NextThingCo, makers of the very popular C.H.I.P. single board Linux computer, have released the latest iteration of their hardware. It’s the C.H.I.P. Pro, an SBC designed to be the embedded brains of your next great project, product, or Internet of Things thing.
The C.H.I.P. Pro features an Allwinner R8 ARMv7 Cortex-A8 running at 1 GHz, a MALI-400 GPU, and either 256 MB or 512 MB of NAND Flash. The Pro also features 802.11 b/g/n WiFi, Bluetooth 4.2, and is fully certified by the FCC. This board will be available in December at supposedly any quantity for $16.
The design of the C.H.I.P. Pro is a mix between a module designed to be installed in a product and a single board computer designed for a breadboard. It features castellated edges like hundreds of other modules, but the design means that assembly won’t be as simple as throwing down some paste and reflowing everything. The C.H.I.P. Pro features parts on two sides, making reflow questionable and either 0.1″ headers or a cutout on a PCB necessary. As a single board computer, this thing is small, powerful, and a worthy competitor to the Raspberry Pi Zero. A C.H.I.P. Pro development kit, consisting of two C.H.I.P. Pro units, a ‘debug’ board, and headers for breadboarding, is available for $49, with an estimated ship date in December.
A $16 Linux module with WiFi, Bluetooth, and no NDA is neat, but perhaps a more interesting announcement is that NextThingCo will also be selling the module that powers the C.H.I.P. Pro.
The GR8 module includes an Allwinner R8 ARMv7 Cortex-A8 running at 1 GHz, a MALI-400 GPU, and 256 MB of DDR3 SDRAM. Peripherals include TWI, two UARTS, SPI (SD cards support is hacked onto this), two PWM outputs, a single 6-bit ADC, I2S audio, S/PDIF, one USB 2.0 Host and one USB 2.0 OTG, and a parallel camera interface. This isn’t really a chip meant for video out, but it does support TV out and a parallel LCD interface. A limited datasheet for the GR8 is available on the NextThingCo GitHub.
Putting an entire Linux system on a single BGA module must draw comparisons to the recent release of the Octavo Systems OSD355X family, best known to the Hackaday audiences as the Beaglebone on a chip. Mechanically, the Octavo chip will be a bit easier to solder. Even though it has almost twice as many balls as the GR8, 400 on the Octavo and 252 on the GR8, the Octavo has a much wider pitch between the balls, making escape routing much easier.
Comparing peripherals between the OSD355X and GR8, it’s a bit of a wash, with the OSD coming out slightly ahead with Ethernet, more RAM and fancy TI PRUs. Concerning pricing, the GR8 wins hands down at $6 per chip in any quantity. That’s significantly less than the OSD355X.
The original C.H.I.P. has been exceptionally well received by the community NextThingCo is marketing to, despite the community’s distaste for Allwinner CPUs, cringeworthy PR, and questions concerning the true price of the C.H.I.P.. The C.H.I.P. Pro will surely see more than a few uses, but the GR8 is the real story here. A jellybean part that contains an entire Linux system has been the fevered dream of a madman for years now. The GR8 makes putting the power of open software into any project much easier, and we can’t wait to see the applications it allows.
Neat device, but 6-bit ADC, why in the name of…
for things like battery management or overtemperature protection, this sounds pretty sufficient. Still, no idea where they got that silicon/IP from…
Ah, found the ADC in the datasheet (“LRADC”, p.9):
> LRADC– Low Resolution Analog-to-Digital Converter
> • Analog to digital converter with 6-bit resolution for key application
> • Supports multiple button press detection
> • Supports single, normal and continuous work mode
> • Sampling frequency up to 250 Hz
Seems to be meant for keypress detections, so throw in a couple of debounced buttons with resistors, and calculate the keypress combination in software (I really don’t see why I’d do that when I have 3x I²C (which each means a lot of I²C addresses) + 2x SPI and, such a lot of GPIO, that, even if I use **every single dedicated digital interface as such** leaves 5 GPIO pins unused, but oh well, chips aren’t made for the most elegant solution; they’re made for average engineers with a small time to market…)
Because resistors are much cheaper than I2C or SPI port extenders. If the precision of this ADC is sufficient then it can distinguish 64 keys on a single port pin. Even if the precision is lower, 16 keys are easy.
how would this key press detection work? Every key is in series with another resistor, so the ADC measures the voltage of a resistive divider?
Keys + resistors are also easier to layout.
You’d have 6 resistors, say 1K 2K 4K 8K 16K 32K. Each one can be bypassed by a switch across it. Put the whole thing in series with another resistor and you’ve got a voltage divider controlled by the switches. Any combination of switches can be detected.
If you want more switches, just use lots of switches in parallel, each one in series with it’s own resistor. Again, use another resistor as a voltage divider, then each switch will give a certain voltage. Pressing multiple keys together will give misleading results. The resistors in this case would each be unique values going up a step at a time, rather than each being twice the previous. As long as the DAC can tell each one apart.
It’s possible to do a combination of both, so you’d have some keys that can be pressed in parallel with a series of others. For, say, a shift key.
Overall it’s a bit ugly but I suppose it’s a way of putting a lot of buttons on a remote on the end of a cable. Means only 2 wires in the cable, where a grid layout would need (rows + columns) wires. Of course there’s other ways of sending control data down just 2 wires, and it might be cheaper to use one of those. Like a separate microcontroller in the remote, or a remote-control encoder chip. I think at some point the price and bother of adding all those resistors becomes a pain.
Some early VCRs used something like this in wired remote controls. Do designers still do that nowadays? I’d think the DAC is more likely, like Marcus says, for things like battery voltage.
“how would this key press detection work? ”
Analog voltage sum. Each key has a trimmed resistance eg. 1,2,4,8,16,32 kOhm and pressing down a combination of the keys will produce one of 64 possible voltage levels.
Or, you have 64 keys whose resistances are 1,2,3, … , 64 kOhm and pressing any one key will produce one voltage out of 64 possible. In this case, simultaneous key presses are not possible without some sort of matrix scanning.
“Do designers still do that nowadays?”
Very common in automotive steering wheel controls, as it requires so few wires. Getting signal lines out of a spinning wheel is expensive.
AFAIK, the LRADC doesn’t work on C.H.I.P. now, so why do we think it will work on the C.H.I.P. “Pro”?
This is a company that’s can’t ship product on time, so beware. I like their stuff, but be prepared to wait MONTHS for it. And the OS still has a lot of bugs in it. For example, PCM recording (like arecord) doesn’t work in version 4.4 (latest).
“• Sampling frequency up to 250 Hz”
It’s not intended for anything amazing (there is a 24-bit ADC running at up to 192kHz used in the audio system), the spec seems to mention “key application” and button pressing
I don’t understand:
> SPI (SD cards support is hacked onto this)
Does the author know more than the datasheet admits? That claims that the SoC has _three_ dedicated SD/MMC controllers (not to mention the NAND flash controller that is probably of far more interest to the low-cost embedded device designer) which, at least by the block diagram on page 13, seem to be completely independent from the SPI peripheral.
In the spec PDF it does claim “Three SD/MMC Host Controllers (SMHC)” and I assume the HaD text was a translation of the spec sheet description: “SD Card-ready SPI”
So, Brian read “SD Card-ready SPI” and interpreted that as “this device doesn’t have dedicated SD card controllers, you have to use SPI to build a software block device on top” (note: considering this is supposed to be supported by Linux, I’d expect the kernel to do that, but then again, I don’t know the quality of the SPI kernel drivers for this hardware)? Seems unlikely!
I’m not entirely sure what Brian was trying to infer on his translation of the original text but, looking through the pdf, I’m not sure what they were trying to say either as there appears to be the pin-out for all 3 SD/MMC controllers and 2 SPI interfaces with no explanation as to what or which the ‘SD card-ready’ SPI is.
Generally any hardware SPI is “SD card ready.” It isn’t a very meaningful phrase if they already said “SPI.”
The datasheet is a joke, like always with Chinese hardware.
Zero description of the peripherals and their registers.
I guess you’ll have to pull that from the binary blobs.
Well, this is a datasheet, not a register map. It’s not uncommon to release these as a separate PDF; blowing up a single PDF to 1000+ pages isn’t that sexy from a customer POV, imho.
by the way, I **thought** NTC was from California, but oh well, you seem to know better.
Even if, there’s far more Allwinner inside this chip than everything else.
Yeah, register map will probably be in the full manual if you can get a copy of it, but I believe it’s almost exactly the same as the A13.
So cynical! Did you try actually googling? On http://linux-sunxi.org/R8 is a 342 page PDF, in English, with all the register descriptions. Sure, it’s not on par with say a Texas Instruments datasheet but combined with the linux driver source it’s perfectly usable. I really get sick of people slagging off Chinese stuff as a kneejerk reaction. I’ve used A20-based boards for ages and it’s pretty great being able to run linux and yet at the same time hit the registers directly if you need to.
Of course they are releasing new models, I only have 3 more weeks till I get the C.H.I.P.s that I ordered back in June or July. I think I will be waiting for the new products to be stable on the shelves instead of funding this time around. Wonder if they would be willing to just upgrade me….
If your intention is to buy, then a crowd funding campaign is the wrong place to look for.
I new the risks of delays going into my funding and I was fine with the first ” were shipping ahead of schedule unless…” delay announcements, I was irritated at finding out that the next delay meant that I might receive forked products (added vendors meant different NAND? and new firmware) but I didn’t cancel cause these things happen and they were working diligently on the fix. What chaffs me now is the timeline on the new product shipping in Qty in December pretty much matches the expected fulfillment of the first campaign raises questions as to their diligence to the first project.
Some more thinking on it, the CHIP pro and GR8 are not crowd funded so just leveraged off the CHIP? And while yes you can buy a dev kit now, it wont ship till the regular production runs ship so… horse in the cart and going downhill?
First and foremost, when ascribing blame in a commercial transaction in which a vendor does not live up to their side is THE VENDOR.
Dont victim blame.. What did you think your comment would be taken as? The wise words of a shrewd man?
NO. It is just 20/20 hindsight victim blaming, and it totally gives the vendor, who’s fault it is, a pass.
Grow up. Buyer beware, yes, but also dont be too cowardly to lay blame where it belongs: on the heads of those taking MONEY from people then not living up to their end of the bargain.
It is super remedial ethics….
I’m not expecting my CHIPs until at least December. At that rate, it means the CHIP Pro probably won’t ship for another year, and by that time, the CHIP Pro Plus Mega will be “shipping next month” ;)
yeah, I’m just hoping that the CHIP hasn’t hit end of life already as far as next thing is concerned, Id hate getting the project I had in mind for this done then have to redo it cause the hardware is no longer available.
[heavy breathing]
i prefer a chip+epiphany (parallella.org)
now epiphany hve 1024 cores
i see on parallella.org the board is still stock to 18 cores (16 of chip + 2 of som). where did you read about 1024 cores?
castellated edges and a double sided load doesn’t make a goddamn lick of sense.
Does if the drilling is all one price and you just want the perforations to break the boards apart.
Put a cutout in the board.
No, it’s you that doesn’t makes sense – just cut a hole in your board. Would you rather the thing was single sided and twice as big?
Not really sure what’s the point of the CHIP Pro. It’s basically a less capable C.H.I.P offered at almost twice the price of the CHIP. The 256MB of RAM and 512MB Flash are simply not enough. The GR8 chip is a really neat integrated soln..but again with 256MB of RAM only, it falls short. One can purchase many an AllWinner H3 board with more interfaces & RAM for $12 and under. Or just stick with the original CHIP.
Not enough for what? Memory and resource needs are dictated by the project/use. 256M RAM and 512M Flash are more than enough for a number of embedded uses
Maybe so, but the last time I saw an Embedded board advertise that it has 256MB of RAM was in 2012 with the first iterations of the Raspberry Pi & Beaglebone.
Even the $5 RPi Zero has 512MB of RAM on it.
The catch is that you can’t buy large quantities of the original CHIP. Yes this board offers less value for money ($9 was an insanely good price) but it’s still pretty decent and – given that they’re obviously not expecting to sell any addons for it (like the HDMI board for the original one which had a healthy markup on the BOM) is reasonable. I’m considering using it for a commercial product; if we want to drive the price down later we’ll spin our own GR8 board and save a few bucks.
I was underwhelmed by the CHIP, but I like the idea of a small module with all the IO I could want running Linux.
$6 dollars for just the chip, pretty amazing that!
There are plenty of SBCs with closed off and poorly documented chips. I’m more interested in open chips. If you make a board with a (silicon) RISC-V CPU, I’m interested. If you release one with a RISC-V chip and the files to produce the chip, take my money!
from day zero, the whiney little shits bitching about transport costs being more than the item have really pissed me off.
here in Oz, a Raspberry Pi has NEVER been less than $50, a Pi Zero is over $40.
Most of the “developed world” has jumped ship for china, but at what cost???
I get this site is extremely right wing and anti union, but have any of you actually thought (ever) about the real cost of “globalisation”???
I don’t see hackaday readers as anti-union or right wing. Being hobbyists mostly they want to build cool projects for as cheaply as possible because they’re spending their own pocket money with little to no intention to monetize or commercialize their designs afterwards. In many cases spending as little money as possible is part of the hack.
Having said I do see where you’re coming from. The need for everything to be so cheap that it’s practically free, makes it very hard to open/sustain businesses in the west (and even in China ) that cater to makers/hobbyists. It’ a race to the bottom where everything can be compromised for the sake of keeping the price of a board/prodect lower than everybody elses. It is very disconcerting.
It’s not really my business if Chinese workers decide to form a union or not.
well it should be!!
folks trying to form unions in China get jailed.
here in the “west” we’ve been yelling “better dead than red!”, yet industry has/is jumping ship for china.
here in Australia, GM, Ford, Toyota are jumping ship to go to asia, just like they have in the rest of the “developed world”, that is thousands of jobs that will go.
once those skills and industries go, it will be a huge, expensive undertaking to get them back.
the whole “greed is good” has been embraced, with a vengeance by the “big end of town”.
our world is broken, in a terrible way
One could argue that a ‘Left Wing’ approach is causing this. If you ask why businesses are leaving for China, it’s likely because your country has made it very expensive to stay. One can argue it’s better to get a bit less out of a business (Taxes) than to wind up killing the golden goose.
Just like people, a business does what’s in it’s best interest. If you want a business to do something, make it in their best interest.
but it didn’t, what the gubernment spent on subsidies they got back in taxes, 4:1, it’s not just the car manufacturing that will go it’s all the component manufacturers as well.
but is that really the point?
what social cost can we put on thousands of folks out of work???
there is way more to life than making an obscene profit over making a fair profit.
and lets not talk about a neuvo rich class in asia being responsible for the extinction of wild rhinos and elephants
Are you saying you pay more than $40 for a pi zero and then more for the shipping? Or is it you pay $5 for a pi zero and $35 for shipping?
I know what you’re saying, people moaning. I’ve got a few zeros, all being used. I just brought them whilst getting other stuff and when the order was > $100 I got free shipping. And so they cost me $5. Not had to pay more……
10.75USD shipped to NSW. Make the most of the ailing GBP. Stock generally available. Annoying restrictions apply.
https://shop.pimoroni.com/products/raspberry-pi-zero
You can be right-wing, anti-union, AND anti-globalization.
Simply because outsourcing doesn’t actually save you any money in the larger scheme, since it unemploys and impoverishes your customer base. Same as with displacing people with industrial robots: if you throw all the workers on the streets, or force them to work for a shirt button, who you gonna sell to?
You basically sell to people who still have money/savings until they too don’t have money. Veil everything under a thick shadow of debt so that true wealth and the value of money becomes very hard to truly determine; making it harder for people to realize that they’re being duped and bamboozled out of their true wealth.
Basically, it’s the biggest heist of all time.
The majority of posters are ultra-right wing – compared to an Oz Wobbly anyway. Left of center in the US. I would guess pretty centrist by EU standards, but that is just a guess. Writers and topics lean left of center towards “doing something for society is better than doing something for yourself”. Whatever it is, the site contents and comments are far from the Objectivism one would expect from extreme right wing anti-union makers. (And the origins of hacking are rooted in utopian Fabian socialism or anarchy for cryin out loud!)
How quaint – you think the democrats are actually left wing, anti-globalist, and support unions..m How old are you? That is what I thought too when I was 19… almost 20 years ago. USA has no party for the workers. None. Look at the last 16 of 24 years.. The left has let the workers down.
And yes, I am a leftist. Most of HackAday is – but the right leaning are here also, and they understandably are testy about having their political opinions scapegoated by sanctimonious left-win-in-name-only sycophants.
Oh I see you are Australian and have no idea what you are talking about.
Although I’d hardly call this site ‘Right Wing’, I wouldn’t call it a political website either, and I sincerely hope it doesn’t become one.
me as well brother!!
but some commenters here seem to be pretty “conservative”
> but the design means that assembly won’t be as simple as throwing down some paste and reflowing everything.
This is pretty basic thinking out of a box actually for a hacker. Cut a hole in your PCB to accommodate the secondary side components, you can throw down solder paste and reflow. This also allow for a much lower profile PCB.than silly connectors stacking.
Really surprised a company with recent layoffs and money problems are actually considering building a platform and device and hoping people build with this. I won’t be betting on this company being around another year or two.
Where’d you hear that? Two folks in my circle just got offers from NTC.
Can you post your evidence for “layoffs”? They have several job postings on their site. I’m more prepared to design the CHIP Pro into a product than a number of other cheap ARM boards (which don’t specifically market themselves for use as a subassembly)
Cortex-A8? That’s a bit… retro. I thought the Cortex-A7 delivered almost the same performance, at much lower power. I guess it allows using a really cheap, trailing-edge manufacturing process.
The Beaglybone code base more than makes up for any age and A8 is in full production by a bunch of companies. Plus, 1 GHz and no heat sinks ever needed.
You said $16 twice in the article, and $6 once. They are actually $6 ea.
RTFA.
LOL. You tell ’em Brian :-)
I hope the Allwinner development environment has improved. A couple of years ago I sent a week in Shenzhen at a tier one support company Allwinner recommend when developing a tablet. They had no idea about the proper way to do software development, it was terrible! Like stepping back to the 1980s.
Yes, they are releasing new models, I only have 3 more weeks till I get the C.H.I.P.s that I ordered back in June or July. I think I will be waiting for the new products to be stable on the shelves instead of funding this time around. Wonder if they would be willing to just upgrade me….http://www.java-tips.org/
Umm… why no µSD interface and instead such a small flash disk? Even the old GNUBlin has a µSD interface and is only the size of a DIL40 chip. I don’t get it…
Think about the use case – as a submodule in an embedded system raw NAND is fine. If you want to add micro SD add your own $0.50 connector offboard. The R8 can boot off the “SDC2” port which is pinned out (or have a bootstrap that then goes to SD in the onboard NAND or… spin your own board with the GR8 chip).
The solution for key bounce is to spend enough money on keys to get the ones that don’t bounce. It’s been 15 years or more and I don’t remember where they came from but the things I put those keys on still work every time with no delays or denounce routines. I think they cost a dollar fifty each so they we-rent cheep but they were cheap compared to the time it took to get less expensive switches to work.
Red
Trust is built in drops and lost in buckets. – KEVIN A. PLANK