Well, not everything for Linux is open source. Drivers and such are one example that springs to mind. If you have some specific gizmo in the mini-PCIe slot or attached via USB, you might need a binary blob to run it.
The 3A thing is a bit of a downer considering the relatively slow speed of the CPU. However, it might be required for the PCIe slot rather than the board itself. Anyone see current draw specs anywhere? (I found some links for more info on Intel's website but they weren't clickable. Nor were they text that could be selected, at least in Chrome. Sheesh. Too hard, I give up. XD)
Grumpy_Mike:
It is the Arduino IDE that is full of restrictions saying you must not disassemble it which is stupid given that they said the source code was going to be available. I think it is just big corporation non thinking that keeps the restrictions on, but you have to click to say you agree before you can down load it.
Asinine. Why is Intel licensing Arduino's IDE anyway? Or did Intel re-write their own? Either way, that sounds like a good way to end up with a dead product in a year's time, when whatever project manager at Intel whose pet this is decides he's too busy or has better things to do.
@Grumpy_Mike
What software was this? Boot-code? A BSP? I would like to know if the software can be gotten around... How can Galileo be "Open Source" if there's a SWLA? For those who might not know, by BSP, I mean board support package, which can include an OS, boot code, and necessary drivers. What's the point of partnering with Intel if it's not really open source? Sounds like a "strategic marketing decision" by Intel, to me. It will be interesting to see if this takes off in any way, shape, form, or "market segment." :~
Inthe10percent: @Grumpy_Mike
What software was this? Boot-code? A BSP?
It was the whole arduino IDE which I assume includes the compiler and any run time package.
Just go to the arduino download web page and follow the links to download the IDE for the Galalieo .
By the way they said the next board is comming soon and it is named after another famous old scientist.
Good news - I think they must have noticed this and changed it, because the SW & FW download site for the Galileo is now showing this Open Source SWLA text (that you have to click & agree to in order to proceed) when you try to download:
(Intel Support)
Download Agreement for Packages of Source Code Under Open Source License
This package contains software licensed under one or more open source licenses.
Consult the COPYING, README, or LICENSE files in the appropriate subdirectory.
Intel does not make any representations or warranties, express or implied, including
without limitation, any warranty of fitness for any purpose, merchantability or non-infringement.
I'm not sure what there is to get excited about here. Just because it's Intel I should be jumping up and down? I'd rather hear there are a pile of GPIOs and a bunch of transceivers I have never even heard of before and differential signaling. Something, anything, special, at least. Something that says you can do things with it you can't do with an Arduino - integrate it with a really high-speed ADC or DAC, software defined radio, something. ANYTHING! That would get me excited. Instead, the first application Intel suggests is driving LEDs. NOT INTERESTING. I can do that with any 8-bit processor.
The following description of the analog pins is provided:
A0 – A5 - 6 analog inputs, via an AD7298 analog-to-digital (A/D) converter (datasheet)
o Each of the 6 analog inputs, labeled A0 through A5, provides 12 bits of resolution (i.e., 4096
different values). By default they measure from ground to 5 volts.
The AD7298 is priced at Digikey at $9.26 each or $4.79 for 3,000 or more. Because the Atom doesn't have a native ADC, in order to get that functionality, they have to include a chip that, in large quantities, costs just about 3 times as much as the 328P does in large quantities ($1.76 for 2,000 or more). The AD chip at 12-bit and 1Msps is more capable than Atmel's built-in 10-bit ADC, but I don't know if this matters, and in this case, it probably doesn't. I doubt the application programmer can ever make use of it.
This all seems to me to be trying to fit a square block in a round hole. Intel takes a very powerful microprocessor and dumbs it down into a microcontroller board. But, since microprocessors don't have a lot of native built-in peripherals or GPIO, like microcontrollers do, they (expensively!) bolt those features on. It doesn't feel natural. It feels contrived. Maybe they should make a microcontroller version of the Atom with proper GPIO and peripherals.
The AD7298 is priced at Digikey at $9.26 each or $4.79 for 3,000
Forget distribuitors like this for pricing. First off 3,000 is a piddling little number and someone like Intel will go direct to a manafacaturer and not bother with a distribuitor. They are likely to pay ten times less for this chip at least. I have worked for a consumer mass market company and you would not beleve the price you can get chips at if you buy enough.
What it can do is have an Ethernet stack and still have plenty of memory, so if your application involves having a complex web server then it beats the normal Arduino hands down. That's all I can think it is good for.
Or maybe the Tre. ... I'm guessing that's close to the truth though. Arduino has the power to create an ecosystem of the dreaded ARM chips! Must do something!!!
Or maybe the Tre. ... I'm guessing that's close to the truth though. Arduino has the power to create an ecosystem of the dreaded ARM chips! Must do something!!!
That happened years ago. With consumers, not hobby developers, though, maybe that is the distinction. And it all happened because Intel didn't, and mostly still doesn't, have anything that competes with ARM at ARM's level of price and power consumption. And apparently still no real microcontrollers. With ARM you can choose from at least ten suppliers, dozens of models each.
The only thing they're bringing to the table over the typical ARM offering is the ubiquity of the x86 instruction set, and mini PCIe. The former is of dubious usefulness (like I said earlier, mostly drivers and the occasional closed-source binary app), while the latter will have appeal to a certain market segment.
Certainly missed the boat on a lot of other opportunities though. Not the kind of innovation I would hope for from the CEO of Intel.
GoForSmoke:
I think that Intel has just a little experience with microcontrollers and not just the 8051.
Yeah, the 8051 was such a bomb, what an embarrassment for Intel.
They got out of microcontrollers because they didn't want to be there because of the price-points. And now they are stuck slapping a full-featured CPU (which lacks many microcontroller features) on a product that would normally be run off a microcontroller because they have no other option.
Do you have any idea how many 8051's were made just by Intel? And they licensed them. 8051's and derivatives are still made today and still bring money to Intel. How embarrassing!
GoForSmoke:
Do you have any idea how many 8051's were made just by Intel? And they licensed them. 8051's and derivatives are still made today and still bring money to Intel. How embarrassing!
We agree. The 8051 is a huge success for Intel and a giant among microcontrollers. I was being satirical, that's why I included the rolly-eyes emote. Maybe I shouldn't do that.
It turns out I am working today with a SILabs C8051F350. I am trying to see if I can get the built-in 24 bit ADC to wow me. The can is an AD590, an analog temperature sensor. The reference is an AD4525 2.5V precision reference. Those three parts are samples. Sadly, the rest are not, though I got a good deal on those 15-segment dual displays, which are "intelligent" and incorporate shift registers and current limiting - they are 3 wire devices (CLK, DI, CS). Jameco used to sell them for a buck each when they had them in stock and I bought 30 of them. They are very useful. But they eat power like the dickens. This thing runs on 30mA until you light up all the segments on that display. Then it's about 1200mA. Which means I am beyond what that regulator can do for long (its dropping 5.0V to 3.3V).
Here is an Atmel 8051 trivial project I did a while back to get my feet wet, so to speak. I posted it as a response to another member's question:
