Serious question: what are people building with these boards? The recurring projects I've seen are controlling lights, doors/locks and monitoring water in plants. I don't find these particularly compelling, am I missing something? How is this the next big thing?
Because they're much friendlier to get started for the novice, most of the widespread examples seem extremely trivial to someone with experience.
Now- I'm DEFINITELY not an expert, probably barely mediocre with the EE fundamentals, but I did take a lot of EE and some robotics at Uni to get my degree in Computer Engineering. I am nothing more than a hobbyist but some of my cooler projects have been:
- Autonomous 4WD robot
- Quadcopter
- A USB nerf gun that shoots at the developer who broke the Jenkins build
- Used an electronic photo frame as a display to show happy birthday tweets and Facebook posts to my wife from all the friends and family on her birthday
- WiFi/Proximity garage door opener (before they were cool)
- Garage freezer monitor (after an accidental unplug that spoiled a season's worth of salmon and elk)
- A vending machine that let you pay with your work badge from 'points' you earned (this involved a Raspbery Pi and a surface tablet as well, but the Arduino was what actually controlled the vending machine)
I've done a ton of random home automation IoT gizmos, mostly just for the fun of it, and a lot of robot builds for the boys. But the reason I use a lot of Arduino is that they are pretty friendly to use, there's a lot of help when I'm stuck, and I often don't need more. When I do, sometimes I just add another Arduino (or 4) and use serial or SPI communications between them. I've got some PIC chips for stuff, and some other random chips laying around, but I get more done quickly with the Arduino.
Now, if I ever got on to an idea I wanted to make more permanent I would look elsewhere, but for a hobby, everything else provides more friction and ramp up in the few hours a month I can free up to work on stuff.
EDIT: I'll add I've gotten some ESP8266 chips/breakouts that I've gotten more used to and I may start using them in a lot of places I've previously used Arduino + WiFi because the WiFi breakouts use a LOT of the available I/O and are usually the most expensive part of a build. I think I have 2 or 3 that I reuse.
I always mean to write up my projects but never do. The gist is really, through trial and error I was able to map developer's desks to coordinates and elevation angles to get to their seat and just made a flat file that had their username, X, Y, and Z values in it.
The main loop was just the standard socket server you'd find in a Ethernet module example and I wrote a Jenkins plugin that would connect on break and send the offending username over the socket.
There were at the time a number of examples showing how to control the USB launcher from Geekwire w/ an Arduino that I pretty much just copied.
You probably could have done some basic artillery calculations.
Then again, you might work in one of those weird offices where trying to measure the distance from your desk to theirs is more awkward than repeatedly attempting to shoot your coworkers while you fine-tuned the numbers.
Actually, having someone sit at their desk while I continued to shoot closer and closer to their face was probably the most fun of the entire project.
I tried some basic calculations, but as you could guess with a nerf dart, building HVAC, and all the other things going on it wound up being faster and easier (and again, more fun) just shooting until I was on target a couple of times.
After the first few desks it was pretty easy to get close on the first shot because you could generally know that 4ft left or right was about an adjustment value of +/- whatever from the last coordinate you had dialed in.
Similar to the nerd question - could you describe your approach to the freezer monitor? I have a cabin I don't live nearby to, and I basically can't keep anything in the freezer. It seems to go out often enough to unfreeze then refreeze the ice cube water.
I was thinking a flat thermistor. Then I started thinking - what if it's not the power, but the freezer itself?
Now I'm getting into a battery to log when power is out, save that state, continue logging temps, then transmit after the wifi (assuming loss of power - although I have a UPS for the modem/wifi that will last a few hours) is available again. This obviously adds complexity.
Would love to hear your approach to that project....
We used to put a marble and water in an empty camera film roll canister (transparant). Freeze in the freezer and flip over. If you come back and the Marble is frozen on the bottom of the canister: don't eat anything! #lowtech
I've always just put an ice cube in a ziplock bag. If it isn't square, that means it's melted & refrozen. The marble trick seems like it's probably more sensitive, but I also don't have marbles and film canisters laying around.
This is a tip EVERYONE should know and I still do this! I use a half filled bottle of water, a small rock, and mark safe/not safe so anyone who's not me knows which way it should be.
Pretty sure it was my great-grandpa who taught me that.
I remember that my parents' old freezer had something like this built-in (using colored liquid instead of the marble, but the principle is the same). Seemed like such an easy and obvious feature.
I dabbled in designing a cooler/freezer monitoring setup for a local farmer not too long ago.
The 2 biggest constraints I found were power and communication from within the freezer.
One promising approach was
- a BLE enabled temp. sensor beacon with a coin lion battery
- a nearby externally mounted unit (RaspberryPi) plugged into a wall socket as a relay
- a small central 'black box' SSD/Wifi/router ground-station to collect, store and push the data to the internet
The biggest question was whether the beacon signal could get through the freezer walls. If not then I was going to look into the idea of some super flat cable to enable a wired sensor connection without compromising the freezer seal (like a meat thermometer in an oven)
Depending on the freezer, putting a hole somewhere and sealing it up really well is just as good as the walls and insulation that were in there before. I did this to a chest freezer that wasn't terribly expensive and I haven't ever noticed that the outside of the hole is any cooler than surrounding areas, nor did I notice that it had to work any harder to keep it cool.
That being said, I didn't exactly exhaustively detail the before and after performance of the freezer either. I used the smallest drill bit possible that would let me get a probe in there and everything else is external of the freezer down by the compressor motor where there is a lot of extra room- and input power to tap in to.
I was evaluating designs from a potential commercial opportunity perspective so keeping things non-invasive was considered a constraint for ease of use and to avoid warranty/liability risk.
I keep approximately 5 minutes of temp readings (used a TMP36 sensor) at 1 minute intervals using an RTC (Real Time Clock) to more accurately time it. I made a small 'power tester' circuit that basically acts like a normally open button and closes when there's no power coming into the freezer. I may or may not have spliced into the wiring of the freezer for this and lied to my wife about it when she asked if I was going to burn the house down.
Side note: Is it just me or does it seem like you get asked this question frequently?
I have the 'button' circuit wired into an Arduino I/O pin just like you would an old normally open button. When the 'button' circuit closes the Arduino triggers a VERY audible siren I cut out of an old water leak detector. If I have 5 minutes above my temp threshold the siren goes off. I upgraded it over time to use a small LiOn battery and got a circuit to switch to it when incoming power is cut for whatever reason, and basically when in battery mode it's also siren time. I got through 10 minutes of siren battery testing before deciding it was good enough and I was tired of hearing it.
We've talked about doing something similar for the propane fridge and an activity sensor in my friend's cabin in remote central Oregon using a 3G breakout for communications but we've run into the issue that the 3G there is really spotty and we just couldn't reliably get a data connection out there. Also debugging electronics with no internet, running off generator and batteries, and in the dark because we were fishing all day doesn't lend itself to really getting the job done.
Since you have WiFi you may have what you need to do it more simply, although the 3G might be a safety net for the safety net once you get your solution done. My biggest problem at the cabin, besides the lack of cell coverage, is the lack of constant power. There's barely any solar power, and it's used to keep the batteries conditioned, and it's often weeks or months between people getting out there. I would think at your place you could pretty much just use my approach and add WiFi, or just use an ESP8266 instead of an Arduino and do everything else the same. The battery switchover stuff was the biggest pain in my butt, mostly just because I've not got a ton of LiOn experience.
EDIT: On the RTC and temp logging, it's really not necessary. I just did it to do it. You can use the oscillation ticks and a counter to get close enough. Hell, you can just bump the threshold up a little and raise the alarm as soon as you see it without logging anything and get a drastically simpler circuit.
Adalight is also cool and one of the very few projects where instead of being able to just do it on a basic stamp instead, it actually benefits from moving up to a Teensy for faster refresh rates. Adalight is TL;DR picture-sensitive ambient monitor / TV / projector backlighting. The colors around the monitor change based on what's on the screen. It's a neat effect that was standard on some Phillips TVs: https://learn.adafruit.com/adalight-diy-ambient-tv-lighting/...
It's not "the next big thing", as much as it's a lot of little things. There's always been a gap between software and the physical world; the Arduino lets people easily bridge that gap. Though the typical projects are trivial as you suggest, the effect of being able to easily construct such projects is significant.
That said, I think there are other MCs aside from the Arduino that are more interesting today, because they have WiFi or Ethernet on-board, and you get into a much more interesting class of project when you add networking as a core feature rather than something that needs to be bolted on.
The ESP8266 [0] is, in my mind, one of the most interesting things in the microcontroller world right now. It's cheap, small, easy to use, and connects to WiFi.
If you go into real computers, the C.H.I.P.[1] is quite an interesting device. It's $9, which is price competitive with the ATMega644 chip (in single quantities), has WiFi, Bluetooth, and runs Linux. It's also got lots of GPIO, like the Raspberry Pi, so it can be used in places where a microcontroller might normally be used.
I'm sure there are others, but these are the two that are on my radar at the moment.
I'm linking to the SparkFun ESP8266 breakout [0] because I found this to be so much easier to get my bearings with than the raw chips/breakouts I'd gotten from Alibaba. Granted, I got my hands on them very early and they were still reverse engineering the documentation on them, but using this dev board has saved me many hours. It's also quite inexpensive for one or two if you're just trying to play around.
Oh yeah, there's no doubt that you'd need something like this to get started. I figured I'd like to the Wikipedia article because there are so many things being built with this chip that it's hard to point to any one of them and say "this is it". It's really about the ESP8266 itself, what companies like SparkFun are building around it, and, finally, what hobbyists are building with those.
I've entirely moved from Arduino to the Teensy, which is pretty much entirely software-compatible (excepting pin number changes), but uses an ARM Cortex M4 with 32-bit architecture at up to 96MHz on a much smaller board. The guy who makes them is also incredibly helpful on the support forums and constantly adding new features.
The peripherals are so much better it's comical. Basically all pins capable of 16-bit PWM, having dual 12-bit ADCs, a real DAC, and removing most of the extra junk? Amazing.
Now I just use it as a plug in brain for my custom boards, since it takes all the hardest to solder parts for an entire SoC and puts them in a form factor that's quite small and appropriate power for anything I've wanted to do so far.
Teensy is great but mostly overkill for the kinds of basic IOT stuff that I find myself doing.
I do want to do a project using a teensy with the audio board for beat detection eventually, but the teensy audio library still doesn't suppert beat detection. Paul posted that he would be adding it a couple of years ago I think, but it currently isn't implemented.
ESP8266 for ~$3 at 160MHz with wifi built-in is such a bargain I can't go past it most of the time, and when I don't need wifi and want to use the Arduino libraries, Arduino pro micros are about the same price and have USB HID support.
It's so great that there are so many great and well-documented options out there nowadays!
Aha, yeah, that does look quite nice. I haven't done a wifi project in a while, but the last one I did [1] I nixed using a wifi chip itself since they didn't have enough PWM outputs. It seems like they've gotten better enough that I could just use the one chip to make my wifi lighting projects vastly simpler.
I messed around with several versions of these and will say they are all sorts of a pain in the ass to get going. The Python stack I was using on them kept crashing when I would put it into power saver mode.
Still, it's pretty cool you can buy one for like $2 and basically have a computer that runs Python with some minimal microcontroller functionality.
I'm a big fan of the MSP430, which is the only ultra-low-end microcontroller that's not Harvard architecture. That is, code and RAM live in the same address space, which lets you load code at runtime without needing to fiddle about with reflashing.
This allows you to port real operating systems to it:
For a similar market, the ARM Mbed platform is very compelling. You can program a wide range of ARM-based development boards from a browser-based IDE. They provide an Apache-licensed RTOS with comprehensive hardware support. There is an enormous range of compatible hardware, from 16MHz M0 up to 400MHz A9.
I've been working on a project that's using Adafruit's Feather line extensively and I've been very happy. The ones with the M0 chip are great to work with. Edit: They're using ATSAMD21G18
Arduino (the platform) has been to amateur electronics what BASIC was to home computing in the '80s. It's the platform with a user community and a huge number of example projects, that lets people learn easily with early results and a minimum of time and cost investment.
The toy projects are a journey, not a destination. Compare them to all those people typing in listings for Breakout games in the '80s computer magazines.
It's also a means of doing "IoT" home automation and monitoring without getting locked into someone else's surveillance ecosystem. I have a Raspberry Pi reading my electricity meter, for example.
The meter has a (standardised) red LED that flashes once for each watt-hour. The Pi has a phototransistor that turns flashes into a GPIO transition interrupt. I've put it on my solar panel generation meter, and plan to build another (slightly improved) unit to monitor the mains input meter.
I've had similar questions from some of my hard core embedded colleagues, here's my answer:
Just think of something you might do with a traditional embedded systems approach, then google "arduino for ________" or "arduino shield for _______" and see what you'll find. Its can be pretty amazing. Its much broader that the surface stuff of LEDs and basic sensor / actuator tasks.
As someone who's done production embedded systems work, what I build from arduino (and the fairly large arduino ecosystem) is not for high volume product, its for proof of concept, small run pilots and show and tell (and fun).
Off topic a little:
One thing I've done to convince colleagues of the a plausible story of arduino POC to Production is collect a bunch of arduino gear (simply through web search and amazon) and build a paper-product with all the cobbled together parts. Then pose the question "if these cobbled together things cost _______ in single quantities, is there a plausible story for them to be 10x (or 100x) less on a cost optimized board with a proper contract mfg"?
If the answer is yes, then you cobble together your arduino thing and go show and tell with customers.
From what I have seen a lot of ppl who wouldn't want to sit down and grok a datasheet [1]. The friction of getting started with one of these is soooo small, assuming the serial driver installs on Windows. Out of the box support on other systems.
I used one to control a pair of candy machines for a Watson demo - there was a laptop with a microphone that did sentiment analysis on your speech, and then it told the Arduinos to give you either sweet or sour candy.
The next version is probably going to have a Raspberry Pi replace both the laptop and the Arduinos, but they were a good starting point to get the project up and running.
RaspPis are startlingly cheap these days, and surprisingly capable.
For less than a tank of gas, I just built a retro ROM console, and the new version 3 boards are beefy enough to run video through Kodi without missing a beat or play any old console games I've tried up through the PS1/N64 era.
They're not really equivalent. The Pi is a full-blown computer, for better and for worse. The Pi Zero draws 70mA at idle, while the Pi 3B draws more like 300mA even with most of the peripherals turned off. With a bit of tweaking, an Arduino will run for months on a coin cell.
That is not true to all countries and regions. For example, in Brazil a raspberry pi 3 will cost a bit more than USD 100 which is a lot of money for many people and prevent it from being acquired by low and middle income classes which are arguably among the ones that would most benefit from access to this ecosystem.
Arduino/Genuinos on the other hand are easily available on our version of eBay and can be purchased as low as USD 30 which is an order of magnitude difference between the prices, and yes, one is a real computer and the other a microcontroller but sometimes you get what you can afford.
this schism between arduino people is completely ignored here and we'll buy the cheapest chinese knockoff if it is available.
Yikes! Is that because the main distributors (RS/Allied Electronics/Newark) charge more to ship it to Brazil, or do they just not ship there at all and make you purchase through more expensive re-sellers? (Or is it something else like tariffs?)
Case one: $35 (product cost) + $40 (international UPS shipping) + $100 (import and administration taxes)
Case two: "I bought one R-Pi from Farnell Brazil (http://www.farnell.com.br/) and I paid R$185,50 ( ~ U$90) with taxes and postage."
I remember trying to help some Brazilian users find the cheapest way to get a hold of one a couple of years ago, and I heard similar stories: high import taxes even when there were official distributors operating in the country, and high shipping charges when there weren't and they needed to order the devices internationally. I don't think that it's resellers trying to fleece their customers. I think it's just a fact of the Brazilian government's policies.
Of the main Pi distributors, only Newark had local operations, and they closed shop last year.
Now the cost is higher due to low volume imports, so you end up paying more for shipping the boards. Also, taxes are higher if an individual is importing stuff, as customs use a simplified 60% tax over the retail price vs using the specific import tax (around 45% for the Pi) over the reseller price.
Taxes and greediness play a strong role in Brazil. For example, I just imported 10 Tessels to start a class on IoT. So they charged 98% taxes, (60% import taxes + 20% icms + 18% of administrative fee from UPS).
Not an endorsement of this, but nothing that I have order directly from China has ever had an honest customs declaration on it. It's always a "gift" of "soap" or "lotion". I'm guessing if you charge 98% in taxes it's worth it to have customs agents who actually inspect things though.
I built lab grade NIR spectrometer around Arduino. At least first two prototypes of it, then graduated to Teensy (32bits compatible board) and will use custom board for production.
The thing is: most of electronic projects needs some way to control things and it is great to take shortcuts. In my case all analog section was custom made to the point of 16bit ADC and used Arduino/Teensy as a way to react to clicks on button, transfer data to computer and stuff like that.
The compelling thing about them is that they're a gateway into hardware design. It's like asking "what serious apps have people made with Scratch?", it's rather missing the point. You start making a rotary mobile phone with Arduino and move on to make your own autonomous drone.
Arduino is a platform (the boards, the IDE, the language, debugging tools, etc) that allows any sort of user new to programming and/or hardware to get started very quickly. It is a great gateway into raw microcontrollers, as it is based on the Atmel/Microchip ATMega line of chips, which are also fairly easy to program. Most people can easily transition into that and build "production-ready" things.
Also, many 3D printers, laser cutting tools, and CNC routing tools use Arduino as their microcontroller of choice. It represents a huge community and is one of the most basic of tools in any new or seasoned maker's toolbelt.
I wired my own custom board for it rather than using a commercial Arduino type board, but I built a guitar effects pedal that runs Arduino code to process a guitar signal using the built in ADC and a pretty basic passive DAC. It works great for effects like bitcrushing, distortion, compression, and I'm trying to get high and low pass filters working on it but haven't made any progress lately.
This is really cool, I'm doing something similar! I'm making synthesizers with the Uno and Due; I haven't tried doing anything with audio input yet, but want to play with that as well. How is the sound quality of it? I did the 8-bit resistor ladder DAC from the instructable linked, but quickly broke the layout when I tried to bring it to a friend's house and haven't tried putting the breadboard back together. I really should go back to that, I liked how crappy it sounded. I also tried making a 1-bit DAC out of it, which worked better than I expected (though it still wasn't good, I just didn't expect anything haha).
Now I'm using the Due and Teensy (Cortex M3 / M4) because of the hardware DAC and 32 bit processor. I haven't gotten filters working yet either, they are my next goal. IIR first, and then FIR if the Due can handle it. But I have gotten a cool resonant filter simulation by using a hard-sync as described here [0]. You get some neat variations by changing the waveshapes used and the windowing function. I might not even have subtractive filters in my final design, just because of how cool it sounds to approach it differently, but I still want to understand how to implement them.
Nothing on Github to share at the moment, unfortunately. I'm too embarassed by the state of the code, lol. But I will be checking out your project this weekend, so thanks for sharing!
Thanks! The sound quality of it depends mostly on the sample rate it is able to achieve, which is somewhere around 9.6khz if all it's doing is reading a sample and then outputting it to the DAC. At 6bit 9.6khz a guitar sounds surprisingly clean through it, aliasing is only really noticable at the highest frets of the high strings. At lower bit depths and sample rates it gets really interesting.
Getting the best sound quality out of it also depends on finding the sweet spot on the input gain to get the best dynamic range from the DAC without clipping, although I designed the preamp to sound pleasant when overdriven, mostly using JFETs.
It's not the next big thing, so much as the next big educational thing, in my opinion. It is a simple enough system to put electronic power in the hands of people who wouldn't otherwise have it. They learn some basics of coding, some basics of electricity, and get something tangible out of it.
I’ve built a few ESP8266[0] (so not Ardunio but same space) based sensors and have them distributed throughout a building so we can see how changes in one part affect the temperature and humidity in other parts.
They’re very cheap to build (we only use the project boards to prototype, final builds are on PCBs) and they can be run on small batteries when they’re just doing simple data collection.
I used KiCad[0] PCB design software and followed a lot of Youtube tutorials. This[1] one was the best tutorial I found, it seems pretty trivial at first glance but touches on all the parts of the process.
I use OSH Park for the actual PCB production[2], they're pretty cheap and good quality boards but be prepared for a bit of a wait. There's also a good price comparison site[3] for alternate suppliers.
I started on KiCad but found the interface very unintuitive.
I got stuck at the point of making my own footprints for custom devices, which I needed to do because the library doesn't seem to include all that much (or maybe I couldn't find devices I was using because I couldn't work out the UI for searching well enough.
Well, the easy answer is the arduino is "easier" to get into than most other embedded development boards.
These days, though, that's becoming less and less true. You can pick up an ESP8266-based board complete with USB-serial built in and motherfucking Wi-Fi for 15 bucks from Adafruit, and it's identical to an Arduino code-wise (i.e. you can use nearly all the nice fuzzy constructs that the Arduino version of Processing provides), plus motherfucking Wi-Fi. I've been prototyping something with one the last couple of weeks, and it's great.
Compared to, for example, my senior design project in college, which predominantly involved writing shitty C code for ATMega-series microcontrollers, it's a breeze. I get C++, a huge standard library, consistent types, a Java-esque wrapper around std::string, and all the register ops to toggle or read pins and enable/disable features are completely abstracted away. It's like comparing writing an app server in Node.JS to writing it using Java EE. Obviously, equally possible in both languages, but one just feels so much nicer than the other.
Oh yeah, the D1 Mini also has a very low quiescent current (25uA) RT9013 regulator, so it should be much better for use on battery power than the nodemcu with the LM1117 which uses 5mA! That's 200x less current wasted!
Also the dropout voltage is only 250mV compared to about 1V for the LM1117. You couldn't use a LiPoly battery to power it with an LM1117 because of the high dropout voltage.
Most of the interesting projects will never be posted online. You are seeing mostly beginners excited to get into hardware and posting what they do.
Arduino is "the next big thing" mainly because of the community that's sprung up around it: tons of libraries (of varying quality...), lots of cheap hardware and online resources make it easy to get started and continue past the beginner stage. Previous attempts (BASIC Stamp, etc) never made it to remotely this level of popularity.
e.g., I'm an experienced electrical/firmware engineer and I do Arduino prototype consulting on the side.
One of my recent projects was a downhole scanner for measuring well diameter. I also have a set of ongoing automation projects that are specific to a particular customer in the entertainment industry. Arduino turns out to be perfect for this kind of work.
This is the CRUD work of the Arduino world: basic automation that's not exciting or even difficult, but fills a real need.
Last week I used a Leostick[1] and an SR04 ultrasonic distance sensor[2] to create a presence detector for my computer. Now, if music is playing whenever I get up and walk away from my computer, it pauses the media player, and when I come back it starts playing again automatically.
I worked on some robotics research not too long ago. We used an arduino for interacting with motors, sensors etc. We ran some semi-interesting filtering on the data before passing it back to a more powerful processor that ran a python program that did the decision making and provided interactivity from the user.
The main reason we used the arduino was compared to all the alternatives (other similar hacker boards, and FPGAS say) it gave us a flexible toolkit, easy deployments to a board, and a very simple way of interacting directly with sensors throuhg a fairly convenient API.
We're just now getting them up and running, so I don't exactly know yet. I'm pretty sure that we can do it though. There are some interesting data analysis techniques (PCA, neural nets) that can be used to get meaningful results out of less than perfect data.
I don't think it's "the next big thing." Not everything needs to be. I'm making a synthesizer with the Arduino Due (and Teensy which is compatible). It's a fun playground for things I would never normally get to do in C++ and asm. I've never gotten to code on a platform with true realtime and no OS restrictions before. Also the barrier to entry is so low, it makes it so much fun. WIthin less than 15 minutes of receiving my Amazon package I had my my first blinking LED working.
Arduino is not the next big thing; it's arguably a few years past its peak. But the thing is, it's really really fast to prototype stuff.
Last week I needed to convert an analog voltage to a pulse width -- I had a potentiometer and I wanted to control an ESC. It took me less than 10 minutes to breadboard and code that on an Arduino clone (Teensy).
The week before, I needed a frequency counter. Again, it took less than 10 minutes.
I've done more elaborate projects, but the point is that the overhead to get started is tiny.
I looked for some time trying to find a good egg timer, but there are really just two options, and neither are that great. You can have a nice, intuitive analog interface (the classic dial timer), but you lack precision and many are very poorly constructed and unreliable (seriously, look at Amazon reviews for this type). There's also no way to fine-tune the alarm; some are horrendous, while other are inaudible. The digital ones solve some of this issue, but have terrible UI requiring many button clicks.
I used a digispark ($1 for ebay clones), two rotary switches set up as voltage dividers, an arcade button, and an addressable LED strip to make a better timer. You input minutes with one knob, and seconds with the other. This has wonderful tactile feedback, is completely precise, and entirely intuitive. The LED strip counts up (and shifts red to green) as the timer passes, does a rainbow swipe/strobe for a few seconds to alert, and then 'cools down' so you can get a sense of how badly you overshot. There's also a buzzer. It's good for the kitchen, but I actually find it best as a rest-period timer for weightlifting. The visual display is great for loud gyms.
It's 100% a better timer than anything else on the market, and I absolutely love it.
1. A mechanical/environmental monitor/display for a grinding/polishing machine for telescope mirrors. It measures the rotation speed of two key machine parts; also measures ambiental temperature and humidity. It displays all these parameters on a front panel.
Regular fish tank, LED lights with timer. The lights would turn on or off suddenly, scaring the fish. :) I used pulse-width modulation and a MOS-FET to gradually fade the lights in and out over 1 hour.
3. PC clock
I've setup time limits on PC usage for my kids. But before Windows 10 the login screen on Windows 7 didn't have the time displayed, so they didn't know when their accounts would actually let them log in. I've found a big LED display that fits almost exactly in a CD-ROM bay, and built a clock with it. It's synced to the PC every time the OS boots up.
The big thing about Arduino is that it makes embedded electronics more accessible to people.
Arduino is a C Framework for interacting with hardware that is portable across microprocessors. Arduino also offers an IDE that's modular and allows vendors to package up their Workflows for programming their Microprocessors and easily integrate it.
What this means is that a naive user can pickup any Arduino compatible board, configure the IDE to be able to program that board via a single URI from the vendor, write some code in C using the Arduino framework, and push it to the board with the click of a button.
So if you have had an idea for a piece of hardware, Arduino makes it ridiculously easy to prototype it. And since it's based on standard microprocessors, if the overhead of Arduino is too much then you can replace it with the microprocessor OEM's workflow and port your code over. It's a very accessible RAD environment.
So as others have said, if you have the ability to do it you're limited by what you can imagine. If you treat the output in the same way you would any industrial automation you're virtually only limited by what you can imagine you want to do with the device. I've set up a Pi as a home media server (though that's extremely common). I've got a friend using Pi for underground sprinkler control.
Depending on what your trade is, you often need prototypes of things that do stuff. Before Arduino is usually cost thousands of dollars to make these (only counting the hardware cost). Arduino drove that price down to a fraction. It's easy to develop pretty much everything in very little time.
Here are couple rather uncommon examples:
- a greeting card with an integrated mobile baseband that calls a certain number once opened ("hey grandma!")
- a tea-timer that automatically pulls out the teabag after a set amount of time
- a UV-dosimeter that measures the actual amount of UV exposure and connects to your smartphone to track the data and warn of potential damage (with respect to your skin-type).
- the DIY-cellphone by David Mellis modified to display reading from arbitrary sensors
I'm not sure it is the "next big thing" as you say.
The Arduino IDE and all the ecosystem of libraries and community could be the actual "big thing", because the Arduino is just a convenient development board; but you can get those from the chip manufacturers themselves (OK, may be now those are nicer because the Arduino boards).
I think the IDE is making the hobbyists able to work with microcontrollers in a very easy way; compared with all the tooling and docs you have to read to build your own dev board and program the MCU yourself.
When they first came out I used one to build a UAV on top of a 6' electric glider. The Arduino simply controlled the PWM's for the rudder, elevator and motor and then fed sensor data over a zigbee. My laptop, connected through a zigbee, actually flew the plane. I had hacked together my own INS from accelerometers and magnetometers, plus a joystick to override.
Was fun until it crashed into a pond after a stupid code update that had the IR sensors invert the sky and ground.
Anecdotally, I have been using the pi and arduino to learn. They have the basics for small projects and easy interfaces for getting started.
The communities are large and it makes it very easy to find beginner and intermediate projects. I like them both and recommend them. They may not Be the perfect base for a robust IoT project but one can do a lot of interesting things and learn a lot doing beginner and intermediate level projects
As a small robotics startup we use Arduinos to control our grippers just because it was easy for the founders to get started and because adding new features is fairly fast. We've actually moved to a board that pretends to be a mega2560 but is entirely custom but we're still sticking with the Arduino libraries.
Some day we might move to a standard uC approach but that'll be when we have many more employees.
I am working on a write up on a project for CAN bus messages to CCD bus messages for old Chrysler vehicles. A mixture of reverse engineering and vehicle bus interfacing.
Note: I am using a Teensy instead of an Arduino, but using the Teensyduino Arduino IDE plugin.
These boards are just low cost "evaluation" bards for whatever chip is on them. If we include larger chips ( say an ARM Cortex A8 on the Beaglebone Black ) , they're great for test fixtures and prototyping systems.
It's neat to have a nice, easily controllable small board, but yeah - I went to a Maker Fair last year and one stand was showing off an "Arduino-powered doorbell". It's possible to take it too far.
I used an arduino to receive weights of coins from a scale and then trigger flippers to remove coins into various buckets depending on their classifications.
We use them all over my office because each team has their own metrics portal and we can just plug one into a big screen TV and have it run that portal. Cheaper than a full computer and just as effective.
I don't think it's really a "big thing", it's mostly for hobbyist projects (since you need to know quite a bit about software development, Linux, etc.), and for that purpose it's quite good.
It's cheap and you can easily connect leds, motors, etc. and control them via Python scripts. It's also frequently used to create home entertainment units for music, films or video games (emulators in particular), since there's an HDMI port that you can connect to your TV.
If this [untold Arduino history](http://arduinohistory.github.io/) link is true, it seems like Massimo Banzi is not only difficult to do business with but that he had a very particular way to give credit.
I was going to post the same link. I have no qualms buying clones what so ever since reading this story. I've had several now and there have been no issues with quality or stability. Not even been hit with any fake FDTI chips either.
This is ultimately the challenge of open source. It is always sad to hear when people go unrecognized like this, it is of course important for the Arduino folks to position history in a way that makes them look like the main guys.
A lesson I learned early on, and this is a good example, is that all events have a number of facts associated with them. Adding and subtracting facts from the event can change its message in any way the narrator wants. You could say "Alice shot Bob dead, in cold blood." and it sounds like Alice is a murderer, if you add "After using a toothpick to open the handcuffs holding her to the post where Bob had abused her. Alice found a gun in the dresser and picked it up.", followed by "Alice shot Bob dead, in cold blood." More facts, different interpretation of right and wrong.
If you read the Wired article on Arduino you think one thing, if you add facts from this document you think another. General advice, keep a journal and corroborating evidence.
I never get when people do something like that. In the long run it can only hurt them. And it costs nothing to praise another person's work that resulted in your success.
Well, I guess the point would be that the whole Arduino/Genuino vs. Arduino spat is about who "owns" Arduino. Which is kinda ironic given it's an open source product built from someone else's original work.
I wholeheartedly disagree. A TI launchpad dev board is sub-10 dollars in bulk, and an ESP8266 (which is almost incomprehensibly more powerful and has built-in WiFi) dev board is 15-ish. And both of those have nice friendly Arduino wrappers. There's really very little reason to buy vanilla Arduino anymore other than the somewhat larger community.
A typical hobbyist doesn't need to buy in bulk. When I started with Arduino when it first came out, I never heard of TI launchpad series. So Arduino was first to the show. As a side: in my simple projects, not one has ever needed Wi-Fi; especially when considering power consumption.
TI launchpads can be bought direct from TI. I don't have a link handy.
As for the ESP8266, Adafruit sells a board called the HUZZAH ESP8266.
For both, the best resource in my opinion for environment set up/build system is platformio (platformio.org). It exposes an Arduino-style API for both boards, so basically the arduino docs will get you 95% of the way there, and the community sourced libraries in platformio will get you the rest of the way.
I really don't see the point of the arduinos when you can buy an esp8266 that is arduino compatible (you can use the arduino IDE and sketches).
You can get the bare esp8266 chips for $1-$2 from aliexpress, or buy a tricked out one from adafruit (the huzzah) for $9. They have a full wifi stack, support micropython, etc.
I think we're in a post-arduino age for microcontrollers. There are simply better options out there, of which the esp8266 is just one of the current better choices.
Depends on what you need. The ESP8266 has a lot of limitations depending on what you want to do. It doesn't have any internal memory, so is dependent on an external flash device. It only has a single adc channel, so if you're doing an sort of measurement operations, you'd need a breakout board of some sort. It requires an external clock crystal, again increasing the BOM. It's also not 5v tolerant, which is useful in some applications. There are definitely a lot more compelling micros out right now, but AVRs still have a lot of tricks up their sleeve that other micros can't meet yet.
I don't understand the comments that ESP8266's can replace Arduinos. Don't you still need to wire the ESP8266 to a microcontroller?
I'm a hardware novice, and I love Arduinos. I'd appreciate any info to help me learn, seems like I'm missing something. To me the ESP8266 is a superior alternative to Arduino wifi shields, not the Arduino itself.
When the ESP8266s were first introduced, they came with a closed firmware that simply responded to AT commands over a serial port. To do anything with the chip, you had to have another micro controlling it.
However, sometime after the hobbyist/maker market started to take notice of the devices, an SDK was released, allowing for code to be run natively on the devices without a separate micro.
Nowadays, there are multiple options for running code on the devices. You can do it natively via C/C++ with the SDK, or you can try out one of the other firmwares that various parties have produced, like NodeMCU (Lua) or MicroPython.
The chips are actually pretty damn powerful compared to the Atmega series micros used with the classic Arduinos. And cheap too. You can get a NodeMCU module that already has a voltage reg. and USB->UART controller on it for around $3. Very hard to beat.
Nope, the esp8266 is a microcontroller featuring a 32-bit Tensilica Xtensa LX106 running at 80 MHz. It can be overclocked to 160 Mhz, but that sucks for power.
What used to be the biggest advantage for Arduino - and still is, though diminishing quickly - is support for a variety of peripherals e.g. TFT displays, servos, analog and digital sensors.
With the proliferation of so many arduino hardware variations and knockoffs it's getting harder and harder to match the correct library with specific hardware.
While it is saved as a wifi module with a basic serial interface to be controlled from, The ESP8266 actually has a powerful microcontroller that can now easily be programmed using the Arduino IDE (and other stacks).
There are different firmwares that allow programming in Lua, Basic, Javascript... or just compile directly from C/C++
Also, don't you need weird hardware to drop the voltage and connect it to the PC? I've only managed to do anything with NodeMCU so far, which is basically ESP8266 glued to the aforementioned weird hardware.
3,3V is a pretty common if not prevalent voltage level with this type of embedded Hardware, nothing weird. Also these USB to Serial converters are pretty common.
Weird from the POV of an electronics noob, who sees ESP8266 recommended as Arduino-killer. One of the two plugs straight into PC and doesn't require a soldering iron to use...
The NodeMCU is absolutely no different than an arduino. The ESP chips are the equivalent of the AVR chips that are on an arduino. When people say "use an ESP chip" they really mean one of the many ESP based boards.
The reason someone serious about embedded development would want to buy Arduino boards is that they're convenient Atmel-architecture microcontroller boards.
The ESP8266 may be "arduino compatible", but a large number of those Arduino libraries running on it had to be rewritten.
If you want convenience and ISA/peripheral-register compatibility with existing low-level Arduino code, you're locked in to atmegas unless you want to do some fun porting work ;)
The next step up was the Arduino Due, which is a low-end ARM system supported by the Arduino toolchain. But it seems to have been retired. (Now I find this out. I built something on a Due because I needed more memory and CPU speed.)
It may be retired, but Sparkfun still sells it[1].
If you're looking for other faster microcontrollers with Arduino toolchain support, there's a bunch of stuff out there. See the Wikipedia page on non-atmega Arduino-supported boards: [2].
Additionally, if you're willing to learn how to use the ARM CMSIS HAL and write raw C++, there's the STM32 series of microcontrollers. Those also support Mbed, if you want to learn that instead (it's much easier than using CMSIS with C++, it's basically the equivalent of arduino for many 32-bit ARM microcontrollers).
It's a brand name people recognize. And some of them don't realize they have options.
I've heard stories of people who had built several Arduino projects, but when someone mentioned microcontrollers and working with embedded systems, responded with "what's a microcontroller?"
The local Microcenter sells Arduino Uno R3 clones under their Inland store brand for $10, currently and frequently on sale for $6. Ideally, I'd like to support Arduino LLC, but given how absurd this whole feud has been, I don't feel especially guilty when I buy a clone. I have bought a few Arduino brand 101 boards from Microcenter as well so I'm not a total sell out!
TLDR: Reseller is pissed that "Genuino" has to be sold as "Arduino" in the US and that they have separate SKUs. Also Genuino seems to have poor B2B/reseller support.
Is this post really that petty or am I missing something?
The new requirements make providing a good customer experience much more difficult, because it passes the retailers the burden of explaining this bullshit no-one cares about to their customers.
For example, what if I visit the shop while logged out, put a Genuino into my basket, then log in and ask for delivery to a US address - should the website replace the Genuino in my basket with an Arduino? What if Arduinos are out of stock, while Genuinos are in stock?
What if it's a european shopping for an Arduino because that's what all the guides online say, and it looks like the retailer is ripping them off by replacing their order for a genuine Arduino with an off-brand clone? And it looks like they can get a "real" Arduino at half the price from AliExpress?
Providing customer service is what a retailer does. It's fine not to want to sell something, but writing rants in response (especially when you suggest your own products instead) isn't very nice. Retailers also take quite a big piece of the pie from smaller manufacturers.
(As far as I understand) Arduino LLC only has the trademark for Arduino in the US, meaning they can only sell and manufacture them in the US. It seems reasonable that Arduino LLC don't compete with their own US manufacturers by selling (under license) Chinese made Genuinos in the US.
People always say they want open source, support, locally manufactured products etc. but then they seem to take pretty much any excuse possible not to support it.
>>People always say they want open source, support, locally manufactured products etc. but then they seem to take pretty much any excuse possible not to support it.
I think you confuse the 2 vocal groups, it is not normally the same people Wanting Locally Manufactured products and wanting Cheap Knock offs, even if both want Open Source
I Love opensource, but I have little respect nor desire to support Arduino LLC, and will happily buy the cheapest board, I will also support projects and vendors by paying a premium if I feel it is worth it.
>Retailers also take quite a big piece of the pie from smaller manufacturers.
Margins are these boards are very very thin, this is why Retailers like SparkFun, Adafruit and other are making their own Arduino Compatible boards because it is more profitable, It is not true that the retailer is taking a "big piece" of the pie selling Aurdinuo LLC licensed boards. this is also why other Platforms are starting to replace the what I consider Overpriced Aurdinuo. I have been starting to use ESP8266 based boards (like NodeMCU) for example. If I want a $40 board I am using a rPI not an Aurdinuo...
I get the impression that they're probably trying to avoid further diluting the Arduino brand in the US with the expectation that they'll eventually prevail in the trademark dispute.
The two groups are headed in different directions, they are just selling the same product now. I'd be annoyed as a customer if I needed to figure out the difference between two companies products and it wasn't clear. They need some way to make the distinction.
I think you're missing the additional costs and logistics of separating out SKUs by country which are placed upon the reseller, simply for "branding" reasons.
It sounds like they were already buying both product variants and surely it shouldn’t be too hard to restrict certain SKUs to certain territories via a shipping rule in their ecommerce system?
Shame they didn’t breakdown how many sales they get from the US (wouldn’t expect it to be many considering they’re UK based).
Perhaps part of it is down to the fact they said the margins were slim already. It seems like even if it was trivial to facilitate separation or SKUs geographically it's just not worth having to work with a supplier with such restrictions.
It sounded like they didn't want to buy the "Arduino" variant to me because it was a separate SKU; they'd been selling "Arduino" in the past before the whole kerfluffle started.
My recommendation is a bit different, instead of that, try looking for a local manufacturer and support them.
Here in Brazil, people go for the cheap aliexpress clones to the point where some of our own manufacturers are closing shop. In these cases, strengthening the local ecosystem might be interesting and wiser in the long run.
Why? Your manufacturers could focus on something more profitable. What's the point in making another Arduino like board? Supporting non innovative shops is not good in short and long run.
It is like China and the shanzai spirit, first you clone then and make it sustainable then you innovate... (this is my personal opinion, I may be wrong).
I believe that supporting local non innovative shops is better than having no local shops at all.
But this is a result of that country politics. You should rather press government to make it possible to compete with Chinese (by for example lowering taxes, lessening the amount of bureaucracy - I heard this is a nightmare in Brazil). Throwing money at economy that is full of flaws will just delay the agony and not solve anything.
There are many clones. Check out "mateduino"[2] which is made in the south of Brazil and also the one from robocore which is made in the southeast. There are others. The ones from robocore are great[1].
FastTech have some called "Arduino Compatible" they are indistinguishable from the original ones except for the text on them. I suspect they come from the same factory.
I bought like 2 of them so far and they both were exactly as expected. The only thing to note is that you should try to find those with CPUs (are those CPUs?) similar to the ones from the original designs. Some (the super cheap ones, < $1) often are identical by design but come with some crappy CPU
Almost all the ones I've seen have used the atmega328P. You might see some slightly different models with different amounts of ROM and RAM, but most people would probably never notice the difference. If you need more than a few KB of RAM/ROM, you should probably be getting something STM32 based instead.
What does change is the Serial > USB converters that they use. The more expensive ones (>£3) tend to use FTDI chips, whereas the cheap ones use a CH340G or a similar part.
On some platforms you may have to install drivers to make these cheaper converters work, but they were just plug-and-play for me on Linux.
> If you need more than a few KB of RAM/ROM, you should probably be getting something STM32 based instead.
Which might not be a bad idea anyways. The Arduino framework's been ported to STM32 (stm32duino.com), and you can get a board and programmer for under $10.
Any of them, I've been using everything from teensy,nano, and rebranded Uno. So far they have all worked like a charm and at a steep price reduction. Example, 5 nanos for $15 from Amazon [1].
Yeah, I've not really had a problem with off-label Arduino chips/boards whenever I've ordered them either, except for one being DOA and another having an I/O pin that just wasn't connected somewhere along the way inside the microcontroller.
So basically the Arduino the Software Company (IDE), and Arduino the hardware supplier broke ties?
I use the ESP8266 which I do believe is an Arduino (Hardware) Killer. However, the Arduino IDE has amazing support for the ESP8266 among other hardware components.
This isn't a bad thing, at least for consumers, since I feel like Arduino (the hardware) is a few cycles behind (eg. Particle, CHIP, bbc micro, pi zero, nodemcu/esp8266... etc)
But I can buy an ADC that communicates over SPI for like 3 bucks. Then I have two chips, but then also I have a total of like 10 dollars worth of electronics, which is far more capable than a 25 dollar arduino.
Yeah, I pretty much threw all my Arduino gear in the trash after I bought an ESP8266. I kept some Micros just for the analog pins, but I haven't had any use for them since.
Agree 100%. Paul provides incredible support and he's a real engineer. Teensy family are amazing little machines, and I've used them for many small side projects. The latest one was a custom fight stick.
Paul even provides support you if you want to make your own version of teensy, once you're out of the prototype phase.
It appears that greed and control issues were present right from the beginning of the Arduino project, given the experiences of Wiring's developer, Hernando Barragan, upon whose work it seems most of Arduino was branded, without recognition or recompense. I therefore would have very little sympathy for Arduino LLC.
This article acknowledges over and over that Arduino/Genuino is following common business practices. They are complaining that arduino has become too big, and it's not nice for them, because pomoroni is also big.
Kind of ironic, really.
Yes, they want to cultivate a new brand while defending the old brand. It's awkward and I'm sure they aren't happy about it either.
Oh, yeah, it took only a few moments to go up to the address bar and pop off "blog.", but it throws up a barrier to conversion. Doesn't matter much to me, honestly, but I had to be more than casually interested in looking at their offerings to make the effort.
I have a Teensy board, which is much cheaper and works pretty much exactly the same as Arduino and Genuino. The Teensy LC is great for small projects and is about 15 bucks IIRC. Never understood why you would stick to a more expensive brand just because of the name.
Indeed, and I've heard of similar stories happening around Polish 3D printer scene few years back. Basically, it seems that when an open source project becomes popular, somebody wants to make money off it, and eventually shit like this happens.
Anyone know of any small, low-cost, battery powered programmable boards for Bluetooth LE? I've got the Pi 3 and CHIP is ordered, but I'd love something smaller if possible. Thanks!
Adafruit Feather 32u4 Bluefruit LE has an Atmega32u4 (same as Arduino), LiPoly charger and BTLE built in for $30.
I'm cheap, so soon I'm going to have a go at reflashing a $3 HC06 module with the RN-42 firmware [1], and using that with a $3 Arduino Pro Micro and a $1 TP4056 Charger board for the same capabilities but messier. I always learn more by messing around though.
It's really not that unusual for resellers to be given geographical territories within which they can/can't sell. Try looking up your favorite electronics on Amazon.com and see if they will ship to Europe - most times the page will say 'for US delivery only'.
It is a shame that the Arduino guys split up but I would argue the situation would be even more of a mess if both new companies were able to sell into the same territories with the same name + similar product.
You don't buy what? They didn't claim that it's impossible to deal with it, just that they won't.
And looking at Sparkfun's site, I see how stupid this is. For one model, both are available: 2 SKUs for technically the same board, one is out of stock. And a notice that buying either might be delayed for a few days due to address verification. Very consumer-friendly.
All others they sell as Ardunios only, and they either make them themselves (and license the US trademark) or buy from the european company, so they mostly sidestep the issue.
Up until recently, at least, ESP8266 boards were far more finicky. Even today, for a simple task that didn't need wifi, I'd probably trust an Arduino a little more than an ESP'.
It may also be partially a software thing - there are a lot of libraries that were written for Arduinos and then adapted to mostly work on ESP's. This too has been improving recently.
And, as others have mentioned, Arduinos have better built-in analog support.
More analog pins, is the only compelling reason I can come up with. In general I think the ESP is a better choice for most "turn the LEDs on and off" type of project though.
For applications that need Wifi, the ESP8266 is indeed a good and very competitively priced product.
However, all the wifi magic on the chip is closed source, and the datasheets / documentation are much less complete than chips like the ATmega328 (which the Arduino Uno is based on). The ATmega328 has a 660-page datasheet, whereas the ESP8266 has a 30-page datasheet.
I can read datasheets perfectly well, and Arduino seems pretty good by the low standards of microprocessor programming tooling.
I mean, you can program in C rather than assembly language! And you don't have to pay £300 a seat for the compiler! It works on windows, linux and mac instead of being windows-only! If you open source your project, other people can contribute and they don't even need a £150 tool to download the program to the chip! Other people can compile your code and have it work without needing to know the details of 'linker scripts' and 'codesourcery toolchains' and all that stuff!
Maybe, but this (often buggy) closed source stuff and not having a proper data sheet causes a lot of trouble for projects creating these nice firmwares (micropython, nodemcu, …) which people tend to use.
Bah! Arduino is way overpriced for the specs anyway. Raspberry pi and others are where it's at now. But I do think there is a massive power usage difference.
I mean, sure, you and I know the difference between an 8-bit micro with some friendly libraries and a 64-bit ARM SOC running a full OS. But for a lot of stuff they accomplish the exact same thing. So why not compare them by what most people do with them rather than their raw technical specs.
AVRs have much better sleep modes (single digit microamps) and predictable performance (to the exact clock cycle) by default. If you need long battery life or hard real-time performance then they're probably a better choice than the Raspberry Pi.
All reasonable responses in this thread to GP, but parent hits the nail on the head.
For non-hobbyist problems where you really need a microcontroller, it is selected by spec (in terms of performance, operational predictability, physical/material properties, power consumption), and then purchased in volume (~ millions). Employing a general-purpose CPU for said problem, while potentially alright when prototyping, would not be satisfactory.
True, however that's AVRs and not Adruino. You very quickly get out of arduino-land when you have any constraints like battery life or performance. By far the vast majority of people contemplating either Arduino or RPi will never know enough to be able to tune things like that.
Heck the last time I looked into doing sleeping on an AVR-based Arduino it basically said go look at the 328's manual to set the registers directly. (Although that was a long time ago.)
One abstraction layer above in fact. It's like saying a farm tractor and a pickup truck are the same thing because they both can move you from point A to B.
They're not the same thing, but unless you have some specific use case that demands a farm tractor and you concretely know why, you should go for the pickup truck.
No, they are not. One has more in common with a FPGA, the other is literally a Linux computer. Different complexity, different skillset, different use case.
..Unless you're talking about something like the Arduino Yun, which actually runs a Linux OS. But as an example, you're probably not going to build a wifi security camera on the original device.
Yeah, but you're still talking about the raw technical implementation of how they work.
All I'm trying to say is that for the vast majority of users a raspberry pi can do anything that an arduino can. Sure you might not be able to so some super tight real-time stuff that you could do with hand-rolled asm that you could do on an arduino, but at that point you're more at the level of 'using an AVR' than 'using Arduino'.
When the problem is "I want to control some RGB LEDs" or something, either can be used. That means that they are of the same category.
A rackmount server can control some RGB LEDs, if wired correctly. That doesn't mean a rackmount server is in the same category as a Pi.
Furthermore, most people using Pis and Ardunios are hobbyists who are necessarily familiar with the raw technical implementation and are choosing the product based on that technical data.
Yeah, but if a rack-mount server was cheaper, easier to use, about the same size, and used a reasonable amount of power compared to an Arduino, I'd probably use a rack-mount server for all my DIY projects.
Digging into it, it looks like the Arduino and friends trump the Pi versus sleep/idle modes in terms of power draw, and that's likely useful for easy battery operation. If you're tapping mains power, it's likely a rounding error.
OTOH, all the folks saying the knockoffs are $2-10, that's cheaper than a Pi.
I think people are getting hung up on verbiage here. Yes, I realize they are not completely identical.
But, the original comment was taking about the price difference between the arduino and an RPi. Sure the RPi is 'overkill' for for controlling brewing beer, but if it's cheaper, then why not use it? I could understand this argument if the more-feature-full thing was more expensive, but not when it costs more for less.
Don't get me wrong, I love Arduino. I use it way more than I use RPis. I just use it on $2 ESP8266 boards straight from China.
All the projects I did with arduino could have been done with the pi. Some even easier to code because I could have used python. So when a full microprocessor based PC is cheaper than a microcontroller why wouldn't you use the better option. Again, I know there are certain things a microcontroller excels at, but most people don't use their arduino for those things.
Now if the arduino was 10x cheaper, the entire value proposition and application space changes.
I witnessed two of my students trying to drive a neopixel led strip using a Raspberry Pi, and after hours of struggling with Python scripts broken by Raspian updates, they just stopped trying.
Before that, they had it working with an Arduino board quite easily (as Adafruit have example scripts and a library available). They are total beginners.
No operating system mean less power but also less complexity.
> It's a real shame that Arduino LLC seem to have lost any of the Maker-vibe it had
What "vibe" would that be? Supply shortage with undependable lead time? Having to piece together documentation from malware infected wikis that may or may not still exist 10 days from now?
Besides the obvious trolling occurring here, I'll answer your question on behalf of OP.
Arduino was (at one time) considered the holy grail of Makers.
They had open source schematics[1] and were very easy to use while quickly prototyping something.
For Makers, once the prototype phase had finished and it was time to move to 'production', the ATMega and ATTiny chips were very easy to buy in bulk and use directly in the product, with the guarantee that the code written on the Arduino would (for all intents and purposes) work on the embedded chip.
Long story short, before they shot themselves in the foot by infighting, they had a respectable standing among tinkerers and makers.
