@kb3001, I agree with your "security soution is too predictable" first point....but again, this is a relatively immauture market. People use what they have to hand to get in market and learn. What is different with this industry wave is that feedback is fast....the innovation cycle enabled with people using Kickstarter, Indiegogo has lowered barriers to entry to young entrepreneurs....what I think will be key to improving power and system cost will be in shifting (dynamically potentially) the processing to the connected PC/Phone or Cloud...make the end device a "thin client" equivalent for the IoT space. Given use cases, there will, of course, be applications where nearly all the intelligence HAS to be in the edge device..but for many, I think the thin client analogy is a good strategy for success.
IoT is about improving the intelligence of systems through connectivity. In some cases, the overall goal is to reduce energy consumption. In other cases, the motivation is different (a connected patch given to you at hospital is to improve diagnosis/treatment, a connected agricultural system could be in place to improve efficiency of using pesticides or water irrigation). The fact you add wired or (more frequently) wireless connectivity to a system means, when compared to a "black box" embedded system, power is higher..today.....howvever, more highly integrated chips are coming.Integration of connectiivty, sensing and processing subsystems will significantly drop power usage of the nodes...and there are other techniques (see my other reply) to improve cost and power too. It is not all doom and gloom!
@fergie1965, I see your point. It seems to me however that this security solution is too predictable (add an ecryption engine to an MCU, big deal!) and that is why the cost is the sum of the parts. What the industry needs is a brand new way of addressing serious issues such as security, using solutions that are compatible with the low cost and low power needed for mass scale IoT. I think the industry is running out of truly inventive ideas to be honest when it comes to IoT.
@kb3001; As I mentioned, price is important. But we cannot get wrapped up purely on the price of the semiconductor component itself. That's my only point. You will see different business models associated with IoT. In some cases, the use case, the "who pays" and the value of data is such that cost is king. In other places, it is sensible to pay more. Been in this industry for 27 years. I could write a book with so many examples of companies that had failed projects as they picked the cheapest part.
It is not a race to the bottom @fergie1965, it is an economic imperative. If IoT is going to be the ubiquitous reality everybody s talking about, we need to bring the cost down. Companies would still make decent money through economies of scale.
1) In my opinion, security is a non negotiable function for connected things. Hardware engines to do this will become increasingly prevalent to improve performance and energy efficiency compared to software only implementations
2) Interesting to see the pricing commentary.....Yes, important.....but I think we have to escape from this "race to the bottom" mentality..If I have valuable data to protect I will pay more..if my supplier (Microchip is recognized as doing a great job with that) can get me into this rapidly developing IoT space quicker and with a more agile solution than others, I value that.