I'm surprised these haven't been mentioned. I saw them on J-List some years back and ordered them from Japan. Since then, I've found JetPens to be a better supplier--it's domestic, so shipping is cheaper and faster (for NA at least).
I've compared the
I'd recommend picking up a
Okay, so a company that works to 'serve the community' produces a CPU card in a PCMCIA form factor (though which is electronically incompatable) chose a very inexpensive Chinese processor for their first project. The CPU is 3x the speed of a Rasberry Pi. It has some GPL code provided by the CPU manufacturer--who seems very cool to the OSS movement.
The schematic and layout are out for this card. There is code coming. There will be boards coming. The BOM is $15, but who knows what the shipping cost will be? With shipping, it might not matter what the BOM and sales price will be.
This could be interesting, but we know way too little to make any meaningful statement at this time.
I recently read MS spends $500 on advertising for each phone sold.
No, that figure is wrong, it's $250, they sold another phone.
According to: http://techreport.com/articles.x/19514 the peak FLOP should be the same between a BD 'module' and a SNB 'core' if the BD is using FMA4/AVX and the SNB is using plain AVX.
To get maximum performance, you're going to have to code in assembly or use a library that's been coded that way. I expect programs like Prime95 will be first adopters of this.
Supposedly, Haswell (the full tick after IVB) will have FMA3/AVX which should double the FLOP rate and surpass BD, but that's some time out, so we'll have to see what BD does in the mean time. By then, we could see a shrink of BD with more 'modules' or clock speed improvements. Best to worry about those eggs at least until they're laid if not hatched.
Your opinion of the sensability of the term changes nothing. That is the term and it has a meaning. Noone asserted that it was a connection between spatially separate 'leaf' nodes, so your point is moot. You're inventing an arguement to support something that wasn't asserted in the first place.
Yes, storage is getting cheaper per unit, lower power per unit, and denser per unit. I can only assume you're not aware of Kryder's Law. Yes, it's not by our man Gordon, but it's the same kind of power law relation. It's inaccurate to say that Moores law has 'absolutely nothing to do with
I'd suggest you learn a bit more about statistics before you make the assertions that you have with regards to the growing diversity of Netflix's customer base and the movies they serve to them. For one, research the term erlang and reassess your comments in that light.
Please stop before you embarras yourself any further.
"Last Mile" is an industry term to mean the connection between an ISP and their customers. It's common usage and not a literal expression. Yes, it's different for DSL and cable, but the point of the term is that the network fans out near the leafs and the cross section bandwidth gets very large. Please stop trying to read anything else into it.
True, caching doesn't eliminate long haul bandwidth, but it can lower it by several orders of magnitude, which is sufficient to make it neglegable. Though space in data centers is expensive, data storage gets cheaper, denser, and lower power with time. See 'Moores Law'. Bits get cheaper to store and transmit with time.
As the supply of movies gets more diverse and so does the demand for them, different layers of the caching will bear the burder, but, the same rules still apply--data gets cheaper to store and transmit with time.
An authority is a person who can tell you more about something than you really care to know.