EmbeddedRelated.com
Forums
Memfault Beyond the Launch

OLinuXino, a serious Rasberry Pi competitor?

Started by Don McKenzie March 7, 2012
On a sunny day (Fri, 09 Mar 2012 13:26:09 GMT) it happened nico@puntnl.niks
(Nico Coesel) wrote in <4f5a042f.3990283062@news.kpn.nl>:

>>I made a little 9-20V to +5 V switcher with USB connector to charge a real Chinese mediaplayer I have, >>that because I have plenty 12V wall warts. >>Adds an other 5$ at least even if DIY. > >You know about Dealextreme.com? You probably could have saved some >money and clear some space by throwing the 12V wall warts away :-).
Oh, I forgot to mention WHY I did the separate charger... The PC USB port would switch itself off because of too much current, IIRC it is only 600mA in the spec? Not sure. If the strawberry pie is indeed 700mA that will promise a whole lot of fun on a whole lot of peesees.
On 03/09/2012 06:07 PM, Jan Panteltje wrote:
> On a sunny day (Fri, 09 Mar 2012 13:26:09 GMT) it happened nico@puntnl.niks > (Nico Coesel) wrote in<4f5a042f.3990283062@news.kpn.nl>: > >>> I made a little 9-20V to +5 V switcher with USB connector to charge a real Chinese mediaplayer I have, >>> that because I have plenty 12V wall warts. >>> Adds an other 5$ at least even if DIY. >> >> You know about Dealextreme.com? You probably could have saved some >> money and clear some space by throwing the 12V wall warts away :-). > > Oh, I forgot to mention WHY I did the separate charger... > The PC USB port would switch itself off because of too much current, > IIRC it is only 600mA in the spec? Not sure. > If the strawberry pie is indeed 700mA that will promise a whole > lot of fun on a whole lot of peesees.
The USB 2.0 spec only allows for 500 mA maximum current after the device has been configured. Unconfigured devices are only allowed 100 mA. For devices that require more than 100 mA, the configuration is only done after the host first reads the current requirement, and has determined that it can still supply it. For instance, a 4 port bus-powered USB hub can only support 4 devices if they have a max 100 mA requirement. Cell phone/battery chargers with a USB type plug are not limited to those currents, and can typically supply more.
Jan Panteltje wrote:
> On top of that it needs 5V *stabilized*. > That is a big problem, it should have been 7 to 20 V unstabilized, > or 2 to 20V unstabilized, > so you could use it in a car or some battery powered system. > the extra regulator is expensive.
> Adds an other 5$ at least even if DIY. I suspect the strategy is to tell people to just go buy a cigarette lighter (12V) to micro-USB power adapter on eBay for literally a buck or two shipped to your door. (I just found one for $0.99 shipped from Taiwan, 5V @ 1A output...)
> What's the form factor? Does it have 0.1" headers so you could add > processor horsepower to a veroboard / stripboard? That would be great > for quick one-off tinker stuff. >
Exactly. What I'm looking for is the Embedded Linux Computer with FPGA connected directly to the CPUs bus and to 0,1" headers, so that my students could deal with different peripherals (implemented in the FPGA) and connected to simple circuits assembled on the breadboard (or on prototype PCB connected via flat cable). Does anybody knows about such cheap boards? -- TIA & Regards, WZab http://www.ise.pw.edu.pl/~wzab/wireless_guitar_system/ Build your own digital wireless guitar system
In article <4f5a0371.3990093843@news.kpn.nl>, nico@puntnl.niks says...
> > Paul <paul@pcserviceselectronics.co.uk> wrote: > > >In article <4f597914$1@dnews.tpgi.com.au>, keith@nowhere.com.au says... > >> > >> On 9/03/2012 8:54 AM, Paul wrote: > >> > In article<4f59230c$1@dnews.tpgi.com.au>, keith@nowhere.com.au says... > >> >> > >> >> On 9/03/2012 3:00 AM, Paul wrote: > >> >>> In article<9rqf87FadnU1@mid.individual.net>, 5V@2.5A says... > >> >>>> > >> >>>> On 08-Mar-12 11:51 AM, Nico Coesel wrote: > >> >>>>> Don McKenzie<5V@2.5A> wrote: > >> >>>>> > >> >>>>>> On 08-Mar-12 10:23 AM, Paul wrote: > >> >>>>>> > >> > >> Most USB wall warts are of the universal voltage type, you just buy a > >> cheap adaptor. I even have one that came with slip in connectors that > >> cover 90% of the power sockets in use in the world. > > > >Reality is especially in classrooms budgets are tight so 20 to 30 of > >these is not prctical causes problems of time wasted, making sure all > >bits are together at beginning and returned at end of class, let alone > >whose got a dodgy one to debug. > > Thats true. I don't think the RP is very suitable for use in a > classroom. Its too fragile. > > >How many wall warts and the USB cables are speced for 5V @ 700mA. > > How many are not? Most are rated 1A and 2A is on its way.
The mahority of USB wall warts people have are for phones in the 100 to 3000mA range to keep costs and weight down. USB Wall warts initially inteneded for particular pseudo USB devices can be anything and most rely on being able to grab 100mA easily from any USB port, so the wall wart replicates that. -- Paul Carpenter | paul@pcserviceselectronics.co.uk <http://www.pcserviceselectronics.co.uk/> PC Services <http://www.pcserviceselectronics.co.uk/fonts/> Timing Diagram Font <http://www.gnuh8.org.uk/> GNU H8 - compiler & Renesas H8/H8S/H8 Tiny <http://www.badweb.org.uk/> For those web sites you hate
In article <4f59dbaa$0$6960$e4fe514c@news2.news.xs4all.nl>, usenet+5@c-
scape.nl says...
> > On 03/09/2012 10:35 AM, Paul wrote: > > >>> A classroom with other computers already there teacher does not have > >>> to hand out an extra thing. > >> > >> If the classroom already has computers, you don't need to get the > >> Raspberry Pi at all. Just let the kids use the regular computers. > > > > Yes you do as IT services will not allow loading of packet sniffers > > practicals with network addressing, subnets and a myriad of other things > > done to those systems. > > > > Bad enough finding accessible places from classrooms for ftp, telnet > > even ping practicals. > > If the IT dept is concerned about that, they would be equally concerned > about Raspberry Pi boards being loaded with packet sniffers and other > software. > > There's no fundamental difference, except that the Pi is smaller, and > doesn't come with a case. > > If you don't want the students to have full access to the computers, > just set up a virtual machine for each student. If you're worried about > the students messing up the network, configure the host machine as a > firewall, or have an external firewall to isolate the entire classroom.
Which all cost money and time, and obviously you have never seen how long it takes to get computer infrastucture changed in large organisations with limited budgets. Normally IT services are staffed by no hopers as the wages are so low, and have diktats because they read about a security problem in the National Inquirer. There are BOFH as in Bastard but more often Brainless -- Paul Carpenter | paul@pcserviceselectronics.co.uk <http://www.pcserviceselectronics.co.uk/> PC Services <http://www.pcserviceselectronics.co.uk/fonts/> Timing Diagram Font <http://www.gnuh8.org.uk/> GNU H8 - compiler & Renesas H8/H8S/H8 Tiny <http://www.badweb.org.uk/> For those web sites you hate
wzab wrote:
> Exactly. What I'm looking for is the Embedded Linux Computer with FPGA > connected directly to the CPUs bus and to 0,1" headers, so that my > students could deal with different peripherals (implemented in the > FPGA) > and connected to simple circuits assembled on the breadboard (or on > prototype PCB connected via flat cable). > Does anybody knows about such cheap boards?
No, but keep in mind here that some Linux-capable CPUs these days are pretty fancy (caches, superscalar, etc.) and have CPU busses that are non-trivial to interface too -- it's a lot more complex than the old days of data, address, and a R/W strobe. Similarly, many CPU busses are fast enough that you might not get away with ribbon cables anymore unless you're running, e.g., every other wire as ground. So just be careful to avoid those for your purposes. :-) (In a modern PC, while things like PCI cards are memory-mapped, when someone writes a line of code as simple as *MyDeviceRegister = 0xdeadbeef, what actually happens is a very complex transaction between the CPU, its local bus, the north bridge, the PCI bus, and whatever chip is on the target PCI device.)
On 3/8/2012 11:07 AM, hamilton wrote:
> On 3/8/2012 10:44 AM, Don McKenzie wrote: >> On 09-Mar-12 2:20 AM, Nico Coesel wrote: >> >>>>> I just had a look at the RPi videos, and they seem to have mounting >>>>> holes and a DC power input plug. >>>>> >>>>> So I wonder what Paul was referring to. >>>>> >>>>> Cheers Don... >>>> >>>> I don't think that it has mounting holes, DC power is via a micro USB >>>> connector. >>> >>> I see no mounting holes: >>> http://en.wikipedia.org/wiki/File:Raspberry_Pi_Beta_Board.jpg >> >> http://www.youtube.com/watch?v=TgR74Kp6Ws4 >> was the first video found by google, and it has a power plug and holes. >> >> Have removed both in later versions? >> >> Cheers Don... >> >> =================== >> >> > I think an opportunity to make a plastic case that captures this board > with holes for the connectors. (won't need holes in the board)
http://elinux.org/RPi_Cases
> > Maybe that's what they were thinking of doing, get another revenue > source flowing. >
 David> too little ram

64MB! (and promises it running Android!)

Even a programming language like Haskell, run from the command line,
needs more core than that. C++ programs can easy need 200MB+ of real RAM
to compile.

Sell it for $10 GST+shipping.inc, and you might find a market. 

 David> Part of the reason is that it is non-profit - it appeals to free
 David> software developers. It will take a lot more to persuade these
 David> people to do the same thing again, only for less exciting
 David> hardware with more limitations, and to benefit a commercial
 David> company that is making these boards for profit.

Exactly.

-- 
Chris,,

Arlet Ottens <usenet+5@c-scape.nl> wrote:

>On 03/09/2012 06:07 PM, Jan Panteltje wrote: >> On a sunny day (Fri, 09 Mar 2012 13:26:09 GMT) it happened nico@puntnl.niks >> (Nico Coesel) wrote in<4f5a042f.3990283062@news.kpn.nl>: >> >>>> I made a little 9-20V to +5 V switcher with USB connector to charge a real Chinese mediaplayer I have, >>>> that because I have plenty 12V wall warts. >>>> Adds an other 5$ at least even if DIY. >>> >>> You know about Dealextreme.com? You probably could have saved some >>> money and clear some space by throwing the 12V wall warts away :-). >> >> Oh, I forgot to mention WHY I did the separate charger... >> The PC USB port would switch itself off because of too much current, >> IIRC it is only 600mA in the spec? Not sure. >> If the strawberry pie is indeed 700mA that will promise a whole >> lot of fun on a whole lot of peesees. > >The USB 2.0 spec only allows for 500 mA maximum current after the device >has been configured. Unconfigured devices are only allowed 100 mA. For
Sorry, but that is pure theory. In reality 99% of the PCs will happily deliver >1A through their USB port. Current regulation circuitry and controlling it adds extra costs and customers might think their PC is broken because the devices are not working or are not getting charged. Manufacturers want to avoid that extra hassle. I just tested a USB port of my PC with a dummy load and it supplies over 1A without a device attached. -- Failure does not prove something is impossible, failure simply indicates you are not using the right tools... nico@nctdevpuntnl (punt=.) --------------------------------------------------------------

Memfault Beyond the Launch