Nanu nanu! Mork calling Orson on VoIP over 2G A Singaporean outfit called Gentay Communications has launched “nanu”, an ad-funded VoIP-over-mobile-data service it says will succeed where others have failed. Ad-funded phone calls are an old idea, but have hardly ever met with success. Last March, for example, free mobile network operator Ovivo went titsup. Gentay's …
And it'll also fail because 2G isn't good enough for 11Kbps down AND up. 2G is good for nothing. It cannot deliver a consistent data rate, and with the number of push messages being sent to mobiles these days, it'll be a very broken-up experience. Now if only someone could crack the voice-over-2G problem, I'm sure people will pay more of a fairly solid reliable and good-ish quality call, that they don't need to install an app for. Perhaps this is a problem that doesn't need to be solved for some reason.
Maybe after they fail, this company will invent a type of copper tubing for water transmission. They could insert this into your home plumbing to run over your existing plumbing system. No need for a new water supply. Or they could invent a sort of rubber sock for car tyres. Or a luminous lampshade.
Why would you spend money developing a proprietary compression technology that carries voice in 11 kbps when g.729 runs at 8kbps and with extensions can get to 6.4kbps?
Either way, on 2G data this isn't happening. With consistent bandwidth g.729 is good enough, but on 2G mobile data latency and jitter will be big issues. Combine that with the ever decreasing cost of just making a call from your phone and there's no incentive.
In most the world, the cost of mobile calls are declining or being bundled as unlimited packages, whereas data is becoming more restricted. That reflects usage. If, in areas where calls are still pricy, there is any sudden shift to VOIP, carriers will just change their pricing structure.
The licensing cost is likely trivial compared to the cost of developing a quality alternate codec that will have hardware decode/encode support that's necessary for decent scaling. But, if you really want to avoid the costs of licensing something that's been tried and tested, use speex which can work at similar bitrates.
I remain skeptical of any need or benefit from developing a proprietary codec.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
