I really don't understand everyone's obsession with mobile computing power. Why aren't we equipping our our laptops with low energy processors and use them to remotely access more powerful stationary work stations via networking? Instead of carrying around 4.8 GHz all the time, I'd much rather have multiple days of battery lifetime. We must have taken a wrong step somewhere. I am still convinced processing outsourcing is the future for all things hardware intensive such as gaming, if it is not for some unexpected milestone in battery technology.
Besides,
> 4. which doesn't need batteries or plugging in
what do you mean by that? Would that include only solar-powered devices?
I was talking of the future, but you're describing the present! There's no reason why local neighbourhood datacenters and ubiquitous high speed mobile networks couldn't be a thing some day.
You don't even need neighborhood datacenters; you could just access your desktop machine in the basement with your lightweight portable mobile terminal in the living room. Just the other night I used mpv on one laptop to stream a video file from the other laptop over Wi-Fi with python2 -m SimpleHTTPServer, and XPra can do the same thing in the same way for remotely accessed applications. (Of course, ssh -X can kind of do that too, but it's a lot less efficient and more insecure.)
In theory we could have much-lower-power wireless communication systems: maybe using lasers, with MEMS corner reflectors on the mobile station to transmit, and a simple photodiode with a dichroic filter over it to receive. Or maybe using submillimeter waves from a phased-array antenna, like the Starlink terminal. Or just time-domain UWB pulse radio in conventional microwave bands, but optimized for low power usage instead of super high data rates or precise ranging.
But, right now, evidently even Bluetooth Low Energy from the leading ultra-low-power silicon vendor costs 10 milliwatts when you have it on. And it's not clear if the technologies I described above will materialize. So the amount of dumb that it makes sense to put into a wireless networked mobile terminal is only about 10 milliwatts of dumb. And 10 milliwatts is not that dumb. Even with a conventional low-power CMOS Cortex-M (300 pJ per cycle, 2 DMIPS/MHz) that's about 30 MIPS or 60 DMIPS of dumb. That's dumb like a SPARC 10 workstation from the mid-90s, not dumb like a VT100 or an analog TV tuner. With subthreshold logic it's more like 600 DMIPS of dumb, dumb like a 450 MHz Pentium II (introduced 01998, mainstream around 02000).
I agree that making computing power mobile makes it enormously more expensive, especially if you consider batteries unacceptable. But making computing power remote means that you need to spend energy on a radio to access it. That's a good tradeoff for some things, but not for others. In my other comment, note that if we believe Ambiq's datasheet, we can get the CPU speed of a SPARC 20 for 1.8 milliwatts.
It turns out the chip also includes a Bluetooth Low Energy 5 radio, so you can use it to remotely access more powerful stationary workstations via networking, as long as you're within a few meters of a base station. The radio costs 10 milliwatts when you're running it, six times as much as the Pentium-class CPU. Normal radios (Wi-Fi, cellphones) cost orders of magnitude more than that.
So constant remote wireless access to more powerful stationary workstations doesn't start to save energy until the amount of computation you're accessing is close to a gigaflop. Maybe closer to a teraflop if we're talking about streaming full-motion video. Intermittent remote access, of course, is a more reasonable proposition.
It's true that gaming commonly uses teraflops or petaflops of computing power, and plugging in such a beast in a server closet is a huge improvement over trying to somehow cram it into your pocket. But there are a lot of day-to-day things I do with a computer — recompiling my text editor, writing stupid comments on internet message boards, chatting on IRC, simulating an analog circuit, reading Wikipedia — that very much do not require gigaflops of computing power.
(Remote wired access of course can use very little power indeed, but if you're in a position to plug into a wire, you might as well deliver power over that wire too.)
If you take a modern cellphone and take almost all the processing power out of it, you still have a 1000-milliwatt radio and a 1000-milliwatt backlit screen. So you aren't going to get multiple days of battery life that way. 1000 milliwatts is enough to pay for dozens of gigaflops of computing power nowadays.
Myself, I have another reason: I travel, though I've traveled very little during the pandemic. But I am often someplace other than at home: at a café, in a park, at the office, in a bus, in the subway, in a taxi, visiting a friend in another city, at my in-laws' house, and so on. All of these places are out of Bluetooth range of my house. I could obtain internet bandwidth from a commercial provider, but that sacrifices privacy, it's never reliable, and I don't consider it reasonable to make my core exocortical functions dependent on the day-to-day vagaries of mere commerce. Personal autonomy is one of my core values.
Besides,
> 4. which doesn't need batteries or plugging in
what do you mean by that? Would that include only solar-powered devices?