I have the chip, esp8266, rpi, teensies, trinkets, arduinos... I studied electronics principles and built various circuits.. yet I have no idea what I can practically use these devices for in my life. They all sit in a box and I have a hard time justifying buying more of them.
Raspberry Pis and CHIPs: good for a home automation hub of sorts or a thing where you want a portable computer capable of working with video, audio, or large volumes of data.
Arduino et al: need to have buttons do more complicated things than buttons can actually do: slowly change light colors, actuate things, CNC, etc.
ESP8266: sensors and remote controls for switches and such. Anything Arduino does above, except the button is now on a network and not necessarily physical.
I have the same problem, but at the same time I have found some use cases. For example an Arduino to remember the height settings for my sitting/standing desk project. Or an ESP8266-based garage door opener on the cheap.
I do think that a lot of the not-Arduino hardware a la teensy is not really that relevant to me. I don't want to study hardware. I want to build shit, and in most cases a Wemos D1 Mini is actually what I need. But there are cases where things like that can be really nice. The MSP430 for example is a really low power chip.
Basically if you don't know what to do with the stuff, donate it to a local hackerspace. Or find a project idea, then dig into the box-o-stuff.
I'm just getting into electronics but have found the Pi indispensable for this purpose: without much effort, I now have Pi Zero running Homebridge with plugins and hardware to control an AC power outlet switch, ambient temperature and humidity sensors, three contact sensors, and another Raspberry Pi 3 also running Homebridge with a light sensor, fan speed controller (PWM) / tachometer, motion detector, and camera.
What's great about the Raspberry Pi is it is a complete computer system, so it can run the Homebridge stack (Node.js, mDNS, etc.) for bridging your custom home automation devices (via GPIO) to the rest of your IoT system (I wrote up my experiences here if anyone is interested: https://medium.com/@rxseger/home-automation-with-raspberry-p... Home automation with Raspberry Pi + Homebridge).
However, having a full OS is probably overkill for most home automation sensors/devices, I'm looking into getting an ESP8266 or other low-power wireless microcontroller next as I continue to build out my automation to other parts of the home. An ESP8266 may be better than e.g. a Pi Zero for many of these use cases, having built-in Wi-Fi support versus an add-on USB Wi-Fi adapter. I also want to put an MSP430 to work on home automation, but need to figure out the wireless story. Instead of a heavyweight Wi-Fi stack, probably a lightweight low-data rate protocol on an unlicensed sub-1 GHz or ISM band, bridging the slower RF connection to the rest of your Wi-Fi or Ethernet network (ala ZigBee/802.15.4 and the Philips Hue bridge).
The Wemos D1 Mini (~$5) looks really nice for an ESP8266 board. I went with a NodeMcu on Aliexpress (~$3), hopefully it works well once it arrives, but at these prices not much to lose, and you can buy many for multiple projects: a computer in every room?! Exciting times, more and more companies are developing expensive home automation products to build an ultimate "smart home", but with the explosion of inexpensive single-board computers/microcontrollers for electronics hobbyists there has never been a better time to get into DIY home automation.
The D1 mini definitely is nice, I have a dozen of them running all over the house, controlling the lights, monitoring CO2, temp, humidity, power/gas/water usage the works. Do get the feeling that they might start congesting my router, but have a pi home server in the works to prevent that.
I wish there was something ESP8266 like with Ethernet, ideally even with PoE. WiFi is ok, but since I need a cable for power anyways it could carry the data just as well...
(and WLAN spectrum is quite cramped here, and part of me wants to isolate everything into nice separate networks, just to be sure)
Thanks for that, I like the compactness of the Teensy solution. If we're going with shields it's probably worth mentioning that the Arduino also has one:
I'd say the Teensy is in the Arduino++ camp there. It uses the same libraries, etc, just has more power processing, a pretty nice price, and some interface bits and pieces that are very nice, like being able to act as a usb device and host. A lot of hardware, I'd argue is better to be treated as a library -- there are libraries and hardware that have definite feature sets, some of which overlap, that allows you to choose the best tool for the job.
The ESP8266/Wemos can run Arduino too. Over the summer I built a wifi clock, but couldn't get any of the native LCD libraries for the ESP to work. I flashed Arduino and used the LiquidCrystal library, initially just to make sure I had wired stuff up correctly. But then it worked, and I decided to stick with it :)
Best feature on the Teensy IMO is usb native, so you can do stuff like emulate a USB keyboard or mouse. Many DIY keyboard projects use them as the brains. https://www.pjrc.com/teensy/projects.html has more ideas. Also works great for larger Ambilight displays than Arduino can handle.
I used a Wemos D1 Mini and a relay shield for this. Soldered leads to the physical button inside my garage and connected that to the relay. Toggling the relay on then off is now equivalent to pressing the button. There is an outlet next to the garage door button, so the project is powered via a USB plug and a short micro USB cable. It runs a firmware that emulates Belmont WeMo (which is really UPnP) so my Amazon Echo recognizes it. Viola! A voice controlled garage door.
I have yet to install the reed switch on the door so the thing is smart enough to actually open and close and not just emulate button pressses.
I don't trust this setup to control mains power (look up Sonoff devices for that), but the garage opener is 12 V so I feel ok about not setting things on fire with that.
For me: because it requires the least amount of hardware and also works with Amazon Echo for voice control. With a different type of radio, if you still need it on the internet, you have to somehow bridge it, which means more hardware.
You mean like going to the super market without having a shopping list, buying everything that attracts your attention which all goes out of date in your fridge because you forgot you had all that food and not having any time to plan your house hold duties?
Sibling comment to mine thinks you're harsh, but I think you nailed it. I have an Arduino and a Raspberry Pi both just sitting around because I never came up with interesting things to do with them.
I also have a lot of food in the fridge and pantry that's expired or near-expiring because of the promises I made myself at the supermarket that I've totally reneged on.
I think that's overly harsh. It's closer to wanting to learn how to make cocktails, taking advice from a friend on how to stock a basic bar, but not really feeling comfortable enough with the domain to experiment beyond fixed recipes.
That's me when I visited Shenzhen. I came back with close to $1k worth of electronics - boards, controllers, sensors, actuators, stuff. Donated half of this to my local Hackerspace though.
I have 4 RPi's, a couple of BeagleBone Blacks, numerous Arduino and Teensy boards.
Disclosure: This stuff happens to be my hobby and profession, and I have extensive experience with embedded systems and microprocessor based electronics design.
The smaller boards have been infinitely useful to me. I do a lot of development, using Teensy boards to create peripherals that I can connect via USB to either a PC or a RPi. These peripherals control things like sensors, motors, etc. The microcontroller code serves as a sort of hardware abstraction layer, letting me test and talk to my gadgets via a terminal program.
Admittedly, the RPi's have been less useful. Because my gadgets talk via USB, it's easy to configure my systems so that I can do my higher level software development on either a PC or a RPi. The same software, in Python, runs on both systems with no code changes. So I tend to use the more "comfortable" system for development, and that's the Windows system with its big screen and quick response.
Bill Gates claims to have predicted that more people would be interested in developing software than hardware, and when the software runs on any platform, it gets harder for the lower performance hardware to justify its existence. Most of my projects are finished to my satisfaction without ever completing the final step of transferring them to a RPi to run autonomously.
Maybe the problem is that cross platform portability ultimately makes platforms less important.
That looks like fun. I haven't played with a FPGA yet. My preference for using a microcontroller, as opposed to bit banging the FTDI, is that the MCU runs in real time and has peripherals such as SPI, timers, etc. Right now my favorite board du jour is the Teensy 3.2, though I just got a 3.6 to try out.
I recently bought a Wifi Modem for my Commodore 64 that is based off the Arduinos board. OK, that sounds embarrassing when I say it, but I do value it nevertheless. It allows me to connect to some BBSes (!) where like-minded crazy people like me can hang.
I think it all has to do with the creativity and/or niche (in my case, a passionate hobby) designers like you can fill.
I've got a little thingy I made myself with a Minimus AVR, to replace a parallel port to BBC Micro cable I bought about 12 years ago, when PCs still had parallel ports. USB cable connects PC to AVR, also providing power, and a bunch of wires connects AVR to BBC.
The AVR pretends to be the original unmodified PC software, translating things to and from the USB protocol that the modified PC software now uses (which was pretty straightforward to do using libusb).
I have a similar rig for my Oric-1/Atmos collection, wherein the processor being used to emulate a disk sub-system for the Oric is itself a ridiculously faster computer than the Oric hardware its supporting .. (Cumulus)
Nice idea to make a Wifi Modem, I've got an ESP8266 I may just wire up to an Oric one of these days, when I open that drawer full of junk and find it..
Always cool to keep the old machines going, no matter what ..
@Level29_BBS, Particles BBS, and Centronian are my favorites. You can connect using Strikelink. (discovered via @duhproject, who is an awesome follow if you're into the 8-bit scene, or want to be)
Media devices or home automation is two big players I would say. But I feel exactly the same as you. I used my rpi as a server for different things and I have used it as a media player.
But now I want to build video surveillance and control it all with my rpi. Though, I live in an apartment so there is not a super large need which makes me prioritize other stuff since I don't really need it.
Another thing I want to do is an app that connect to different lights in my house so when my alarm goes off in the morning I want it to light up the hallway so that I can wake up. I live in Sweden so here it's dark as fuck most of the year at 06:00-07:00.
You can get RF Controllable Wall Light switches that replace the existing switch. The ones in the UK commonly use 433mhz and a configurable static code to turn them on/off. The benefit of them are that you can still control them from the switch itself so you won't be looking around for the bloody remote, The in socket types have to have the wall switch on all the time so if someone turns the light off from the switch then your automation goes along with it. And the 433Mhz signal is normally easily decoded.
If you don't have a SDR (a cheap RTLSDR will do) or an oscilloscope you can wire the remote that will come with them to your soundcard on your computer (And you can normally use a cheap USB soundcard if your paranoid about frying the sound card in your pride and joy, I never had an issue when I used to do it) and use an audio suite such as audacity to record the signal and then manually decode the signal.
Once you have the signal you basically then do a replay attack to control the switch using your pi an a cheap 433mhz transmitter.
The downsides to these are that they have no security if someone knows the codes to turn on/off your lights they can but they are limited by the range of their transmitter. You also don't get feedback from these style of switches so the software doesn't know if the switches are actually on or off. I would send my codes a number of times just incase they switches didn't pick up on them the first time.
You can get WiFi switches too but they are more expensive, add complexity to the project (if they are not well documented) and personally I would put them on their own wifi without access to the internet. (But you could write your own bridge so you can control them over the net.)
Openhab[1] is a nice Home automation suite that runs on the Pi and has a lot of modules to control lots of different things already written for it. One of the things it can do out of the box is calculate sunrise/set where you are. So I would use it to turn on my outdoor lights 15 mins after dusk and switch off at a set time.
Just a word of advise to help you making the same mistake I did with it at one point, make sure you write the log's to ram otherwise you can burn though SD Cards pretty quickly (well it took 6 months for one of mine to die)
EDIT: You could also use Philip Hue's (or their compatibles or anything using ZLL) and hook up a zigbee to your pi, create your own "hub" and control the lights directly as the ZLL Master key has been leaked. But these also suffer from the "in socket" issue where if someone turns off the switch on the wall then you lose control of the bulb.
I was in the same boat as you until I stumbled on a real, money-saving project. My PetSafe invisible fence wired transmitter blew out, and I wasn't keen on dropping over $100 on a new one. After dragging my feet for a few weeks and doing nothing about it, I realized that I could probably replicate whatever the transmitter was doing with my RPi. Sure enough, somebody online had hooked their transmitter up to an oscilloscope and figured out the correct signal pattern to send over the wire.
If I have time tonight, I'm going to finish the code and test it out.
When I'm done, I should have a new transmitter that I can control via my phone for significantly less money than the junk PetSafe sells.
Set up an automated indoor garden. I'm currently working on a simple ebb and flow hydroponics garden with a dashboard to monitor health stats etc.
Pump on/off, lighting and sensing some stuff.
I'll try my hand at tomatoes and maybe chilli peppers. The initial goal was to grow as much as I can which goes into home cooked chilli :)
I used an RPi for my hydro system, I grew a few generations of really good green "tomatoes" and had quite amazing yields... satisfied my geek side and my gardening side all at once... :)
The basic ebb and flow tech is simple. I'm prototyping it with a simple hardware timer. For now my plan is to simply add a programmable switch and some sensors once the system is up and running and go from there (maybe a webcam).
exactly what most people were talking about computers before.
think of those things as hobby computers and Intel/amd/arm as the huge inaccessible ivory tower mainframes. you know how you have zero access to any device on your laptop or phone? those things will eventually solve that. but they are hobbyist grade for now.
Aha, thanks for the perspective. I was about to lament my lack of motivation to fart around Arduino'ing & CHIPping stuff but this actually puts a bit more wind in my sails!
Edit: Pun not intended but retroactively affirmed between "fart" and "wind."
As a muggle, every time I see these things I can't help but wonder:
WHY do we not have a good open-source gaming handheld based on one of these yet?
Smartphones have all but killed the traditional physical-buttons Game Boy/DS/Vita model. Gaming on glass sucks, but people generally don't want to carry around another device in addition to their phone. I'm looking forward to the Nintendo NX, but I'm afraid a proprietary platform won't be able to bring handhelds back again — it will just be too limited and locked-down.
Here's my wish/suggestion list for a modern handheld, for those of you that can build these things:
- Open-Source OS. (Android?)
- Bundled with emulators for all classic systems (C64/GB/NES/SNES/etc.) out of the box.
- A game store.
- Two joysticks, D-Pad, 4 buttons, 2 Triggers.
- 720p Touchscreen.
- AirPlay-like ability to play on my TV screen, or on a computer screen via a macOS/Windows/Linux app.
- SD card storage.
- WiFi.
- Ideally, Swift toolchain with a high-quality games SDK :-) (or at least Unity support.)
Don't need a camera. Don't need "apps" like browsers and media players or whatnot. Got 'em on the phone. Probably don't require Bluetooth either. Not sure about a microphone — may be needed for multiplayer chat.
The price point would need to be cheaper than any other handheld and most smartphones.. I wonder if the tech is there to make and sell something like this for less than a 100 bucks?
Probably because that would be a legal nightmare. You'd have to get approval from all the companies that made the systems you're emulating and the companies that made those games (sometimes they're the same company). And I doubt they'd be interested in supporting a system that will eat into their own sales.
I still have the wish to build a board for a Raspberry Pi or similar that, through the GPIO pins would be able to drive an Apple II as its console (using the keyboard and 80-column display) by directly reading from and writing to the 6502's address space. A tiny bit of firmware could initialize it as a normal Apple II peripheral and, with the right software running on the Pi, it'd be possible to emulate a number of different cards in software.
If I could lay my hands on an Apple ///, which had a color-capable text mode, it'd be even more usable.
My real wish, however, is to be able to actually work from the Apple II :-)
I've actually moved a tiny bit on from where I started. I have a 40 x 4 monochrome display working with an rpi, correctly displaying keyboard input. Just working on scrolling now which obviously requires actually remembering what is typed (I'm adapting Antirez's Kilo editor).
I posted a reply on an online forum earlier today where I said that 90% of Arduinos, etc. are sitting in someone's desk drawer gathering dust and the other 10% are doing some very trivial task. Then I felt I was being too negative, so I deleted the post.
You can make C.H.I.P into an airplay receiver (or DLNA) . This was the cheapest option I was able to figure out for doing this, and while I'd use e.g. a Chromecast audio if I could, it can't replicate the functionality I want. Namely, to use my phone as a remote control for my home audio system, not a streaming source.
I've also made some nice, tactile (e.g. big switches and clicky knobs) electronics for simple things like kitchen timers, etc. with Arduino style boards.
It sounds like you shouldn't buy any more of them, though. If you're not inspired to make something, more boards certainly won't help that.
Learning and prototyping are key uses. Both scenarios are not about end products, but can lead to great outcomes in the future. Especially if the users are young people that end up building things like satellites that can warn you about upcoming hurricane or pinpoint your location on the GPS smartphone map and other great inventions of humanity.
The software world metaphor: It is like your first programming manual that has examples and maybe even projects - they are not to be 'used' but 'learnt' from.
I left one RPi2 at my mother's apartment to work as a NAS; second RPi2 is powering a skeleton for home automation system at my current apartment; a RPi3 sits in the drawer and waits for better days...
Same feeling here... I hardly get excited with all the pointless/because-I-can projects based on these boards. It's fun, OK, but that's all.
Anyway I use Arduino a lot to teach electronics in high school - really useful there - and a Cubieboard (think RPi with SATA and decent eth) as webserver (small sites via home DSL).
While not as big as security and media devices that were already mentioned, seems like a lot of effort is going into DIY 3D printers and CNC machines. Having a lot of those kind of machines around could enable the rise of guilds that produce a lot of useful things.
Write a driver to offload interrupt processing of your Ethernet stack or something like that to them. Let's you get more done. Alternatively, the monitor the health or security status of your system with recovery option.
I've been thinking about the odroid c2 http://odroid.com/dokuwiki/doku.php?id=en:odroid-c2 (Hardware details: http://odroid.com/dokuwiki/doku.php?id=en:c2_hardware ) but they strongly recommend an eMMC module which costs a pretty penny in my opinion compared to the board itself which holds me back. I'd rather just buy a 64 GB micro SD card (which costs ~ $20 even if I got a fancy U1 samsung one that I have on my phone).
it would be a pretty neat way to use my 1680 x 1050 LG tv as well I think and the Ethernet port is a major plus.
It's great actually (modulo binary blobs). With the eMMC modules compute, storage, and communication feels fast enough to be actually useful. I use the default Ubuntu installation. The 2 GiB of DRAM is really valuable to me.
Even though the Raspberries are cheaper, their only advantage to me are the vast assortment of good [value] accessories, such as the Official 7" Touch screen.
I'm intrigued by the CHIP Pro, but am so disappointed by the lack of DRAM.
Got a Pi controlling my Christmas decorations (Overkill? Yeah. But I wanted to easily push updates to the patterns so why not ssh into my decorations). Got one to play media from network (built in DLNA player on that tv isn't great) and runs livestreamer for me to get my Twitch fix. One is also the base for my home automation with a number of radios attached so I can control zigbee and 433mhz stuff. Heard of people use them as honey pots, Larry Pesce seems to have an addiction to them.
I was working on a project called TwitchPlaysKitten as I had a new kitten and thought it would be funny. A few Servos with cat toys attached to them connected to a Pi and a webcam and allow twitch chat to control the kitty toys. But Twitch shot me down talking something about Terms of Service and staff saying "you trust twitch chat with physical control of items in your home? Are you fucking nuts???" Screw you twitch, it was going to be glorious :-p.
Got plenty of ATMega's dotted around the house. Started with them with the Arduino and the ATMega168 and progressed from there. Used them in stuff like energy monitoring using Current Transformers, more Christmas decorations (My mom wanted some new lights for her tree and she has saw my decorations the year before and roped me into to knocking her up a set) based of a ATTiny. Also used them to repair failed electronics.
Programming AVRs can be done many ways. One of the go to methods is SPI programming and making your own programmer is cheap. You can make a SPI programmer from an Ardunio or use the SPI port on a raspberry pi and a modified version of AVRDude to use true SPI of the Pi instead of bit banging it and getting faster programming times.
Got a few esp's dotted around as I liked the built in wifi, slapped one in my cat flap so I can easily see if the cat is outside or just asleep under the bed upstairs, also lets me set timers and override the cat flap functionality from my phone/desktop without having to go over to the cat flap, clean the labelling around the manual dial to see with direction is which option (Sureflap Dual Scan cat flap - https://sureflap.com/en-gb/pet-doors/dualscan-microchip-cat-... it's based of a Pic16f iirc).
Which leads me onto Microchip. Fell in love with the PIC's some time around the PS1 era. It completely had noting to do with someone releasing code that allowed a PIC to convince the PS1 it was booting a valid game, honest guv... But mainly use Microchip for their other components until the rapidfire mods for the 360/PS3 kicked off.
I personally use AVR more often then PICs simply because I fell love using using gcc and avrdude and programming newer PIC's required getting a PicKit3 and at at the time there wasn't any clones of them and I was broke. Until then we could program them using a simple homemade serial or parallel programmer which suited my broke ass's budget. But I have to love Microchips sample program, they have a generous sample program where they will give hobbyists and students freebies (ATM it's 12 parts per month iirc).
Also having a love affair with Nordic Semi's NRF51's atm but that is mainly cause I wanted to teach myself ARM.
So what's the point of the wall of text? To say that I'll prob get a couple when they become readily available because I'm sure I will find or make a use for them, even if it's just to play with them to see if they could be useful in the future.
I can see your point where you say that for you they end up just living in a box gathering dust but some of us can just come up with ideas to use these. It's not for everyone but I'll happily "waste" an evening wiring some WS2812's to a ATMega32u4 and write some C and C# (my current goto language for hacking things together on Windows) just because I want to put some lights in my PC case and I wasn't happy with anything I found on the market - https://youtu.be/IhsALOwoszk (yeah, I know. I'm 33 going on 14...).
