Intel Launches 'Galileo,' an Arduino-Compatible Mini Computer 130
MojoKid writes "Although Intel is Chipzilla, the company can't help but extend its reach just a bit into the exciting and growing world of DIY makers and hobbyists. Intel announced its Galileo development board, a microcontroller that's compatible with Arduino software and uses the new Quark X1000 processor (400MHz, 32-bit, Pentium-class, single- core and thread) that Intel announced at the IDF 2013 keynote. The board makes use of Intel's architecture to make it easy to develop for Windows, Mac, and Linux, but it's also completely open hardware (PDF). Galileo is 10cm x 7cm (although ports protrude a bit beyond that), and there are four screw holes for secure mounting. Ports include 10/100 Ethernet, USB client/host ports, RS-232 UART and 3.5mm jack, mini PCIe slot (with USB 2.0 host support); other features include 8MB Legacy SPI Flash for firmware storage, 512KB embedded SRAM, 256MB DRAM, 11KB EEPROM programmed via the EEPROM library, and support for an additional 32GB of storage using a microSD card."
more the better (Score:5, Interesting)
Re: (Score:2, Flamebait)
Re: (Score:1)
Amen. Intel has far too much power on the computing landscape, much more than IBM ever had in the 70s and early 80s, and that's saying something, since I'm old enough to have lived through that period, and using computers at the time. Probably the worst fear at Intel is that AMD stops producing x86 compatible processors, they would have more power on the hardware landscape than Microsift ever had on the software, and could no more dispute that they are a monpoly and could no more resort to all the dirty tri
Re: (Score:3)
I generally stop reading anything when I encounter the word "hater" in it.
It's just so stereotyping a term to use. I hope a lot of us will disregard comments that use that word to describe others, too.
It's hateful and it shrieks cultishness. Just stop.
I generally stop reading when anything when I encounter the word "hateful" in it. I hope a lot of us will disregard comments that use that word. It immediately identifies the author as a hater. The text that surrounds it is generally useless except to the author.
Re: (Score:2)
Have you seen the maximum speed you can change an I/O pin at? All the pins are accessed via a multiplexer on the SPI bus - really slow.
Also the pins are very limited current capacity. 10mA max, less if you turn on several pins.
Re: (Score:2)
current capability is problem on any microcontrollers..
but.. is the multiplexer fast enough? how about interrupts?? what I am asking is if it can replace an arduino in any project??
Re: (Score:2)
current capability is problem on any microcontrollers..
True, but Arduino owners are used to having a lot more than 10mA.
what I am asking is if it can replace an arduino in any project??
No, not even close.
It's obviously aimed at completely different market than the Arduino Uno, et. al..
Sounds.... Expencive (Score:1)
Re:Sounds.... Expencive (Score:5, Interesting)
Though, on the downside (similarly not-yet-confirmed) reports are that the arrangement Intel uses to support the GPIO is pretty limited, compared to much cheaper parts that do GPIO closer to the metal, in terms of the speeds at which it can bit-bang the assorted oddball peripherals (those cheapie LED strands for instance) that many arduino projects end up bit-banging to communicate with. Having a real ethernet and SD interface, not SPI hacks, is nice; but if those reports are to be believed, your project had better be doable without extensive bitbang interfacing.
Re: (Score:3)
When they say single thread, they mean not hyper threading. It's an x86, it can context switch like any other x86.
Re:Sounds.... Expencive (Score:5, Informative)
The "someone" mentioning 230 Hz is INTEL, in their Galileo FAQ.
http://www.intel.com/support/galileo/faq.htm [intel.com]
The question is near the end, specifically "What is the maximum rate at which GPIO output pins can be updated?"
The answer, which you'll see if you click that link and expand the question to see the answer, is:
The GPIO output pins on Intel® Galileo are provided by an I2C Port Expander that is running at standard mode (100 kHz). Each I2C request to update a GPIO requires approximately 2ms. In addition to software overhead, this restricts the frequency achievable on the GPIO outputs to approximately 230 Hz.
Re: (Score:2)
Re: (Score:3)
Erm... an ENC28J60 module with RJ45 jack, magnetics, crystal, and all you need to connect an arduino (or any other MCU) is $3.50 on ebay. Less than the price of the ENC28J60 chip alone!.
Re:Sounds.... Expencive (Score:4, Interesting)
Intel didn’t announce pricing for Galileo,
Aaaaand I'm instantly not interested.
Seriously, you can throw as much hardware as you want at a problem, it's all just a matter of price. We could shove an iphone everywhere we want compact processing capabilities. (And god knows enough people actually do that).
Also, it really helps if it's open. The raspberry pi is neat because it's specifically useful as a full-fledged computer that DAMN cheap. It runs Linux so there's a lot of leeway with what you want to do with it. (Quickly, without having to develop your own RTOS and windows manager) But it IS questionable about what sort of long-term legs it has because the broadcom chip on it is very much closed. I don't care how awesome the hardware is if I can't even blink an LED without asking mother-may-I from some corporate whore.
Re: (Score:3)
Software types cling to their thick padding of abstraction. I like coding up from the reset vector.
I like not having to re-invent quicksort, atoi, hexToDec, etc etc etc every god-damn time. Or having to whip out the scope to find out what quirk this SPI implementation is doing. I like libraries where I know how they work, I can plop them down, and instantly have known capabilities. And no, you don't need big engines for everything. But if you need an Ethernet connection, even though the speed constraints are so lax that even a 8051 could handle it, it doesn't mean that you should piss away weeks re-inv
Why not single chip? (Score:3)
Looking at the picture of the PCB they used, first question that strikes me - why not simply make it a single chip ASIC? I counted at least 7 chips on board. It would seem that a single chip w/ all the functions, and connections running out to all the ports - PCIe, USB, Ethernet, SPI and so on would enable Intel to minimize on chip cost, and let the rest of the cost hinge on the peripheral interfaces.
If that would be too expensive, Intel could make things cheaper by going as far back to a Pentium I core
Re: (Score:3)
To me the CPU is meant to be used in embedded systems, where you don't necessarily need ethernet or USB or something else.
What's announced here is a low cost general purpose and development board.
The integrated 512K of special RAM means it can maybe be used without external memory chips. It's like having a PC that can boot DOS without memory DIMMs.
Re: (Score:2)
Re: (Score:2)
Can't say how much of the funtionality is onboard, but they claim that Quark is a SOC.
There's a note about compilers to the effect that it's 586.
I think part of the point is to provide a design that uses the Quark.
pricing? (Score:3)
Re: (Score:2)
Re:pricing? (Score:5, Informative)
Several articles have appeared claiming "under $60".
For for free if you're one of about 50000 students or apparently about 400 people who attended a talk at Maker Faire last weekend in Rome.
However, if you check out Intel FAQ, there are a number of Arduino compatibility caveats. Probably the main on is the I/O pins are controlled by an I/O expander with approx 2ms latency. That's pretty slow compared to Arduino's slow digitalWrite() function, which run about 4us on 16 MHz AVR, or direct AVR register access, which takes 125ns.
The processor runs Linux and Arduino sketches are compiled to native Linux userspace programs, so it probably will open up a lot of possibilities.
Re: (Score:2)
Inigo Montoya... (Score:5, Informative)
Re: (Score:1)
Re: (Score:2)
Re: (Score:1)
Re: (Score:2)
Re: (Score:1)
Yes, if you know how to use it. Signal and power on one wire pair and the wire pair can be really, really long if necessary. Not too shabby.
Re:Inigo Montoya... (Score:4, Informative)
Yeah... I was wondering how to hook up the teletype and the terminal servers.
Through the Ethernet port, using SSH protocol. Ultimately, anything with preemptive multitasking and virtual memory can fill the role of a minicomputer. An 8-bit MCU is a micro, but a VAX or i386-family PC is a mini.
Re: (Score:3, Interesting)
From Wikipedia: "In a 1970 survey, the New York Times suggested a consensus definition of a minicomputer as a machine costing less than 25 000 USD, with an input-output device such as a teleprinter and at least 4K words of memory, that is capable of running programs in a higher level language, such as Fortran or Basic."
It would seem this board meets the definition, as long as you connect it to some I/O device.
Mini computer?? (Score:2)
Re:Mini computer?? (Score:5, Funny)
Hey, everybody! An old man is talking!
Re: (Score:1)
I'm the operator with the pocket calculator.
I'm adding, and subtracting.
When I press this special key, it plays a little melody.
because 1985 (Score:4, Insightful)
The 1960s - 1970s minicomputer was gone by 1985. Thirty years later, there's no confusion and therefore no reason not to reuse the term.
Re: (Score:2)
To summarize, "History is bunk" [phrases.org.uk].
Usually spoken by someone who believes themselves immune to Santayana's Law: "Those who cannot remember the past are condemned to repeat it."
Re: (Score:2)
Not even close, man. The concept of the phrase "mini computer" has far more in common now with small form factor computers than that old device from the 60s with a similar name.
I don't think anyone could reasonably confuse the two and assume there is a relation.
Re: (Score:2)
Re: (Score:1)
Draw me a line plz (Score:2)
Lots of little boards (Score:2)
There are lots of little boards available. With reasonable CPUs and amounts of memory. Ardunos, with 2K or 8K of RAM, were just too limited.
On the other hand, having to run bloatware like Windows or Linux on an embedded board has its own headaches.
Re:Lots of little boards (Score:4, Interesting)
Re: (Score:2)
Re: (Score:1)
Re: (Score:3)
Re: (Score:1)
I like Linux on small devices like this, but I hate Busybox. Saving a couple of MB was a big deal back when you had 16MB of flash storage and that was it, but these days it's not unusual to have 16GB. Saving those few bytes on a version of Bash that barfs on a lot of common scripts is just dumb.
Then compile up bash and add it to your rootfs. That's what we do, for exactly the reasons you mentioned; I wanted to run more modern shell scripts andd busybox didn't support them. But that's ok, just because you start with busybox doesn't mean you have to stop there.
Re: (Score:2)
This has 8 MB flash.
(FYI, with Busybox 1.20.2, I've not found scripts that make it barf to be common.)
Re: (Score:2)
You don't have to run Linux on it. There are any number of OSs that can run on an x86.
2-4 MB for Linux is bloated? (Score:3)
I run a Linux appliance with 4MB of RAM as a VPN endpoint for my kvm, ipmi, pdu, etc. I don't consider 10 cents of RAM "bloated".
Sure for some things you don't need an operating system, but if having Linux saves five minutes of development time it may be worth the extra $5 of hardware.
Obviously if you plan to sell a million units of a particular design, omitting 10 cents worth of RAM from each saves you $100K. For hobbyists, 4MB of RAM to run Linux is very often worth it.
Re: (Score:2)
With my first Arduino project, I'm finding the Uno's 16mhz clock speed to be more limiting than 2k of RAM.. all I'm really doing is twiddling an RGB LED strip with generated values.
I'd personally be happy if there was an Arduino with a 64 mhz clock, but Atmel doesn't make such as chip as far as I know, the fastest AVR is 32mhz, and even that would be a huge improvement.
Re: (Score:2)
Re: (Score:2)
I am, but my sketch is populating and manipulating a 40x4 array of uint8_t values with a simple fire algorithm, it doesn't use many library calls. I've made a few optimizations, but I don't think I can get more than 30 loop() interations per second even in this preliminary stage on an Uno. I'll be able to achieve more than I originally planned by using a Tensy 3.0 instead.
Re:Lots of little boards (Score:4, Interesting)
Well, there is an Arduino with 84 MHz clock, called Arduino Due. It's 32 bit ARM, not 8 bit AVR. It sells for $49.
My little company makes an Arduino compatible board called Teensy 3.0, which is technically spec'd 48 MHz but overclocks to 96 MHz without any trouble. It sells for $19.
There are also other less compatible alternative boards, like ChipKit, Maple and Fubarino, with clocks speeds in the 50 to 80 MHz range, and attractive prices. Their compatibility isn't as good, which might be a factor if you're using libraries or code from websites. If you're wring all your project's code, that's less of a concern.
These boards also tend to have more RAM and other built-in resources.
I'm going to totally date myself here, but, (Score:1)
Wow, imagine a Beowolf Cluster of these!
Re: (Score:1)
Re: (Score:2)
I'm going to totally date myself here
It's okay. No-one will judge your life choices here.
The most important features (Score:3)
Ummmm, what about the most important features of the arduino: digital I/O pins, analog input and PWM output? It looks like there might be some in the picture, but the specs don't mention anything at all...
Re: (Score:3)
The datasheet, linked from this Slashdot article, shows a full-page diagram on page 3. On the left side are the usual 6 analog inputs. On the right side are the usual 14 digital pins, with 6 clearly indicated as PWM capable.
On page 4, it says:
14 digital input/output pins, of which 6 can be used as Pulse Width Modulation (PWM) outputs;
o Each of the 14 digital pins on Galileo can be used as an input or output, using pinMode(),
Re: (Score:2)
Look at the post by pjrc above your original one, smartass. Most of what you've said is outright wrong, which implies you didn't even look at the datasheet before going off on your own little rant about what it does or doesn't support.
Is the JTAG Port Open? (Score:2)
Are the specifications for communication protocol over the JTAG port open? Will projects like OpenOCD (http://openocd.sourceforge.net/) have enough information to support this?
Sure they might have used a standard connector. But the devil is in the details.
OCD (Score:3)
Will projects like OpenOCD
How many times do you have to wash your hands and open and close the door [wikipedia.org] before you can use an Open On-Chip Debugger [sourceforge.net]?
Possibilities (Score:3)
Re: (Score:2)
Open? Hardly (Score:1)
Re:Open? Hardly (Score:5, Insightful)
And try to make an Arduino without buying an Atmel microcontroller.
Re: (Score:2)
FreeDOS? (Score:3)
Um... 1971 wants you to know Intel=Hobbist (Score:3)
http://en.wikipedia.org/wiki/Intel_4004 [wikipedia.org]
The 4004 gave rise to the z80, the 8008, 8080, and 8086 chips that before the IBM PC came along were mainstays in the hobbyist community. It was all hobbyist, all the time back then, and heady days. So wouldn't it be fairer to say that Intel is going back to its roots rather than "reaching just a bit" in the DIY and hobbyist arena?
Re: (Score:2)
Arduino Tre is a better board (Score:5, Informative)
Sorry for Intel, but the just announced Arduino Tre is far better from any point of views.
http://blog.arduino.cc/2013/10/03/a-sneak-preview-of-arduino-tre/?utm_source=Arduino+World&utm_campaign=9f14cc4ca3-MakerFaire_World_201310_2_2013&utm_medium=email&utm_term=0_69a7d1abe4-9f14cc4ca3-76843037 [arduino.cc]
* Run faster than the Intel solution: An Atom core yield the same code execution speed as an Cortex-A8 core at the same frequency, so 1GHz A8 will easily catch on a 0.4GHz ia32).
* Cheaper and simpler to design on a custom board: just look at the chip package and at the PCB routing...
* Simpler power supply design, again just look at the schematics and at the PCB.
* HDMI output.
* More I/O, and all are integrated directly into the two CPUs, not using peripheral chips with low bandwidth.
* Already supported by larges communities, for the two processors.
Intel is just trying to enter a new market with a big buzz, but there actual solution still far away from the concurrent solutions. There just don't understand that in the embedded market nobody is bounded to the ia32 instructions set. Integration is the key and there Quark X1000 don't bring anything new on the table.
Re: (Score:2)
The Arduino Tre looks interesting, but it basically is an Arduino Uno bolted on top of a Raspberry Pi, while the Intel Galileo is a Raspberry Pi (sans HDMI) emulating an Arduino Uno.
I think neither will be much of a success because they will be too expensive due to the cruft they carry around to ensure a compatibility that is IMHO not needed.
Re: (Score:2)
The AM3359 is a far more flexible chip than the BCM2835 of the Raspberry Pi, just look at his datasheet: http://www.ti.com/lit/ds/symlink/am3359.pdf [ti.com]. And at USD 23, the AM3359 is a very good deal.
Re: (Score:2)
Re: (Score:2)
I do embedded systems since early 1990 and while I have see some machines running DOS at this time, I never see for at least the full last decade a new design based on DOS-like OS. And a simple Linux driver allow direct access to the hardware on any platform probably even more easily as you can do on a DOS-like OS because Linux provides already a lot of services to make drivers as simple as possible, witch is really not the case of a DOS-like where you basically have to write everything by yourself. Now if
Re: (Score:3)
Which all completely misses the point of the Arduino. It was never about performance, it was about making things easy, and having a community build around a common platform. The people who use it just want to write some logic that glues libraries and shields together.
There are many, many better options if you need more power or flexibility. All these spin-off devices are fine but always need more knowledge to understand and use. Traditional Arduino users aren't at that level, and people who are don't need t
Re: (Score:3)
Yes Arduino is about making things easy and the hug success of the project prove that it fit this goal very well. Now there is no reason to not bring this feature to more powerful platforms. Actually, this is still not as easy to code basic hardware interaction on Linux. There is no a standard and simple API to do on Linux what you can actually do on a Adruino. There is project like Comedi, but there are not focused on SoC. There is almost a different IO API for every SoC on Linux, even it there is ongoing
Why did Intel made this Arduino compatible? (Score:5, Interesting)
While I like the idea of having an Arduino compatible board running Linux to do some more advanced projects, I don't understand what drove Intel to force this board to be Arduino compatible. The Quark processor is not designed for this sort of stuff as it has neither a sufficient number of GPIO pins nor any ADCs. It sure has a lot of interfaces (SPI, I2C, PCI-E, SD-Card, Serial etc.), but it lacks the things that are useful for a hacker project.
So they had to include a separate GPIO extender chip (over a slow I2C interconnect) as well as an separate ADC. The Quark SoC has some 15 GPIO Pins, the extender another 40. But of those 55 Pins only 20 GPIO pins are actually available on the Arduino shield pins -- the rest is used for all the Muxes to switch pins between the ADC, the GPIO Extender and the Quark SoC to emulate the flexibility of the Arduino AVR processor.
While I haven't looked at the actual PCB schematic, I think the board layout is also strange. The ADC is on the opposite side from the analog input pins, meaning that all analog signals have to travel a long distance in the vicinity of some high speed digital signals. And the GPIO Extender Chip is on the opposite corner from all the digital output pins.
This, together with the BGA devices (SoC, RAM), seem to indicate that this is at least an 6 layer board which will make it hard to clone this design -- something that IMHO has contributed to the success of the Arduino. The Schematic [intel.com] for this board has 27 pages compared to the single page of the Arduino Uno [arduino.cc]
It seems that this Board is designed more as a competitor to the Raspberry Pi than to the Arduino, both in price and in features.The Arduino compatibility is just some marketing thing which makes the board overly complex and more expensive than it needs to be.
But hey, it sure must be fun to employ a few million transistors and a full blown operating system to run the Arduino Blink demo :-)
Hahaha, it's not GPL. (Score:2)
analysis of the Quark and Galileo (Score:4, Interesting)
i did an analysis of the Quark X1000 based on the Galileo schematics, and the assessment isn't good:
http://lists.phcomp.co.uk/pipermail/arm-netbook/2013-October/008979.html [phcomp.co.uk]
the key failure is that there's absolutely no I/O multiplexing. given that intel actually designed the PXA series of ARM processors before selling them to marvell you have to wonder what was going through the minds of the engineers behind the Quark X1000.
the main points of the above link which automatically and very unfortunately make the Quark X1000 a complete failure are:
1) there's no video outputs, and the only options are USB2 (DisplayLink with no 3D capabilities and too slow to do video), SPI (for character-based LCDs) or PCIe. to match a 0.4 watt processor with a 20 watt 3D PCIe Graphics card is completely insane. there are therefore no good options for video display of *any* kind.
2) there's no "industrial" or "embedded" style GPIO. no CAN bus, no PWM, no ADC, no DAC. there's also no audio. there's not even I2S and there's certainly no SPDIF. so to make up for that lack you'd have to add something like a Cortex M0, M3 or M4 embedded controller... and given that those usually come with built-in Power Management, NAND Flash and SDRAM, for the majority of purposes where you'd need to use an embedded controller with a Quark as a GPIO expander you'd be better off, cost-wise, with... just the embedded controller.
overall then there really aren't *any* markets that this chip could be useful for. if i'm wrong about that, and anyone can actually think of good uses for it, please do speak up.
Re: (Score:1)
Re: (Score:2)
Re: (Score:2)
They're talking here about the BIOS flash that goes into the motherboard. On an average, it used to be 4Mb, so 8MB is rather high for firmware. Unless they are planning to contain more than that - say the OS kernel, such as FreeDOS, Linux, Minix & so on. The interfaces too have evolved over time - at one time, it used to be 2Mb parallel flash, then it went to an Intel standard called Firmware Hub, a market that Intel exited as margins eroded, and now it uses SPI.
I think that a good model is that
Re: (Score:2)
8MB is considerably bigger than the BIOS chip on my 3GHz dual core PC, maybe 32x bigger (I don't know exactly what chip it is. I think the BIOS image is well under 100KB anyway).
Re: (Score:2)
The only one of those missing from my old $85 Intel dual-core Atom mini-ITX board is the flash.
Re: (Score:2)
Missing are free 3D drivers, too. In fact, on a dual core Atom with PowerVR GPU and ubuntu 13.04, basic xv video output wasn't working properly so you barely have a 2D driver.
Re: (Score:2)
This is the original dual-core Atom, which did have 3D drivers for Linux. I presume this new chip will too, to whatever extend it can do graphics.
Re: (Score:2)
What you're looking for is 22nm Atom, or maybe a board with a 9W Kabini. Or for a small board thing, the Minnowboard.. no 3D graphics driver (it's got a PowerVR GMA 600) and 1GB ddr2. You'll have to wait for a Minnowboard 2.