What's Your Pi Plan??

I'm going to rig it to solar powered fans and try to combat global warming .Hoho .

No seriously , if it does what it says on the box , I'll purchace a batch for my local

school and hope they get as much enjoyment as I did with the Z80 base , bring

real programers back into the fold , good luck ,tecs.

Wish there were people like you that would buy these for our school.  Would love to learn all this at school instead of everything by myself.  Although, leraning by myself is fun.

Wish there were people like you that would buy these for our school.  Would love to learn all this at school instead of everything by myself.  Although, leraning by myself is fun.

I was planning to fund a complete lab for a local school but with all the issues and type of responses from RPF

the Raspberry Pi is not yet (don't know when) ready to be deployed in mass at schools.

So, while I'm waiting I'm looking at alternatives that may work better and won't be a big headache for teachers.

-J

Lots of Pi plans have been dashed by the faults, there can be little doubt of that.  Most Pi applications probably require a board that works, and works with most common devices and data 100% reliably.  The Pi is currently not that board.

At the risk of generalizing, most faults can probably be fixed one way or another in time, maybe even deep-seated problems in the BCM2835 hardware can be overcome by rewriting drivers in a way that restricts what is attempted and reduces performance if necessary, but as a trade-off eliminates unreliability by keeping allowed operations within the known safe area.  Sometimes the remedy needs to be harsh to overcome a difficult problem, the strong medicine that tastes bad as per the metaphor, but it achieves the desired end goal.

I'm trying to be optimistic about the problems there, but it's harder to be optimistic in another area:  human resources.  Because of RPF and Broadcom's obsession with closed documentation and development, the Pi's future is conditional on a very restricted set of non-community developers doing an immense amount of work, apparently without pay nor official duties but just on loan from Broadcom.  The company may be contributing manpower "for free" to combat its reputation being tarnished by the BCM2835 difficulties, but that resourcing may dry up once the incentive of adverse PR disappears.

This may become an issue once the obvious bugs are fixed (hopefully) and what remains is just software development of no further interest to Broadcom.  For example, Pi users have many times requested the following quite unambitious facilities to be provided:

• Hardware-accelerated X11.  Given that the primary purpose of Pi is as a development platform for education, this is extremely important.  People won't be writing Hello World and its descendents forever, and as their work becomes more ambitious, the slowness of unaccelerated X11 will become felt more and more.

• Hardware-accelerated media beyond the mostly unusable omxplayer.  The Pi is allegedly an extremely competent media player, yet you wouldn't think so by looking at the available set of media applications that actually work properly at reasonable speed.  Essential Linux stalwarts like mplayer and vlc don't work at all at full resolution because the lack of hardware acceleration makes the experience unusable.  RPF's choice of OpenMAX was instrumental in making Pi incompatible with the majority of media applications, and now the weight of that decision falls on RPF+Broadcom's shoulders to remedy because there is little outside interest in this corporate-backed standard for interfacing to closed media facilities.

• DSP and OpenCL functionality.  RPF regularly extoll the great power of the DSP and media core, but it's an empty promise unless people can use these things for development (the point of Pi), not just for acting as passive media consumers.  Cuirrently there is zero ability to harness that power in programs, and it won't become available until the closed RPF developer group provide it.  The open source community is powerless to help without documentation nor assistance from the "secretive inner priesthood".

These are manpower resourcing problems of RPF's own making by their choice of being closed and tied to Broadcom.  It's hard to see how the situation will get better rather than worse over time, since there would appear to be little prospect of Broadcom adding further unpaid resources as opposed to withdrawing them.

Morgaine.

how about uputer pi? it is now a little high in expense now (price on web is 69$) but sure when order quantity increase price will decrease.

Would you be willing to offer some sample units for review ? I'll be willing to pay shipping.

-J

that is the link http://www.quickembed.com/Tools/Shop/Solution/201208/260.html

please contact to apply for sample.

Yea, if you wouldn't mind, I will pay for shipping as well.  I'm a  student at a high school and some of the engineering teachers were looking at small devices like this to help teach.  I would love to try it out and let the teachers play with it a bit.

Would not mind getting one on loan to review for the linux community such as TheLinuxLinkTechShow aka TLLTS.

>Lots of Pi plans have been dashed by the faults

Lots ? Could you name a few just for the record ?

>Most Pi applications probably require a board that works

The only problem I've had is reseting when a USB peripheral is plugged in. Very annoying but easy to work around (plug it in before you power it up).

>and data 100% reliably.  The Pi is currently not that board.

When you invent this "board" the world will beat a path to your door. As of this moment in the real world it does NOT exist.

I wanted to make a wifi extender, but it simply wasn't working stable with 2 wifi sticks.

My dlna audio player project failed from the poor analog out quality.

My thin client alternative at work failed because the Pi doesn't work with our 4 serial port ftdi devices.

All embedded projects will require a way to proper shutdown in case of power failure. Needless to say it's a lot of work with the Pi.

I wanted to create gpio modules with an usb interface, so that they could be used on pc's and arm hardware like the pi.

Doesn't look like a good idea at the moment for the Pi either.

It's not all bad. I used a linksys with tomato software as access point and the Pi as wireless client for my wifi extender.

I hope it's not unpowered 2 much as there is no sd card corruption prevention at the moment...

I could succesfull connect an usb scanner to the Pi and use it's driver on a windows pc using usb2net.

A flashing led on the gpio works fine as well.

But,so far, more failure stories than successfull ones. Am I expecting 2 much?