Ivor Jones wrote:
> "The Invalid" <the_invalid@hotmail.com> wrote in message
> news:nc8qg29aaf6rpvjrd8gtrukvpf2lhr2k2f@4ax.com
>
>>On Fri, 15 Sep 2006 11:46:25 +0100, "Ivor Jones"
>><ivor@despammed.invalid> wrote:
>>
>>
>>>>My ear
>>>>That tells me good or bad quality
>>>
>>>It is also impossible to prove. I am basically a
>>>scientist, I require proof before I accept something as
>>>fact.
>
>
>>'basically' ?
>
>
> Oh do put a space in, it's not that hard.
>
> Basically. It's a word, it's in the dictionary. Look it up.
>
>
>>>Either you can give me that proof or you can't.
>>
>>You don't believe that Skype quality is better and I
>>doubt you would accept any proof so try it yourself, But
>>I bet you still wont believe yourself
>
>
> I will believe it when you can provide me with proof. If that proof is
> scientifically obtained it will be verifiable so I will have no choice but
> to accept it.
>
> I won't hold my breath waiting.
>
> Ivor
>
>

Skype's protocols are not "open" but they have been
subjected to scientific examination. Computer scientists
from the University of Columbia, NY published a paper 2
years ago describing their experiments.

Amongst the results, they were able to measure a
pass-through of frequencies between 50Hz and 8,000Hz - twice
the bandwidth of a typical telephony codec sampling at 8KHz.
This is why a Skype call will normally sound much clearer
than other telephony (including the PSTN). Wideband codecs
of this clarity are normally only found on ISDN terminals
for the broadcasting market.

Bandwidth requirements for a call were measured at 5
Kbytes/sec (combined directions) for a number of uncongested
network conditions. This compares favourably with the
normal, low-bandwidth codecs - the PSTN G.711 codec uses
8kbytes in each direction. Intelligibility would be reduced
if there is congestion.

If you want to read the paper yourself, search for the
authors - Baset and Schulzrinne. They decribe their test
methods, so it should be possible to replicate the
experiments.