NETduino | Hacker News Reader

NETduino

netduino.com

45 pointsshawndumas20 comments

In the past, I've had some projects that I'd like to have tried, if I knew what I was doing with circuitry. I have a feeling an Arduino (or Netduino) would have taken a lot of the pain out of it. But eventually, I'd probably want to make a version that didn't start with a $35 part, possibly to sell or let others make.

Is there a pretty clear path from starting with a 'duino and moving to custom set of components (assuming we're not talking about anything crazy complex like making a smartphone lol) or would it be essentially re-learning everything from scratch?

For example, the other day I wanted to plug a GC controller into a USB port. I'm pretty sure I could wire it up to a programmable chip and make that happen. I'm also pretty sure an 'duino would be easier. But would making one version make it easier to make the other?

It all depends.

Systems like this are (IMO) great for doing proof-of-concept projects and development things where you are concentrating on other factors. Ex: if I enable my TV remote to operate my garage door, what should I use the Mute button for...

With these platforms (Arduino, BASIC stamp, etc.) you create a lot of "glue" between different devices and protocols, including getting samples from sensors, and so on without a lot of effort.

The down-side, as you note, is that for many many reasons they are unsuitable to build any kind of semi-commerical product around.

Going from Arduino prototype to real product is too vague of a question though. If your needs are simple, you might be able to use a handful of discrete components and 74xx series logic chips, or you might need to add in a PIC, or an Atmega chip, or whatever. But in almost every case you can put together a purpose built piece of hardware once you have your basic requirements fleshed out that is cheaper and easier to produce than a product built on top of the Arduino.

Most of the "pain" with starting from scratch goes away with just about any board-level product. I like the Teensy because it's cheap ($19 to start) and has an onboard bootloader so it programs (and is powered!) directly from USB with no other hardware needed. You can get helloworld.c running in less than an evening.

If at your entry point you're already thinking of a commercial or hobbyist product, it's probably better to skip the Arduino and either go straight to the chip (maybe get an Atmel Dragon development kit) or buy a cheaper board that has a bootloader and sample code.

At some point you're going to have to get comfortable with hardware, so Arduino won't buy you anything.

If you want to retain Arduino sketch compatibility for one reason or another (many identical builds but not at commercial scale? Who am I to question?) you can build Arduino clones from components and perfboard for under $10.

Edit: Nix that, just buy the Ardweeny: http://www.makershed.com/ProductDetails.asp?ProductCode=MKSB...

You can pop the ATMega chip off of the development board and use it in place of the complete arduino. ATMega chips are pretty cheap (edit- I said $1 before but that may have been optimistic).

If you just want to plug a Gamecube controller into your computer this is probably a better option: http://goo.gl/Y0nP9

Actually, I've read bad things about that particular adapter, but I did have an alternative in mind. But thanks!

I didn't fully understand the ATMega bit, but I'm assuming it means I can even use the same C++ program and not have to rewrite it for a new chip? That would be an amazing way to go, if so. I had assumed I'd at least have to recompile for a different chip.

Edit: Looks like http://www.arduino.cc/playground/Learning/AtmegaStandalone explains it for me. Thanks for the tip! That's exactly what I needed to push me further.

Building your own Arduino clone is pretty straightforward - it's half a dozen components, under ten bucks. Once you get the hang of it you can prototype your design using an actual Arduino board, then simply integrate the components into your actual project and continue using the same firmware. I've done this many times.

The only gotcha with this is that you'll need a way to put the Arduino bootloader on new AVR chips, or pay a small premium to buy them preloaded with Arduino firmware.

I just need to correct myself. That adapter is the one I had picked out. It was the OTHER adapters that I read problems about. :)

>> "Netduino is also pin compatible with Arduino shields."

I think this is an awesome idea. Writing code in the original Arduino C dialect was a bit frustrating (for a Java guy). Being able to write in a higher level language that interops with already available Arduino accessories is just fantastic.

I saw a demo of a Netduino a .NET developers group in Raleigh, and was very impressed. It was extremely cool to see someone step through C# code in Visual Studio that was running on the board.

IIRC, the SDK runs on the Express (free) version of Visual Studio as well, so you really can get started for $35.

now I want iDuino... an objective-c based open source embedded system. It's a pretty good startup idea actually, considering everything i-based product can charge more...

Very nice. I've wanted to do an Arduino project for a while (or an Android ADK, or this) but I can't think of any project ideas. I don't want to create a remote controlled car, or a gmail notifier. I want to create something real, something useful.

Are there any NETduinos that can handle more than 20 Gs of force out of the box?

Depends on the duration.

I know a guy that is basically looking for an MS Stack solution to a problem he is having. He makes displays for fighter jets and he would like to know if these can solve them. The only problem is that it is all military spec stuff, ergo question.

Whether the item works is the least of your problems when it comes to military jets. Trust me, if it's UK-mil it's a bureaucratic nightmare (expensive) to develop software that can be proven to be of high-integrity. To reduce costs you use stuff that has been proven safe before (we still use 8051s and wire-wrap).

Put simply, if the pilot makes decisions on how to fly (in flight or on later analysis) based on an electronic piece of equipment, it'll have cost shitloads. This fact is why I want to get out of mil-aero in the UK; developing embedded soft/elec is mostly about jumping through hoops.

Does anyone know if .NET micro supports F#?

I actually asked a guy at OSCON, and no, F# is not supported. The CLR on the Netduino isn't complete and doesn't support every .NET language.

I've played with a netduino extensively. My only comment is that they are HIDEOUSLY SLOW and expensive compared to other similar platforms such as AVR+GCC and PIC. They are however convenient and easy to debug, which are the only positive things I can say.

I spend 70%+ of my time with .Net/C# happily so this is not a langauge/Microsoft bias.

Reply on news.ycombinator.com