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...).
Considering the fact that they missed delivery date of the previous one by at least 150% (I am still waiting) announcing new one with "Available Dec 2016" sounds like tasteless joke.
My CHIP still not delivered. They re-scheduled 3 times the delivery. The last was 3 Oct. Today I received the "happy" news this will be shipped at the beginning of Nov. The initial shipment was June when I ordered in March.
I ordered a TextBlade (small-form factor bluetooth keyboard with pretty nifty technology) on March 2015 for delivery on June 2015 (not a Kickstarter, mind you) and I'm still waiting (actually all purchasers are, except for a beta testing group).
Actually, I did not participated in the Kikstarter campaign, but I purchased from the web site. So, my assumption is that if the company is trying to do regular business it must act like regular business.
The rule of thumb seems to be if they did a kickstarter, don't touch the company in any way till there's proof that a significant % of the backers actually have the product in their hands.
I have one Chip and one PocketChip as a backer; my lifepartner has another PocketChip. Last February I preordered five Chips (the max) and another PocketChip. My understanding is that it's actually the PocketChip that holding up shipping my order.
I was an early order of TextBlade too, cancelled after waiting a year for a device that was supposed to be in production and ready to ship. I got tired of letting people hold my money who wouldn't tell me the truth. The TextBlade forum is now being purged of anybody who complains; having originally created a "Rants" section, ranting is no loner allowed. For uncensored information, go to the TextBlade Reddit.
The TextBade mobile app, required for firmware updates and adavanced keyboard mapping has been in the iOS app store since last December (it's needed to be a beta user). Only mock-ups (at best; the video could well be fake) of the Android app have been shown; previously WayTools said they would not begin development on Android until after they started shipping product. They basic attitude has been that they don't give a crap about Android users but like their money.
Oh, is that you Maggie from the forum? I'm still vaguely able to complain there, or so it seems, but I still keep it low key. I also cancelled after a year, but then reordered when the beta testing thing started, since it seemed kind of imminent to get one. How wrong I was.
Checking your Scala on Chip deck, I'm pretty deep down the Scala rabbit hole since a couple of months ago.
A TextBlade? Well, there seem to be too many beta testers in the wild for it to be a scam of any kind, and too much money has been dropped on it, I suspect, so, yes, I expect to get it at some point. Of course I'm pretty pissed at having to wait that much, though.
Hang in there, this is par for the course with a lot of these boot-up hardware projects, the guys behind CHIP are not yet mass-market vendors of these things, it is always bumpy for first adopters of this brave new world of quaint hardware ..
That said, I have my PocketCHIP. It is a great little Linux box, the perfect thing to .. indeed .. stick a pencil in, ssh into, and use .. well, like .. any other Linux workstation .. like, omfg, could this actually BE a decent Linux desktop? Answer, immfo, is this: YES! Worth the wait!
Ditto. I had planned on porting FreeBSD to it, but with all of the delays and more interesting boards coming out since it was announced, I've lost all interest at this point.
I got mine, ordered not in the kickstarter but later, arrived on time. Since then I did not really use it, so far other stuff to do and my existing ARM homeserver is working just fine.
If you can use it do to some serious work like porting stuff to it, I'd at least lend it to you. Send me an mail (in profile), I'm in Europe.
I've got mine. the pocketchip ts pretty fun. installed python and idle and messed around. quake 3 would not run. but I played the shit out of nethack like I never have before.
I haven't figured out what exactly I want to do with it but it's been fun so far having a little midget debian box with a tiny screen. It forces you to use the console more which is healthy. like eating your vegetables.
I've already borked the OS like 3 times messing with packages and trying to see what window managers run... So at least flashing it is easy with the chrome app.
My only complaint so far is that the video drivers suck.
If the new ones are only $16 and they have the video driver sorted sign me up.
Also do yourself a favor and make sure to change the default passwords ASAP because it's just 'chip/chip' for everything.
Same here, I got my CHIP in January from the Kickstarter. However, my VGA adapter that was supposed to arrive in March didn't get here until June, long after I'd lost interest in the platform. Honestly, the RPi3 is a better deal overall, if you consider the value of the respective communities and much broader range of operating system support on the Pi.
I will say this: I wish there was something like the Pocket CHIP for the Pi, without having to build your own from the board up.
I ordered mine when pre-orders were first available and got them probably about two months ago. I would say it was worth the wait and have been really happy with the build quality and software. Hopefully they can figure out their distribution problems because they do have a good product.
I coordinated a group buy of 40 Chip computers and they are all delivered on time in June 2016.
We all had to wait. We bought them in November 2015, but we (like many others) were told from the start that shipping would have begun from June 2016, in order of purchase time.
I order the original CHIP at end of May, and I still haven't received it today, it was supposed to ship by June. They kept getting delayed. Latest update is that the shipping is for November. Can't help but suspect they are doing this on purpose to wait for price drop on the components to sustain the business.
just to provide a counterpoint, i have received mine on time (original promised date). although, they're still sitting around waiting for my next big idea(tm)!
"Powered by a chip you can actually buy" - lets hope so.
Assuming thats a reference to the rpi Zero which, as of November, has been out a full year and still must be bought one at a time. Seems like we'll never be able to build more than one or two of something based on the zero.
Do wish nextthing would rebrand it because its impossible to Google "chip" and hope to find anything related to this. How about "NESOC" - pronounced "knee sock" and short for nextthing SOC?
c.h.i.p. is probably not that great of a name, but 'getchip' always returns the relevant information for me. I do like the "white-label" implications of this name: If one builds a product using chip, that fact would be hard to recover from a cursory web search.
How is it optimized for SMT? It seems like the PCB that you mounted it to would need to have a hole or milled pocket in it to accommodate the backside thickness of the CHIP. Not sure how feasible a pocket is, but a hole would mean you then lose all your layers/backside in that area.
Milled cut out in the middle of the mainboard seems like an annoying additional cost. Headers that raise the module off the board would be acceptable, except when vertical space is a constraint.
yeah, it just seems like if it were really optimized for smt, they would have milled the back side of the CHIP and placed the flash et al components in there so that you wouldn't have to find the extra real estate on your own board.
If you were really optimizing for cost, you wouldn't have a $16 monolithic board in the middle with extra components you don't need. Compared to that, the extra cost of milling the board you're placing it on is minimal.
Seems like a good enough prototyping kit for open source hardware.
I wonder why the 'no-price-scaling' per unit for bulk orders is considered a feature though.
I have the PocketChip; great device. Recommended for sure. Together with come CHIPs you can do nice things. Sure for the same price as the PocketChip you can pick up a far more functional Android / Windows 10 tablet but this has a keyboard and you can replace anything in it yourself and do interesting hacking for both software and hardware.
Edit: (I bought the PocketChip for the introductory price; it's more expensive now but still cheap I find, especially as I would really like more people to make things like this... Is there anything else than the Pyra & this for handheld devices with a keyboard that you can actually order now and be quite sure you'll get it instead of throwing money away?)
I know it's cheap and all but the keyboard is hardly anything I would recommend for typing anything for more than 5 minutes. It's probably among the worst things out there to type on.
When I was googling the PocketCHIP I was surprised to find some random forum posts warning people NOT to peel the clear plastic coating off the keys, as apparently they hold them in place! That's certainly something I can imagine doing.
Oh yes, definitely heed that advice. I almost ripped mine off before I realized what it's there for. A very unfortunate omission on their part - there should be a warning in screaming caps on the device, or at least somewhere in the packaging.
Great hardware, but do you really want to run a full Linux system on low-end IoT devices? That's what made the recent massive DDOS attack possible - lots of little machines with way too much network-side functionality.
> While it's not possible for us to investigate all the attacking devices, it is fair to say that these attacks came from Internet-of-Things (IoT) category of devices.
> There are multiple hints confirming this theory.
First, all of the attacking devices have port 23 (telnet) open (closing connection immediately) or closed. Never filtered. This is a strong hint that the malware disabled the telnet port just after it installed itself.
> Most of the hosts from the Vietnamese networks look like connected CCTV cameras. Multiple have open port 80 with presenting "NETSurveillance WEB" page.
> The Ukrainian devices are a bit different though. Most have port 80 closed, making it harder to identify.
> We had noticed one device with port 443 open serving a valid TLS cert issued by Western Digital, handling domain device-xxxx.wd2go.com suggesting it was a hard drive (Network Attached Storage to be precise).
So its not Linux, its the apps. The insecurity of the boxes has nothing to do with their choice of OS, but rather their shoddy apps. Its a general pattern for IoT devices to be configured with default passwords and internet-facing admin pages etc, and using another OS doesn't directly address any of this.
ADDING: I'm a big fan and watcher of everything from OpenBSD to QubesOS to the Mill CPU to CHERI and all the rest, and I'm reluctant to say that right now the OS is immaterial to security :( ;)
Exactly. The best-selling home IP camera on Amazon [1], from a Chinese company, has the following "features":
- Default Wifi setup is done through their phone app through the "cloud" (let me tell you how much I trust that one)
- Has a webserver listening on port 80 with default u:p admin:admin (to be fair, their instructions are clear that you should change it, but it's not a mandatory part of the setup process)
- Has an RTSP server listening on port 554
- All of this a disaster waiting to happen because of UPnP (ugh, how many home routers have this enabled...)
- Sends outbound TCP traffic to amcrestcloud.com and amcrestview.com every few seconds (cannot be disabled on the device) [2]
- Sends a continuous stream of UDP data to 52.91.189.219:8800 (cannot be disabled on the device) [2]
The only way to prevent this device from a calling a CNC server is with a hardware firewall or an isolated LAN segment (I suppose this idea isn't at all specific to this camera). I bet fewer than 0.01% of their customers do that.
> All of this a disaster waiting to happen because of UPnP (ugh, how many home routers have this enabled...)
The problem is shitty UPnP implementations rather than UPnP itself. If you're pwned you are fucked one way or another, if an online device is vulnerable it's going to be vulnerable wether it exposes itself through UPnP or if it's manually forwarded.
And in the end if you don't like it and want to do your own manual forwarding in a home router, you're free to disable it.
Agreed. Bare metal programming is vastly preferable for many, if not most, realistic embedded applications. I hope they have a convenient gcc-arm-none-eabi setup available. It's not just about security; it's also a lot more work to do hard realtime systems with any sort of OS, even an RTOS.
We're even starting to see a bit of this on the high-level non-embedded side of things; projects like HaLVM and Mirage OS allow us to write bare-metal non-embedded code using extremely high-level languages. The idea is that you write a program that runs on bare metal (no OS or any OS-provided features) and you use standard library abstractions to get features that you would normally get from the OS. Then, if you want to run a bunch of these things, you run them in efficient hypervisor systems like Xen. This approach has many interesting benefits (and costs) compared to traditional OS-based high-level programming.
No. It would be more accurate to say that is is a lot more work to reimplement a hard realtime embedded RTOS from scratch, when you have a nontrivial application that needs features like tasking, synchronization, IPC, memory virtualization etc.
That is probably not Linux, but going bare target every time you have a new application and platform with hard realtime requirements would be a huge mistake.
> you use standard library abstractions to get features that you would normally get from the OS
They don't come ex nihilo. There are a lot of "standard library" APIs that are dependent on what would ordinarily be called an OS.
> Then, if you want to run a bunch of these things, you run them in efficient hypervisor systems like Xen.
This is often, if not usually, the wrong level of abstract for either application architecture or efficient resource usage.
There are intermediate steps between bare metal and Linux, and they're widely used in the embedded world. See Wikipedia for a long list.
I just noticed that there's now CapROS for ARM.[1] This is a capability-based secure operating system for small machines. It's a successor to EROS and KeyCOS, which had a good track record for security and reliability. Unfortunately, there's very little documentation. Somebody could bring that up to usability for ARM embedded devices and make money selling support to IoT companies.
The hobbyist/non-technical market can get on board better with Linux.
For one it has a GUI.
Also you don't have to write firmware and you dont have to cobble all the pieces together from scratch or from a miscellany of out-of-date examples.
Instead you can pipe together existing software and/or write your own in the high-level language of your choice.
No, you don't. You want your devices to be simple and talk to a bridge that runs e.g. Linux and handles networking. But at the moment this is where the ecosystem is at.
...Now if only we unwashed, non-kickstarter-using masses could actually buy a chip+pocketchip. I'd pay good money for it, but with orders this backed up, I'm not putting money in for a ticket to a waiting list that may never end.
That's just the developer breakout board, I doubt they intend people to use it in production devices. Most people are going to design their own boards that this solders into as a module.
There's a post[1] on their forum for more detail. TL;DR they are announce 2 new hardware
* GR8 - a 14mm x 14mm System-in-Package, combines 1GHz R8 SoC with 256MB of DDR3 memory - $6 - available in any quantity in December 2016
* C.H.I.P. Pro - GR8 + 512MB NAND + WiFi/BT + ..., 76% smaller than C.H.I.P. - $16 - available in volume December 2016 and Dev Kits are on sale today for $49 shipping in December 2016
Having some hobby project with CHIP and loving it, but I have to say this naming of Pro sounds misleading, I'd rather call it Air :)
Previous CHIP has twice the RAM (512MB vs 256MB) and eight times more flash (4GB vs 512MB), and it is much cheaper (9$ vs 16$). The "Pro" CPU has same specs (1GHz ARM v7a). The only advantadge is smaller components, and probably less power consumption.
In my opinion, the 8$ CHIP is better than the 16$ "CHIP Pro". Also, for 15$ you can get a Pine64 [1] with much better specs (512MB RAM, 4 ARM64 Cortex A53 cores, ethernet, USBs, HDMI output, etc. -without flash/WiFi/Bluetooth-).
I guess that the 9$ price for the original CHIP was too low, being the reason of shipping delays (needing to compensate the lower board price with overpriced add-ons, with the risk of if not selling enough add-ons, running out of budget (?)).
1) A 5V 2A power suply is required (cheap to buy e.g. on Amazon or eBay). It boots with 5V 1.5A, too, but less stable if you are using the two USB ports at once with devices requiring notable power (e.g. wifi adapter).
2) Requires a 1080p HDMI monitor/TV, at least, for the images I tried (you can change that, editing a file in the SD card image, I knew that afterwards).
3) The OS image I use (Debian) have a bug in the network driver so I have to boot with network cable unplugged, and after booting, plugging it, and stop/start the network in the network icon (top right Debian desktop bar). It may be already fixed, but because it works as it is, I have no problem to click twice to restart the network after infrequent reboots I do for maintenance (e.g. once a month). I'll update it when I see someone reporting that it got fixed.
Regarding temperature, I put small metallic heatsinks (fanless) over the SoC and over the I/O chip. Running flawless on 100% load (where I live maximum temperature is about 35ºC in Summer, and I had not problems at all running the board 24/7 as ARM64 build server).
It is true that some provided images have issues, e.g. lack of 3D hardware acceleration on Debian.
TL;DR: for ARM64 development and GPIO stuff the Pine64 board is great (1 and 2GB model, the 512MB for head-less Linux), as "Linux desktop home cinema", it is not, because of lack of hardware-accelerated video and 3D support on provided Linux desktop disk images. However, using the same board (models with >= 1GB) you can run Android, with hardware acceleration, if I recall correctly (haven't tested Android myself on that board, yet).
Yeah I'm lost too. The Pro is almost twice the price of the original chip, and afaict the only difference is that it's a little smaller (and has no USB port).
Huh?
I've got a handful of the original CHIPs, and they're great for what they are. But I don't get this "Pro".
Seems this is not the same form factor or pins as the CHIP.
So it can't be swapped out to upgrade PocketCHIP, etc. They seem to be missing an opportunity here to create a standard. Pity, I had assumed that was the goal.
(Of course, maybe there was a good reason to throw away the first attempt at a standard..)
1. It is now SiP, not SoC, because the GR8 chip incorporates 512MB RAM! System-in-Package.
2. They made changes here and there so all of the GR8 is open compared to the previous SoC. Most notably I expect the video decoding/encoding in hardware should be unable now.
3. Much smaller form factor.
4. Now has very fast NAND memory, so flash once and forget.
5. Overall they got input from developers and implemented what is needed.
6. Most important, same day Linux kernel support for the GR8 in Linus's tree. The commit was accepted a few days ago.
A chip with a design more suitable for embedding in a device. After you make the initial outlay, SMD components become much more easy to deal with than pth, for example.
You can use any SoC as the starting point for an open source smartphone. I don't think this is appropriate, though, given the inability to break out high-bandwidth busses from the board, e.g. for a video display or LTE networking.
3g is plenty fast, for the next few years at least. The only advantage of LTE is latency, and there's only LTE when you're in the middle of a city and probably have wifi anyway.
Owning a phone that came out like 3 months before 4g got built into everything (from high-end to cheap stuff), I can say I've never missed it. I'd rather they spend money on improving 3g reception, which is pretty good but could still be a lot better.
Yeah but that doesn't make 3g slow. Like I said, 3g is plenty fast, meaning it's fast enough for mobile browsing and even watching videos (though videos over mobile data is just stealing money out of your own wallet, but that's a separate issue). That 4g/LTE is faster doesn't make 3g any slower. The only slightly noticeable advantage of LTE is latency or downloading large files (which you wouldn't do over a metered connection if you had a choice).
> Not in the US. Every marginally populated area has it.
Looking at coverage maps of the Netherlands, I guess I'm outdated. There's little difference between 3g and 4g anymore. I remember not long ago when 3g was nearly everywhere (so close to 100% coverage) and then they started putting down 4g everywhere. 3g advancement stopped altogether. Right now it looks like they just upgraded the 3g spots to 3g+4g and forgot about advancing the network at all.
No idea. I'm not familiar with the CHIP Pro, just quickly read the datasheet. The SOC contains Mali400 GPU, so in theory it should capable for video output, but I can't find any mentioning whether the video capability is enabled in hardware/software level in CHIP Pro (i.e pins wired & Linux driver).
Raspberry Pi has similar problem: although it uses a SOC designed for mobile applications, there are no open source solutions to connect generic smartphone displays to DSI port, and HDMI is very bulky/over-engineered solution for transferring video signal over few centimeters. (Blame Broadcom, not the Foundation.)
Hoping to put this or RPi3 into foldable card cases and connect to power, vga/lcd monitor, keyboard and mouse for schools in developing world. Would the chip or chip pro be better than RPi3 please?
Well, CHIP Pro is not suitable for this use case. I would recommend RPi3 because it have more CPU power, more USB ports (you can connect kb and mouse without using a hub), built in Ethernet and HDMI port, so you can connect HDMI and DVI monitors directly (DVI with passive adapter). CHIP have only composite video out of the box (as well as RPi) and in order to connect HDMI or VGA you need to purchase additional "dip" board. So, to round this up, RPi is better except you need VGA, in this case CHIP + VGA DIP is probably a better choice, but it would require an USB hub.
But on second though, CHIP have a 4GB on-board storage, and or RPi you need to use removable micro SD card. I'm not sure if kids and removable micro SD's is a good combination, it may be ok, may be not, I just don't know.
Every chip vendor I've dealt with, from Amlogic to Allwinner to Freescale, requires a signed NDA for full documentation. Public documentation is usually heavily redacted.
The GR8 seems to be a rebadged Allwinner chip, and the no-NDA datasheet seems to be next to useless, to be honest.
One of the good things with the Chinese companies is that the datasheets will usually make it out into the public if the chips get popular enough. A byproduct of the "Gongkai" ecosystem:
It could be, but reading up on the features of the C.H.I.P. Pro it supports mainline Linux. I assumed this meant without blobs. Is there mainline support for any Allwinner chips?
I've been running bananian with the 4.4 kernel on my bpi-R1 (A20 SoC) and everything seems to be working fine. Although I don't use it for anything graphical (it's a networking/storage/server device.) Can't say how much it's relying on binary blobs though.
The GR8 is an AllWinner R8 with RAM in a single piece of resin. They are targeting product builders, see the section:
> Powered by a chip you can actually buy.
> R8 SoC + 256MB DDR3
> 1 package. 1 price.
> $6 in any quantity.
(that's $6 for their rebadged chip.) TBH this looks like not a bad solution for someone looking to board-down a linux capable product. I wonder if the GR8 needs a power mgt chip or if that's integrated into the SoC as well.
Back to your original question, NTC (among others) worked with AllWinner I believe to mainline some of the R8 code and reduce (or possibly eliminate?) binary blobs.
What do you mean by "full documentation" and "NDA"?
Go to NXP.com, sign up for an online account, and you can download full reference manuals for most of their products. They may hold errata and more low-level/non-user documentation closer to the chest, if your sign-up data is obviously fake or appears to be North Korean your account may not be activated, but it's not the case that you need to have an NDA signed in blood to get manuals.
Many companies. Almost too many to list. This is probably a dig at Broadcom who makes the chip in the Raspberry Pi and usually requires NDAs for their chips. Hell, Broadcom won't even let you buy many of their chips without their permission!
Pretty much every SOC manufacturer. To protect their intellectual property, they control access to the documents that describe the behavior of their products. Also, if you leak any of their intellectual property, the NDA allows them to sue their lost profits (due to clones of their product) out of you.
Do you know if these manufacturers then suffer any more from IP theft because of the more open policy? Or are the NDAs here used more as a 'serious customer' vetting tool so they can concentrate their support efforts towards these customers?
The first number, "33" is the pin number of the CHIP Pro, "CSID4-SDC2-CMD" are the functions, and "PE8" is the actual pin name from the GR8 module. The key on that diagram is confusing.
