InvestorsHub Logo
Followers 5
Posts 807
Boards Moderated 0
Alias Born 08/24/2013

Re: Dmcq post# 138328

Friday, 11/21/2014 5:04:24 PM

Friday, November 21, 2014 5:04:24 PM

Post# of 151791

I see no problem with making competing products on the leading foundry if they are charged an amount that would make the same end profit as Intel's own designs.


Sure, but I doubt IoT parts will be produced on leading edge foundries. It certainly depends on the details of the product, which is what I meant by variety, e.g. if you take some microcontroller with a communication stack and a modem attached while controlling some sensors, this can be done with a simple microcontroller. That usually has Flash and SRAM integrated which would be an issue to do on leading edge fabs, since those are mainly logic processes. Flash options usually are only available for older/mature processes, just as an example. The same is true for mixed signal stuff, which also makes sense from a cost, power and size perspective. To make it short: Many of those IoT chips will be produced on older processes on depreciated fabs.

ARM produce what they call R series for real time as opposed to M for straightforward embedded.


Cortex M certainly are real time capable. Cortex R, which by the way is a nice concept from ARM, basically provides a compromise of both worlds: tightly coupled memories for real time software and caches plus high speed interfaces for external RAMs. Real time and caches just don't go together. This way you can run complex user code together with some (smaller) realtime code. I don't know if Intel can match that one either.

ARM can certainly be beaten at generic application processors. For embedded applications, they are very well positioned and I don't see much reason for anyone to use Intel processers instead.
Volume:
Day Range:
Bid:
Ask:
Last Trade Time:
Total Trades:
  • 1D
  • 1M
  • 3M
  • 6M
  • 1Y
  • 5Y
Recent INTC News