The First High-Yield, Sub-Penny Plastic Processor (ieee.org) 47
IEEE Spectrum reports:
For decades, hopeful techies have been promising a world where absolutely every object you encounter — bandages, bottles, bananas — will have some kind of smarts thanks to supercheap programmable plastic processors. If you've been wondering why that hasn't happened yet, it's that nobody has built working processors that can be made in the billions for less than a penny each.... The problem, according to engineers at the University of Illinois Urbana-Champaign and at British flexible-electronics manufacture PragmatIC Semiconductor, is that even the simplest industry-standard microcontrollers are too complex to make on plastic in bulk.
In research to be presented at the International Symposium on Computer Architecture later this month, the transatlantic team presents a simple yet fully functional plastic processor that could be made at sub-penny prices. The Illinois team designed 4-bit and 8-bit processors specifically to minimize size and maximize the percentage of working integrated circuits produced. Eighty-one percent of the 4-bit version worked, and that's a good enough yield, says team leader Rakesh Kumar, to breach the one-penny barrier.
"Flexible electronics has been niche for decades," says Kumar. He adds that this yield study shows "that they may be ready for the mainstream."
Thanks to Slashdot reader Iamthecheese for sharing the article
In research to be presented at the International Symposium on Computer Architecture later this month, the transatlantic team presents a simple yet fully functional plastic processor that could be made at sub-penny prices. The Illinois team designed 4-bit and 8-bit processors specifically to minimize size and maximize the percentage of working integrated circuits produced. Eighty-one percent of the 4-bit version worked, and that's a good enough yield, says team leader Rakesh Kumar, to breach the one-penny barrier.
"Flexible electronics has been niche for decades," says Kumar. He adds that this yield study shows "that they may be ready for the mainstream."
Thanks to Slashdot reader Iamthecheese for sharing the article
I would think the communication would matter (Score:2)
more than the processing to bring big benefits from something like this. As the amount of processing you want from your banana is quite limited..
Re: (Score:2)
I won't be satisfied until I can run a Beowulf cluster of bananas.
Re: (Score:2)
I want my smart banana phone!
Power? (Score:1)
Re: Power? (Score:3)
Alkaline button cells are about 0.9 cents USD in bulk. and Zinc air button cells are 1.2 cents, maybe less if you look hard enough and are flexible on specs. Lithium coin cells are around 4 cents. For a disposable circuit I would prefer the zinc-air as it is the least toxic and uses easily repenished materials.
They'll land in the ocean (Score:5, Insightful)
And sooner or later, you're eating AI sushi.
Re: (Score:2)
That's sick, it could be sentient.
"God Goo" (Score:1)
No, the AI sushi will be eating you.
Fingers crossed (Score:4, Insightful)
Wouldn't it be wonderful to swamp the Earth and its soil, oceans and air with yet another kind of ubiquitous, single-use, impossible to collect and recycle plastic thingy? We'll have them in food we eat and in our own bodies real soon now
Re: (Score:3)
These processors could easily more than earn back the couple grams of plastic they might weigh through efficiency gains in distribution, reductions in waste of perishable goods, leaner supply chains, or simply displacing conventional processors in low-end niches.
Re: (Score:1)
Re: (Score:2)
Really? You want to judge people for being made happy by more useless disposable crap that just ends up in landfills?
FTFY.
Re: Fingers crossed (Score:4, Interesting)
Re: (Score:3)
Re: Fingers crossed (Score:2)
Re: (Score:2)
And a a "normal person" would just know all that by just looking at the item/wound ...
Re: (Score:2)
Remote sensors: Internal temperatures on thermos containers, fluid levels in sealed containers, water sensors for individual plants, fluid levels for hummingbird feeders, airflow monitors built into furnace filters, airflow monitors built into individual vacuum bags, voltage monitors built into batteries, pressure monitors on asthma inhalers, balance sensors on fans
Coupled with e-ink displays: display drivers for magazine adverts, individual price tags on merchandise, conference name tags, cheaper RFID dev
Re: (Score:2)
fluid levels in sealed containers,
If the container is sealed, the fluid level is exactly the same it was, before it got sealed.
Re: (Score:2)
I guess it depends on how we define "sealed". How about I redefine it as "opaque". How much fuel remains in a portable propane cylinder? How much spray paint remains in a pressurized can? A small display on the spray nozzle connected to a cheap processor could eliminate the guess work.
Yeah, some of those examples are perhaps a stretch of what is possible, but at less than a penny per processor and a nickle for a display, perhaps some of them might be marketable.
Re: (Score:2)
Would probably work if the cpu only measures what is going through and the display is displaying the remaining part. It is not very easy to have a thick pressurized metal cylinder and a measuring device inside that somehow reports through the metal to the outside.
But interesting ideas.
Re: (Score:2)
How does that help with plastic pollution?
Re: (Score:2)
Re: (Score:2)
When did everybody become a technophobe even on /. ?
These processors could easily more than earn back the couple grams of plastic they might weigh through efficiency gains in distribution, reductions in waste of perishable goods, leaner supply chains, or simply displacing conventional processors in low-end niches.
Or perhaps solve a major problem if they can help recycling machinery automatically identify the exact type of plastic present in a piece of waste so it can be sorted properly.
Cost vs. benefit (Score:3, Informative)
If you've been wondering why that hasn't happened yet, it's that nobody has built working processors that can be made in the billions for less than a penny each....
Well, that and the added value of smart bandages, bottles, bananas is apparently less than a penny.
Re:Cost vs. benefit (Score:5, Funny)
I don't need a CPU on my banana to tell me when it's good to eat -
if it's green, wait a couple of days.
If it's all yellow it's good to go,
if its got a few black spots, eat it immediately
if it mostly black chuck it or make banana muffins
if it's green and hairy, make a note to look at my bananas more often.
Re: Cost vs. benefit (Score:2)
Re: (Score:2)
That is not how shops work with bananas, lol.
They buy them - and simply put them on the shelf. What else would you do?
Re: (Score:2)
-1, TMI.
Waste disposal? (Score:1)
Re: Waste disposal? (Score:2)
Inactive yeast and dry agar agar. once it gets damp with moderate heat it actives. Make sure the polymer for the circuit substrate is something that your yeast strain can break down. Nothing weird about adding yeast in a landfill either.
I thought this was about plastic recycling. (Score:3)
As for the actual topic, do they also have penny battery and at least a penny LED or speaker so it can blink or beep for four cents? Woops it might need a timing device too uh oh thats another penny. That toy is now five cents.
NFC handles that for under a penny (Score:2)
Power and IO can be over NFC or similar. The required hardware is a loop of thin wire.
digital designers can't do math ? (Score:2)
From TFA
"It hasn’t been for want of trying; in 2021 Arm reproduced its simplest 32-bit microcontroller, the M0, in plastic, but even this couldn’t hope to meet the mark."
They knew they wanted to hit sub-penny. They know what the yield is. Still, they put "simple" 32-bit processors on a wafer and then found out that they weren't small enough and the cost was too high ?
WTF ?! Rule # 1 of cost on ICs, the smaller the cheaper, duh. Once they knew the price target they new about how big the die cou
Re: (Score:2)
If they'd been able to make the M0 work, they wouldn't have had to spend the time and money to create a new, simpler design, and their customers would've been able to reuse existing hardware and software that are compatible with it. But yeah, a 32-bit chip that can monitor a sensor and emit a signal based on the sensor's value seems excessive.
Design, not material. (Score:2)
The most important element to actually achieving reliability was the actual CPU design.
* Use 4-bit instructions
* Segment instruction memory from data memory.
* Eliminate pipelining entirely. It's 100% clock synchronous.
* Minimize the number of gates as much as possible.
Ultimately, they use 2,104 "semiconductor devices" which suggests they are not mere transistors.
Re: (Score:2)
4 bit *instructions* ?
That would be *extremely* limiting. Even the 4004 used 8 bit instructions.
For that matter, I'm not aware of any processor actually put into use that had 4 bit instructions, although Popular Electronics had a project in the late 70s or early 80s that built a 4 bit computer with 4 bit address space and 4 bit instructions.
hawk
Re: (Score:2)
Popular Electronics had a project in the late 70s or early 80s that built a 4 bit computer with 4 bit address space and 4 bit instructions.
That means there are only 256 possible programs which could run on the device. It does not look very useful. Anyway it is an interesting bit of info.
Re: (Score:2)
indeed--it was a demonstration project to be interesting in a day when 8 bit machines were beyond the reach of many readers.
But there are more possible programs than that--16 four bit words is 64 bits, so 2^64 theoretical combinations of program and data.
I finally lucked out on a search. Here's the June, 1978 issue:
https://worldradiohistory.com/... [worldradiohistory.com]
it has an add, but no subrtact, and no "if" . . . but it was a nice demonstration of how things work.
I never built it, but rather an 1802 from scratch (wire wrap
Tracking ? (Score:2)
A 4bit processor (Score:1)
If they can print electronics onto flexible media (Score:2)
Another Data Amassing Tool (Score:1)