I have a modified Seagate Dockstar[1][2] and I'm using it with an external HDD, but I think it would be cheaper using a modified NetGear Stora[3] that has 2 SAS ports than using more than one USB external HDDs (with a dockstar or a Raspberry Pi).
I think more interesting applications are in those field in which you can take advantage of the video HDMI port and the IO ports..
I had one of those. It's a very limited device, it runs some custom embedded OS (certainly not Linux) and craps out if you try to access it from two workstations at the same time, etc. etc.
Could be worth it for $50. But I think in the next few years our own smartphones will be used as consoles. All you'll need is a controller, and for some games you will only need the phone itself (mostly the accelerometer based, etc)
I've tried playing some games with an iPad2 + Airplay. There was almost always a lag that made playing from (i.e. looking at) the television screen feel very off. Still it worked to let other people watch the game as I played it by looking at the iPad. Maybe your experience was different?
I haven't used AirPlay for games yet (waiting for the iPad 3). But from my experience with a PS3, HDMI and TV post-processing can add a lot of lag to the output. You'd thought a modern standard would've taken this into consideration...
I've been playing GTAIII on my Transformer Prime using a wireless PlayStation DualShock3 controller for a week or so now. The Trans'Prime and the controller connected via Bluetooth just fine with no work from me. I could do the same with my Galaxy Nexus, though from what I can tell, it requires some messing about with root access and a "Sixaxis Controller" app. (I'm happy Asus spared me that minor annoyance with the Trans'Prime.) On his Youtube channel Christian Cantrell plays GTAIII on his Galaxy Nexus using a keyboard. Smartphones as game consoles need not be a few years off, they are already here for the geeks with a little spare time, they just haven't be polished up and made dirt simple for the masses.
Yes, MAME cabinet etc. Thanks for that pixel-kit: I own a lot of 8, 16 bit retro machines + joysticks (and mice), so I would like to get them working. I made something myself for that years ago, but it didn't work totally so I just bought the old machines themselves as that's cheaper/easier (and I think they look great :)
GuruPlug type wall-socket based server for home use. This would require a casing that takes wall-socket power and modifies it for the USB-based power the unit requires, and an adapter for ethernet-to-wifi, or to support ethernet-over-power.
Also, keep in mind that an Arduino with networking capabilities (i.e., Arduino + Ethernet shield) is $30 + $45 = $75, three times the cost of the RPi; this is the first hackable device (that I know of) to have networking capabilities for <$40.
True. Moreover, with this price and features it is stupidly trivial to attach a computer source to your TV now. No excuses not to. Attach a cheap wireless dongle (<10$) and you got yourself a versatile streamer for less than 40$[1].
[1] AFAIK this does not include the essential SD card
How many monitors support HDMI? And can you see anything with A/V resolution? No, so just make that just "ready to be connected to a newer TV or an expensive computer monitor".
Also, how many people build code to run on ARM? Not quite as many as Intel-based/AMD. Maybe 3-4%? No, maybe 2%? No, how about 1%? Not even. Oops. Fail.
A lot of the uses listed are cool, and seem worthwhile - but I thought the point of Raspberry Pi was to provide affordable computers for educational purposes?
NAS, Gaming, Home automation, and other GuruPlug-type solutions are awesome. These would be great projects offering a lot of fun and inspirational documentaries. Nonetheless, I think a focus on the educational side is appropriate. How about:
* Sites offering RPi-specific hardware instruction and troubleshooting
* StackExchange-like RPi forums for different programming languages and/or use cases.
* Inexpensive hardware bundles with input/output devices and kits for working with the bundles.
The problem isn't going to be finding something to do with a tiny computer, it's making said computer as useful as possible for children and educators. Keep it useful and easy for them, that's a problem for a startup to address.
It's a valid point of view, but I don't think that should be restrictive. They said in the blog that the intention is to subsidize and enable educational uses by making it available for everyone and all kinds of projects. The educational package is coming in September.
True, and thanks for pointing it out. I'd just note that they seem to have their hands full at the moment, and I think documentation and teaching utilities will be lacking.
The motivated and inspired will still be able to thrive, but the under-funded school may only be able to use what is easily available - and it seems like a good hacker-crowd-sourced batch of instructions could easily dwarf whatever they are planning to provide.
What you say is good and true, but it's important to remember that even behind the subsidization aspect others allude to, more uses of RasPi -> more things the child receiving it can do -> more appealing the device will be in his eyes -> dedicate more time educating himself using it. (him/her of course)
At least that was the general theme of my computer usage as a kid. Games were fun, but they also drew me to spend more time and educate myself about "controlling the beast".
I definitely agree, and I probably didn't qualify my statements enough. I just wanted to highlight some caution because a lot of times hackers want to really push hardware and technology to its limits. This is great, but to a newcomer, I think some of these projects can be so challenging that they end up driving users away.
It's a valid point that having example uses/projects that cover the widest possible array of options is ideal for inspiring the most use.
Cheap thin clients / terminals to be left 'scattered' around a uni or large office block, etc.
Sailing boat logging / navigation computers.
House / office automation.
A 'smart data-projector' box that you plug into the back of a regular projector, and can install in a ceiling-mounted projector which you can then drop files on, or pull from a server.
Warehouse / storeroom management.
Audio recording / streaming systems for churches and other conference rooms.
Car music server, which when you're home syncs with the house to download any new music (over wifi?).
Quite honestly, I'm looking forward to getting a good cheap replacement for my Logitech Revue. The device is the closest thing I've gotten to a good media center device under $100, but it falls short due to a buggy OS and massive stuttering in 1080p. I'm hoping this board + XBMC will be smooth enough to be an improvement. A Debian bundle that includes XBMC, which works out of the box on a RasPi could be the kind of simplicity that pushes XBMC into the mainstream. Yes, I'm aware of Boxee but honestly I (and many others) don't like their application either.
If you want to use the actual Rasberry Pi in a product, I think the most practical choice is a range accessorized hobbyist platforms. This would be much like the arduino and its shields except that it is possible to exploit multimedia capability.
I want a Raspberry Pi in an AppleTV-like black case, and I'll pay $99 for that, delivered in a beautiful box.
Apple has shown that a box like that sells like hotcakes but they won't do it for a low-price computer device or will lose the nice profits of the mac mini.
So here, a R-Pi in a beautiful case and box for $99.
I hope it convinces other large manufacturers like TP-Link that there is a market for little things like this that are open and we see a few more jump in.
Being able to run apache or lighttpd on it, python, ROR, etc but still staying cheap enough to "waste" one to stick in your washing machine's panel just to tell you when your socks are clean will open up "internet of things" programming to a huge number of people.
If you're doing any form of traning, put the complete development/whatever environment on a computer and give it away as part of the course. Make a business setting up pre-packaged environments for various courses and fulfilling orders for training companies. etc etc.
You could build small supercomputers with clusters of Raspberry Pis.
Design a chassis for several dozen Raspberry Pis, and package them with software for doing numerical analysis, statistical analysis, graphics rendering, OLAP cubes.
There's a huge enterprise market for small supercomputers that run SAS for data crunching. You could tap into this market, packaging your units with comparable open source software.
Relatively hack-proof internet interface for wifi security cameras. If someone breaks into the house and steals all my gear (including the cameras), at least I wouldn't be losing another $2-400 PC.
I think it can spawn off a series of "sensor" based applications:
1. Remote electricity monitor
2. In bed constant monitor for patients under constant supervision
3. In home continuous surveillance unit
etc.
55 comments
[ 3.9 ms ] story [ 128 ms ] threadAt around $50, might be cheaper to go with this. Unless you are implementing the NAS just because it can be done :-)
I think more interesting applications are in those field in which you can take advantage of the video HDMI port and the IO ports..
[1] http://ahsoftware.de/dockstar/
[2] http://jeff.doozan.com/debian/
[3] http://www.openstora.com/
* Quadrotor autopilot
* NAS
* RFID Home security device
* Webcam based security cam
* Very powerful autopilot (orders of magnitude faster than an arduino)
* Home control station connected to TV
* 3D printer controller
* Some Kinect roboter without the need for a laptop or tablet to control the thing
* Everything we use hacked wireless routers for now, but with TV-out ;-)
Have not thought about how to implement it yet, but it's the first weird idea that poped up.
Off the top of my head:
* Easy[1] support for USB devices
* Ready to be connected to TV/monitor
* 1080p output (and HDMI includes audio)
* Of course, more applications and libraries are available, but what new use cases does this enable us?
[1] as long as there are compatible Linux drivers
[1] AFAIK this does not include the essential SD card
Also, how many people build code to run on ARM? Not quite as many as Intel-based/AMD. Maybe 3-4%? No, maybe 2%? No, how about 1%? Not even. Oops. Fail.
* Home automation (hooking it up to lights, etc) with an in-browser controller
NAS, Gaming, Home automation, and other GuruPlug-type solutions are awesome. These would be great projects offering a lot of fun and inspirational documentaries. Nonetheless, I think a focus on the educational side is appropriate. How about:
* Sites offering RPi-specific hardware instruction and troubleshooting
* StackExchange-like RPi forums for different programming languages and/or use cases.
* Inexpensive hardware bundles with input/output devices and kits for working with the bundles.
The problem isn't going to be finding something to do with a tiny computer, it's making said computer as useful as possible for children and educators. Keep it useful and easy for them, that's a problem for a startup to address.
The motivated and inspired will still be able to thrive, but the under-funded school may only be able to use what is easily available - and it seems like a good hacker-crowd-sourced batch of instructions could easily dwarf whatever they are planning to provide.
At least that was the general theme of my computer usage as a kid. Games were fun, but they also drew me to spend more time and educate myself about "controlling the beast".
It's a valid point that having example uses/projects that cover the widest possible array of options is ideal for inspiring the most use.
Sailing boat logging / navigation computers.
House / office automation.
A 'smart data-projector' box that you plug into the back of a regular projector, and can install in a ceiling-mounted projector which you can then drop files on, or pull from a server.
Warehouse / storeroom management.
Audio recording / streaming systems for churches and other conference rooms.
Car music server, which when you're home syncs with the house to download any new music (over wifi?).
...
Apple has shown that a box like that sells like hotcakes but they won't do it for a low-price computer device or will lose the nice profits of the mac mini.
So here, a R-Pi in a beautiful case and box for $99.
Being able to run apache or lighttpd on it, python, ROR, etc but still staying cheap enough to "waste" one to stick in your washing machine's panel just to tell you when your socks are clean will open up "internet of things" programming to a huge number of people.
Finally, restful control of my cat feeder.
How else is it going to tell Obi Wan that he's our only hope?
/ld
Design a chassis for several dozen Raspberry Pis, and package them with software for doing numerical analysis, statistical analysis, graphics rendering, OLAP cubes.
There's a huge enterprise market for small supercomputers that run SAS for data crunching. You could tap into this market, packaging your units with comparable open source software.
Example here: http://www.falcon-nw.com/desktops