🚀 go-pugleaf

"Carl Zwanzig" <zbang@radix.net> wrote

> What are stations using these days for wireline STL? Direct T1? ISDN? Any 
> IP-based systems? (Would you trust "the network"?) What about backup 
> links? Anyone still have analog lines?

Well, apparently in some parts of the country where Verizon (former General 
Telephone) has wireline service, they want to eliminate ISDN.  This is 
causing engineers to scramble who are dependent on ISDN for things like STLs 
and remote broadcasts.

The big issues is the latency and flakiness of IP.  ISDN works fine for 
remotes when you can set up one each way and the announcer at the remote can 
hear the station's audio in realtime.  But you can't do that with IP.  The 
packets are simply delayed too long on all but the most robust connections.

I have no idea what got into Verizon but I guess they figure they're not 
making enough money on it to maintain its infrastructure.

On 3/27/13 10:01 AM, Carl Zwanzig wrote:
> I've been out of this for a while...
>
> What are stations using these days for wireline STL? Direct T1? ISDN?
> Any IP-based systems? (Would you trust "the network"?) What about backup
> links? Anyone still have analog lines?

It's a mix of Analog-FM and digital microwave on 950, T1 lines (though 
those are less well maintained by the carriers as time goes on), and 
some other stuff.  ISDN is only used for backup facilities as far as i 
know.

The posher places are using digital microwave AFAIK.  KFJC is still on 
analog FM.  KKUP as well.

There is one local company that just set up a private broadband digital 
microwave (TDM-plus-packet) from HQ to all their sites, with the hub on 
a prominent mountain in the area.  The carrier-based T1 lines will be 
all gone when that goes live.

In article <kiu3pq$tsf$1@dont-email.me>, Carl Zwanzig <zbang@radix.net> 
wrote:

> I've been out of this for a while...
> 
> What are stations using these days for wireline STL? Direct T1? ISDN? Any 
> IP-based systems? (Would you trust "the network"?) What about backup links? 
> Anyone still have analog lines?

At my little station group, we will shortly be using our own digital 
microwave network to carry programming to main transmitters and 
boosters. No AT&T. No Verizon. No crappy Moseley toy UHF gear or lame 
digital STLs. We're talking the real thing: 300Mbps, bi-directional 
(symmetrical, of course), IP and native TDM, and it is all ours. No 
monthly charges, on licensed bands that are far better coordinated than 
any Part 74 nonsense. 

I'm not at all sorry to kiss all those ancient means goodbye. The 
equipment is already in place and the licenses granted. We're just 
taking care of last minute details regarding our new LAN.

-- 
John Higdon
+1 775 253 3838

On 4/6/13 9:12 AM, John Higdon wrote:
> In article <kiu3pq$tsf$1@dont-email.me>, Carl Zwanzig <zbang@radix.net>
> wrote:
>
>> I've been out of this for a while...
>>
>> What are stations using these days for wireline STL? Direct T1? ISDN? Any
>> IP-based systems? (Would you trust "the network"?) What about backup links?
>> Anyone still have analog lines?
>
> At my little station group, we will shortly be using our own digital
> microwave network to carry programming to main transmitters and
> boosters. No AT&T. No Verizon. No crappy Moseley toy UHF gear or lame
> digital STLs. We're talking the real thing: 300Mbps, bi-directional
> (symmetrical, of course), IP and native TDM, and it is all ours. No
> monthly charges, on licensed bands that are far better coordinated than
> any Part 74 nonsense.
>
> I'm not at all sorry to kiss all those ancient means goodbye. The
> equipment is already in place and the licenses granted. We're just
> taking care of last minute details regarding our new LAN.
>
Two questions, purely for curiosity:

1. Is it fully redundant? Not just the gear, but also the pathing, such 
that there's no single point of failure that can leave one or more 
locations dead in the water?

2. What's the payback period, and when you hit it, what's the 
anticipated savings?

In article <51606ce1$0$52777$742ec2ed@news.sonic.net>,
 Neil <weiserguy@sonic.net> wrote:

> 1. Is it fully redundant? Not just the gear, but also the pathing, such 
> that there's no single point of failure that can leave one or more 
> locations dead in the water?

Yes, we have redundancy but it is a fall-back just to keep things from 
going silent. But that redundancy constitutes last resort backup. The 
system itself is hardened to a great degree against weather, electrical, 
and single radio failure. It's a pretty tough bird to kill.

But are you aware that not only is any failure in the system far less 
likely than the failure of a telco T1 circuit, but the repair of any 
failure can be accomplished far faster than it takes to get telco to 
restore a downed T1. If we could live with telco, we will thrive with 
this when it comes to reliability.

> 2. What's the payback period, and when you hit it, what's the 
> anticipated savings?

The payback period is around two years. After that, the monthly cost 
(for rent at one relay site) is down in the noise, so to speak.

-- 
John Higdon
+1 775 253 3838

I see the topic is STL and local delivery but what about long haul
real time content delivery?

If ISDN is going away what about those hosts who do their shows
remotely...often times from another state.  How will they get their
content back to HQ if not by ISDN?  Are there national hosts doing
their show via IP?








On Sun, 7 Apr 2013 01:31:17 EDT, John Higdon <higgy@kome.com> wrote:

>In article <51606ce1$0$52777$742ec2ed@news.sonic.net>,
> Neil <weiserguy@sonic.net> wrote:
>
>> 1. Is it fully redundant? Not just the gear, but also the pathing, such 
>> that there's no single point of failure that can leave one or more 
>> locations dead in the water?
>
>Yes, we have redundancy but it is a fall-back just to keep things from 
>going silent. But that redundancy constitutes last resort backup. The 
>system itself is hardened to a great degree against weather, electrical, 
>and single radio failure. It's a pretty tough bird to kill.
>
>But are you aware that not only is any failure in the system far less 
>likely than the failure of a telco T1 circuit, but the repair of any 
>failure can be accomplished far faster than it takes to get telco to 
>restore a downed T1. If we could live with telco, we will thrive with 
>this when it comes to reliability.
>
>> 2. What's the payback period, and when you hit it, what's the 
>> anticipated savings?
>
>The payback period is around two years. After that, the monthly cost 
>(for rent at one relay site) is down in the noise, so to speak.

In article <nhh6m85ffbtm8jtphuormj7bvlq07ocqea@4ax.com>,
 joewo <noemail@noemail.org> wrote:

> I see the topic is STL and local delivery but what about long haul
> real time content delivery?
> 
> If ISDN is going away what about those hosts who do their shows
> remotely...often times from another state.  How will they get their
> content back to HQ if not by ISDN?  Are there national hosts doing
> their show via IP?

Long haul is routinely done via IP, and yes, there is content regularly 
distributed via the Internet that ends up being broadcast.

-- 
John Higdon
+1 775 253-3838

"joewo" <noemail@noemail.org> wrote

> I see the topic is STL and local delivery but what about long haul
> real time content delivery?

Everything -- EVERYTHING is moving to IP.  Think about it.  Every piece of 
equipment ever made and ever thought of that needs any kind of intelligence 
has/will have its unique IPv6 addresss.  Name it: printers, traffic signals, 
intestinal probes, gas spectrometers, cell sites, cell phones, cameras, 
sound and video recorders, automotive air pollution test stations --  
everything.

In time the IP infrastructure will get to the point where packets move 
faster and through more direct paths so that delays become less and less and 
are wiped out enough that the ear won't hear them.

Buildings today are being built and remodeled for ethernet and/or fiber, not 
for telco lines.

While a lot of broadcasting is still routed via ISDN, it's being replaced 
daily.  And as said here or elsewhere (maybe I said it?) Verizon (the wired 
company, ex-GTE) is not selling anymore ISDN in some of its service areas 
right now.

On Apr 8, 11:39 pm, "David Kaye" <sfdavidka...@yahoo.com> wrote:
> "joewo" <noem...@noemail.org> wrote
>
> > I see the topic is STL and local delivery but what about long haul
> > real time content delivery?
>
> Everything -- EVERYTHING is moving to IP.  Think about it.  Every pie
ce of
> equipment ever made and ever thought of that needs any kind of intelligen
ce
> has/will have its unique IPv6 addresss.  Name it: printers, traffic sig
nals,
> intestinal probes, gas spectrometers, cell sites, cell phones, cameras,
> sound and video recorders, automotive air pollution test stations --
> everything.
>
> In time the IP infrastructure will get to the point where packets move
> faster and through more direct paths so that delays become less and less
and
> are wiped out enough that the ear won't hear them.
>
> Buildings today are being built and remodeled for ethernet and/or fiber,
not
> for telco lines.
>
> While a lot of broadcasting is still routed via ISDN, it's being replaced
> daily.  And as said here or elsewhere (maybe I said it?) Verizon (the w
ired
> company, ex-GTE) is not selling anymore ISDN in some of its service areas
> right now.

I don't recall "network congestion" affecting ISDN, but it does affect
IP.

In article <kk0a6i$tu0$1@dont-email.me>,
 "David Kaye" <sfdavidkaye2@yahoo.com> wrote:

> While a lot of broadcasting is still routed via ISDN, it's being replaced 
> daily.  And as said here or elsewhere (maybe I said it?) Verizon (the wired 
> company, ex-GTE) is not selling anymore ISDN in some of its service areas 
> right now.

The date for that cutoff of new ISDN orders is in May (as was discussed 
right here at the NAB Convention today). We're talking about dial-up BRI 
service, not point-to-point PRI service, which many broadcasters still 
use for STL duty. Ultimately, that service is going away, not because 
telcos stop offering it, but because broadcasters ultimately will not 
tolerate the cost and the increasingly poor reliability of these 
circuits.

We will continue to employ PRI circuits (otherwise referred to as TDM) 
for the foreseeable future. However the service will be provided by our 
own equipment, owned and maintained by us. 

Telco no longer cuts the mustard. By the way, this state of affairs is 
something I publicly predicted more than five years ago. A lot of people 
were surprised to find that no, Ma Bell will not be providing acceptable 
communications service forever. A lot of srambling is now happening as a 
result.

-- 
John Higdon
+1 775 253 3838

"spamtrap1888" <spamtrap1888@gmail.com> wrote

> I don't recall "network congestion" affecting ISDN, but it does
> affect IP.

Agreed, but ISDN is a real circuit and IP is a virtual circuit.  Just about 
everything you can think of is moving from real to virtual.  Two-way radio 
networks have moved from real channels to virtual channels (trunking). 
Telco phone numbers haven't reflected real phone lines in several decades. 
It's simply more cost-effective to switch data via trunking than to use 
dedicated lines.

IP will get better because customers will demand it.  Network congestion 
will eventually be gone.  Comcast has upped its throughput at least twice in 
the 7 years I've been using them for internet.  Used to be that a ping from 
SF to bbc.co.uk took about 300ms.  Now it takes about 159ms.  That's almost 
double the speed.

There are several STL units on the market using IP.  There, of course, are 
various streaming codecs designed for broadcast.  They'll both fall flat 
with today's IP because of congestion.  But give it time.

John Higdon <higgy@kome.com> writes:


>The date for that cutoff of new ISDN orders is in May (as was discussed 
>right here at the NAB Convention today). 

Are there any cites for that? 
-- 
A host is a host from coast to coast.................wb8foz@nrk.com
& no one will talk to a host that's close........[v].(301) 56-LINUX
Unless the host (that isn't close).........................pob 1433
is busy, hung or dead....................................20915-1433

In article <kk43jr$her$2@reader1.panix.com>,
 David Lesher <wb8foz@panix.com> wrote:

> John Higdon <higgy@kome.com> writes:
> 
> 
> >The date for that cutoff of new ISDN orders is in May (as was discussed 
> >right here at the NAB Convention today). 
> 
> Are there any cites for that?

Nope, just something heard in passing from people who knew. It isn't a 
topic of enough interest to me to have compiled a bibliography.

:-)

-- 
John Higdon
+1 775 253 3838

On 4/10/13 9:42 AM, David Kaye wrote:
> ...Telco phone numbers haven't reflected real phone lines in several decades.
> It's simply more cost-effective to switch data via trunking than to use
> dedicated lines.
>
Well, not exactly. The telephone numbers that we dial are one level of 
abstraction from the actual circuit numbers that are used to identify 
the physical path, much like a URL gets translated into an IP address 
through a DNS inquiry. Someone who has a hard-wired telco circuit all 
the way to the CO could just as easily still have that circuit 
identified by the tel#, but someone (like me) who's been transitioned 
from copper to the CO over to a remote terminal (copper to the RT and 
then fiber between the RT and the CO) can have the change made largely 
in software. (Yeah, I do realize there are a couple of hard wiring 
changes needed too.)

Further, this system facilitates number portability. If you want to drop 
telco for a cell phone and port your number over to (say) VZW, this 
makes it easy to do, compared with your phone number identifying one 
specific hard circuit.

> IP will get better because customers will demand it.  Network congestion
> will eventually be gone.  Comcast has upped its throughput at least twice in
> the 7 years I've been using them for internet.  Used to be that a ping from
> SF to bbc.co.uk took about 300ms.  Now it takes about 159ms.  That's almost
> double the speed.
>
But only a part of that is the local loop to Comcast's headend. Much 
more of it is the infrastructure behind the scenes, the backbone 
networks, the core routers, the Beeb's infrastructure of network and 
servers, etc. (All facilitated by the Great Infrastructure Buildout of 
the late 1990's, followed by the Dot Bomb bust.)

"Neil" <weiserguy@sonic.net> wrote

> Well, not exactly. The telephone numbers that we dial are one level of 
> abstraction from the actual circuit numbers that are used to identify the 
> physical path, much like a URL gets translated into an IP address through 
> a DNS inquiry.

Well, not exactly.  When I was in the call center business eons ago, we had 
DID (direct inward dial) lines, which were trunks.  We had 15 or so of these 
lines but thousands of phone numbers.  Dial a number in a block that was 
assigned to us and it would come in on the next available trunk.  I can't 
remember the exact blocks, but I believe we had 415-995-2500 through 2699 
and 2800 through 2999.  And then 415-267-3000 through 3399, and I forget the 
rest.  So, though we had thousands of numbers we had only 15 actual talk 
paths.

Our customers had either DCF (delayed call forwarding where the call would 
forward from their number to ours after X number of rings, or they used 
simple call forwarding into the number we assigned them.

This was over 20 years ago and it wasn't even new technology then.

On 4/10/2013 9:42 AM, David Lesher wrote:
>
> Are there any cites for that?
>
I received the info on a list server for broadcast engineers.  Several 
engineers on the East Coast say Verizon in New York and Maryland will 
discontinue taking orders for ISDN this May.  Apparently this is cited 
in Talkers Magazine, although I do not have a copy and it doesn't appear 
to be online anywhere.

So no solid documentation that I can provide, but it's being discussed 
by credible individuals.  We switched from ISDN last year and never 
looked back:

http://www.thebdr.net/articles/audio/codecs/RI-ZIPOne.pdf

We get around the congestion problem by increasing the buffer size.  I 
realize that's not practical for some, but it works fine for our broadcasts.

Dave B.

On 4/10/2013 9:42 AM, David Lesher wrote:
>
> Are there any cites for that?
>
Aha.  Found it.

http://www.talkers.com/2013/03/28/verizon-no-longer-taking-orders-for-isdn-service-in-northeast-starting-may-18/#more-17845

Dave B.

On 4/10/13 10:49 PM, Dave Barnett wrote:

> We get around the congestion problem by increasing the buffer size.  I
> realize that's not practical for some, but it works fine for our
> broadcasts.

It gets to the point where you have to do the remotes as a "blind 
broadcast" and go by the clock or countdowns (or talk on the cell 
phone).   Not as handy as full-interactive but it gets the job done.

On 4/11/2013 12:33 PM, Eric Weaver wrote:
>
> It gets to the point where you have to do the remotes as a "blind
> broadcast" and go by the clock or countdowns (or talk on the cell
> phone).   Not as handy as full-interactive but it gets the job done.
>
Indeed.  With profanity delay and/or HD radio the luxury of monitoring 
live is gone anyway.  Our programmers have become quite adept at 
countdowns.  The other method is to use bumper music from the studio and 
let the remote people talk over it for the first couple of minutes, at 
which point the board op just fades out the music and the show runs from 
the remote site.

Dave B.

On Thu, 11 Apr 2013 15:33:17 EDT, Eric Weaver <weav@sigma.net> wrote:

>It gets to the point where you have to do the remotes as a "blind 
>broadcast" and go by the clock or countdowns (or talk on the cell 
>phone).   Not as handy as full-interactive but it gets the job done.
>
When we did remotes in the '50s - before Marti units - we had two
phone lines - one for the program and an order wire to the studio.

Phil Kane
Beaverton, OR

On 4/11/13 1:58 PM, Phil Kane wrote:
> On Thu, 11 Apr 2013 15:33:17 EDT, Eric Weaver <weav@sigma.net> wrote:
>
>> It gets to the point where you have to do the remotes as a "blind
>> broadcast" and go by the clock or countdowns (or talk on the cell
>> phone).   Not as handy as full-interactive but it gets the job done.
>>
> When we did remotes in the '50s - before Marti units - we had two
> phone lines - one for the program and an order wire to the studio.

And indeed even well into the Zephyr-using days I have ordered up 
dial-tone lines to go along with the ISDN, for order-wire doubling as 
last-ditch backup (Winter Music Conference remote for Energy comes to 
mind).  Couldn't run a DJ live mix on that but could at least do 
interviews.