Catch up on stories from the past week (and beyond) at the Slashdot story archive

 



Forgot your password?
typodupeerror
×
Hardware

Raspberry Pi Beta Boards Unveiled 161

First time accepted submitter anwe79 writes "Those of you who have been wishing for a Raspberry Pi this Christmas will sadly not get your wish granted. However, you may be happy to hear that populated beta boards have now been produced. Beta of course means the boards still have some more testing to undergo. But, if all goes well, those inclined should be able to get their hands on production boards in January!"
This discussion has been archived. No new comments can be posted.

Raspberry Pi Beta Boards Unveiled

Comments Filter:
  • I've been let down before.
    • by Anonymous Coward

      I'll wait till it boots myself. I've been really let down before.

      • by Zerth ( 26112 )

        I'll wait until it runs 3 months without the power supply melting. I've been burned before.

        *cough* Globalscale sucks.

    • by Threni ( 635302 )

      Go away. If you've been following the website you'll see they're around 3 weeks behind schedule but have finished the whole design process and are now testing the hardware before mass production. Do you even know what vaporware is?

  • by Osgeld ( 1900440 ) on Thursday December 22, 2011 @11:16PM (#38467910)

    Of course I am still under the "it doesn't exist until I can blow it up my self doing something dumb" crowd but it's making good progress

    • by Frosty Piss ( 770223 ) * on Thursday December 22, 2011 @11:31PM (#38467970)

      Of course I am still under the "it doesn't exist until I can blow it up my self doing something dumb" crowd but it's making good progress

      It *is* making "good progress". But where these types of projects usually hang up is when they finally get to the stage where they need to put together the infrastructure to source parts, manufacture, and market the *product*. At this point, they generally realize that they just don't have the organization and resources necessary, and the sub-$100 price point is out-the-window unrealistic for the volume they can realistically project to move...

         

      • Reality is coming (Score:4, Insightful)

        by jmorris42 ( 1458 ) * <{jmorris} {at} {beau.org}> on Friday December 23, 2011 @12:10AM (#38468156)

        Dunno, I was in the same camp, no way they would actually ship at the stated prices, expect a doubling which would make it too expensive to be interesting. Or at least less interesting than the many other similar project computers and/or microcontroller products actually shipping. But if they are expecting to begin shipping next month and still holding to the original price they are either really going to pull it off or are truly idiots with zero business sense. I'd give em even odds at this point. :)

        But why is it front page news every time these guys pass gas? If they ship it, that is news. Heck, when they auction off these guys I'd guess that would be news too. But d we need a story every month even when there isn't any actual news to report?

        • by bcmm ( 768152 ) on Friday December 23, 2011 @08:17AM (#38470116)

          But why is it front page news every time these guys pass gas?

          Mostly because they're being very open about the development process on their blog, meaning you see stories about stages which wouldn't be announced publicly in comparable projects.

      • Re: (Score:2, Informative)

        by Anonymous Coward

        Perhaps you missed it, but Broadcom is selling them the silicon by tacking it on to larger production runs, so they've got as much as they want at quantity pricing.

        They've already bought the other parts so sourcing isn't a problem(for the first 10k anyway).

        http://www.raspberrypi.org/archives/302 [raspberrypi.org]

        And clearly they've got the marketing down, otherwise you wouldn't be discussing it:)

        • by jimicus ( 737525 )

          Perhaps you missed it, but Broadcom is selling them the silicon by tacking it on to larger production runs, so they've got as much as they want at quantity pricing.

          I had missed that. I don't know much about the semiconductor industry so perhaps someone could confirm - does that mean that while they get them at quantity prices, that pricing is subject to Broadcom receiving sufficient orders for the same part so there are larger production runs to tack it onto?

          • by boley1 ( 2001576 )

            Some of the key volunteers have an "In" with Broadcom. Broadcom is being very supportive, with inside information on which parts make the most production and cost sense.

      • by DrXym ( 126579 ) on Friday December 23, 2011 @05:05AM (#38469420)

        It *is* making "good progress". But where these types of projects usually hang up is when they finally get to the stage where they need to put together the infrastructure to source parts, manufacture, and market the *product*. At this point, they generally realize that they just don't have the organization and resources necessary, and the sub-$100 price point is out-the-window unrealistic for the volume they can realistically project to move...

        I think Raspberry Pi's price goal is pretty ambitious but at the same time it's not outrageous. It's basically running the same parts you'd find in any cheap ass media player. You can pick up media players for less than $100 and if you cut out the case, packaging, power supply, application software, optional software licences (e.g. AC3, Dolby), reseller margins, and just ship the barebones product you could do it for the price they're proposing. Or if not exactly then not far off it.

      • by Alioth ( 221270 )

        Sourcing the parts and getting it made really isn't a problem. I've done it for electronics with even more limited demand than this for a similar sized, similar part count board. There are assembly shops that can do runs of assembled boards right from just one example to hundreds of thousands, they do all the parts sourcing for you, you just give them a BOM and they organize it all. Most the parts on the Raspberry Pi will be common parts (resistors, capacitors, standard connectors, various ICs that get put

        • Sourcing the parts and getting it made really isn't a problem. [...] I've managed it *as an individual* with no organization backing me using these services.

          If you did it for under $100 then you must value yourself cheaply. $100 doesn't even buy an hour of my time.

          • Sourcing the parts and getting it made really isn't a problem. [...] I've managed it *as an individual* with no organization backing me using these services.

            If you did it for under $100 then you must value yourself cheaply. $100 doesn't even buy an hour of my time.

            Most people aren't consultants!

  • Design flaw? (Score:5, Interesting)

    by Keruo ( 771880 ) on Thursday December 22, 2011 @11:33PM (#38467984)
    I think that surface-mount usb power connector will fail eventually since the images seem to show it not welded through-board.
    Maybe they'll fix it on later models.(or it is already, but I'm not seeing the throughwelds from the pictures)
    • Re:Design flaw? (Score:5, Interesting)

      by fnj ( 64210 ) on Thursday December 22, 2011 @11:43PM (#38468026)

      The surface mount USB on my Beaglebone fell right off. The glue holding it failed with hardly any stress. There are big lands to solder it to, but they didn't use these. They only used glue. What the heck is the attraction of these stupid mini and micro USB connectors anyway? Give me a soldered-through full-A connector any day.

    • Re:Design flaw? (Score:4, Interesting)

      by Gourou ( 123409 ) on Friday December 23, 2011 @12:36AM (#38468248)

      From the earlier post of the bare boards (http://www.raspberrypi.org/archives/402) the holes are there for the micro-usb, and the project has been geared towards clumsy hands plugging and unplugging the ports a lot so I'd expect a robust connection.

    • All your worries can be solved with the right enclosure. The USB connector will be as sturdy as you make it. For example, encase this entire board in resin... nothings coming off it. That's the simplest solution, and it might cause overheating, but if you sit down and think about it, there are a lot of other ways of making that connector more secure.
    • by bcmm ( 768152 )
      Are you by any chance an N900 owner?
  • by cachimaster ( 127194 ) on Thursday December 22, 2011 @11:59PM (#38468106)

    This board is perfect if you want to learn to program ARM assembly or cross-compiling but the ARM architecture it's one of the most closed and patent-restricted technologies out there. Teaching ARM is the equivalent to teaching Visual Basic Programming, common but very closed architecture.

    So it's not really open, even if the PCB design is open.

    A truly open system would be OpenRISC [openrisc.net], there are dev. boards out there like this [orsoc.se] one (I'm not affiliated to OpenRISC in any way). They are more expensive because are made with are FPGAs, but that's what you should learn in school.

    Wait until work to learn proprietary stuff.

    • by Anonymous Coward

      You could use these boards to teach Linux (operations, programming, etc.), cross-compiling techniques, embedded programming and lots more. Sure OpenRISC might be pure openess but it does not hurt to learn and teach practical stuff that could be put to use making a living.

      • It is not about pure openess, it's about knowing what the f*** is happening inside a CPU. I bet none of you know.

    • Re: (Score:3, Insightful)

      by Anonymous Coward

      What other $25 Linux boxes are out there for teaching Unix, web programming, and other high level stuff? I don't think teaching ARM assembly was high on their list.

      • Re: (Score:2, Interesting)

        by Frosty Piss ( 770223 ) *

        What other $25 Linux boxes are out there for teaching Unix, web programming,...

        Show me WHERE I can buy one of these for $25?

        Remember the OLPC project? Weren't those supposed to be sub-$100? How much did they end up being?

        These are *not* in production (or anywhere near production) yet.

        When (if) these make it to production, expect the price to be more than $25.

        • by Anonymous Coward
          If you take a look, they have been pretty open about how they plan to achieve that price - it looks reasonable. Not saying it's 100%, but it's entirely possible.
      • by jmorris42 ( 1458 ) *

        Stop with the $25. That is just marketing, I doubt they actually expect to move the first one of those because they are pointless. You need a keyboard and mouse and the Model A only has one USB port. Good luck finding a hub cheap enough that it doesn't make more sense to just spend the extra $10 for the Model B and get a network port as a bonus. As if network is optional these days.... unless you are going WiFi but price that out... along with a hub. Not to mention that running much of anything modern

    • most closed and patent-restricted technologies out there. Teaching ARM is the equivalent to teaching Visual Basic Programming, common but very closed architecture.

      So it's not really open....

      Take a look at Qt on Pi...

    • God forbid you might like to use the board to teach C, C++, perl, python, pascal, BASIC,..... this thing is designed for kids in a classroom and for use at home on the TV.It is designed to be dirt cheap so breaking one isn't going to be a problem.

      • You can also use OpenRISC for that, and if the kid want to design his own CPU he can do that too, unlike with this board.

    • by bcmm ( 768152 )

      Teaching ARM is the equivalent to teaching Visual Basic Programming, common but very closed architecture.

      Uh, what?

      No; teaching people to program C on an ARM Linux machine is the equivalent of teaching people to program C on an x86 Linux machine: the CPU is proprietary, but who cares?

      The cross-compiling thing is a red-herring too; you'd just run GCC on the RaspberryPi and for educational purposes it would be plenty fast enough.

    • by josath ( 460165 )
      RMS, is that you? The world is never going to be your communist open source utopia where closed source and profiting off of software is illegal. People don't need the netlist for every IC they use. This board is more closed than most, but it's a compromise I'm willing to make as the price per performance is better by a mile than anything out there.
  • by dbc ( 135354 ) on Friday December 23, 2011 @12:37AM (#38468260)

    They appear to be bending the USB spec quite seriously. A USB device is allowed to draw up to 100mA before enumeration, and up to 500mA after being enumerated and negotiating for high power. They talk about using up to 700mA with networking connected -- it's not clear to me how it could enumerate without booting first -- so they seem to be giving the middle finger to the USB specs. I predict unhappiness when people find that only some USB power sources are going to tolerate the load.

    Is it so hard to put a couple of holes in the board to solder wire to?

    • so they seem to be giving the middle finger to the USB specs.

      Like the iPad?

    • Is it so hard to put a couple of holes in the board to solder wire to?

      Is it so hard to provide screw holes holes for mounting?

      Also, it's usually considered a good idea to put all the connectors on the same edge and line them up flush so you can put the thing in a box.

    • by harrkev ( 623093 )

      Well, I would imagine that this is so that you can use any old USB power supply that you have on hand. You know, the same ones used by all decent modern phones, the Nook, some cameras, etc. If you shop around, you can score a 120V to USB power supply for well under $5, and most everybody has a micro-USB cable lying around.

      Some monitors even have USB hubs built-in. Easy enough to hook a cable from the monitor to the Raspberry PI for power, and another cable from the Raspberry Pi to the monitor for USB con

    • by Nursie ( 632944 )

      That spec is all but irrelevant these days, as many, many mini- and micro-USB chargers exist that give far more than this with no negotiation at all.

      Perhaps the spec needs changing to reflect the real world usage?

      • by Anonymous Coward

        USB 3.0 allows 900 mA, the battery charging spec 1.5 A.

    • As someone who's built USB devices I can tell you it's a gamble anyway. I've seen busses give 500mA without negotiation and I've seen busses that won't put out 500mA after, and I've seen busses where the manufacturer realized people wanted to charge things so they put out something like 1500mA.

      As for devices that plug in to USB but require more than 500mA to run check out the BeagleBoard - it requires more than 500mA to use most peripherals (network) but if you run on anything more than 500mA the thing star

    • by redelm ( 54142 )

      Half my 6 USB chargers claim 1+A output, the other half 500 mA (older). Who knows what they really do?

      I strongly expect Raz went for USB power to avoid all the national electrical approvals necessary for wallwarts. Remember, this is a shoe-string outfit. Just get a phone charger with someone else' approvals. They probably chose micro- over mini- because the former are more likely to make 700+ (iPod & smartphone draws).

    • Re: (Score:2, Informative)

      by Anonymous Coward

      The charging specification does not require enumeration.

      http://www.usb.org/developers/devclass_docs
      Battery Charging Specification, Revision 1.2
      Section 1.4.7

      Try to keep up.

    • by bcmm ( 768152 )
      The choice of microUSB rather than USB-B or something is probably because there is a spec for high power over microUSB. It's used in the new standard mobile phone chargers. IIRC, a dumb charger can short the D+ and D- pins; indicating that it is too stupid to operate a real bus or do enumeration, but a device is welcome to draw as much current as it wants (with voltage possibly dropping if it draws to much; like a battery).

      Perhaps somebody more knowledgable could correct some details or link us to the spec
  • If this cheapo pc made for TVs gains any traction, we'll start seeing them built in to the TVs. Surest way to commoditize this stuff.

    • by Nursie ( 632944 )

      Most modern tvs probably have more power than this already.

      My Samsung plasma (a couple of years old now) already has a UPNP/DLNA network media player, youtube interface, app-installer type thing and a variety of other stuff built in. Meaning it's already got a general purpose embedded computer in there, it's just feature limited at the moment.

    • That's true. I think there'll still be a place for these as separate boards though; for the educational and hobbyist markets (which I think is what they're targeting and expecting to be popular with) it's quite important to be able to easily replace broken devices and to be able to incorporate them into other designs.

  • by fredan ( 54788 ) on Friday December 23, 2011 @08:15AM (#38470110) Homepage Journal

    does it blend?

  • Seriously, I would install a dozen of these type B boards in a case, only use a single power supply, a Ethernet switch and make a low power blade server. I think the power / speed / price ratio would work out. Add a NAS for storage, and you could have a fairly powerful blade for a fraction of the big boys. BOM works out to 12 x 35 = 420. Add a case / PS, Switch. Boot from SD and store everything on a NAS (add extra cost for storage). There's a lot to like about these boards. I think they could be a game changer.
    • I'm thinking about adding monitors to my pc with raspberry pi and symergy2.source forge.net. I hope I can do network over the USB power

The Tao is like a glob pattern: used but never used up. It is like the extern void: filled with infinite possibilities.

Working...