3 ms·

I hope they succeed but making micro/nanoscale structures with humanscale machines has always been a hard thing, even for those with better funding than the hob

I hope they succeed but making micro/nanoscale structures with humanscale machines has always been a hard thing, even for those with better funding than the hobbyist. I recently learned about DNA-directed crystal growth and was excited by the idea that it might be a more tractable approach to being a big thing and making a small thing (like an integrated circuit). I'm not sure how one would go about it in their garage, but programming the fine-control-needed steps into the chemical rather than into the machine feels like a win.

Joel_Mckay 2y ago
One has to admire their efforts, as the upfront costs of a fab lab are ridiculously high. As with any technology the Metrology becomes the predominant problem domain. i.e. Answering the question "Where do we get the repeatable precision?" There are low-volume lab processes around that can hit below <234nm feature sizes, but a clean room must be considered part of the machine... And it can take years to figure out how to maintain atmospheres and gas mass-flow-control. Pretty cheeky selling community designed hardware without citing the original hobbyists. Nothing they posted looks remotely new or novel. Meh =3
nzrrsm 2y ago
I've been thinking about this a bunch lately and very much agree. We really need a way to build these nanodevices that doesn't involve lithography. Using something like DNA to carry information to the surface seems way easier and more effective/robust as we continue to scale down and out.
- __MatrixMan__ 2y ago
  I wouldn't go so far as "way easier". It's just difficult for a single reason that's easy to imagine a solution to rather than 100 different reasons which are not. I'm not in the field, but I'm under the impression that our ability to encode our wishes into DNA is closer to: > make me a 1um tall pyramid out of nanoparticles which are all the same type ...than it is to: > make me a NAND gate out of this solution of nanoparticles, some are doped in different ways than others. Most miracles involving DNA are heavily inspired by nature, and there's not a lot of natural code available for the construction of logic gates. I'm not sure which of these is harder: 1. make biochemical systems which construct arbitrary solid state circuits 2. abandon solid state and do our computing in living tissue What I like about the former is that there are a lot of other things that investment in that area might produce, things which aren't a hassle to keep alive, even if a general purpose circuit constructor happens to be out of reach.

Joel_Mckay 2y ago

One has to admire their efforts, as the upfront costs of a fab lab are ridiculously high. As with any technology the Metrology becomes the predominant problem domain. i.e. Answering the question "Where do we get the repeatable precision?" There are low-volume lab processes around that can hit below <234nm feature sizes, but a clean room must be considered part of the machine... And it can take years to figure out how to maintain atmospheres and gas mass-flow-control. Pretty cheeky selling community designed hardware without citing the original hobbyists. Nothing they posted looks remotely new or novel. Meh =3

nzrrsm 2y ago

I've been thinking about this a bunch lately and very much agree. We really need a way to build these nanodevices that doesn't involve lithography. Using something like DNA to carry information to the surface seems way easier and more effective/robust as we continue to scale down and out.

__MatrixMan__ 2y ago
I wouldn't go so far as "way easier". It's just difficult for a single reason that's easy to imagine a solution to rather than 100 different reasons which are not. I'm not in the field, but I'm under the impression that our ability to encode our wishes into DNA is closer to: > make me a 1um tall pyramid out of nanoparticles which are all the same type ...than it is to: > make me a NAND gate out of this solution of nanoparticles, some are doped in different ways than others. Most miracles involving DNA are heavily inspired by nature, and there's not a lot of natural code available for the construction of logic gates. I'm not sure which of these is harder: 1. make biochemical systems which construct arbitrary solid state circuits 2. abandon solid state and do our computing in living tissue What I like about the former is that there are a lot of other things that investment in that area might produce, things which aren't a hassle to keep alive, even if a general purpose circuit constructor happens to be out of reach.