On a tangential note, are they ever going to update this thing with a faster SoC? Its been a year since the design was finalized. I just saw an android dev board that blows this thing away for not much more. Not sure what the allure here is at this point. What's the Pi's roadmap, if any?
In your efforts to RE the GPU, have you uncovered any hope of getting faster I/O? I'd love to use all that GPU power, but the applications I have in mind require I/O to match (at the very least, solid 150 mbits/s USB2.0 and 25mbits/s Ethernet).
The only options I can think of for fast IO would be exploiting the MIPI CSI and DSI interfaces. I think you can get 2 Gbps input on CSI and same on DSI for output. You'd need some fairly specialist interface chips.
What would be nice in a future SoC would be say 1 Gbps Ethernet or USB3 right into the core of the beast. The current USB situation is rather disappointing - I would imagine the team would be avoiding the current IP in a hurry on future devices.
I've seen a number of comments like yours, mostly on Slashdot, and they always end with the the same "for not much more". But never "for the same price or less". The RPi foundation is focusing on price first, and to cram as much into that budget as possible. Which is exactly why these things are flying off the shelves as fast as they can make them -- you don't see the same demand for the "not much more" boards.
Oh I don't know, they have the best marketing/word of mouth.
Yes, the price is excellent, but like all things in tech, it should be advancing. When is the newer SoC coming out for this, if ever? Its a tempting device but currently a bit too slow for what I'd like to do with it. The newer SoC's are out there and perform well. Heck, that ARM11 chip on there has a 2004 release date!
For reference, they were taking pre-orders in Aug of 2011 and its going to be 2013 in two weeks. So, I don't think its unreasonable to ask if they're going to replace that long in the tooth broadcom SoC soon.
Also I question the utility of being bottom basement. The ODROID-U is $69.99 retail. Is $34 more worth it for tons more performance? I'd say yes, but to each his own.
To me $35 is far more 'disposable' than $70. This is why I bought a Raspberry pi a few days ago. I have no real need for it but I can imagine some (likely permanently hooked up to my TV).
At $35, if the device is never used for more than a few minutes to tinker with, it is OK by me which would not be true if the device was $70 (and twice as fast).
From what I've read, the device is capable of doing full 1080p which alone is a huge feat and opens up a lot of possibilities. One of my earliest PCs was a Pentium 2 at 400Mhz with a video card that I doubt could even do 720p video and cost $1600. The Pi is running at 700Mhz and does full 1080p so I don't feel the need for anything faster for what I need at this time.
Agreed. $35 is impulse-buy territory and $70 is "let me think about it" territory. I'm leaning towards the ODROID now for a fancy little HTPC. Figure an extra $35 bucks won't hurt me, but slow performance drives me crazy.
They're targeting education. The whole point is being able to get all the kids the same computer so that dev tools can be standardized and you can worry about learning something useful, rather than learning about driver incomptabilities.
This argument comes off as "how dare someone make a product that doesn't match what I think would be cool to have".
