6 ms·

I do think that RISC-V will open the door to open source implementations as well. It will take a long time, but we might reach a day when people are 3d printing

I do think that RISC-V will open the door to open source implementations as well. It will take a long time, but we might reach a day when people are 3d printing their own chips. Why does everyone always put so much pressure on new technologies to change everything over night? Maybe RISC-V's best days are 10-20 years from now.

tyingq 6y ago
"we might reach a day when people are 3d printing there own chips" Hah! Imagining what die size a 40μm process might need to make a CPU. The Intel 8008 (1972) was 10μm.
- Koshkin 6y ago
  I don't know why it would make sense, even (apart from having fun doing it). We do not print books at home even though we all have printers, we use print shops for that.
  - vidarh 6y ago
    We do not print books, but we do print plenty of photos and documents and newsletters that it'd take a professional photo lab or large-scale offset printers to print just a few decades ago.
    - oblio 6y ago
      The market for people home printing photos and documents and newsletter is probably 1 million times bigger than the one for people home printing CPUs, though.
      - vidarh 6y ago
        Yes, and yet plenty of tools and products target that market anyway. Especially since while a good chunk of it is hobbyist use, another large chunk is small-run electronics where relatively high price tags can easily be justified. But even in the hobbyist space, while the people who would actually use the tools are few in numbers, many hobbyist projects get built in the 100's or even thousands - whether by end-users themselves or by people selling them to end users. FPGAs are often a good chunk of the cost of projects like that.
  - bluGill 6y ago
    Maybe you don't print books at home, but the home publishing does exist. Mostly for small businesses who need to give each customer a slightly different book (if only because of updates). If you will sell 1000 books you should get a publisher to print the book, but if you are looking at only a couple per month print on demand at home makes sense as you can ensure everyone gets the latest corrections. The above is moving to electronic, but paper still has advantages over electronics so it isn't dead.
tambourine_man 6y ago
3D printing high performance (competitive with state of the art TSCM) chips is probably not for this century. I’m hoping for an open source implementation that can be backed by a company or crowdsourced, manufactured at scale and auditioned by the community. This could happen this decade and would be transformative.
- vidarh 6y ago
  Something that could allow people to print chips competitive with stuff you can run on present-day FPGAs would already be transformative (and yes, I recognise 3d printing is far off achieving anything like that as well, but the point is it doesn't need to compete with the state of the art to be highly useful for many use-cases).
  - glhaynes 6y ago
    What use-case would that satisfy better than a current FPGA?
    - lallysingh 6y ago
      You want more than one and don't want to pay FPGA money for each one.
      - vidarh 6y ago
        While much smaller in volume, there's also a whole sub-culture of people opposed to "emulation", a subset of which sees FPGA's as borderline "cheating", and/or who would otherwise love to print pin-compatible replacements for old chips. Heck, even being capable of printing replacements for 6581/8580 (SID - sound chip from Commodore 64) or 6526 (IO chips used in the C64 and others that were notorious for burning out) would get a whole lot of retro-enthusiasts excited, as there are a bunch of FPGA reproductions of the SID which are way overkill. Doing that would be possible by matching processes that were getting dated by 1980...
jcranmer 6y ago
The sizes of parts of modern transistors are measured in the number of layers of atoms (monolayers) that comprise them. Getting that stuff built requires some pretty beefy, expensive technology, and is not anywhere near the realm of 3d printers. There's a reason you can count the number of cutting-edge semiconductor fabrication companies on one hand (and the suppliers of some of the machines in the fab on one digit).
- postalrat 6y ago
  Just give things 30 years and see how it all develops.
- philipswood 6y ago
  positioning a single atom is doable with a STM, right? And hobbyists are making STMs, right? Beating the fabs at their own game? Not likely. But maybe another game, probably.
  - nynx 6y ago
    It seems that bonding atoms in 3D using an STM is very difficult, if not impossible. We'll probably have self-assembling molecular processors before then.
    - philipswood 6y ago
      I don't think you'd need to bond them. I imagine you could put down a "wire" of metal atoms without too much trouble. Something like a capacitor shouldn't be too difficult if you could put down a dielectric. Not sure about how to do a transistor - maybe there could be an alternative "switch". Yeah, I'm hand-waving. And maybe the metal substrate an STM uses is a non-starter. But still, atomic precision seems to be in hobbyists reach and I imagine SOME approach to low volume tiny circuits are possible that would not be economically viable in terms of RnD and fab for "big silicon".
      - nynx 6y ago
alted 6y ago
Previous discussion: https://news.ycombinator.com/item?id=24119102 https://news.ycombinator.com/item?id=24119102 ("3D Printing Integrated Circuits: What's Possible Now and in the Future?")
lallysingh 6y ago
Right now it looks like POSIX and some good minix equivalents. It's waiting for a Linux to change the game. As for manufacturing, I think it's harder but not crazy harder. Small batch fabs making small vendors' customized implementations seems reasonable. It's really about reducing requirements for big scale. ISA is one.

tyingq 6y ago

"we might reach a day when people are 3d printing there own chips" Hah! Imagining what die size a 40μm process might need to make a CPU. The Intel 8008 (1972) was 10μm.

Koshkin 6y ago
I don't know why it would make sense, even (apart from having fun doing it). We do not print books at home even though we all have printers, we use print shops for that.
- vidarh 6y ago
  We do not print books, but we do print plenty of photos and documents and newsletters that it'd take a professional photo lab or large-scale offset printers to print just a few decades ago.
  - oblio 6y ago
    The market for people home printing photos and documents and newsletter is probably 1 million times bigger than the one for people home printing CPUs, though.
    - vidarh 6y ago
      Yes, and yet plenty of tools and products target that market anyway. Especially since while a good chunk of it is hobbyist use, another large chunk is small-run electronics where relatively high price tags can easily be justified. But even in the hobbyist space, while the people who would actually use the tools are few in numbers, many hobbyist projects get built in the 100's or even thousands - whether by end-users themselves or by people selling them to end users. FPGAs are often a good chunk of the cost of projects like that.
- bluGill 6y ago
  Maybe you don't print books at home, but the home publishing does exist. Mostly for small businesses who need to give each customer a slightly different book (if only because of updates). If you will sell 1000 books you should get a publisher to print the book, but if you are looking at only a couple per month print on demand at home makes sense as you can ensure everyone gets the latest corrections. The above is moving to electronic, but paper still has advantages over electronics so it isn't dead.

tambourine_man 6y ago

3D printing high performance (competitive with state of the art TSCM) chips is probably not for this century. I’m hoping for an open source implementation that can be backed by a company or crowdsourced, manufactured at scale and auditioned by the community. This could happen this decade and would be transformative.

vidarh 6y ago
Something that could allow people to print chips competitive with stuff you can run on present-day FPGAs would already be transformative (and yes, I recognise 3d printing is far off achieving anything like that as well, but the point is it doesn't need to compete with the state of the art to be highly useful for many use-cases).
- glhaynes 6y ago
  What use-case would that satisfy better than a current FPGA?
  - lallysingh 6y ago
    You want more than one and don't want to pay FPGA money for each one.
    - vidarh 6y ago
      While much smaller in volume, there's also a whole sub-culture of people opposed to "emulation", a subset of which sees FPGA's as borderline "cheating", and/or who would otherwise love to print pin-compatible replacements for old chips. Heck, even being capable of printing replacements for 6581/8580 (SID - sound chip from Commodore 64) or 6526 (IO chips used in the C64 and others that were notorious for burning out) would get a whole lot of retro-enthusiasts excited, as there are a bunch of FPGA reproductions of the SID which are way overkill. Doing that would be possible by matching processes that were getting dated by 1980...

jcranmer 6y ago

The sizes of parts of modern transistors are measured in the number of layers of atoms (monolayers) that comprise them. Getting that stuff built requires some pretty beefy, expensive technology, and is not anywhere near the realm of 3d printers. There's a reason you can count the number of cutting-edge semiconductor fabrication companies on one hand (and the suppliers of some of the machines in the fab on one digit).

postalrat 6y ago
Just give things 30 years and see how it all develops.
philipswood 6y ago
positioning a single atom is doable with a STM, right? And hobbyists are making STMs, right? Beating the fabs at their own game? Not likely. But maybe another game, probably.
- nynx 6y ago
  It seems that bonding atoms in 3D using an STM is very difficult, if not impossible. We'll probably have self-assembling molecular processors before then.
  - philipswood 6y ago
    I don't think you'd need to bond them. I imagine you could put down a "wire" of metal atoms without too much trouble. Something like a capacitor shouldn't be too difficult if you could put down a dielectric. Not sure about how to do a transistor - maybe there could be an alternative "switch". Yeah, I'm hand-waving. And maybe the metal substrate an STM uses is a non-starter. But still, atomic precision seems to be in hobbyists reach and I imagine SOME approach to low volume tiny circuits are possible that would not be economically viable in terms of RnD and fab for "big silicon".
    - nynx 6y ago

alted 6y ago

Previous discussion: https://news.ycombinator.com/item?id=24119102 https://news.ycombinator.com/item?id=24119102 ("3D Printing Integrated Circuits: What's Possible Now and in the Future?")

lallysingh 6y ago

Right now it looks like POSIX and some good minix equivalents. It's waiting for a Linux to change the game. As for manufacturing, I think it's harder but not crazy harder. Small batch fabs making small vendors' customized implementations seems reasonable. It's really about reducing requirements for big scale. ISA is one.