There is no way the design took two years and required supercomputers
But they had to crunch the numbers
Oh shit I take it back
The math is so complex that research into Pringles with ridges was considered a national secret and it was classified. DARPA was rumored to have provided partial funding.
It took 3 seconds, and it was AI what figured it all out.
AI doesn’t figure out shit.
Lol, you’re both wrong. AI absolutely can assist, and sometimes be the main part in finding patterns/discoveries, but in the case of the hyperbolic plane, it was created by humans.
Its design took 2 years and required the use of supercomputers
[citation needed]
It is quite hard to track down but here’s it being reported by the head of modelling at P&G in 2006
https://www.hpcwire.com/2006/05/05/high_performance_potato_chips/
Though worth saying that the link suggests the computing was used for aerodynamics for ensuring production wouldn’t destroy them not. For the shape as such. I’ve also seem it said that the can is part of that too.
Is this real or am I being trolled
I’m pretty sure it’s real. I met someone once who worked in materials research for food and they said that modelling was big there because the scope for experimentation is more limited. In materials for construction where they wanted to change a property they could play around with adding new additives and seeing what happens. For food though you can’t add anything beyond a limited set of chemicals that already have approval from the various agencies* and therefore they look at trying to fine tune in other ways.
So for chocolate, for example, they control lots of material properties by very careful control of temperature and pressure as it solidifies. This is why if chocolate melts and resolidifies you see the white bits of milk that don’t remain within the materia.
*Okay you can add a new chemical but that means a time frame of over a decade to then get approval. I think the number of chemicals that’s happened to is very very small and that’s partly because the innovation framework of capitalism is very short term.
This meme is wrong and likely based on a Reddit post that is itself wrong.
“TIL that in the '50s P&G used a supercomputer for designing Pringles…”
The only source I found referencing pringles association with a supercomputer was a 2007 article with this sentence:
Pringles potato chips are designed using [supercomputing] capabilities – to assess their aerodynamic features so that on the manufacturing line they don’t go flying off the line," said Dave Turek, vice president of deep computing at IBM.)
Pringle’s didn’t exist until 1968. Why would they waste a decade’s worth of supercomputing time (per the Reddit post that they were designed in the “‘50s using a supercomputer”) to design a potato chip?
It does not state that the chips were designed in ‘68 with a supercomputer. It directly states that “today’s supercomputers”…”are creating potato chips”, so their current design was done that way for the purposes of expedited manufacturing processes.
The Reddit posts even links to the article stating that the reference for supercomputer usage in Pringle’s design is modern.
Pringles aren’t legally chips because they’re made from mashed potato dough (as opposed to slices of potato), hence why marketing calls them crisps even in the US.
With the stuff about ‘super computers’, this seems more like a shitpost than a science meme.
That article is useless. Any other reference?
I mean, you can just google “pringles super computers” They did use a supercomputer in modeling the thing, most articles are useless cuz they’re not going to go into details about how or why.
or here’s a CNN article about it. the blurb is just
“Pringles potato chips are designed using [supercomputing] capabilities – to assess their aerodynamic features so that on the manufacturing line they don’t go flying off the line,” said Dave Turek, vice president of deep computing at IBM. Personally, though, you’re missing out on this gem
Keep in mind, pringles were around since 1968, apparently. modeling and simulating dynamic airflow was probably barely in it’s infancy. (and even to day, computational fluid dynamics is some exceedingly hard math to rock)
wonder what kinda of supercomputer they used in 1968
A Cray would be my guess. Possibly even a CDC 7600 running at a whopping 36 MHz if they were really on the bleeding edge (it came out that same year).
