Sometimes the end of a product’s production run is surrounded by publicity, a mix of a party atmosphere celebrating its impact either good or bad, and perhaps a tinge of regret at its passing. Think of the last rear-engined Volkswagens rolling off their South American production lines for an example.
Then again, there are the products that die with a whimper, their passing marked only by a barely visible press release in an obscure corner of the Internet. Such as this week’s discontinuances from Intel, in a series of PDFs lodged on a document management server announcing the end of their Galileo (PDF), Joule (PDF), and Edison (PDF) lines. The documents in turn set out a timetable for each of the boards, for now they are still available but the last will have shipped by the end of 2017.
It’s important to remember that this does not mark the end of the semiconductor giant’s forray into the world of IoT development boards, there is no announcement of the demise of their Curie chip, as found in the Arduino 101. But it does mark an ignominious end to their efforts over the past few years in bringing the full power of their x86 platforms to this particular market, the Curie is an extremely limited device in comparison to those being discontinued.
Will the departure of these products affect our community, other than those who have already invested in them? It’s true to say that they haven’t made the impression Intel might have hoped, over the years only a sprinkling of projects featuring them have come our way compared to the flood featuring an Arduino or a Raspberry Pi. They do seem to have found a niche though where there is a necessity for raw computing power rather than a simple microcontroller, so perhaps some of the legion of similarly powerful ARM boards will plug that gap.
So where did Intel get it wrong, how did what were on the face of it such promising products fizzle out in such a disappointing manner? Was the software support not up to scratch, were they too difficult to code for, or were they simply not competitively priced in a world of dirt-cheap boards from China? As always, the comments are open.
Header image: Mwilde2 [CC BY-SA 4.0].
Between this and Intel dropping mobile Atom, I expect Intel may never get design wins outside PCs every again. At least AMD has shown they’re willing to do semi custom SOCs and support them. They may never get back to winning performance in PC but maybe they can cross some of those bridges they haven’t burned into future markets.
Aside from the lack of documentation and poor quality of what was available, what stuck out to me as hurting Edison was that it didn’t have any built in video output, and none of the common ways of connecting any video hardware to it.
Apparently Intel envisioned Edison as a ‘black box’ to which the user would Telnet and issue instructions – an embedded processing block controlling other stuff.
The RasPi and all the other FruitPi knockoffs, the Beagles and so many other little SBCs – they have some kind of easy to use built in video. They also are easy to interface hardware to, using common and dirt cheap connectors possible to hand solder with a cheap pencil iron.
If Intel had included both of those with Edison, even if the connection was just a dumb adapter to a bare pin header, then it might have gained a market. Two versions would have been better, one just like what they shipped, and one with built in video. Combined with the connection header adapter, prototyping and experimenting would have been a huge amount easier.
Then for the final product, specify the tiny connector to plug in the video-free Edison module, if your product didn’t need video or would make it from some other part of the electronics.
It’s amazing that any company making a product with the intention of generating mass appeal and adoption of it goes to great pains to make it *unacceptable* by the very people they want to buy it. It’s as though they’ve never heard of the concept of making a shitton of money by selling huge quantities at a small profit margin.
Just look at pizza and pepperoni, or any other product class that has one option that sells far more than any other. Is the price kept low (or it is nearly always on a special sale) because it’s so popular? Nope! Exactly the opposite. It sells well because it’s cheap. If the pizza industry swapped pricing and marketing strategies on ‘Canadian’ Bacon and pepperoni, then ‘Canadian’ Bacon would soon become the most popular pizza topping.
I cal this 3P or Pepperoni Pizza Principle. ;) Price + Promotion = Performance. You have to price the product right, you have to tell people (constantly) that the product exists. Do both properly and it will have good market performance – if your product isn’t total crap.
It doesn’t matter how right-priced and how good your product is if you fall down on promoting it. Witness the 2004-2006 Pontiac GTO. General Motors marketing ignored it almost totally. Pontiac introduced an all new model line for 2004, sent out large numbers of fancy booklets titled “Meet the new Pontiacs”. Absent from that was any mention of the (built in Australia by Holden) V8 rear drive GTO, but the similar sized (front wheel drive and American built) G6 was prominently featured.
TV shows and movies get the same treatment. Success of visual entertainment is closely tied to how much effort marketing is willing to put behind a property to push it. It usually works if the show isn’t an absolute stinker that nobody likes. We end up with a lot of pepperoni shows and movies while better ones don’t get the viewers they really deserve.
Its the payroll issue…
When you look at the millionaire salaries senior Intel people get, they have to price their products through the roof. Who was going to pay $ 80 to $ 200 for a computer board, when R-PI Zero sells for $ 5 to $ 10.
The innovators are those who work harder than anyone else, for far less pay, but lots of love and partnership with a great community. Intel lost that 30+ years ago. Now they milk the cow at the expense of shareholders.
IBM is going the same way…
One of my customer insisted on using the Edison in its product (not successfully released yet…), so I had to stick with it. If it was for a personal project, I would have just tossed it away.
The wifi quality was really terrible. Both the Rapsberry pi 3 and a super cheap A5-V11 had better download speeds. And the ping was bad too. But the icing on the cake was really the fact that you couldn’t use certain GPIO at the same time as the wifi, of course Intel wouldn’t tell you, they would let you discover it.
Concerning the support, all the comments here show how bad it was. One more example : There was no clear documentation on how to upgrade the board by pushing the Firmware (FW) button on the board.
Those guys spend time adding a dedicated button on the board, labeled it ‘FW’, yet it was not possible to flash the firmware by pressing it.
In the past I’ve developed small add-on board for Edison (25×35 mm same size as Edison), which had USB-Serial converter, IMU, Barometer, LiPo charger and power regulator. Even though the connector was pain to deal with, it was pretty functional w/ on-baord BT/WiFi, Flash memory, running embedded Linux. HW-wise it was a nice little board, SW and documentation simply wasn’t there. It would’ve been much more successful if it was open-source. I once burned the PMIC chip, but the TI part was not an off-the-shelf component and datasheet wasn’t available, due to the NDA reasons. Anyway, R.I.P Edison Board.
[img]http://i.imgur.com/he2taz0.jpg[/img]
Nice little package, battery on bottom , then Edison an your board on top. Now why intel could not wrap it like that, in the first place? As mentioned above it lacks a display connection…
By the way what is battery capacity an how long does it run the board (assembly)…
To view the pix above I had to take out the first i an the dot …
later
Sorry for the wrongly formatted imgur link (http://i.imgur.com/he2taz0.jpg).
Battery was 400 mAh, and it could run continuously for two hours, while logging and streaming IMU data over WiFi.
Next week we will all see the news that Intel has discontinued working on driverless cars and drone swarms as well.
Intel proves it is still running scared like a lost child.
Good riddance, at least to Galileo. I got one for free when I was in grad school and tried to mess around with it. I was using Arduino a lot at the time, and the idea of a pin-compatible board with much greater computing power would be great.
My first plan was to load it with a sketch to drive a TFT touch screen since the Uno wasn’t fast enough to draw the graphics I needed. Turns out Galileo didn’t support SPI. I also spent a solid 3 hours trying to get Intel’s awful little toolchain to actually compile because, apparently, the latest version of their libraries wasn’t compatible with my specific unit and there was documentation on anything.
Well okay, I’ll wire up my little robot’s H-bridge and… no PWM because the Arduino emulator running on the internal Linux OS is too slow.
Surely I can use the on-board Ethernet adapter to… not supported. So why is it there?
Okay, I’ll plug in this mini-PCIe wireless card so that I can… of course the bare-bones Linux install has no drivers for it. Is it even usable?
The Galileo was mostly compatible with Arduino, except in all the ways that Arduino is useful.
The Galileo was a powerful single-board computer, except in all the ways that a computer is useful.
A miserable failure of a development board design, as far as I can tell. I never could figure out what the target application was.
Arduino and Raspberry PI won because of function vs price, period. The Intel x86 bloodline should have blown them away, BUT their overblown price just didn’t make it for the controller market. The top end of the market was 35.00 and when the esp8266 came out not even the network became an obstacle for the Arduino boards. Plain and simple, they read the IoT hobby market wrong.
Wow, they were just flogging their Joule at the last Maker Faire.
Take a look at Digi’s ConnectCore 6UL. May be a valuable alternative… https://www.digi.com/products/embedded-systems/system-on-modules/connectcore-for-i-mx6ul
Well, Intel is pretty paranoid as most have heard. Jump on the wayback machine, and there I am attending a conference on “The Paperless Office” down South in San Diego. (I think is was S.A.I.C. hosting) Anyway, a full crew of Intel engineers were attending as well. During a lunch break, I asked one of the Intel guys did he have any source for cheap ram. I never mentioned the words “Beg” “Steal” “Borrow” “Loan” or anything like that..All I wanted was a source I could buy from.
Well, they completely freaked. You would have thought I had asked for keys to the main office.. Damn.. It sure got chilly there that afternoon.. Go figure..
If your Edison just sits on the shelf, please send me on the post ☺ I will use this powerful Edison for bicycle share system. I know a lot of case, how I can use Edison for my global task.
As a long time Intel guy, now thankfully happily retired, my opinion is that Intel has lost their way and simply can’t figure out what to do for their long term business model. Yes, this sounds strange, but it used to be so simple when we couldn’t get enough performance and the quest was clear. Now, performance is not the main issue as it’s really a balance between performance, power, cost, and usability. Intel will continue making the fastest processors (even if they are x86), but their forays and exits into mobile (cell phones), embedded, and now IoT are just emblematic of the confusion and mismatch that exists between their design, marketing, and technology arms.
They will never be an RPi replacement not because they can’t do the hardware, but rather because the finance guys will demand 90% margins and the sales guys will require volumes to make it worth their while. This would require a huge change in their company practices and culture that I can’t imagine them making in their current state.
That and their current CEO is a process donk who has little understanding of design or marketing and makes extremely poor hiring decisions gives little hope for what was once a great company under Andy Grove. The current guy in charge of all their design is a retread from Qualcomm and is a joke.
Hi, We were working on the project with Intel Curie but whatever happened recent days were a shocking news for us. anyway, we have to move ahead. Can anyone suggest me what will be the replacement of Intel Curie other than Qualcomm because Qualcomm charges money for even testing purpose. If anyone can suggest me some other device than these than that would be highly appreciated. Thanks
Had to stop by and see what the whine was about, my dog was barking… We did a project using an Soc from China that was actually missing the part (GPS) inside the Soc. Found out the hard way that Chinese Soc’s don’t support SPI standards, no slave.
Well this commentary has been a big heads up for me on there new development kit…
https://click.intel.com/intelr-realsensetm-robotic-development-kit-2438.html
I hope as a programmer I don’t have all the problems you hardware guys had.
I found this for $5 at my local thrift store, when I brought it home the preflashed windows iot failed to boot, I tried flashing sd nothing would work, and it keeps restarting over and over despite being in bios, I’m just starting so I keep having to question myself if I don’t know what I’m doing or this device is literally impossible, I cant reflash bios, I cant ssh/serial into it, nothing works!