How far can custom asics and hand coding take us?


#1

When Onlive did their 2-3$ Armada based custom asic with some hand coding it could handle streaming real time games better than a high end PC. I know it might be hard to comment but how far could such an approach take MaidSafe?

I get that PCell is asymmetric and not a good candidate to augment mesh. I also understand more and more that current mesh approaches simply aren’t good enough to straight cut the cords or even support what we’d consider good quality core info sharing applications- the latency is killer. If I’ve read the claims right a single PCell wall mount antenna can serve many devices with 1mw of power across 4 square miles with sub millisecond latency. For phones that would include regular LTE toll pipe upload so maybe that lower latency is for devices that are wired to the internet but even there sub millisecond latency seems too low as the upload alone, even on wired, would be much higher than sub millisecond. PCell apparently uses a powerful PC to drive this. End result is Artemis pitches PCell as wireless fiber.

Still, maybe an interference radio or distributed MIMO could be made symmetric (cut out the wired upload and skip the current internet) and all the hardware kept in the phones themselves? Maybe such a custom asic and hand coding might be able to do the same for MaidSafe in the same units? Point is we need to cut all the cords now and do it with a device like the Indie phone, backed up by a new secure internet software approach like MaidSafe. Artemis may well have skipped trying to do a custom asic and hand tuning because it wasn’t planning on packing the processing into phones. MIT’s Media Lab predicted better mesh would happen over a decade ago and lead to cutting the cord. Maybe some predictable political BS got in the way. If this tech worked all that telcom and cable investment would be toast, but thing is it needs to be toast. I don’t want to be knave but invest in evil and fall on your sword- who willingly funds cable, seriously?

  1. Even if such tech could work would maybe it wouldn’t with out a density of phones in an area presenting a chicken and egg problem?

  2. Maybe all the spectrum is accounted for or it would be checkered. To me PCell seem set up to pander to the current cord regime. Is this because spectrum was a lost cause or because its parent is a for-profit?

  3. Maybe this leads to conflicts over spectrum. At this point these companies spectrum licenses look like BS that needs to be liquidated along with their fraudulent data cap sponsored data schemes- for instance if a firm can only reliably deliver 5mbs then let that be the advertised unlimited rate, not 20mbs with a cap. I would guess I am not alone in that assessment on liquidating licenses, especially if these firms will insist on using their spectrum in the least efficient way to be monopolistic. I thought PCell resolved this by using tiny slices of spectrum but maybe not as their approach seems to need to snuggle up to spectrum licensees.

  4. This situation is intolerable as it would seem to lead to a situation of being drown out by the megaphones of the loudest bidder again, which the internet was temporarily addressing. It reduces to a: I have more money than you so I get to tell you what to do situation. No political voice, then no rights. The point of democracy is power sharing which requires a voice to retain rights, and possibly the prime utility of power sharing is not being over a barrel over wealth.