8 ms·

Moore's law happens when the output of the system can be used to improve the system that produced the original output. In other words some recursion is required

Moore's law happens when the output of the system can be used to improve the system that produced the original output. In other words some recursion is required. It also requires that there is an economic need for more output. So yes if somehow putting engineers and labs in space allows us to iterate faster or spaceship designs then the first requirement is achieved. Then there's the question if the masses want to travel to space. Unlikely unless it just so happens to speed travel and reduce cost for getting from Earth point A to Earth point B. We'll see.

phreeza 5y ago
> Moore's law happens when the output of the system can be used to improve the system that produced the original output. This is a thing that possibly facilitates exponential progress, but it is by no means a prerequisite. I have a hard time believing that compute power has been the bottleneck in chip design for many years now. Other current counterexamples include solar cells and batteries. They have been improving exponentially, but no recursion to speak of comes to mind.
- throwaway803453 5y ago
  My understanding was Moore's law implied a form of doubling at at least every 24 months. Otherwise why not just say {qualifier} exponential growth.
- atq2119 5y ago
  > I have a hard time believing that compute power has been the bottleneck in chip design for many years now. Considering that many of the bread and butter NP-hard problems are actually inspired by chip design, this is blatantly untrue. Though I do believe progress is currently held back more by the closed software ecosystems and secrecy around design rules.
  - phreeza 5y ago
    My completely amateur understanding of chip manufacturing progress is that it is not bound by routing problems or similar (which I am guessing you are referring to), but by lithography physics, chemistry, etc. I might be wrong though, are you saying that faster chips are actually becoming possible because more compute power has become available through the previous generation of chips, at this point?
    - atq2119 5y ago
      The lithography has obviously been much more salient, and still is, but both are limiting factors. To draw a parallel: as the improvement of computing hardware performance slows down, one would expect more attention to be paid to software performance optimization, so that more of the potential of the hardware is actually used. In the same vein, as lithography improvements slow down, one would expect more attention to be paid to e.g. the routing problem (but also placement of gates and buffers/repeaters, logic optimizations for timing closure, and others) so that more of the potential of the lithography is actually used. So: > are you saying that faster chips are actually becoming possible because more compute power has become available through the previous generation of chips, at this point? Yes. This is obvious when you look at it in the extreme: chips like the Intel 8080 were designed by hand. As in, the shapes to be created via lithography were drawn by hand. Doing this for a modern CPU core is plain impossible. We can only design those because older CPU cores already exist and can run automated tools that do this job for us. It's less obvious but still true for generational improvements. The chip design problems are becoming computationally harder in each generation, for two main reasons. First, the designs are simply becoming bigger (more gates, more wires). Second, the design rules are becoming more complex.[0] If you kept the chips that you use to create the design constant, then each design generation would take more wall-time than the previous one to go through the automated design tools, until at some point everything just becomes infeasibly slow. Using faster chips to run the physical design tools on allows the designs to be processed faster, which leads to a faster feedback loop with logic designers (the people who write Verilog), which then leads to a faster feedback loop with architects, which ultimately leads to faster/better designs for the next generation of chips. This may be surprising for software developers -- we get grumpy when our feedback loops take minutes. Chip designers have to deal with feedback loops that take days in some cases for the physical design. [0] For example, at the lowest routing layers you can't just draw wires wherever you like, but have to satisfy some pretty byzantine rules about things like minimum metal area, distance to neighboring wires, allowed shapes of wires, allowed shapes of holes between wires, and so on. These rules exist because of limitations in the capabilities of the lithography: the process engineers are able to make features smaller, but at the cost of forbidding certain shapes that would cause problems for physical or chemical reasons.
Robotbeat 5y ago
SpaceX has a pretty strong economic motivator in the form of Starlink. Starlink is also facilitating Starship launches by providing telemetry during launch (or at least that’s the goal). Eventually, Mars (or other places like the Moon or asteroids) may become a base of operations for launch due to lower gravity. This could, in principle, make things like space based solar feasible. Such large structures would probably be best built by mechanized astronauts just like Starships on Earth. …and see experiments with EVA assembly of telescopes and other structures NASA did in the 1980s and 90s… orders of magnitude faster than space robotics, held back by insanely high cost of human spaceflight. So if you made human spaceflight cheap, that’d make large space structures cheap as well. Mars or the Moon or asteroids could provide lower cost and lower ecological footprint propellants, enabling more space travel and potentially reducing terrestrial high-impact industry.
- gspr 5y ago
  Starlink has little chance of being the cash cow that the muskrats think. Look, it's low latency satellite internet. It's a giant "meh" for anyone not in very rural locations. That rural market exists, but is quite small — as you can easily see from the revenue of the 3-4 established providers in the field. Yeah, Starlink will differentiate itself by providing a low latency service, and the market for that might be slightly larger, but it'll never be enough to pay for the launches and equipment and then be the cash cow the muskrats dream about.
  - MPSimmons 5y ago
    > That rural market exists, but is quite small Curious about your estimate on the market size here. How many potential customers are there, worldwide, in your estimation?
  - JumpCrisscross 5y ago
    > rural market exists, but is quite small — as you can easily see from the revenue of the 3-4 established providers in the field The existing providers suck. There are plenty of villas, chalets, et cetera in places where Starlink would be the winning offering. These may not represent a large number of people. But they do represent a lot of dollars. This is before tapping the moving target market, e.g. yachts, planes, backpackers, et cetera.
fallingknife 5y ago
> Unlikely unless it just so happens to speed travel and reduce cost for getting from Earth point A to Earth point B. I doubt that anyone I know would pass on a ticket if it was affordable.
- throwaway803453 5y ago
  Many people would pass, including me and most of my friends, due to the environmental impact and radiation exposure. It also seems like a dumb way to die if you aren't contributing to a mission in someway. It's cool to die as a researcher, explorer, or pioneer like Branson. But as a popcorn eater who can afford to overpay for a roller coaster ride ? On second thought, if the marketing was good and I had kids, I admit I be tempted.
mrfusion 5y ago
> Moore's law happens when the output of the system can be used to improve the system that produced the original output. Actually I’ve never understood how this was true for microchips. Anyone know? I understood it as the industry just Moore’s law for planning iterations?
- throwaway803453 5y ago
  At the very least when you design a new microprocessor you design and simulate it using software running on a computer. Then when that new processor releases, you build a computer with it and then you can simulate and design an even more powerful microprocessor. The tools and production equipment in the fab are also likely designed and partially simulated using CAD tools as well so there's an additional benefit there.

phreeza 5y ago

> Moore's law happens when the output of the system can be used to improve the system that produced the original output. This is a thing that possibly facilitates exponential progress, but it is by no means a prerequisite. I have a hard time believing that compute power has been the bottleneck in chip design for many years now. Other current counterexamples include solar cells and batteries. They have been improving exponentially, but no recursion to speak of comes to mind.

throwaway803453 5y ago
My understanding was Moore's law implied a form of doubling at at least every 24 months. Otherwise why not just say {qualifier} exponential growth.
atq2119 5y ago
> I have a hard time believing that compute power has been the bottleneck in chip design for many years now. Considering that many of the bread and butter NP-hard problems are actually inspired by chip design, this is blatantly untrue. Though I do believe progress is currently held back more by the closed software ecosystems and secrecy around design rules.
- phreeza 5y ago
  My completely amateur understanding of chip manufacturing progress is that it is not bound by routing problems or similar (which I am guessing you are referring to), but by lithography physics, chemistry, etc. I might be wrong though, are you saying that faster chips are actually becoming possible because more compute power has become available through the previous generation of chips, at this point?
  - atq2119 5y ago
    The lithography has obviously been much more salient, and still is, but both are limiting factors. To draw a parallel: as the improvement of computing hardware performance slows down, one would expect more attention to be paid to software performance optimization, so that more of the potential of the hardware is actually used. In the same vein, as lithography improvements slow down, one would expect more attention to be paid to e.g. the routing problem (but also placement of gates and buffers/repeaters, logic optimizations for timing closure, and others) so that more of the potential of the lithography is actually used. So: > are you saying that faster chips are actually becoming possible because more compute power has become available through the previous generation of chips, at this point? Yes. This is obvious when you look at it in the extreme: chips like the Intel 8080 were designed by hand. As in, the shapes to be created via lithography were drawn by hand. Doing this for a modern CPU core is plain impossible. We can only design those because older CPU cores already exist and can run automated tools that do this job for us. It's less obvious but still true for generational improvements. The chip design problems are becoming computationally harder in each generation, for two main reasons. First, the designs are simply becoming bigger (more gates, more wires). Second, the design rules are becoming more complex.[0] If you kept the chips that you use to create the design constant, then each design generation would take more wall-time than the previous one to go through the automated design tools, until at some point everything just becomes infeasibly slow. Using faster chips to run the physical design tools on allows the designs to be processed faster, which leads to a faster feedback loop with logic designers (the people who write Verilog), which then leads to a faster feedback loop with architects, which ultimately leads to faster/better designs for the next generation of chips. This may be surprising for software developers -- we get grumpy when our feedback loops take minutes. Chip designers have to deal with feedback loops that take days in some cases for the physical design. [0] For example, at the lowest routing layers you can't just draw wires wherever you like, but have to satisfy some pretty byzantine rules about things like minimum metal area, distance to neighboring wires, allowed shapes of wires, allowed shapes of holes between wires, and so on. These rules exist because of limitations in the capabilities of the lithography: the process engineers are able to make features smaller, but at the cost of forbidding certain shapes that would cause problems for physical or chemical reasons.

Robotbeat 5y ago

SpaceX has a pretty strong economic motivator in the form of Starlink. Starlink is also facilitating Starship launches by providing telemetry during launch (or at least that’s the goal). Eventually, Mars (or other places like the Moon or asteroids) may become a base of operations for launch due to lower gravity. This could, in principle, make things like space based solar feasible. Such large structures would probably be best built by mechanized astronauts just like Starships on Earth. …and see experiments with EVA assembly of telescopes and other structures NASA did in the 1980s and 90s… orders of magnitude faster than space robotics, held back by insanely high cost of human spaceflight. So if you made human spaceflight cheap, that’d make large space structures cheap as well. Mars or the Moon or asteroids could provide lower cost and lower ecological footprint propellants, enabling more space travel and potentially reducing terrestrial high-impact industry.

gspr 5y ago
Starlink has little chance of being the cash cow that the muskrats think. Look, it's low latency satellite internet. It's a giant "meh" for anyone not in very rural locations. That rural market exists, but is quite small — as you can easily see from the revenue of the 3-4 established providers in the field. Yeah, Starlink will differentiate itself by providing a low latency service, and the market for that might be slightly larger, but it'll never be enough to pay for the launches and equipment and then be the cash cow the muskrats dream about.
- MPSimmons 5y ago
  > That rural market exists, but is quite small Curious about your estimate on the market size here. How many potential customers are there, worldwide, in your estimation?
- JumpCrisscross 5y ago
  > rural market exists, but is quite small — as you can easily see from the revenue of the 3-4 established providers in the field The existing providers suck. There are plenty of villas, chalets, et cetera in places where Starlink would be the winning offering. These may not represent a large number of people. But they do represent a lot of dollars. This is before tapping the moving target market, e.g. yachts, planes, backpackers, et cetera.

fallingknife 5y ago

> Unlikely unless it just so happens to speed travel and reduce cost for getting from Earth point A to Earth point B. I doubt that anyone I know would pass on a ticket if it was affordable.

throwaway803453 5y ago
Many people would pass, including me and most of my friends, due to the environmental impact and radiation exposure. It also seems like a dumb way to die if you aren't contributing to a mission in someway. It's cool to die as a researcher, explorer, or pioneer like Branson. But as a popcorn eater who can afford to overpay for a roller coaster ride ? On second thought, if the marketing was good and I had kids, I admit I be tempted.

mrfusion 5y ago

> Moore's law happens when the output of the system can be used to improve the system that produced the original output. Actually I’ve never understood how this was true for microchips. Anyone know? I understood it as the industry just Moore’s law for planning iterations?

throwaway803453 5y ago
At the very least when you design a new microprocessor you design and simulate it using software running on a computer. Then when that new processor releases, you build a computer with it and then you can simulate and design an even more powerful microprocessor. The tools and production equipment in the fab are also likely designed and partially simulated using CAD tools as well so there's an additional benefit there.