Imagine for a moment that you are back in 2015. Radio Shack are going to the wall, Heathkit returning from the dead, and Arduino spliting into two warring Arduinos. And someone has announced a tiny Linux-capable microprocessor board called the C.H.I.P. that will cost only $9. We all thought that last one was pretty cool at the time, didn’t we. Then Heathkit’s new products turned out to be pretty lacklustre, the warring Arduinos merged, and the C.H.I.P? The consensus was that $9 was a tall order for that BoM at the time, and then the Raspberry Pi people gave away a free Pi Zero on the front of a magazine before selling it for £5 ($6.30). It didn’t matter that the C.H.I.P. had a nifty all-in-one screen and keyboard combo called the Pocket C.H.I.P. which was a significant object of desire, the venture lasted for three years before finally hitting the rocks last year.
Now the C.H.I.P. is back, in a crowdfunding campaign fronted by one of its original engineers. It’s been renamed the Popcorn, and it comes in three variants. The Original Popcorn is a compatible C.H.I.P. by any other name, while the Super Popcorn is a much higher-spec machine that comes in quad and octacore variants with AmiLogic SoCs. All three have 32 GB eMMC on board, and the specs are suitably impressive but not out of the ordinary for a 2019 single board computer. Prices are $49, $69, and $89, which takes away that optimistic $9 price tag that made the original so attractive. There is no Pocket C.H.I.P. which is a shame because for us that was the only reason to buy a C.H.I.P, but there is a companion board called the Stovetop that provides Raspberry Pi-style desktop and display interfaces.
We wish them well, but it’s difficult to escape the conclusion that the hardware world has moved on and the window of opportunity has closed. It’s not that these boards are not good ones, more that they now join a plethora of others which come a lot closer to the low price of the original. Still, there remains a C.H.I.P. community still out there, so perhaps that will save the day for them.
We interviewed the C.H.I.P.’s creators back in 2015, and marked its passing last year.
Thanks [Rose] for the tip.
I feel its always a bit uncharitable to just compare the C.H.I.P. to the Raspberry Pi in price and power, because the C.H.I.P. came with some features the Raspberry Pi lacked, like included LiPo charging, onboard 4GB NAND storage and dual channel wifi, just at the top of my head. $9 gets you a adafruit powerboost, which nearly every pi projects seems to use. Not to mention an additional SD card. That’s not comparable to the Pi Zero still today. So I do think, the value proposition would still hold true.
Of course, as we have seen, features are not all in the road to success and at $48, I don’t see why anyone would want this.
One may say that it’s a bit uncharitable to just compare business which grows with business which went bankrupt. Still I’m an advocate of having a choice, thus I feel sad that C.H.I.P. is dead.
You’re right, the CHIP was a way better piece of hardware than the Pi Zero. But it’s as much a question of the market as it is the spec, in that in a tiny Linux SBC most people care little for the details of its spec so long as it is cheap and does what they want.
I have a feeling I’ve written elsewhere about how the Pi boards are never the best ones spec wise but represent the best choice because they have the Raspbian ecosystem and support.
Part of the problem is that sites like Hackaday always told people the Raspberry Pi was the option that would do what they wanted, regardless of whether it would. I remember in particular one article mocking the C.H.I.P for being $5 more expensive than the Zero whilst calling a full-size Pi competitor DOA because the Pi 3 had WiFi – thing is, onboard WiFi mattered less on the full size boards because they had Ethernet and plenty of USB ports for dongles but was a major pain on the Zero. HaD played up this missing feature when it barely mattered but helped promote the Raspberry Pi, whilst ignoring it as a downside of the Pi Zero even though it was a real headache when trying to use it for almost anything.
I refer again to my final point. The Pi rarely has the better hardware. But you’re not buying the hardware, instead you’re buying the ecosystem. I’m sure plenty of people have been burned by boards with awful buggy Linux distros and no support or community.
my experiance with the pi leaves me wondering how it got this reputation, its always got buggy distro’s and horridly out of date support, with the usual linux tude
I have never once had a simple pi project, its always a mish mash of bad outdated obsolete information broken software and a constantly moving target of an OS
In early 2016, I bought two ZTE Speed smartphones for $10 each. They had 1GB RAM, 8GB Flash, 64 bit quad core CPU, and pretty horrible software support. But they worked great for Perk mining, exactly why I bought them.
Considering all the things even the cheap smartphone has that the C.H.I.P. doesn’t (battery, LCD, case, etc.), $9 can definitely be done if the volume could be ramped up enough to get production costs down. Pretty incredible how cheap modern technology can be.
Unfortunately the “and the software taketh away” issue seems to be pretty much universal(once you get beyond parts too simple to have software or which don’t speak some well established USB device class, and even some of those are quirky as hell).
I realize that the vendor who does minimum effort drivers is going to have a cost advantage over the seller of similar hardware who actually cares; but it’s too bad that there hasn’t been more interest from, say, ARM in making investments in the ability of their ecosystem to run something other than awful effectively proprietary Android BSPs.
(Compare to, say, Intel and AMD, who invest nontrivial effort in keeping Linux on x86 playing nicely because having it as an alternative to the classic Unixes is part of that buried most of them, and their CPUs(POWER being close to the sole survivor).
Online flashing tools…
That’s a big fat NOPE for me.
Online toolset + track record including insolvency = bad news
They burned us once, not willing to get burned again
I got two of them and the keyboard thing. Its pretty good.
I had the pocketCHIP, I managed to sell it just in time, but I’m not falling for this again.
I bought a Pocket Chip and if it was open source, why can’t he still support it? The online support last time was dismal so how is he going to do a better job by himself? Why can’t the community still get chip? I asked last time for support and didn’t get it so why is the same person going to be different?
This product makes you dependent on the online flashing tools.
Unless they give away the files to make the circuit board and components to build it yourself along with the source code and online flashing tool, you are still dependent on them in some way.
Apparently they didn’t make enough money on a $9 chip to make it worth their while. How can you put a price on a relationship with the customer? Chip ditched the customer. Try making it right with me, tell me how you are going to do it, what steps you will take and when because that was your relationship with me the customer.
Simple. Charge a price that more closely reflects cost instead of appealing to people’s “something for nothing” nature. e.g. Google.
Google’s really not a good example there… Google delivers what it says it will.
So did NTC. You got a wifi-enabled, LiPo charging SBC with an unreal price of $9.
Financial strategy aside, that is what they did deliver on. I couldn’t imagine how they made that work financially, and apparently neither could they and now they’re gone.
God damn people are salty over $9 losses.
My guess is NTC hoped to make up for the initial lo$$ with owners continually renewing $ub$cription$.
I bought two handhelds which were more than $9 dollars. I bought the chip, audio cables, and there were vga and hdmi versions.
Some of the kickstarter buying options were: $9, $19, $24, $49, $64, $93,$ 98 and $150 and there were accessories.
I tried to flash it using the instructions and it did not work. I asked for help online and the help did not work.
It is not because it was a $9 product that I won’t do business with them anymore. It is a rule of business that we don’t extend credit to whom credit is due because if the manufacturer can’t afford to pay their bills (in this case documentation and help) then extending credit to whom credit is due will just put us deeper into debt with someone who did not pay all of their bills. The whole point is we are consumers and some of us have aspirations to do things greater with the products because we want to be makers and not just consumers so we have dreams and desire so being let down is a real downer because the makers ruin our dreams, take our time which is more than money and ruin our planning.
My college ordered two side cars and equipment from Commodore Amiga and the lab manager knew Commodore was going out of business because Commodore accepted payment and many months later never delivered. If you don’t have the money then you don’t have the resources to properly develop, support and market the product.
Chip was great. It came wrapped in a box but wrapping does not help me with support.
It’s not just the money we spent but as customers we have expectations. The previous product was great but the support did not meet expectations. In order to have repeat sales, the product has to sell itself and the previous support did not meet expectations.
Instead of blaming myself for a loss, I’m going to manage my time better and move on to something else.
No, not all the people are salty about $9.
Some are salty at the countless hours spent to configure the thing for a specific application, just to find themselves with a brick in their hands.
Because of the crappy quality flash chip used. And because of the crappy web tools, which promptly disappeared just like the company.
Now they say “pledge $9 for access to the new web-flasher”. No thanks.
Not even the decency to allow the same people they dumped for a free web update.
I realize they’re not fully related/associated to NTC, but still, this is just too much.
Not going to invest money in another “hipster’s dream”, as somebody said, unless Kickstarter steps up, and makes archival of all tools and documentation mandatory for all participants.
$49, $69, and $89. Ah, I see. Named after the price of small, medium, and large popcorn at the movies.
Hmm, when you buy the large bucket of popcorn, you get a free refill…
Ha ha, exactly.
“We wish them well, but it’s difficult to escape the conclusion that the hardware world has moved on and the window of opportunity has closed”
Moved on to what exactly? CHIP was not significantly different from the wide variety of Allwinner-based Orange Pi or Banana Pi devices, which are still popular. There’s not been any significant technical advancements in the last 5 or so years.
Opportunity is strong for *LOW-COST* ARM devices. If they can get the price lower ($50 SBC have been around for awhile, but there’s little interest.
The original CHIP was interesting, but they had very high costs of shipping to other countries. I would have bought one/some if the cost was something like US$ 3 or 4 for shipping, but it was more in the lines of $12-$15, making the whole thing unsuitable.
And then they removed the first model, which was the best for my intentions ( a lot of pins brogut out ), and replaced it with the other models, with less features. No more advantages in them from that moment on.
If he could make it expansible, but not populate all the optional features, and had a model better priced ( say, $20 including shipping ) , it would be much more interesting. If the board has the necessary pads, people that want can solder an ethernet jack. Those that don´t need it or don´t want the trouble, just buy a model with it already populated.
If you are looking for the lowest prices, best features, and exhaustive support from kickstarter campaigns by not-established companies you simply are looking in the wrong place. The original CHIP campaign is emblematic of that.
I agree with others that this product is a wash, though. Is it compatible with the original PocketCHIP? Doesn’t seem so…
Not looking for the lowest price/best features/support(?). But if they had chose to be honest, advertising it as a “$25 shipped” sbc, not a “$9 plus shipping” sbc, I would have more consideration for them. The way they did it looks simply another kickstarter scam, hiding the difference in cost in the obligatory shipping prices.
The cost was not $3-4 shipping to the US. I looked back at my CHIP kickstarter and I paid $10. That is not unreasonable for shipping from China. I don’t know why you would choose the one thing that is nearly impossible to find wiggle room on to get upset about.
I would have bought some CHIPs as well, in fact on multiple occasions I’d stocked up the ‘shopping basket’ and seen that the cost of shipping was in the order of $20, it just increased the cost out of the ‘expensive but nifty toy’ range, especially as I’d have to pay import tax on arrival into the country as well.
What use is 100% open-source Hardware if it uses a 100% closed source web based toolchain?
Wait, derp, learn to read noob:
“If you prefer not to or can not use a web-based tool, we are planning to launch a cross-platform compatible off-line tool for flashing, back-up and cloning. ”
So ignore my comment
I respect a person that can own up to a mistake and correct themselves. Cheers!
Ditto!
Not so fast. “we are planning to” is Kickstarter-speak for “we’ve noticed that people don’t like something, so we’ll say we’re working on that.”
This is very much like the MANY projects whose developers say “we’ll release the CAD files/source code/API/whatever once we’re in production,” but somehow it never happens.
Or in the more general case, “the check’s in the mail.”
Yesterday I searched for a board in a box and found 4 of these low cost computer boards (the router based kind similar to CHIP) and though to myself they were a waste of money. I never ended up using them for real, just for playing. Somehow the low cost routers always won: nicer box with plenty of space to add components, free power supply, available NOW.
I imagine the break-out is better on C.H.I.P.
yes, maybe. But if the only connection you need is the serial port, then there is no advantage.
Usually I need anything but COM or USB.
But I got a few rPi zeroes back when supply were momentarily higher than demand.
They make for handy stand-alone interfaces for i2c and spi.
R8 based C.H.I.P or C.H.I.P Pro clones look interesting, but not at that price point.
AmLogic also has questionable mainline linux support, so i’d be wary of that.
Voder or dashbot took my money and ran. I can’t trust anything from Kickstarter anymore.
From their own words….
“We’ve spent the last 18 months building and strengthening our supply chain to be able to handle the production of a product like this. Our partners are committed to helping out. Also, this is not our first rodeo. We’ve had two successful Kickstarter campaigns before this. We know what it takes and we will deliver!
<3
all of us at NTC
Learn about accountability on Kickstarter"
Just posted about that but couldn’t remember name. Voder was the thing I backed. Never again C.H.I.P, never again.
A lot of makers came out with computers but went broke on support because they were not sophisticated enough to know what to charge for support or know what kind or how much support customers would need…
Remember the Commodore Reference Guide? Who puts out manuals anymore? They are almost all online or in electronic format.
Ever hear of Oric or Tangerine Computers? They produced more computers and don’t exist anymore.
https://en.wikipedia.org/wiki/Tangerine_Computer_Systems
Byte magazine doesn’t exist anymore but it was for makers who desired to build their own computers and everyone failed except for the computer models that are still sold today. Most of the Computer wars were over in the 80’s and 90’s.
When choosing a microcontroller, do you want a flash pan experience with the newest and greatest or do you want a chip or platform with some staying power that might be around in the next 10, 15 or 20 years?
If chip matters then they need to product quantity of support and if it doesn’t matter that I wasted more than $9 dollars and chip doesn’t matter.
Oh no. Not again. this stupid named computer burned me twice already. not going for a third time (firts with eth availability of the chip, then with the availability of the chip pro. I’ve settled on the FriendlyArm Nanopi Neo air. this one has usb on pin headers, so using it as a computer on a bigger board is no cable hassle with usb pigtails or pogopins like on the chip or raspberry flavored things.
If it weren’t for their proclivity to run on fucking FIRE, I’d use more OrangePi Zeros if I wanted a sub-$10 Wifi enabled SBC.
The PocketCHIP was the real star. The fact that it has a tactile keyboard, 4:3 screen, and exposed GPIO makes it perfect as a benchtop terminal. I’ve yet to see anything approaching it.
I grabbed a few of the boards and flashed them while the website still existed as backups. If this thing actually comes out (no way I’m backing the kickstarter), I’d pick up a couple.
I bought a CHIP and I loved it at $9. For me, it was at a sweet spot, more capable than a Pi Zero W at $10 but less than a Pi model B at $40, all in a compact size. I think it could still be viable at $20 but not at $49.
Not sure it was the CHIP folks or some spinoff but I had kickstarted this device that looked like KITT with lights that sat on your dash and would give you direction arrows and take commands like SIRI. I backed at $49 and that was the first thing I every kickstarted or crowd supplied that just failed. No update, nothing. I’m not supporting anything C.H.I.P related again.
The C.H.I.P. timeframe is close. Many other boards are on the market and play quite good. C.H.I.P. can rest in peace with many others dreams of the past.
I got one of the CHIPs because of the low price. At this new price, you can get a Raspberry Pi or even a Libre for way less!
I found a better and more powerful board on Kickstarter to tinker with:
“The UDOO BOLT is a quantum leap compared to current maker boards: a portable, breakthrough supercomputer that goes up to 3.6 GHz thanks to the brand-new AMD Ryzen™ Embedded V1000 SoC, a top-notch, multicore CPU with a mobile GPU on par with GTX 950M and an integrated Arduino™-compatible platform, all wrapped into one.” And it works with Grove kits from Seeedstudio.
https://www.kickstarter.com/projects/udoo/udoo-bolt-raising-the-maker-world-to-the-next-leve
http://linuxgizmos.com/udoo-bolt-is-first-ryzen-v1000-based-hacker-board/
You can talk about the successor to Chip being open source all you want but unless you get the “Bill of Materials, schematics, 3D design, binaries and other resources” like the UDOO BOLT gives you, the design is not 100% open source.
The other thing I like about the UDOO BOLT is the creator(s) did more work.
I found it after finding an article called:
Introduction to catalog of 125 Linux hacker boards
http://linuxgizmos.com/introduction-to-catalog-of-125-linux-hacker-boards/
In other words, shop around. Don’t buy the first board that you see like Pocket Chip or Chip. See what other single board computers have to offer. Just because you saw the successor to Chip doesn’t mean you are stuck with only one board. Shop around.
These boards don’t do anything for new users unless there is enough documentation to get started because there is a learning curve and not everyone gets it or is experienced enough because there are several disciplines to master which means that non-existent support is a “no no”. If you are a beginner, you might want to start with Arduino and leave the others alone.
The appeal of the C.H.I.P. was the $9 price point. Of course UDOO can put in more effort into making a polished product with all that extra room for margin. You could get 25 C.H.I.P.s and a cup of coffee for the cost of the entry level UDOO BOLT.
Udoo Neo Basic is a $50 board so it is in the price range of the Popcorn board.
The C.H.I.P was not the greatest SBC, but it cost only $9 ! This was its saving grace.
These Popcorn SBCs are not low cost at $49-$89. If I wanted to buy an ARM-based SBC at this price, I’d consider the new ODROID N-2, Jetson Nano or the new Beaglebone AI. This popcorn SBC gets a hard pass
Or many of the ST Micro developer boards…
Since the Pocket C.H.I.P was a Single Board Computer (SBC) with a Microprocessor (MPU) that runs Linux, I was only considering other SBCs capable of running Linux at the $50-$100 price range. Not Microcontrollers/ Microcontroller dev boards, which is what ST Micro seems to be most successful at making and selling.
ST Micro did release the Cortex-A7 based STM32MP1 family of MPUs (and boards) earlier this year. Unfortunately I’ve yet to see any reviews or projects based on these parts/boards.
The two SBCs capable of running Linux and are comparable to the pocket C.H.I.P in price and overall specs are the Raspberry Pi Zero W and the Onion Omega 2+. There’s also a couple of low cost AllWinner based nanopi SBCs. But I’d avoid any SBC based on AllWinner SOCs.
The website for Popcorn computers is hosted on another site called Medium which has paid access to articles.
My employers won’t put anything on the web that represents them unless they pay a premium so I question the longevity of this venture especially after seeing how the last one failed. In other words, I want to see them invest in longevity because they could easily pick up and leave again.