Hacker News Reader

Comment by smrtfckr | Hacker News Reader

IOTA is built for the internet of things and more specifically for sensors operating in the field. Perhaps with limited acces, perhaps with long intervals between service periods. Generally, you would't care about an 8 fold efficiency increase when you're hooked up onto a powerline but remote sensing equipment does.

View on HN

Such devices are using ARM Cortex M0, M1, or M3 processors. Maybe even lower power Atmel stuff like in the Arduino. Such devices are connected via either expansive uplinks or slow uplinks like with LPWAN.

For all these applications IOTA is not feasible. It is not feasible because the ternary logic needs to be emulated on a processor that itself has not even enough beef to do cryptographic functions by itself. If they have cryptographic support like the e.g. the SE series from STM with a Cortex M0 as used in the Ledger Nano S, then this is optimized for standard cryptographic algorithms and not for some newly invented stuff.

Also if we talk about IOT sensor networking with LPWAN (LoRa, SigFox, ...) than bandwith is a real issue. That means collecting transactions from others and doing PoW on that before sending of your own transaction is completely unrealistic. The remote sensor just wants to transmit its data. But that would mean the LPWAN Gateway needs to be a full node doing the PoW. But LPWAN works, that everybody can listen to transaction request and so we are back to the collisions problem.

Great, someone else quoting 8x efficiency. Maybe you know about the claim? We understand the motivation, which keeps getting mentioned, but what I think we all want to know is:

  8x efficiency _of_what_compared_to_what_?

Reply on news.ycombinator.com