Federico Musto Of Arduino SRL Shows Off New ARM-based Arduino Boards

I caught up with Federico Musto, President and CEO of Arduino SRL, at the 2016 Bay Area Maker Faire. Their company is showing off several new boards being prepared for release as early as next month. In partnership with Nordic Semi and ST Microelectronics they have put together some very powerful offerings which we discuss in the video below.

arduino-primo-core-alicepad-star-otto-lcdThe new boards are called Arduino Primo, Arduino Core, Arduino Alicepad, and Arduino Otto.

The first up is the Primo, a board built to adhere to the UNO form factor. This one is packing an interesting punch. The main micro is not an Atmel chip, but a Nordic nRF52832 ARM Cortex-M4F chip. Besides being a significantly fast CPU with floating-point support, the Nordic IC also has built-in Bluetooth LE and NFC capabilities, and the board has a PCB antenna built in.

On an UNO this is where the silicon would end. But on the Primo you get two more controllers: an ESP8266 and an STM32F103. The former is obvious, it brings WiFi to the party (including over-the-air programming). The STM32 chip is there to provide peripheral control and debugging. Debugging is an interesting development and is hard to come by in the Arduino-sphere. This will use the OpenOCD standard, with platformio.org as the recommended GUI.

The same nRF52 microcontroller is present on the Arduino Core and the Alicepad, which are targeted at wearable electronics. The circular form factor of the Alicepad mimics the familiar sewable form of the Lilypad.

Arduino’s other offerings are where the horsepower really gets crazy. The Otto board boasts a gigantic STM32F469: a 169-pin ARM Cortex-M4F clocked at 180 MHz. The chip has a ridiculous assortment of built-in peripherals, and you’re not likely to run out of either pins or CPU cycles. It’s also got a hardware graphics accelerator, so it’s no surprise to find that the Otto has a DSI-IF connector on the back that is designed to plug into the LCD screen also being demonstrated at the event: a capacitive touch 480×800 display. The Otto also includes an ESP8266 to provide WiFi (why not, right?).

There are a few question marks in my mind on this one. First off, the Otto and the LCD have a product-family designator of “Star” which will be assigned to all the boards that feature the STM controllers. This seems a bit confusing (Star Otto, Star LCD, etc) but I guess they want to differentiate them from the “normal” Arduini. But are these devices becoming too complex to bear the Arduino name? Maybe, but the UNO is always going to be there for you and the new boards give you access to newer and more powerful features. Whether or not this complexity can be easily harnessed will depend on the software libraries and the IDE. After all, I think Donald Papp made a great point earlier in the week about the value of Arduino comfort in custom electronic work.

The Lawsuits

Finally, I asked Federico if there is any news about the Arduino versus Arduino trademark litigation. He spoke with us almost a year ago on the topic, but he had no new information for us at this point. (The US court case may be ruled on as early as July of this year, so there’s probably not much he could say, but I had to try.)

Federico spoke a little bit about the conflict between the two Arduinos, and said that it was brewing inside the company long before he got there. And it does appear that both companies calling themselves Arduino are trying to outdo each other with new boards and new initiatives, and going in different directions. If there is a bright side, it’s that this competition may end up building us better hardware than a single company would, because both are making bets on what will put them out ahead of the game.

61 thoughts on “Federico Musto Of Arduino SRL Shows Off New ARM-based Arduino Boards

  1. I think a split is a good thing. Once they get past who gets the Arduino name, We will have two companies trying to out shine the other. Competition normally means consumers win.

        1. I could understand that if I was rolling PCBs in my basement on a budget – but I assume SRL is doing this on a scale of 100s of thousands if not millions. Point being – I can’t throw that in an enclosure as-is. The cost then rolls on to the user who needs to buy a bigger enclosure and install an extension + connector to make it flush – or re-solder x amount of boards to correct the angle before installation. We use jigs in our lab all the time to correct issues like this – I don’t think it would be too hard to place the unfinished PCB in a fixture prior to soldering to ensure the connector angles are proper. /2cents

          1. I agree. Professional job would put slots where needed. Besides, it’s a pcb, not sheet metal. And I hope that connector gets fixed or was accidently twisted, left a bad taste.

          2. If only the arduino users had access to some technical stuff like a soldering iron to fix that if they thought it so important.
            But that’s of course way out of their reach right?

            Fortunately they can still buy apple products and then they also don’t have to worry about controlling their hardware, yay!

          3. Slots are out if you are running to a tight budget but your comments re: soldering jig seem like a good idea if it can be implemented.

            I rarely see slotted holes even on more expensive boards so I assume cost is *the deciding* factor. The smaller USB connectors have the same issue.

        2. There are version of those connector designed for round holes and I have actually ordered them from digikey.
          The slotted hole is done by a router bit. You already need it for making a panel or that curved shape on the right hand side of the board, so there isn’t much of a cost saving not to use it.
          If you insists on doing round holes for slotted pins, you can offset the holes to pull the part straight.

          1. I agree with comments by the most part but I would like to point out a machine to *rout* a board is a lot different to a machine to *drill* a board and would be a lot different in cost.

            True you have to rout the edges of the board in the final stages but these slots have to be routed before the though holes / vias are plated so you need to use a more expensive machine in the early stages *as well*.

            I expect this would be reflected in cost.

            As for the angle – all the boards with round holes seem the same to me no matter where they come from. I rarely see a flat pin in a round hole that isn’t a bit squiffy.

          2. BTW look at the shape of those pads around the barrel connector… Look closely at the area around the width of the pins. Pretty sure that it is slotted. If it were a round hole, you would see part of the hole outside of the package.

    1. That is right. But I don’t like the split of a company due to right management, because one of them is a thief, and one thief is earning money. Another company from an uncorrelated person instead would make the competition right (clones, not counterfeits, are considerable too).

      Btw, here in Italy, in Turin, there is an headquarter+fablab of Arduino.cc (LLC one). The majority of people which works for Arduino before this split, still works there. To me, says a lot to where the heart of this movement belongs.

      1. Maybe you are right, however I have no idea who the real Arduino are, I have heard from both sides and both present a good case. I also read some history about the history of Arduino and it’s based on a students work stolen by their professor so I have no loyalties to either side anyway. I hope it is resolved soon.

        1. Yes, I have heard that the professor you are talking about (Banzi maybe? Not sure) have used the students project. It was not properly stolen, because was open source, but there was no the proper attibution. Moreover, in the Arduino IDE, there’s no mention on the original developers (of the Wiring). This happens in many universities, too.

          A lot of “famous” people are based on something bad, sadly.

          Maybe what happened to Arduino was just the Karma. :)

          If I need to buy a board, for this reason I don’t know from where to buy. Maybe I’d made it by my self (I have the luck of being able to do that).

  2. This will be interesting. A while back Codebender published a chart of projects using Arduino. It was IIRC over 90% wer variations on the 328 boards – Uno, Redboard, Metro, Nano, Pro, etc. And a tiny fraction of ARM versions. As complication goes up, usage drops off very precipitously. I wonder if we will see this with these new boards?

        1. Arduini is perfectly fine in Italian. “Paciocchi ancora con gli Arduini?” – “do you still fiddle with Arduino?” It is correct, but sounds weird as Arduino is a brand name and isn’t pluralized. A special case were it should be used is when two different kind of Arduino exist; “the legal war between Arduini rages!” is a better example. Arduinos sounds too spanish, if you really have to add a plural “Arduins” is cooler.
          As a bonus Raspberry pi is often shortened to “Raspino” and ESP8266 is “EspOttoDueSeissei”.

  3. Does this mean that the Maple code base can merge with the arduino code base? Will the Otto contain an on-board ST-Link? Will the arduino IDE finally get a debugger inside!?!?

    1. The initial codebase is supposed to be built on the base of the Maple-code, just being extended to support the STM32F4’s peripherals and such, but they’re planning to switch to CubeMX-based codebase in the long run. The initial codebase for the Otto-board should be available on Arduino.org’s github in a month or so, they say.

  4. Love STM32F4 one.
    But..Arduino is simple, arduino is fun…how they will incorporate pins configuration nightmare, DMA transfers and RTOS into somewhat childish environment?
    Those things are needed to unleash power of this beast….

    And I don’t ask how much it will cost…certainly too much.

    Meanwhile ST launches new nucleo boards :)
    $30
    http://www.st.com/content/st_com/en/products/evaluation-tools/product-evaluation-tools/mcu-eval-tools/stm32-mcu-eval-tools/stm32-mcu-nucleo/nucleo-f429zi.html

    1. Why do you think providing DMA-support or such would be an issue? What’s stopping them from just adding new functions for that? Like e.g. in the Stm32duino we have separate functions for transferring data over SPI with the help of DMA.

      I mean, it’s not magic. The IDE may be limited compared to the more professional tools, but there’s nothing stopping anyone from writing as complex code as they may want to.

    2. I think the overall code structure is not taking full advantage of the parallel nature of the DMA as much as a RTOS would. The CPU could be doing something else or in power saving mode instead of getting stuck in a polling loop.

    1. BGA packages especially the fine pitched packages requires a lot more layers to break out the inner balls than the old QFP. For a simple board like this, a QFP can be easily be routed in 2 layers or 4 layers if you want better signal quality with a power/ground plane.

    2. With a QFP, all the traces are on a single side of the board. In theory, you could get by with a double sided board.
      Once you go BGA, you are looking at 4 or more layers.

      Then there’s the BGA pad spacing. Larger old chips, like the Xilinx Spartan series, have 1mm between pads. Newer BGA’s have even less. With each reduction in size, tolerances get tighter and the cost of the PCB goes up. What is more important – saving a few dollars, or overall board size?

  5. The Otto board seems very counter intuitive for the Arduino brand.

    The native Arduino language (Sketch) cannot support the advanced or extended features of this chip. I would expect to see many libraries written in C but that isn’t what Arduino is traditionally about.

    Arduino was always about a *simple* platform where beginners could *easily understand* the code base. That is NOT going to be the case if they are confronted with complex C libraries.

        1. With my limited understanding of the Arduino language it seems that it is simply a variant of C with some pre-compile processing that removes some of the more tedious requirements of C by automating the process.

          In the above context the Arduino environment (IDE) does not actually have a language BUT …

          The Arduino website promotes a “language like” environment that has what would otherwise be called primitives like analogRead(), digitalWrite().

          By having such a limited set of pseudo primitives, the Arduino environment has excluded itself from more complex tasks that would, or might, require more processing power such as the board I mentioned.

          So my point was that the more capable board does not fit in because the pseudo language does not support basic (pseudo) primitives that can take advantage of the extra functionality.

          And that only eave pure C and that was never what Arduino was about.

  6. Let’s see what China makes of this… :-D
    At least we get a good amount of faster boards now. I just hope the IDE can keep up.
    It will also be interesting to see people use the flash / RAM of the ESP8266 for the main core, because it seems the ESP8266 can have more of both (and is dirty cheap).

  7. This is why i read/watch HaD, great interview, nice that you asked about the whole dividing of the company and the guy was actually trying to answer that (for as far as he could) For that matter, this guy seems like he has the right idea for Arduino, ima make sure all my future Arduinos come from him.

  8. ARM-base boards have been around for a while but so far they’ve all used Atmel AT9S series ARM microcontrollers. This would be the first time ARM-based Arduino boards have been made with something other than Atmel chips.

Leave a Reply to OlivierCancel reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.