Ask HN: Thoughts on memristors?

11 points·deepnotderp··4 comments

I'd love to know the HN crowd's thoughts on memristors.

How far off they are from production?

Technical problems to be overcome?

Economical problems to be overcome?

Compatibility with CMOS?

Scaling capabilities (ie can it match Moore's Law?)

Anything in general :)

View on HN

Some interesting resources about DIY scale memristors...

The coherer is a memristor:

https://arxiv.org/pdf/1306.0942.pdf

https://www.youtube.com/watch?v=3ZRIPdr1lug

A natural memristor: https://www.youtube.com/watch?v=LpHJARlKT1o

It'd be awesome if someone manufactured DIY friendly, through-hole memristors. All these big companies are focused on making memristors for integrated circuits, but it really makes sense to have some human-friendly memristors available for makers and tinkerers.

I imagine that this could be a viable business for some smart, creative folks who understood the potential for simple, useful applications of these things.

"The main problem with memristors is no one has figured out how to make large numbers of them that are reliable enough for commercial electronic devices. Researchers continue to puzzle over the best materials to use and the most effective way of manufacturing them"

From: https://www.scientificamerican.com/article/hpe-debuts-its-ne...

They are _in_ production, available in 16-DIP packages. They are hella expensive ($20ea for a 'burn and learn' off-spec unit, hundreds for a spec one).

You can buy them here: http://knowm.org/product/

Memristors are an interesting concept theoretically. There was a thread about this on HN recently.

https://news.ycombinator.com/item?id=14562399

Reply on news.ycombinator.com