Actions
Forum Thread Details
  • State Verified Answer
  • Replies 47 replies
  • Answers 2 answers
  • Subscribers 471 subscribers
  • Views 5299 views
  • Users 0 members are here
Related

Choosing a Microcontroller Brand

screamingtiger


I understand that choosing a specific microcrontroller/processor/SOC depends on the requirements and what I am doing etc..

 

I've started a couple discussions but this one is at a critical time, I just used up my last 2 ardunio minis so I need to order a new stash.   I am considering moving away from them and to something more advanced, even if it is just using AVRs directly.

 

This question however is more about choosing a general brand of micro controllers.  In my opinion, for me, I don't have time to learn all of them.  I don't have time or $ to invest in IDEs for different platforms nor do I want a slew of them installed on my PC.

 

So help clear my ignorance here.

 

I am looking AVR, they have a wide range of products.  So does ARM.

 

If I were to say "I only want to use ARM" would I benefit from the following:

-A wide range of products from low power consumption (<100mA) all the way up to full multi core processors ( I can build small embedded projects and then go bigger if needed without having to switch brands)

 

-If I learn the assembler for one ARM product, moving to another ARM product will not result in as steep of a learning curve as the register names, instruction pnuemonics  and paradigm will be similar (subsets).

 

I could as the same question about sticking with AMTEL or PIC (microchip)

 

Does it make sense what I am trying to do?  I just want to cut down the cross system learning curves and stick with a specific set of chips that are very similar in terms of programming.

Parents
  • 0

    I use Silicon Labs, but that's because the place I work uses Silicon Labs exclusively. In general I offer a different opinion than everyone else so far, and that opinion is: it doesn't matter. I find it's the same question as should I program in assembly or in C. Use C, since it will move across platforms easily. Think about the microcontrollers the same way. Understanding the big picture of what a microcontroller does and how it does it will significantly reduce any learning curve. As for the IDEs, they all do the same thing as well. A quick look through a getting started guide is usually enough to get up and going. The number of application notes for each device and sample code makes it quick to setup hardware peripherals easily and quickly. Depending upon the embedded system, you'll probably spend more time developing an interface or a communications protocol than actually coding the micro's peripherals. This is just my opinion though, so take it as that.

  • 0 in reply to michaelwylie

    I would slightly disagree with you.  It seems that different chips may implement things differently.  As in my example with I2C.  The PIC has issues that have to be resolved OR they just have a slightly different way of doing things.

     

    I spent considerable time getting the Pic to talk to the Pi via I2c but the AMTEL did it perfectly.  Some research showed why, but my goal to avoid doing this type of thing again.   The Pi  needed to be configured for repeated starts.

     

     

    I want to pick something, and learn it in depth.  I cannot do that will all the options out there unfortunately.  Sure C will move across platforms but you are going to run into issues and differences that can severely set back a project.  For example, lets say you usually use a PSoC and this time you want to use a Pi for fun.  Sure you can use C, but that is not going to get you very far.

     

    All of this came to light with my Sci Fi your Pi project.  I am using items from Xtrinsic, Microchip, Arduino and the raspberry Pi.  All 4 have different issues to solve and I am using C on all 4.  I wish knowing C was enough to make them all work but they are so different in the way the implement I2C it might as well not even be a "standard".

     

    In the end I want to use sensors, so getting the chip I chose to jump through the hoops needed to use the sensors is most of the work.  That will require different knowlege with AMTEL, PIC and Arm

Reply
  • 0 in reply to michaelwylie

    I would slightly disagree with you.  It seems that different chips may implement things differently.  As in my example with I2C.  The PIC has issues that have to be resolved OR they just have a slightly different way of doing things.

     

    I spent considerable time getting the Pic to talk to the Pi via I2c but the AMTEL did it perfectly.  Some research showed why, but my goal to avoid doing this type of thing again.   The Pi  needed to be configured for repeated starts.

     

     

    I want to pick something, and learn it in depth.  I cannot do that will all the options out there unfortunately.  Sure C will move across platforms but you are going to run into issues and differences that can severely set back a project.  For example, lets say you usually use a PSoC and this time you want to use a Pi for fun.  Sure you can use C, but that is not going to get you very far.

     

    All of this came to light with my Sci Fi your Pi project.  I am using items from Xtrinsic, Microchip, Arduino and the raspberry Pi.  All 4 have different issues to solve and I am using C on all 4.  I wish knowing C was enough to make them all work but they are so different in the way the implement I2C it might as well not even be a "standard".

     

    In the end I want to use sensors, so getting the chip I chose to jump through the hoops needed to use the sensors is most of the work.  That will require different knowlege with AMTEL, PIC and Arm

Children