I just paid 184 Euro for an i5-2500k CPU, add 100 for a board and 20 for two SO-DIMMS and you're get a nice and fast small footprint server or desktop machine, but certainly not a Raspberry PI or Beagleboard etc.pp.
Pictures and details of a stripped-back ultra-compact desktop computer from Intel have hit the web. Packing a Core i3 or Core i5 processor socket on a 10cm-by-10cm board, the hand-sized Intel Next Unit of Computing (NUC) doesn't feature a touchscreen nor a keyboard, although it is one of the smallest complete x86-compatible …
What he said! If anything this is more akin to Via's Nano-itx (at 12cmx12cm) than the rasberry pi.
The main selling point of the pi is that it is so cheap, cheap enough that you could take risks with it that you would not want to do with something costing 10x as much. Basically I see it as an experimental computer (something between a fully fledged PC and micro-controllers), something you can use to teach yourself, or stick it in difficult environments without risking a lot of money.
The Intel does not compete with it, they are in different leagues completely.
Agreed Ogi. Definitely different beasts. I plan on using the Pi in Big Gun R/C Model Warship Combat to replace the more conventional R/C guts while adding new functionality (like cameras, sensor readings and such that I don't have now). Some things are hard to tell from 50ft away with a 5-6' model ship on the water. I *suspect* the other robotic combat guys (ala the Robot Wars folks) will start doing the same.
The Pi ($25 for regular Pi and $35 for Pi+) can also get used in quantity with a lot more laissez faire attitude if your experiment doesn't work out. (and by not work out, I mean a release of magic blue smoke) Definitely a hobbyist board. Kinda like the Altair of old in that regard. I don't see doing that with the Intel board.
Penguin cuz that's what Pi runs innit it?
That's pretty cool! You could get a gun's eye view :D Does the boat actually fire? (I presume the combat bit actually involves actually firing stuff at each other?). You could have some fun implementing a ballistic calculator into it, perhaps with range estimation :)
I have so many ideas for the pi, primarily because I can risk it in environments and experiments where the magic smoke/water damage is more likely. Places where I'd never dare stick an expensive PC.
Good luck with your project, sounds awesome :)
Ogi, well, things move too fast to get much use from a turret's eye view. But for an after-action "report", that kind of thing would be nice to have and see just how bad a shot I am. :)
But yeah, CO2 powered cannons in turrets firing bb's to 1/4" ball bearings (or in Australia 3/16" to 1/4" ball bearings due to gun laws). WW1 to WW2 era ships (ala the Big Gun era). Been around as a hobby since 1979. Regular R/C has its pluses for easy entry, but if can get the code together for a Pi system for stuff like that (and have it available for other folks), it would add some extra features/functionality not available with a regular R/C controller. Combined with GPS, could have automated convoy's to shoot at. :)
Anyway - My point originally being that uses for the incredibly cheap Pi's are quite nearly endless, and that was just one example I don't think the Pi designers expected. :)
This post has been deleted by its author
Imho this shows how intel are floundering around having realized that arm are going to eat their business from the bottom up.
The raspberry pi must have alarm bells ringing. This could be the thing that finally gets linux adopted firstly by youngsters. They learn linux, they don't need intel boxes to run it. Another 10 years from now they start to be the influence in business with what they buy and use.
I cant help but feel that ms don't like the pi for the same reasons, and unlike the netbooks you can't shoe horn windows onto the pi. Therefore the only alternative is get a supplier to make something to counter the pi. A little pressure never hurts, especially if they (intel) believe ms may drop them as a supported platrom down the line anyway. Its a great time to rebuild a bridge if such an opportunity arises like this.
Then again, I may just be reading too much into it....
I don't think so. I think Intel are horrified at a whole new generation of kids growing up with ARM development. Just like Microsoft will do anything it can to ensure Windows is the one and only OS kids grow up with.
Intel are trying pretty hard to get rid of ARM in phones and I'm pretty sure there's been some calls to Microsoft about their ARM version of Windows. Intel has managed to kill off competition in the past, remember Transmeta and how Sony suddenly dropped the Transmeta chip?
Giles, you could easily be right. Like I said, this *could* be the next Altair which ultimately changes the world. Or maybe it isn't... I just don't know if Intel or MS have noticed Pi *YET*. They're already surrounded by embedded devices running (most of our lives now) on something other than Wintel, so a cheap Hobbyist level board wouldn't normally be perceived as a threat. (or would it?) We'll see. :) Should be fun to experiment with in any event. I got plans for this baby!
What?! The Raspberry Pi uses an ARM chip?! Yeeeeuuuukk! After buying two arm powered tablets, which were by the way, worthless and weak, I have no need for such a limp wristed processor. ARM processors belong in toys, not computer hardware. If these damned manufacturers would quit nickel and dimeing us, maybe human kind could get ahead. AMD has some strong low power consuming processors. why not use them?
I don't think he is.
Look around you. Look around at home, in the office, in the car, in the datacentre.
Lose the need for Windows, and where's the big incentive to use x86? Any consumer electronics or professional embedded system builder can answer that one, most already have, and you can see their answers all around you. You've got examples of their decisions in your home, in the car, in the office, and in your pocket. You can also see the answer in some of the more sensible datacentres.
More and more smartphones, more and more tablets, more and more products that Wintel can't address, and that remind people that there is life without Windows.
Windows and x86. They're inseparable.
Away from Windows, who (other than the certified Microsoft dependent ecosystem) gives a monkeys about x86?
When did Intel last have a commercial success outside the Windows/x86 market? [I class Dell as a Wintel vendor, and most of HP too]
There are lots of embedded PC formats. Apart from the fairly popular Mini (and Micro)-ATX there is also a 3½ inch "biscuit" format and a 4 inch PC104 size, which seems to be about the same as this new offering.
These are intended for embedded use - as is the Pi, though given the power and heatsinking requirements of Intel's latest attempt, I'm not convinced that's the market they're aiming for.
It's definitely not a Pi-worrier.
It might be a Pi worrier, I think.
I'm absolutely taken with the RaspberryPi; I'd love one. But they're not happening! I looked on the website, and you can now again pre-order for a delivery date of maybe later...
This is going to pi-ss people off. If Intel gets this on the market (and I think they have more of a chance to put out volume than the Pi foundation) and offers higher specs, then people might just be willing to pay a higher price for something they can buy, than for something they can only again and again wait for.
This delay is killing the RaspberryPi in my opinion. It gives any competitor too much time to get to the market where some Chinese makers can probably push them out by the 10Ks on short notice and won't worry too much about upsetting a non-profit foundation with a lot less clout than a commercial brand like Intel or Apple or such.
Not sure how a device that cost $200 is a worrier to a device that costs $35.
Pi's are being delivered right now. There is a huge backlog, 350k people, but production is ramping very fast and that expected to be dealt with in 2 to 3 months. If you order now, September should be a good bet - then October for order and fast delivery. If all goes to plan with RS and Farnell who are doing the manufacture.
Well, I thought that a $25 Huawei phone could never be outsold by a $400 Apple one. And yes, I know that the Apple one has higher specs - so has the Intel board.
About the delivery, your very reasoning in your post is putting customers off. 'Huge backlog', 'expected',' should' and 'good bet' are not confidence-inducing terms.
As much as I love the RaspberryPi (don't get me wrong, I really do!), I think it's suffering from marketing and production failure. Sadly more vapour than ware.
-'tbf kiosks do need to be fairly snappy in use'
Yep, my only recent experience is using a Kodak photo-printing Kiosk... was a bit slow, wasn't sure if it had registered a finger press etc The less irritating the kiosk, the more I would be inclined to browse my SD card and spend more money on prints. Maybe a case for Intel.
-'Ergo if the fan doesn't work, your whole system's down.'
Not quite. More recent Intel chips will throttle themselves first, then resume normal service. (Prove this to yourself with a Core2 laptop, some duck tape and a recent game demo : D). In a Kiosk there should be plenty of room for a nice big passive cooling system.
But yeah, an ARM solution would appear to be the better option in many cases. It seems Android on ARM is capable of everything that would asked of most kiosks.
Exactly. Ergo if the fan doesn't work, your whole system's down. Which is a nuisance when what you're looking for is a relatively low-spec item with relatively high uptime. A lower spec processor with no fan will be a better candidate for interactive displays, PoS systems etc.
Then glue it to a massive passive heat sink.
IIRC Intel chips just slow down when they reach their maximum operating temperature, So even without adequate ventilation it shouldn't be a showstopper. (What kind of embedded system would run all the CPU cores at 100% anyway?)
"Exactly. Ergo if the fan doesn't work, your whole system's down."
EVERYONE PANIC !!!
"A lower spec processor with no fan will be a better candidate for interactive displays, PoS systems etc."
Horses for courses ? Is that what you're suggesting ?
And, really, how often do modern fans fail ?
It would seem that some people can't help but to find any "fault" - instantly - with any innovation, instead of immediately appreciating and recognising a viable use for a great product.
Smaller, faster, better connected = fail ?
-And, really, how often do modern fans fail ?-
About as often as non-modern fans since the normal method of failure was filling up with dust and junk. Passive heatsinks can get to the point of being so dirty that they fail, it generally takes longer then fans.
Fans also use power, increasing the cost of operating the device and adding to the heat load generated by the device.
Another factor in fans is they make noise, not so much of a problem in some cases, but in POS systems it really sucks when fans raise the noise floor.
That said, this device will allow the use of already designed and operating software on existing platforms so a lot of the development work is already done. Software makes the device, just ask Apple and Microsoft.
And yes, we (as a forum) can find fault in whatever we like... I do believe that is why we have the ability to post, so we can bitch about things.
"And, really, how often do modern fans fail ?"
Few companies give MTBF figures, but those that do rarely say more then 40-50k hours, which is about 4-ish years. At at the MTBF (if correctly given and a constant rate) only 37% are working by then.
For always-on devices they are *significantly* more likely to fail than semiconductors.
"For always-on devices they are *significantly* more likely to fail than semiconductors"
So, to recap, only a complete and utter div would use a NUC mobo +FAN in a POS environment with an intended constant run time of circa 4 years ?
Well, that's the best reason I've heard today not to buy this motherboard for any other intended purpose. For that, I thank you.
Avoiding fans is a good idea if you want reliability over a long time. Some example info of bearing types vs. temperatures can be found here:
As usual, if you can run cool (and probably lower the fan speed as well) it lasts longer. Of course, if you need a fan, then running cool is probably not a typical case...and if you want it quiet (e.,g. media player) the no fan is best, or a really big low RPM one (acoustic noise is related to the 8th power of air speed!)
Buuuuut.. You can see the components.. And it's small. So they must be exactly the same.. Right?
Consumer electronics devices as any fool knows, are made of the casing, and specially encapsulated unicorns. Which is why you can't open them. Because the Unicorn might escape.
Please bear in mind, the people who seem most offended by the Pi stuff seem to be the least able to understand exactly what it is. Which is perhaps why they are so willing to expose their ignorance when screeching against it. I've lost count of the times a Pi board was compared to an Arduino by some twit on a forum somewhere.
I think you are being a tad harsh on the Raspberry Pi Foundation there. The Pi IS available now, albeit in limited numbers.
I just found out that Element 14 / Farmell have debited my account for the £29.46, and delivery is expected next week, maybe even sooner. The next batch are due in a fortnight...
I'm a really happy (and smug) bunny. ;o)
It's a Mac mini with an external power supply. I bet it'll cost nearly as much too. I doubt Intel is too scared of the Rasberry Pi but I can't help but think Microsoft is scared of any system that cannot run Windows. I wonder if they will scale down Windows 8 RT to run on the Pi?
100 for the board alone? Add a case, power supply and cpu and you've hit normal desktop pricing. It is just a desktop board at an average price with laptop-class pcie and graphics.
That would be nothing to write home about. The inclusion of thunderbolt is nice though. It would make a handy stb device, with enough CPU for transcoding.
Pi references are forum-bait.
It is a very interesting move on their part and whether aimed at competing with some of the ARM based boards like the Raspberry Pi SBC or not, it could very well end up being an attractive alternative to some people, more specifically those in North America.
At this point in the Raspberry Pi project, given some unfortunate flaws in the launch due to a miscue on how popular the idea was, manufacture errors and now shipping delays to as far as first day purchasers in other countries outside of the UK still having a status of back-ordered to August, the reliability of supply comes into question as far as being a worthwhile platform to be used in the classroom.
With a "possible" more reliable supply from a large corporation at a price between a Raspberry Pi and a Beagleboard, something like this could become an option for education.
....the Pi is exciting because it's a truly great idea. It is made by a small startup outfit, budding enthusiasts... It has ideals and principles... it's cool and exciting because it's small, and designed/conceived by a very small business.
Intel doing it just has no magi or meaning, IMO.
Your arguments are no recommendation. Great idea? Really? A little embedded computer thingy? Bollocks - it's no more an idea by Pi than the wheels on a car are Ford's idea. The Pi appeals to people here for three reasons :
It's not Intel! Therefore automatically great!
It runs Linux! Therefore automatically great!
It's cheap! (This is actually a good reason to like something from any point of view, designer or end user. Surprised you didn't add that to the embarrassing idealist stuff. It would've added some credibility for the more realistic of us. About 5 on these forums, sadly.)
The problem was with the network sockets, in that if someone plugged it in unfixed it would not work and could even fry the attached device(s).
This required completely reworking the boards, which takes time and can't be done with a machine.
Read, a row of people armed with soldering irons and braid.
All the boards also had to be retested again to make sure the repair hadn't damaged anything else.
This post has been deleted by its author
Clearly RasberryPi is not going to compete head to head with Intel. For a start thay are a non profit organisation. However just as with smartphones the original eeePC netbooks (I think it's fair to say that these were the devices that started this) what this does show is a trend towards the end of the PC-as-we-know-it whether as desktop or indeed laptop.
In future almost every reasonably complex consumer electronic device will have a 'PC' in it but it very likely wont be much of a multifunction device - more of an embedded media player/tv/centralheating clock/ car dashboard/ phone type of thing that most likely wont be running windows (at least as-we-know-it) either, due to cost/size/power either. The more this happens the more likely that the more complex 'PC' devices will evolve into something else also - see phones/tablets for reference.
Intel obviously have a sharp eye on this game and are preparing for it. Consider how easy it would be to embed either this type of device or indeed Rasberrypi-type-of-device into a modern large screen TV. Oops - where'd the PC go?
IMHO it's unlikely that Intel will completely lose the game to ARM although it's likely they will lose some more market share
What this means for MSFT and indeed linux is the other question. But one thing is likely: more embedded devices = less MSFT = More Linux
This is all good - I look forward to the day when the market for computing devices is a competitive as it is for cars.
You can get a Brazos HTPC in that form factor already:
Seems more likely what they are aiming for - the ultra-compact HTPC arena.
Raspberry Pi just ain't something they can compete with at this level - not enough profit in them at the prices the education/hacker market would want to pay (IE well below £100)
Biting the hand that feeds IT © 1998–2021