I advocate the use of wideband codec because it is more robust against packet loss (contributed by line error, delay and jitter buffer starvation). So in my thinking, the fact that PSTN is only 3 kHz audio is immaterial. I didn't know (no confirmation) they are using iSAC. Since iSAC is rated for maximum 32 kbps, bandwidth consumption can not be the reason.
By the way, wideband codec was planned for Narrowband ISDN as early as 83-84. The switching fabric was the impediment at that time.
I advocate the use of wideband codec because it is more robust against packet loss (contributed by line error, delay and jitter buffer starvation). So in my thinking, the fact that PSTN is only 3 kHz audio is immaterial. I didn't know (no confirmation) they are using iSAC. Since iSAC is rated for maximum 32 kbps, bandwidth consumption can not be the reason.
By the way, wideband codec was planned for Narrowband ISDN as early as 83-84. The switching fabric was the impediment at that time.
The statement that Skype is using iSAC is in the article I linked to in the last paragraph. Though GIPS has a white paper from 10/14/2004 on their website saying that Skype is using iLBC (a full 4kHz codec).
It seems to me that the packet loss concealment algorithm is more important to resilience against packet loss than the bandwidth of the codec.
The statement that Skype is using iSAC is in the article I linked to in the last paragraph. Though GIPS has a white paper from 10/14/2004 on their website saying that Skype is using iLBC (a full 4kHz codec).
It seems to me that the packet loss concealment algorithm is more important to resilience against packet loss than the bandwidth of the codec.
Commenting by HaloScan
