SI
SI
discoversearch

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor.  We ask that you disable ad blocking while on Silicon Investor in the best interests of our community.  If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.
Technology Stocks : Nokia (NOK)
NOK 6.910-0.9%3:59 PM EST

 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext  
To: Eric L who wrote (12895)6/20/2001 9:18:17 PM
From: 49thMIMOMander  Read Replies (1) of 34857
 
Eric, some "dirty secretes" you might want to explore.

One is the same as for ISDN connections, if one makes
an ISDN lonng distance connection, to a different operator,etc there is this chance (and was a rule earlier)
that the networks were not synchronized.

This was one thing which delayed long distance ISDN
deployment for maybe 10 years..

What happens is that with two networks "slightly out of
tune", asynchronous, "my" network might feed 63,999 bits
into "your" network, which needs 64,001 bits during
the same second.

So called "elastic buffers" (now just called buffers)
were included, but every minute or two one bit was
lost or one added, or even worse things happened.

For voice this had no importance, just a barely-audible
click every 2-3d minute.
(except for sprint who tried to use ADPCM)

Anyway, slowly larger and larger netwroks got synchronized,
but there is still this problem between, for example,
scandinavia and main europe, athough internally
synchronized.

the same is true, to a larger or lesser extent, for GSM networks, and probably especially even between
local operators!!

If i remember correctly, voice GSM has some special stuff
to deal with this, might cause a small chirp for the
voice, but main thing that the connection isn't dropped.

Packet networks (ATM,etc) bypass this problem, but add all
the other problems, latency, lost packets, buffering,etc

My point, going back to this "error correction" thing is
that in general a circuit switched connection has a
low, or minimum, delay, good for end-to-end error correction.
Long delays (like ping times for internet) makes
end-to-end error correction difficult, but also redundant,
as packets should be locally error corrected, ping-ponged
until correct.

Packet networks solve the thing by basically having
long latency (delays), time outs for lost packets,
all that thing which happens when "internet is a pain"

Ilmarinen

This is what goes on with GPRS networks (and handsets) now,
just like with WAP-SMS, those who didn't make sure they
minimized delayes,etc are now being judged as "Crappy".
Report TOU ViolationShare This Post
 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext