7 ms·

> it still is a much better starting point than coming up with your own bespoke ISA 5 years ago I would have agreed with this but now I'm not so sure. We live

> it still is a much better starting point than coming up with your own bespoke ISA 5 years ago I would have agreed with this but now I'm not so sure. We live in an era where you can tell a robot "Here's some C code. Design a 64-bit ISA, write the Verilog to implement it in an FPGA, write a C compiler for it, and use it to compile the C code I showed you earlier." And now your ISA and your compiler are part of your moat. I can just see the VCs salivating.

creato 1mo ago
If that's really all it takes, then it is not much of a moat.
- pyvpx 1mo ago
  In the set of all possible working implementations, there are one or more that are novel enough to become a moat legally or otherwise. If everyone has the same power (number of tokens) then capability (experience and understanding) becomes the differentiator to “find” that moat first.
  - inigyou 1mo ago
    If everyone has equal ability to build a moat, it's not a moat.
    - pyvpx 1mo ago
      If every one has a hammer and a saw and identical wood planks, everyone can build a frame that holds up 100 years? Think it through. This is the HN comment section after all.
      - inigyou 1mo ago
        Yes. That's the point of AI. Anyone can make anything just as easily and well as anyone else.
        pyvpx 29d ago
        That’s just not how things work. Some one with experience and knowledge will be measurably more productive and build a better whatever no matter how much spicy autocomplete they have at their disposal. Even if you eventually figure out all the right questions to prompt, the experienced person is long done
    - just6979 1mo ago
      I think you mean "if everyone has the ability (and resources) to build a boat, then it's not a (good) moat". If everyone can build a moat, there's just a lot of moats.
water-drummer 1mo ago
Coming up with any ISA and coming up with a good or better ISA are two very different things.
adrian_b 1mo ago
Designing a good 64-bit ISA, much better than RISC-V, is easy. There are thousands of people who could design such an ISA in a couple of weeks, without any AI assistance. The hard part, which has always been the moat of RISC-V, is writing all the required support software for a new ISA, i.e. all the utilities from binutils (assembler, static linker, ELF/DWARF utilities), compiler backends at least for gcc and llvm, debugger (at least a port for gdb server), dynamic linker and standard C library, possibly some parts of the standard libraries for other programming languages. Previously this could have taken years and it is the only reason that has always justified the choice of RISC-V for minimum cost, despite how bad the ISA is. If today the porting of all these software support applications to a new ISA could be accelerated with AI assistance from a couple of years to a couple of months, that would certainly enable the design and use of custom ISAs, and RISC-V would lose its appeal.
- trollbridge 1mo ago
  The effort to support architectures is now minimal with AI assistance. I made a hobby architecture (based on Intel, but with some changes; I was making an "alternate history" as if a few decisions in the past had been different) and it was pretty much trivial to spit out support not just in gcc and llvm, but I also, for fun, made WATCOM backends and a few other things. With that said, RISC-V is a nice baseline for designing another architecture. Start with RISC-V, and go from there.
  - skydhash 1mo ago
    > and it was pretty much trivial to spit out support not just in gcc and llvm, but I also, for fun, made WATCOM backends and a few other things No links? You may have dreamt about it instead.
    - trollbridge 1mo ago
      It was vibed-level stuff definitely not worth sharing.
  - jcranmer 1mo ago
achierius 1mo ago
How is that a moat, if anyone else can do the same thing? And how would you get this all into Clang? Nobody wants to use your custom compiler. It's certainly not going to be as fast as Clang!

creato 1mo ago

If that's really all it takes, then it is not much of a moat.

pyvpx 1mo ago
In the set of all possible working implementations, there are one or more that are novel enough to become a moat legally or otherwise. If everyone has the same power (number of tokens) then capability (experience and understanding) becomes the differentiator to “find” that moat first.
- inigyou 1mo ago
  If everyone has equal ability to build a moat, it's not a moat.
  - pyvpx 1mo ago
    If every one has a hammer and a saw and identical wood planks, everyone can build a frame that holds up 100 years? Think it through. This is the HN comment section after all.
    - inigyou 1mo ago
      Yes. That's the point of AI. Anyone can make anything just as easily and well as anyone else.
      - pyvpx 29d ago
        That’s just not how things work. Some one with experience and knowledge will be measurably more productive and build a better whatever no matter how much spicy autocomplete they have at their disposal. Even if you eventually figure out all the right questions to prompt, the experienced person is long done
  - just6979 1mo ago
    I think you mean "if everyone has the ability (and resources) to build a boat, then it's not a (good) moat". If everyone can build a moat, there's just a lot of moats.

water-drummer 1mo ago

Coming up with any ISA and coming up with a good or better ISA are two very different things.

adrian_b 1mo ago

Designing a good 64-bit ISA, much better than RISC-V, is easy. There are thousands of people who could design such an ISA in a couple of weeks, without any AI assistance. The hard part, which has always been the moat of RISC-V, is writing all the required support software for a new ISA, i.e. all the utilities from binutils (assembler, static linker, ELF/DWARF utilities), compiler backends at least for gcc and llvm, debugger (at least a port for gdb server), dynamic linker and standard C library, possibly some parts of the standard libraries for other programming languages. Previously this could have taken years and it is the only reason that has always justified the choice of RISC-V for minimum cost, despite how bad the ISA is. If today the porting of all these software support applications to a new ISA could be accelerated with AI assistance from a couple of years to a couple of months, that would certainly enable the design and use of custom ISAs, and RISC-V would lose its appeal.

trollbridge 1mo ago
The effort to support architectures is now minimal with AI assistance. I made a hobby architecture (based on Intel, but with some changes; I was making an "alternate history" as if a few decisions in the past had been different) and it was pretty much trivial to spit out support not just in gcc and llvm, but I also, for fun, made WATCOM backends and a few other things. With that said, RISC-V is a nice baseline for designing another architecture. Start with RISC-V, and go from there.
- skydhash 1mo ago
  > and it was pretty much trivial to spit out support not just in gcc and llvm, but I also, for fun, made WATCOM backends and a few other things No links? You may have dreamt about it instead.
  - trollbridge 1mo ago
    It was vibed-level stuff definitely not worth sharing.
- jcranmer 1mo ago

achierius 1mo ago

How is that a moat, if anyone else can do the same thing? And how would you get this all into Clang? Nobody wants to use your custom compiler. It's certainly not going to be as fast as Clang!