It's great to see such projects being created even while so few have access to the beta boards. I'm very interested to see what the community produces once the first batch are sold and in the hands of users.
I'm really disappointed there wasn't more information on what he was running, this is something I'd love to set up on my own, with or without the Rasperry Pi.
On the theme of a very rewarding change of perspective and exciting new possibilities, I'd heartily recommend everyone try their hand at developing for a small, embedded microcontroller platform. I don't know why sequencing and flashing LEDs is so satisfying.
That being said, I am excited about the possibilities of Raspberry Pi's, and hope they can meet their ambitious price point and stay well stocked. The design is impressive, and I hope they find equal success with the logistics and supply chain management side of the equation.
I'm slowly working on my first electronics project. I decided to go big though and I now have 200 RGB LED modules (from 5 strings of GE15 Christmas lights) hot glued into a 20x10 matrix. I'm still nervous about blowing something out though, so I'm hesitant to wire up the power. Eventually it will be a connected to a Arduino Uno with an Ethernet shield. :-) Fun hands-on project for this software guy.
I don't understand why the rpi is so special. Any half-decent hardware engineer could have built this. They just optimized the BOM to get it to $25 bucks? Someone explain this to me.
You have projects like http://beagleboard.org/ (beagle xm $150 and $89 beagle bone), and http://leaflabs.com/devices/ ($45 maple), which are all ARM based devices, that have been around for a while. I just don't understand how these guys got it so cheap, which is why it seems like they just optimized the bom to get it to where it is. I'll hand it to them though, that is an impressive feat.
Part of being so cheap might be that the main SoC is subsidized by Broadcom. I have no problem with that, TI is doing the same for the Pandaboard (http://www.pandaboard.org)
The big thing going for them price-wise is that due to them being a charity, Broadcom is selling them the SoC at prices you would normally only get when buying millions of them, rather than the thousands that they will actually end up buying. Plus I think a lot of the engineers etc donated their time, again because it is an educational charity, thus their R&D was probably very cheap.
Thus we end up getting a device that is smaller and more capable than the Beagleboard XM (CPU- and graphics-wise) at one sixth of the price.
See I didn't know this. Now it makes sense why they were able to get it so cheap. Getting a deal like that w/ Broadcom is pretty difficult to do imagine.
Well, for one thing, they have the advantage of launching their product later.
For another, they have access to good pricing on a capable SoC that isn't usually available in small volumes at any price. In other words, they knew the right people.
And yeah, they've been very aggressive about other costs too. For example, the alignment of the various connectors on the board are all funkey, because it was cheaper that way.
"just optimized the bom" is really easy to say and hard to do. If it was so easy to cut down the bill of materials of what was previously $150+ (Beagle) to $35 (Raspberry), wouldn't TI have done so with the Beagle?
Also, it's like saying that DarkShikari et al. "just optimized the code" to make x264 the best easily-available encoder of H.264 on the planet.
Finally, the Raspberry Pi is tremendously more powerful than the Maple.
24 comments
[ 5.0 ms ] story [ 111 ms ] threadThis is the related code: https://github.com/albertz/shairport/
snos has the correct link above, from the beginning of the video, it looks like he's running airplay-nmt: http://code.google.com/p/airplay-nmt/
That being said, I am excited about the possibilities of Raspberry Pi's, and hope they can meet their ambitious price point and stay well stocked. The design is impressive, and I hope they find equal success with the logistics and supply chain management side of the equation.
Thus we end up getting a device that is smaller and more capable than the Beagleboard XM (CPU- and graphics-wise) at one sixth of the price.
For another, they have access to good pricing on a capable SoC that isn't usually available in small volumes at any price. In other words, they knew the right people.
And yeah, they've been very aggressive about other costs too. For example, the alignment of the various connectors on the board are all funkey, because it was cheaper that way.
Also, it's like saying that DarkShikari et al. "just optimized the code" to make x264 the best easily-available encoder of H.264 on the planet.
Finally, the Raspberry Pi is tremendously more powerful than the Maple.
This is one of my favorite videos: http://www.youtube.com/watch?v=f9iIgQN5uZE