Greatings

Here's an unusual request. I'm currently working in the acoustics
engineering group for a large cell phone company. We have lots of
industry formal test requirements that most all phones pass. This
seems to have little direct correlation to percieved quality either
recieved or Sent audio. I'm interested in the opinion of trained ears.
On GSM mostly but CDMAs OK. Please If you find the time just find a
stationary known good RF spot at a non busy hour time and reply with
your type/model and opinions. I'm interested in audio not RF
performance so minimum BER/FER is needed.

Thanks
Kevin Tracy

On 5 May 2005 07:29:15 -0700, "KGT" wrote:
Here's an unusual request. I'm currently working in the acoustics
engineering group for a large cell phone company. We have lots of
industry formal test requirements that most all phones pass. This
seems to have little direct correlation to percieved quality either
recieved or Sent audio. I'm interested in the opinion of trained ears.
On GSM mostly but CDMAs OK.

I'm not sure this has anything to do with your request, but I'm
curious about it. My carrier (US Cellular, Northern California) just
switched us from TDMA to CDMA 1xRtt. On TDMA the RF performance was
dismal, but I couldn't tell from the audio whether my Ericcson T18d
was using TDMA or analog - both sounded decent on voice and music.

On the new CDMA system, with a Moto V710m, RF performance is vastly
improved. Voice is much more intelligible, which I think is due to
better AF, transducer, and ergonomics rather than any effect of the
coding scheme. But music is unbearable - sounds raspy, and like it is
bubbling up from underwater. I find myself hanging up when I'm put on
hold so I don't have to listen to it.

The bult-in MP3 player sounds amazingly good, so the horrible music
performance clearly is the coding scheme. I can run a measured 62K of
computer data through the 1xRtt link, so how can it be that they can't
do a better job of reproducing music on hold with that much bandwidth
available? I'm probably on the most lightly loaded cell in the system,
in a remote area no other company even bothers to serve, so it isn't a
matter of oversubscribing the hardware.

Loren

Loren

The CODECs in CDMA (and others technologies) are optimized for speech
at the cost of non speech. They don't like tones, music etc. This is
the price for squeezing that much bandwidth and features out of an 8k
(EVRC) coder. or for that matter a 13k.

Kevin Tracy

In article . com writes:

Here's an unusual request. I'm currently working in the acoustics
engineering group for a large cell phone company. We have lots of
industry formal test requirements that most all phones pass.

What a pity!


--
I'm really Mike Rivers )
However, until the spam goes away or Hell freezes over,
lots of IP addresses are blocked from this system. If
you e-mail me and it bounces, use your secret decoder ring
and reach me he double-m-eleven-double-zero at yahoo

Ok Mike

But given my options (daughter in college NJ mortgage/taxes/roots) and
20 years invested in Bell labs I could have landed from my 1 year in
downsized telecom hell in a lot worse places Also This was a sincere
attempt to get skilled opinions from this group.

PS I'm still a serious weekend warrior musician/engineer as I have been
for 30 years.

KT

Loren Amelang writes:
[...]
I'm probably on the most lightly loaded cell in the system, in a
remote area no other company even bothers to serve, so it isn't a
matter of oversubscribing the hardware.

Rappaport [1] states that QCELP13 is a higher-datarate speech codec
for IS-95 systems. You can also read some interesting history about
the CDMA codec evolution at

http://www.commsdesign.com/design_co...cleID=16500249

Regarding your situation, perhaps you're at the distance limits of the
cell, or the cell is operating at reduced power, thus they may be
dropping to a lower speech codec rate in order to perform more channel
coding.

Have you tried a GSM phone? Most of the GSM networks now support AMR
(Adaptive Multi-Rate), an improved variable-rate codec system.

--Randy

[1] @BOOK{rappaport,
title = "{Wireless Communications: Principles and Practice}",
author = "{Theodore~S.~Rappaport}",
publisher = "Prentice Hall PTR",
edition = "second",
year = "2002"}

--
Randy Yates
Sony Ericsson Mobile Communications
Research Triangle Park, NC, USA
, 919-472-1124

Randy

To put this back on track.. what model GSM phone do you use as a
benchmark of Good downlink SPEECH quality? Do you have one that you
rate poor?

Loren
No currently deployed CELP coder (I believe AMR Is CELP?) can reproduce
music well. Its meant for speech! No matter what the RF FER is.

All
I'm still interested in RAP opinions on phones that have surprized you
with very good or very bad SPEECH.

Thanks
Kevin Tracy

Randy

To put this back on track.. what model GSM phone do you use as a
benchmark of Good downlink SPEECH quality? Do you have one that you
rate poor?

Loren
No currently deployed CELP coder (I believe AMR Is CELP?) can reproduce
music well. Its meant for speech! No matter what the RF FER is.

All
I'm still interested in RAP opinions on phones that have surprized you
with very good or very bad SPEECH.

Thanks
Kevin Tracy

"KGT" writes:

Randy

To put this back on track.. what model GSM phone do you use as a
benchmark of Good downlink SPEECH quality?

Our Z500 sounds pretty sweet.

http://www.sonyericsson.com/z500a/

--
Randy Yates
Sony Ericsson Mobile Communications
Research Triangle Park, NC, USA
, 919-472-1124

On 06 May 2005 11:40:59 -0400, Randy Yates
wrote:

Loren Amelang writes:
[...]
I'm probably on the most lightly loaded cell in the system, in a
remote area no other company even bothers to serve, so it isn't a
matter of oversubscribing the hardware.
...
Regarding your situation, perhaps you're at the distance limits of the
cell, or the cell is operating at reduced power, thus they may be
dropping to a lower speech codec rate in order to perform more channel
coding.

I can see the tower, I get "five bars", and I'd be surprised if there
is anyone else using this particular cell at any given moment. I guess
they truly have optimized away any hope of bearable music performance.

Have you tried a GSM phone? Most of the GSM networks now support AMR
(Adaptive Multi-Rate), an improved variable-rate codec system.

As I said, there is literally no alternative here. Any other company
goes to "no service" about ten miles back toward civilization.

Loren

On 6 May 2005 10:49:03 -0700, "KGT" wrote:

Randy
To put this back on track.. what model GSM phone do you use as a
benchmark of Good downlink SPEECH quality? Do you have one that you
rate poor?

Loren
No currently deployed CELP coder (I believe AMR Is CELP?) can reproduce
music well. Its meant for speech! No matter what the RF FER is.

All
I'm still interested in RAP opinions on phones that have surprized you
with very good or very bad SPEECH.

Thanks
Kevin Tracy

Kevin,
Pardon me for persisting here, but my mention of unbearable music
performance is not a trivial aside - it is a serious complaint about
the performance of my phone. I understand you are charged with
optimizing speech performance, but since you are questioning the
existing performance standards, why not question that goal of "speech
only" as well?

My "audiophile ears" can tolerate FM radio multipath, AM radio skip
fading, and even just plain cheap distorted sound. But there is
something about the distortion of music and background sound through
the new phone that borders on pain - and I don't seem to be learning
to accommodate to it. Can I be the only one who notices this?

Try an experiment. Use your cellphone as your only phone for awhile,
not just for quick speech messages. Start noticing how often you get
stuck with music on hold, or talking to someone who is listening to
music or watching television. Ignoring music performance will limit
acceptance of cellular as an alternative to wired phones.

Loren

Loren

Wow where should I start. In my 8 years of cellular Voice quality You
may just be the first person who have verbalized that you
A) care alot about music over cells
B) dont realize why this is not practical/possible given current
network capacity/bandwidth demands.

This is not a bandwidth limited analog network where both voice and
music are similarly degraded. This is digital with VOICE codecs that
have been squeezed to there last ounce of
compression. These are codecs not A to D / D to A converters. There
only tools are to save your speech as mathematic functions that
synthesize the basic mouth resonant filters /tone generation. This is
why music sounds bad because the resynthesis is working with functions
that are limited to specific mouth/ speech recreation.

As to this slowing the sucess and universal use of wireless/ devices..
do a google on market penetration. I'm more worried about keeping up!

Hope this helps
Kevin Tracy

Randy Yates wrote:
Loren Amelang writes:
[...]
I'm probably on the most lightly loaded cell in the system, in a
remote area no other company even bothers to serve, so it isn't a
matter of oversubscribing the hardware.

Rappaport [1] states that QCELP13 is a higher-datarate speech codec
for IS-95 systems. You can also read some interesting history about
the CDMA codec evolution at


http://www.commsdesign.com/design_co...cleID=16500249

Regarding your situation, perhaps you're at the distance limits of the
cell, or the cell is operating at reduced power, thus they may be
dropping to a lower speech codec rate in order to perform more channel
coding.

Have you tried a GSM phone? Most of the GSM networks now support AMR
(Adaptive Multi-Rate), an improved variable-rate codec system.

When I got my GSM phone almost a year ago, my connectivity dropped to about
half what it is now (Nashville--Cingular). My trusty old Nokia 51xx used to
connect reliably in my home. My new Sony-Ericsson: much less so. It
matters not if I have 'bars' or not. The night I got it, I experienced an
inability to connect, as did my wife, with the same model.

I booted up the Nokia and got a connection in the same place (no actual
'service' of course, only a recording).

jak

--Randy

[1] @BOOK{rappaport,
title = "{Wireless Communications: Principles and Practice}",
author = "{Theodore~S.~Rappaport}",
publisher = "Prentice Hall PTR",
edition = "second",
year = "2002"}

KGT wrote:

Wow where should I start. In my 8 years of cellular Voice quality You
may just be the first person who have verbalized that you
A) care alot about music over cells
B) dont realize why this is not practical/possible given current
network capacity/bandwidth demands.

I can sort of understand him. Now, personally, I refuse to use cellphones
because the sound quality just drives me up the wall... I have a real hard
time telling what people are saying, and I can hear HF SSB stuff just fine.
Maybe it's the lack of sidetone that irritates me.

But I spend much of my day listening to music on hold, and I could see
that being a bad thing on a cellphone.

My wife often calls me on her cellphone from musical events, and the music
behind her turns into weird squealing and beeping on the phone, which
makes it that much more difficult to understand her voice... I think if
the music were not so weirdly altered that it would be less irritating.

But then, I have a WECO 500 set on my desk right now, which probably says
something about my attitude toward telephony. The 1A2 system is waiting
to get installed still.
--scott



--
"C'est un Nagra. C'est suisse, et tres, tres precis."

jakdedert wrote:
Randy Yates wrote:
Loren Amelang writes:
[...]
I'm probably on the most lightly loaded cell in the system, in a
remote area no other company even bothers to serve, so it isn't a
matter of oversubscribing the hardware.

Rappaport [1] states that QCELP13 is a higher-datarate speech codec
for IS-95 systems. You can also read some interesting history about
the CDMA codec evolution at



http://www.commsdesign.com/design_co...cleID=16500249

Regarding your situation, perhaps you're at the distance limits of
the cell, or the cell is operating at reduced power, thus they may be
dropping to a lower speech codec rate in order to perform more
channel coding.

Have you tried a GSM phone? Most of the GSM networks now support AMR
(Adaptive Multi-Rate), an improved variable-rate codec system.

When I got my GSM phone almost a year ago, my connectivity dropped to
about half what it is now (Nashville--Cingular). My trusty old Nokia
51xx used to connect reliably in my home. My new Sony-Ericsson: much
less so. It matters not if I have 'bars' or not. The night I got
it, I experienced an inability to connect, as did my wife, with the
same model.

Meant to say (above) 'about half what it was before...'

I booted up the Nokia and got a connection in the same place (no
actual 'service' of course, only a recording).

jak

--Randy

[1] @BOOK{rappaport,
title = "{Wireless Communications: Principles and Practice}",
author = "{Theodore~S.~Rappaport}",
publisher = "Prentice Hall PTR",
edition = "second",
year = "2002"}

(Scott Dorsey) wrote in
:

But then, I have a WECO 500 set on my desk right now, which probably
says something about my attitude toward telephony. The 1A2 system is
waiting to get installed still.

And assuming the phone company still supports pulse dialing in 50 years, it
will still work and sound better than current digital phones.

"jakdedert" writes:

Randy Yates wrote:
Loren Amelang writes:
[...]
I'm probably on the most lightly loaded cell in the system, in a
remote area no other company even bothers to serve, so it isn't a
matter of oversubscribing the hardware.

Rappaport [1] states that QCELP13 is a higher-datarate speech codec
for IS-95 systems. You can also read some interesting history about
the CDMA codec evolution at


http://www.commsdesign.com/design_co...cleID=16500249

Regarding your situation, perhaps you're at the distance limits of the
cell, or the cell is operating at reduced power, thus they may be
dropping to a lower speech codec rate in order to perform more channel
coding.

Have you tried a GSM phone? Most of the GSM networks now support AMR
(Adaptive Multi-Rate), an improved variable-rate codec system.

When I got my GSM phone almost a year ago, my connectivity dropped to about
half what it is now (Nashville--Cingular). My trusty old Nokia 51xx used to
connect reliably in my home. My new Sony-Ericsson: much less so. It
matters not if I have 'bars' or not. The night I got it, I experienced an
inability to connect, as did my wife, with the same model.

I booted up the Nokia and got a connection in the same place (no actual
'service' of course, only a recording).

jak

jak,

Which model Sony Ericsson are you using?
--
% Randy Yates % "My Shangri-la has gone away, fading like
%% Fuquay-Varina, NC % the Beatles on 'Hey Jude'"
%%% 919-577-9882 %
%%%% % 'Shangri-La', *A New World Record*, ELO
http://home.earthlink.net/~yatescr

Carey Carlan wrote:
(Scott Dorsey) wrote in
:


But then, I have a WECO 500 set on my desk right now, which probably
says something about my attitude toward telephony. The 1A2 system is
waiting to get installed still.


And assuming the phone company still supports pulse dialing in 50 years, it
will still work and sound better than current digital phones.


I certainly hope they do. I've got a nearly mint AE 50 that I'm going
to hang on the wall at the hotel.

My Motorola V60 sounds better than any of the Ericsons I've had. I'm on
AT&T/Cingular service.

I have no idea what the oher stuff you asked about is as I use it as a
phone and require nothing else from it.

Eric

"KGT" wrote:

All
I'm still interested in RAP opinions on phones that have surprized
you with very good or very bad SPEECH.


LG5450 on Telus CDMA in Vancouver, BC is bad. I assume it's the phone
and not the network because:

1. I had a Motorola T7something before that sounded better (assuming you
had the earpiece positioned EXACTLY right -- otherwise you'd hear
nothing at all). Still bad, but better

2. My wife has the same LG phone I have, and intelligibility is even
worse when we're talking cell-to-cell (as opposed to cell-to-landline).

Is bandwidth really so limited that we have to put up with such awful
sound? Even worse, the constant noise gating that makes conversation
more like using a CB radio than full-duplex?

--
"It CAN'T be too loud... some of the red lights aren't even on yet!"
- Lorin David Schultz
in the control room
making even bad news sound good

(Remove spamblock to reply)

Randy Yates wrote:
"jakdedert" writes:

Randy Yates wrote:
Loren Amelang writes:
[...]
I'm probably on the most lightly loaded cell in the system, in a
remote area no other company even bothers to serve, so it isn't a
matter of oversubscribing the hardware.

Rappaport [1] states that QCELP13 is a higher-datarate speech codec
for IS-95 systems. You can also read some interesting history about
the CDMA codec evolution at



http://www.commsdesign.com/design_co...cleID=16500249

Regarding your situation, perhaps you're at the distance limits of
the cell, or the cell is operating at reduced power, thus they may
be dropping to a lower speech codec rate in order to perform more
channel coding.

Have you tried a GSM phone? Most of the GSM networks now support AMR
(Adaptive Multi-Rate), an improved variable-rate codec system.

When I got my GSM phone almost a year ago, my connectivity dropped
to about half what it is now (Nashville--Cingular). My trusty old
Nokia 51xx used to connect reliably in my home. My new
Sony-Ericsson: much less so. It matters not if I have 'bars' or
not. The night I got it, I experienced an inability to connect, as
did my wife, with the same model.

I booted up the Nokia and got a connection in the same place (no
actual 'service' of course, only a recording).

jak

jak,

Which model Sony Ericsson are you using?
The cheapest one I could get--free actually--with a 2-year agreement. It's
the T226. I researched the handsets Cingular would give me for the price of
re-upping my service, and this one had the best reviews...sound quality,
features, range etc. Since then I've used others' phones on Cingular and
have found a few with better (all the above).

jak

jak

On 6 May 2005 13:35:57 -0700, "KGT" wrote:

Loren

Wow where should I start. In my 8 years of cellular Voice quality You
may just be the first person who have verbalized that you
A) care alot about music over cells
B) dont realize why this is not practical/possible given current
network capacity/bandwidth demands.

Kevin,

I'm familiar enough with coding to understand the choices the industry
has made, and the reasons for them. But those reasons are evaporating,
and I'm questioning the current direction of industry priorities. As I
said, even here in "Northwest Nowhere", I can use a 64K data pipe via
this same phone. If you believe the portable music hype, that is
capable of CD quality stereo music. In the city the cellular carriers
are heavily advertising their new full-motion video servies.

If there is now enough cellular bandwidth available for people to
watch video on demand, why can't we redirect a little bit of it toward
making the inevitable music during a voice call somewhat less toxic? I
bet if you gave your customers a "voice quality" option, they would
choose that long before the ability to watch an itty-bitty TV picture.

Loren

Loren Amelang wrote:

If there is now enough cellular bandwidth available for people to
watch video on demand, why can't we redirect a little bit of it toward
making the inevitable music during a voice call somewhat less toxic? I
bet if you gave your customers a "voice quality" option, they would
choose that long before the ability to watch an itty-bitty TV picture.

I don't think so. I suspect very few people would actually pick good
quality voice and reliable services over fancy new features.

And that, in short, is why the whole industry has fallen apart since
deregulation. The Bell System, which built itself on reliability and
quality, has ceased to be relevant in a world where users choose poor
reliability and quality.

After all, the current analogue cellphone infrastructure is pretty reliable
right now, and sounds pretty good. But when was the last time you saw anyone
walking around with a bag phone? People don't want the trouble and expense
that come along.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

Loren Amelang wrote:

As I
said, even here in "Northwest Nowhere", I can use a 64K data pipe via
this same phone. If you believe the portable music hype, that is
capable of CD quality stereo music.

I think I'm confused. (Ain't the first time.)

16 bits x 2 (channels) x 44,100 / 8 (bits to Bytes) / 1000 = 176.4


What am I missing?

--
ha

In article . com,
says...
Here's an unusual request. I'm currently working in the acoustics
engineering group for a large cell phone company. We have lots of
industry formal test requirements that most all phones pass. This
seems to have little direct correlation to percieved quality either
recieved or Sent audio.

Kevin,

I don't know if this is quite the type of feedback you were looking for,
but I find that latency is a much bigger annoyance than, say, sibilance.
I can understand speech on nearly any phone, but some CDMA phones (LG
4400/4500 come to mind) have a noticeably longer latency than others. In
conversations, this turns into awkward pauses, and especially interferes
with normal social turn-taking cues; the conversation ends up like one
of those four-way stop signs, where nobody can tell who's going next, so
everyone moves in fits and starts.

I haven't done any formal experiments, so I can't tell if the perceived
latency comes from the time required to switch from "listen" to
"transmit", or the lack of continued background noise from the other end
as you talk, or the codec itself, but it does feel similar to the
problems encountered on a speakerphone, which is surprising - I always
assumed that cell phones were full-duplex.

I actually did a Google search on this a while ago, expecting to find
all sorts of sociological dissertations on turn-taking, but came up
blank. Is this a recognized issue in the industry?

--
Jay Levitt |
Wellesley, MA | I feel calm. I feel ready. I can only
Faster: jay at jay dot fm | conclude that's because I don't have a
http://www.jay.fm | full grasp of the situation. - Mark Adler

In article eW5fe.50145$tg1.49908@edtnps84,
says...
Is bandwidth really so limited that we have to put up with such awful
sound? Even worse, the constant noise gating that makes conversation
more like using a CB radio than full-duplex?

Noise gating.. of course. That explains the weird phenomena I mentioned
in my other post, where conversations feel half-duplex. I find this is
the #1 annoyance with cell phones for me.

--
Jay Levitt |
Wellesley, MA | I feel calm. I feel ready. I can only
Faster: jay at jay dot fm | conclude that's because I don't have a
http://www.jay.fm | full grasp of the situation. - Mark Adler

In article . com,
says...
Please If you find the time just find a
stationary known good RF spot at a non busy hour time and reply with
your type/model and opinions. I'm interested in audio not RF
performance so minimum BER/FER is needed.

I assume you've seen Steve Punter's page (yes, that Steve Punter) on
perceived codec quality?

--
Jay Levitt |
Wellesley, MA | I feel calm. I feel ready. I can only
Faster: jay at jay dot fm | conclude that's because I don't have a
http://www.jay.fm | full grasp of the situation. - Mark Adler

This is interesting, as I can maybe actually finally contribute something to
RAP rather than asking all the time.
A few years ago, I started a successful mobile music services company in
Asia that delivered across networks in Taiwan, and China. Mostly GSM
networks at 1800 and 1900. One PHS system, and one ss7 network. We developed
several products, one of them a product called "Songmail" which was
essentially a realtime interactive music recording/messaging/delivery system
through wireless networks on handsets. You called a short code or Wapped to
a web page, selected from a variety of material, the background track would
start the streaming music, and you could sing along and
record/listen/erase/rerecord until you got what you wanted. If you didn't
know the words, you could select to sing along with the words, which were
then removed when you sent your message. Our system recorded your
voice/message in time to the background music, after measuring the latency
on any particular switching network, so the end user received a fairly
decent quality personalized recorded music message delivered to ther handset
as a voice mail message, or if they had fancier handsets, as a flash or
animated MMS, and we even sent to email like an email card which was popular
with kids.

Here's a link:
http://www.ericsson.com/network_oper...7-0030474E2F8A

My comments on your poll:

1) The audio is good enough to create a business with several hundred
thousand satisfied customers, and even allow record companies to have talent
contests for singers.
2) POTs = 64k, so yes, it sounds better.
3) GSM - 13.3k, so yes it is compressed, but the codecs are not bad, as long
as you compensate accurately for the amound of bandwidth you are sending,
and your networl isn't over crowded as most are. (or you just call at 2am)
4) GPRS et al - supposed to be 128k + using, but that's a joke, as it's
split into sub-channels, and no carrier offers all channels at one time. The
widest b/w we ever measured in a working environment was about 30k for
voice, can do up to 56k for data.
5) Full duplex - Well, sort of but not really. Audio can enter the
microphone while other audio is playing but it's a feedback loop, and you
have to wonder where the audio goes to that you've added. That was a
complicated issue. The full duplex they should offer is audio/data. You
should be able to be one a conversation, and also send/receive data
transmissions. This allows for all sorts of cool apps, but, no, the networks
don't support it, too much switching and routing = $$.
6) 3G - waiting....
7) Handphones: Nokias and Siemens in general sounded better, but that now
depends on model. Better earpieces maybe. It's much better with stereo
headphones. We tried testing different handsets, but gave up after 2002 due
to over 200+ handsets on the market.
8) working at an "acoustics engineering group for a large cell phone
company" - the Docomo research center in Yokohama had awesome studios for
testing and researching. My favorite was the cut-in-half dummy with a
handphone strapped to his head, that was placed in all sort of different
environments. Audio quality is pretty good when you're not walking down the
streets of any Asian city with 80db+ of noise around you. They should take
the dummy to an intersection in Shanghai with squeeling brakes and blasting
horns and trucks and contruction work.










"Jay Levitt" wrote in message
...
In article . com,
says...
Here's an unusual request. I'm currently working in the acoustics
engineering group for a large cell phone company. We have lots of
industry formal test requirements that most all phones pass. This
seems to have little direct correlation to percieved quality either
recieved or Sent audio.

Kevin,

I don't know if this is quite the type of feedback you were looking for,
but I find that latency is a much bigger annoyance than, say, sibilance.
I can understand speech on nearly any phone, but some CDMA phones (LG
4400/4500 come to mind) have a noticeably longer latency than others. In
conversations, this turns into awkward pauses, and especially interferes
with normal social turn-taking cues; the conversation ends up like one
of those four-way stop signs, where nobody can tell who's going next, so
everyone moves in fits and starts.

I haven't done any formal experiments, so I can't tell if the perceived
latency comes from the time required to switch from "listen" to
"transmit", or the lack of continued background noise from the other end
as you talk, or the codec itself, but it does feel similar to the
problems encountered on a speakerphone, which is surprising - I always
assumed that cell phones were full-duplex.

I actually did a Google search on this a while ago, expecting to find
all sorts of sociological dissertations on turn-taking, but came up
blank. Is this a recognized issue in the industry?

--
Jay Levitt |
Wellesley, MA | I feel calm. I feel ready. I can only
Faster: jay at jay dot fm | conclude that's because I don't have a
http://www.jay.fm | full grasp of the situation. - Mark Adler

On Sat, 07 May 2005 20:28:39 GMT, (hank alrich)
wrote:

Loren Amelang wrote:

As I said, even here in "Northwest Nowhere", I can use a 64K data pipe via
this same phone. If you believe the portable music hype, that is
capable of CD quality stereo music.

I think I'm confused. (Ain't the first time.)

16 bits x 2 (channels) x 44,100 / 8 (bits to Bytes) / 1000 = 176.4

What am I missing?

"If you believe the portable music hype"... Personally, I don't.

iTunes reports a real CD rip as "1411k", to put it in the same units
with 128k MP3, or 64k AAC or MP3pro or WMA. The best I could say about
the compressed formats is that some examples are inoffensive. Of
course not all CDs sound that good any more either, but I clearly
prefer hearing the additional bits.

But the general public is paying real money for iPods and compressed
music...

Loren

Yes good page
& Ive done lots of 13 & 8k FER vs Voice quality analysis.

Jay

For CDMA I use to see ~.25sec round trip. If is gets over ~.5sec
problems like overtalking/ cueing start. Sounds like your GSM system
had massive delay. AFAIK The type phone ( given the same codec ) is
not likely able to significantly change the delay .

Kevin Tracy

Jay

Jay

sorry i placed this incorrectly latter in the thread

For CDMA I use to see ~.25sec round trip. If is gets over ~.5sec
problems like overtalking/ cueing start. Sounds like your GSM system
had massive delay. AFAIK The type phone ( given the same codec ) is
not likely able to significantly change the delay .


Kevin Tracy

In article .com,
says...
For CDMA I use to see ~.25sec round trip. If is gets over ~.5sec
problems like overtalking/ cueing start. Sounds like your GSM system
had massive delay. AFAIK The type phone ( given the same codec ) is
not likely able to significantly change the delay .

I think that's no longer true; I did some informal tests against my
phone and a friend's, and IIRC hers was noticeably shorter. Both were
Verizon digital phones, which implies that both were CDMA.

--
Jay Levitt |
Wellesley, MA | I feel calm. I feel ready. I can only
Faster: jay at jay dot fm | conclude that's because I don't have a
http://www.jay.fm | full grasp of the situation. - Mark Adler

In field tests can be decieving. If the network routes calls thru
different switches/ echo cancellers
etc different delays occur. Also mobile to mobile has the logest delay
and (double coding) worst audio quality.

Kevin Tracy

"jakdedert" writes:

Randy Yates wrote:
"jakdedert" writes:

Randy Yates wrote:
Loren Amelang writes:
[...]
I'm probably on the most lightly loaded cell in the system, in a
remote area no other company even bothers to serve, so it isn't a
matter of oversubscribing the hardware.

Rappaport [1] states that QCELP13 is a higher-datarate speech codec
for IS-95 systems. You can also read some interesting history about
the CDMA codec evolution at



http://www.commsdesign.com/design_co...cleID=16500249

Regarding your situation, perhaps you're at the distance limits of
the cell, or the cell is operating at reduced power, thus they may
be dropping to a lower speech codec rate in order to perform more
channel coding.

Have you tried a GSM phone? Most of the GSM networks now support AMR
(Adaptive Multi-Rate), an improved variable-rate codec system.

When I got my GSM phone almost a year ago, my connectivity dropped
to about half what it is now (Nashville--Cingular). My trusty old
Nokia 51xx used to connect reliably in my home. My new
Sony-Ericsson: much less so. It matters not if I have 'bars' or
not. The night I got it, I experienced an inability to connect, as
did my wife, with the same model.

I booted up the Nokia and got a connection in the same place (no
actual 'service' of course, only a recording).

jak

jak,

Which model Sony Ericsson are you using?
The cheapest one I could get--free actually--with a 2-year agreement. It's
the T226. I researched the handsets Cingular would give me for the price of
re-upping my service, and this one had the best reviews...sound quality,
features, range etc. Since then I've used others' phones on Cingular and
have found a few with better (all the above).

Ah yes, the T226 - one of our most popular models. I personally did a lot
of work on this model, but in the MIDI area - not RF/codec. (I also did
some voice path work, so I'm very interested to hear comments.)

Actually the T226 has one of the most sensitive receivers SEMC /
Ericsson has put out in years. Can you be a little more specific on
"my connectivity dropped...", e.g., are you able to place a call but
the voice quality is so horrible you can't stand it, or do you get
muted voice channels (either uplink or downlink)?

Did you say you can see the tower? Maybe the problem is that you're
TOO close and the receiver is generating intermods because of its high
sensitivity. However, we're right under a tower at Sony Ericsson and
I've never had a problem. There's so many parameters it's hard to tell
what's happening.
--
% Randy Yates % "Midnight, on the water...
%% Fuquay-Varina, NC % I saw... the ocean's daughter."
%%% 919-577-9882 % 'Can't Get It Out Of My Head'
%%%% % *El Dorado*, Electric Light Orchestra
http://home.earthlink.net/~yatescr

Randy Yates wrote:
"jakdedert" writes:

Which model Sony Ericsson are you using?
The cheapest one I could get--free actually--with a 2-year
agreement. It's the T226. I researched the handsets Cingular would
give me for the price of re-upping my service, and this one had the
best reviews...sound quality, features, range etc. Since then I've
used others' phones on Cingular and have found a few with better
(all the above).

Ah yes, the T226 - one of our most popular models. I personally did a
lot of work on this model, but in the MIDI area - not RF/codec. (I
also did some voice path work, so I'm very interested to hear
comments.)

Wow...really cool to get someone--especially on usenet--with your
qualifications directly related to my issue.

Actually the T226 has one of the most sensitive receivers SEMC /
Ericsson has put out in years. Can you be a little more specific on
"my connectivity dropped...", e.g., are you able to place a call but
the voice quality is so horrible you can't stand it, or do you get
muted voice channels (either uplink or downlink)?

In my home--with usually at least three bars of sig strength indication--I
often (usually?) have difficulty connecting unless I hold the handset 'just
so,' or move around the house. I'm aware that the antenna is internal, and
so hold the unit with just thumb and middle finger on each side of battery
compartment, with index finger balancing at the top (in difficult
conditions).

The issue presents as either A) a complete inability to connect; ie a series
of short beeps indicating no connection; or B) difficulty in holding the
connection, with dropouts on either end (intermittant inablility to hear
other party or them to hear me).

If I carry the phone in my pocket, I sporadically miss calls.

Did you say you can see the tower? Maybe the problem is that you're
TOO close and the receiver is generating intermods because of its high
sensitivity. However, we're right under a tower at Sony Ericsson and
I've never had a problem. There's so many parameters it's hard to tell
what's happening.

I don't know where the nearest Cingular tower is located. I'm on pretty
high terrain for the neighborhood. I often have overload problems with
other types of receivers, probably due to an antenna farm (I don't know what
type...several different, including commercial FM, but no cells, AFAICT) on
a hill within line of sight.
A topo map of my neighborhood can be found at:
http://terraserver.microsoft.com/ima...646&Y=4998&W=1
(I'm right at the hill near the 'S' in St. Bernard's Academy)

I've always thought this was a Cingular problem, but since both our GSM
phones are identical, there's no way to tell unless someone on the same
network--different phone--were to come to my home and experience the same
symptoms or lack thereof. The exact symptoms present on my wife's T226. If
I can't connect, she can't either.

I really like the phone. It otherwise works well for a budget model; good
battery life, small size, decent display (difficult to read in sunlight).
Minuses: menu heirarchy is all bas ackwards, and the operating system has a
lot of lag when switching from one function to another. The sound
'quality'--not intelligibility--on recieve 'could' be better; but I imagine
that's result of the decision to make the speaker small...sounds better with
a headset.

Rant mode: ON
I do wish all the manufacturers who devise proprietary connectors would
settle on the common 2.5mm headset plug. The S-E connector is difficult to
insert correctly in the dark. The accessory manufacturers (including S-E)
should mold some tactile indication as to the correct orientation of the
plug. As you know, there's only a small icon which indicates 'up' ie face
up. You can't see it in low light--can't feel it all; and thus it's almost
impossible to plug anything into the accessory interface by feel alone. It
makes the phone hard to connect in the car. I prefer to use a headset; but
the only one Cingular offers has one of those stupid 'dangling' inline mics.
In any case, if I don't hook it up when I get into the car--and a call comes
in--I'm gonna be juggling....
Rant mode: OFF

Sorry to rant at you. It's so unusual to actually connect with someone who
was involved in the design process of a piece of gear that I own, and
'perhaps' have some input into the next one. I do appreciate the personal
attention, something that's sorely lacking in industry these days....

jak