being an open source architecture gives everyone (China, Russia, Iran, North Korea and basically everyone else the US doesn’t like, which apparently is most of world) the freedom to innovate at a fast pace, this is what I mean by security flaw. My thought that they could somewhat try to slow down the development is based on the rational thinking that 1. They are actually leading the chip develpment and 2. if someday everyone gets high performance chips (which is still not the case with RISCV yet) than everyone can get better defense industries, better intelligence systems, better military equipment. I’m not implying that they are actively doing it but that it might be in their interest to do so to maintain some kind of military egemony over their enemies, or at least to never be in a significant gap with anyone.
wat dat mean
Year of the RISCV Linux Desktop
Bout to overtake Windows with this one!
This seems to be positioning Ubuntu as a data center OS. There are several RVA23 chips due out but they are all for the data center (tenstorrent, Alibaba, ventana, etc).
There is the SiFive P870 but I do not think anybody has licensed that so it may never get made.
I have also heard rumours of Expressif chips but I do not know the details.
…except there’s no hardware to run it on. They’ve chosen an ISA profile that’s not been decided on for long enough.
Exactly. The article says that 90% of hardware doesn’t run on it, but in reality, 100% of hardware doesn’t run on it. Only Qemu supports it, which is an emulator (and very slow to emulate RiscV in my experience – latest version we tried with my husband on a very fast PC).
The Orange Pi RV2 was the perfect introductory Risc-V SBC, everyone is going gaga for it, for being a good middle of the road solution for those who want to try Risc-V, and yet, Ubuntu won’t support it (and even the current implementation is done by the Chinese, not by Canonical, so I wouldn’t touch it).
So I’m not sure what they’re thinking. My own conspiracy theory is that EITHER Canonical, OR Raspberry Pi (which are close geographically), are preparing RV23 hardware, so they want to undercut the competition that way.
Nothing else makes sense in that decision.
When compatible hardware is available, it’s expected that having packages built for RVA23 will have a big impact on performance. You can already see a big part of that with the vector (V) extension: running programs built without it is akin to using x86 programs without SSE or AVX. RVA23 is the first RVA profile that considers V mandatory rather than optional.
You might see a similar performance impact if you target something like RVA22+V instead of RVA23, but as far as I know the only hardware systems that’d benefit from that are the Spacemit ones (OPi RV2, BPI-F3, Jupiter) while that’d still leave behind VisionFive 2, Pioneer, P550/Megrez, and even an upcoming processor UltraRISC announced recently. The profiles aren’t exactly intended to be used for those kinds of fine-tuned combinations and it’s possible some of the other RVA23 extensions (Zvbb, Zicond, etc.) might have a substantial impact too.
Hardware vendors want to showcase their system having the best performance it can, so I expect Ubuntu’s aim is to have RVA23 builds ready before RVA23 hardware so that they’ll be the distro of choice for future hardware, even if that means abandoning all existing RISC-V users. imo it would’ve been better to maintain separate builds for RV64GC and RVA23 but I guess they just don’t care enough about existing RISC-V users to maintain two builds.
