About the return on investment, the methodology is interesting, and I’m surprised that a hand crank to nuclear would increase so little in efficiency. But although the direct comparison of EROI might be small, I wonder about this part from that article:
“It is in part for these fully encompassed systems reasons, that in the conclusions of Murphy and Hall's paper in 2010, an EROI of 5 by their extended methodology is considered necessary to reach the minimum threshold of sustainability,[22] while a value of 12–13 by Hall's methodology is considered the minimum value necessary for technological progress and a society supporting high art.”
So different values of EROI can yield vastly different civilizational results, the difference between base sustainability and a society with high art and technology. The direct energy outputs might not be thousands of times different, but the information output of different EROI levels could be considered thousands of times different. Without a massive efficiency increase, society over the last few thousand years got much more complex in its output. I’m not trying to change terms here just to win an argument but trying to qualify the final results of different capacities of harnessing energy and technology.
I think this gets to the heart of the different arguments we’re making. I’m not in any way arguing that these old architectures are more common in total quantity than ARM. That difference in production is only going to increase. I wouldn’t have known the specific difference, but your data is great for understanding the scope.
My argument is that projects meant to make technology that has been manufactured for a long period of time and has been widely distributed more useful and sustainable are worthwhile, even when we have more common and efficient alternatives. This doesn’t in any way contradict your point about ARM architecture being more common or useful, and I’d be fully in favor of someone extending this kind of project to ARM.
In response to some of the other points: using an external crystal is just an example of how you could use available parts to maintain the Z80 if it needed fixing but you had limited resources. In overall terms, it might be easier to throw away an ARM microcontroller and find 100 replacements for it than even trying to use an external crystal for either one, but again I’m not saying it’s a specific advantage to the Z80 that you could attach a common crystal, just something that might happen in a resource-constrained situation using available parts. Better than the kid in Snowpiercer sitting and spinning the broken train parts at least.
Also, let me clarify the archive.org part. I wasn’t trying to demonstrate the best process for getting info. I just picked that because they have lots of scanned books to simulate someone who needed to look up how to program a part they found. I know it’s using ARM, but the reason I mentioned that had to do with the distribution of paper books on the subject and how they’re organized. The book I linked to starts with very basic concepts for someone who has never programmed before and moves quickly into the Z80, all in one old book, because it was printed in a simpler time when no prior knowledge was assumed.
There are plenty of paper books on ARM too, and probably easier to find, but now that architectures are becoming more complicated, you’re more likely to find sources online that require access to a specific server and have specialized information requiring a certain familiarity with programming and the tools needed for it. More is assumed of the reader.
If you were able to find that one book, you could probably get pretty far in using the Z80 without any familiarity with complex tools. Again, ARM is of course popular and well-documented, but the old Z80 stuff is still out there and simple enough to understand and even analyze with your bare eyes in more detail than you could analyze an ARM microcontroller without some very specific tools.
So all that info about ARM is excellent, but this isn’t necessarily a competition. It’s someone’s passion project who chose a few old, simple, and still-in-production technologies to develop a resilient and translatable operating system for. It makes sense to start with the earlier technology because it’s simpler and less proprietary, but it would also make sense to extend it to modern architectures like ARM or RISC-V. I wouldn’t be surprised if sometime in the future some person or AI did just that. This project just serves as a nice starting point for an idea on resilient electronics.
About the return on investment, the methodology is interesting, and I’m surprised that a hand crank to nuclear would increase so little in efficiency. But although the direct comparison of EROI might be small, I wonder about this part from that article:
“It is in part for these fully encompassed systems reasons, that in the conclusions of Murphy and Hall's paper in 2010, an EROI of 5 by their extended methodology is considered necessary to reach the minimum threshold of sustainability,[22] while a value of 12–13 by Hall's methodology is considered the minimum value necessary for technological progress and a society supporting high art.”
So different values of EROI can yield vastly different civilizational results, the difference between base sustainability and a society with high art and technology. The direct energy outputs might not be thousands of times different, but the information output of different EROI levels could be considered thousands of times different. Without a massive efficiency increase, society over the last few thousand years got much more complex in its output. I’m not trying to change terms here just to win an argument but trying to qualify the final results of different capacities of harnessing energy and technology.
I think this gets to the heart of the different arguments we’re making. I’m not in any way arguing that these old architectures are more common in total quantity than ARM. That difference in production is only going to increase. I wouldn’t have known the specific difference, but your data is great for understanding the scope.
My argument is that projects meant to make technology that has been manufactured for a long period of time and has been widely distributed more useful and sustainable are worthwhile, even when we have more common and efficient alternatives. This doesn’t in any way contradict your point about ARM architecture being more common or useful, and I’d be fully in favor of someone extending this kind of project to ARM.
In response to some of the other points: using an external crystal is just an example of how you could use available parts to maintain the Z80 if it needed fixing but you had limited resources. In overall terms, it might be easier to throw away an ARM microcontroller and find 100 replacements for it than even trying to use an external crystal for either one, but again I’m not saying it’s a specific advantage to the Z80 that you could attach a common crystal, just something that might happen in a resource-constrained situation using available parts. Better than the kid in Snowpiercer sitting and spinning the broken train parts at least.
Also, let me clarify the archive.org part. I wasn’t trying to demonstrate the best process for getting info. I just picked that because they have lots of scanned books to simulate someone who needed to look up how to program a part they found. I know it’s using ARM, but the reason I mentioned that had to do with the distribution of paper books on the subject and how they’re organized. The book I linked to starts with very basic concepts for someone who has never programmed before and moves quickly into the Z80, all in one old book, because it was printed in a simpler time when no prior knowledge was assumed.
There are plenty of paper books on ARM too, and probably easier to find, but now that architectures are becoming more complicated, you’re more likely to find sources online that require access to a specific server and have specialized information requiring a certain familiarity with programming and the tools needed for it. More is assumed of the reader.
If you were able to find that one book, you could probably get pretty far in using the Z80 without any familiarity with complex tools. Again, ARM is of course popular and well-documented, but the old Z80 stuff is still out there and simple enough to understand and even analyze with your bare eyes in more detail than you could analyze an ARM microcontroller without some very specific tools.
So all that info about ARM is excellent, but this isn’t necessarily a competition. It’s someone’s passion project who chose a few old, simple, and still-in-production technologies to develop a resilient and translatable operating system for. It makes sense to start with the earlier technology because it’s simpler and less proprietary, but it would also make sense to extend it to modern architectures like ARM or RISC-V. I wouldn’t be surprised if sometime in the future some person or AI did just that. This project just serves as a nice starting point for an idea on resilient electronics.