Comment by kjs3
19 days ago
I'm fairly confident there's more than 1% of people who care about these things
If there were an economically viable number of people who cared about those things (and it would need to be significantly more than 1%), we'd be running SPARC or POWER or maybe SuperH derived systems, all of which have open source, royalty free implementations.
For example, OpenSPARC is something like 20 years old, and covers SPARC v8 through t2. SPARC LEON is a decade older, and is under a GNU license, and has been to space.
And that doesn't consider going the Loongsoon route: take an existing ISA (e.g. MIPS), just use it, but carve off anything problematic (4 instructions covered by patents).
It's a pretty inescapable fact on the ground that in the 'processor hierarchy of needs', an open source license is of no consequence in the actual market.
I hesitate to say this as you seem very knowledgeable but you are missing some pretty massive facts that destroy your argument here.
There are already literally billions of RISC-V chips in the wild. Qualcomm alone has shipped a billion or more. They wrote an article back in 2023 where they disclosed that they had already shipped 650 million of them by that point. Andes Technology has said that there are 2 billion chips using their IP. A recent industry report suggested that RISC-V could represent 25% of the global SoC market by 2030. That is based on growth trajectory, not speculation.
RISC-V is not some obscure ISA that cannot get any traction.
There are a dozen or more credible competitors designing modern 64 bit RISC-V CPUs. Most of them have shipped silicon. Some have shipped multiple generations. Has any ISA ever had so many independent companies independently creating core designs (not designs from a single source like ARM)?
Tenstorrent alone likely made $500 million dollars in 2025. Easier to confirm is that they closed a $650 million funding round.
NVIDIA has announced CUDA support for RISC-V. I do not remember them doing that for SPARC, or POWER, or SuperH.
The current RISC-V standard, RVA23, includes advanced instructions for things like vectorization and virtualization. Many large, important industry players are involved in designing future extensions as well.
RISC-V is an officially supported platform in many mainstream Linux distributions including aggressively commercial ones like Red Hat Enterprise Linux but also foundational ones like Debian and its derivatives (like Ubuntu).
GCC and Clang have excellent support for RISC-V. FFMPEG just released hand-written vector optimizations for RISC-V. Again, can we say this about any of the platforms you mentioned?
It's a pretty inescapable fact on the ground that RISC-V has an absolute mountain of support in the industry. And starting this year, multiple vendors will be shipping cores faster than you can license from ARM.
Honestly, what universe are you living in?
Honestly, what universe are you living in?
The one where I actually read what I'm replying to.
I never one single time said RISC-V wasn't successful. Not even implied it. What I did say, should you ever climb of your apparently thinking-averse, pre-conceived notions is that its license isn't the overriding reason it's successful, because the world is full of open source ISAs that never gained any traction. Something you might be aware of if you took a brief break from furiously jerking off over RISC-V and paid attention.