Cingular Connection Manger tweeks

For the past three weeks, my Bluetooth tethered V557 was having terrible
connection speeds using Cingular's Connection Manager and Windows DUN
for both Accelerated and Non-Accelerated connections.

A CNN page would take several minutes to load, FTP and NG would time
out. Watching the Windows DUN data counter, you could see my speeds were
less than 10 KBps (or 1Kbps roughly) or it would just sit there with
absolutely no data getting downloaded. Sometimes I would get decent
speeds (60 Kbps+), but usually it was much slower 90% of the time,
regardless of the time of day or night (no it wasn't a cell loading
issue), nor was it location related across the county.

A call to Cingular laptop tech support two weeks ago only had me check
my settings and she pushed something down to my phone. No improvement
and it only got worse by now.

Today, support had me change two things in the Cingular Connection
Manager. Under Tools and Settings:

1) Advanced Networking tab > Change from "Use default OS value" to
"Optimized Wireless Data value" Don't reboot yet.

2) Acceleration tab > Click Uninstall. Now you can reboot.

I'm now getting 60 Kbps to 80 Kpbs speeds with an occasional drop in
speed, but otherwise a vast improvement.

When I use the Windows DUN, the problem comes back. Apparently there IS
a difference in using the Cingular Connection Manager and Windows DUN.

Of course, this improvement would not apply to the San Francisco Bay
area as users there never have any problems.

---

› See More: Cingular Connection Manger tweeks

[POSTED TO alt.cellular.cingular - REPLY ON USENET PLEASE]

In <[email protected]> on Sat, 25 Mar 2006
01:00:44 GMT, DecaturTxCowboy <[email protected]> wrote:

>For the past three weeks, my Bluetooth tethered V557 was having terrible
>connection speeds using Cingular's Connection Manager and Windows DUN
>for both Accelerated and Non-Accelerated connections.

Presumably you're using Data Connect, since MEdia doesn't support Acceleration
-- correct?

>A CNN page would take several minutes to load, FTP and NG would time
>out. Watching the Windows DUN data counter, you could see my speeds were
>less than 10 KBps (or 1Kbps roughly) or it would just sit there with
>absolutely no data getting downloaded. Sometimes I would get decent
>speeds (60 Kbps+), but usually it was much slower 90% of the time,
>regardless of the time of day or night (no it wasn't a cell loading
>issue), nor was it location related across the county.
>
>A call to Cingular laptop tech support two weeks ago only had me check
>my settings and she pushed something down to my phone. No improvement
>and it only got worse by now.
>
>Today, support had me change two things in the Cingular Connection
>Manager. Under Tools and Settings:
>
> 1) Advanced Networking tab > Change from "Use default OS value" to
>"Optimized Wireless Data value" Don't reboot yet.

That simply increases the Windows global RWIN (TCP/IP receive window) from the
default (16K in Windows XP) to 64K. Unfortunately, that will tend to be
counterproductive in most situations, since speed and latency of EGPRS(EDGE)
aren't high enough to warrant such a large value, and since that will increase
the time for detection and recovery when a packet is lost. (See
<http://cable-dsl.home.att.net/#WhyItMatters>) Also, this change will affect
*all* Windows TCP/IP networking, not just the Cingular Connection Manager.

> 2) Acceleration tab > Click Uninstall. Now you can reboot.

That eliminates Acceleration, which can be a significant boost to speed with
Data Connect. Also, since Acceleration can be turned off, this tweak
shouldn't matter unless the Cingular software isn't working properly.

>I'm now getting 60 Kbps to 80 Kpbs speeds with an occasional drop in
>speed, but otherwise a vast improvement.

As I've noted above, these tweaks shouldn't really matter, much less help.
If they do, then either something must have really been screwed up (e.g.,
improper configuration settings, software not working properly) or Cingular
data service must be really bad in your area (e.g., very high latency).

>When I use the Windows DUN, the problem comes back. Apparently there IS
>a difference in using the Cingular Connection Manager and Windows DUN.

Indeed there is, but there is also interaction between them, as noted above,
and as shown in my controlled tests with MEdia (not Data Connect) over Windows
XP SP2 Bluetooth to a Motorola V551 using the Intel Broadband Speed Test
<http://www.intel.com/personal/resources/broadband/speedtest.htm> to measure
throughput in kilobits per second (10 trials each):

1. Windows DUN (port speed 460000 bps):

Mean 78.66
Median 85.05
Minimum 61.9
Maximum 88.3

2. Motorola Phone Tools 4.0.4b (default settings):

Mean 70.14
Median 65.75
Minimum 52.2
Maximum 90.1

3. Cingular Connection Manager:

I tried versions 5.1.10 and 5.2.19, and both resulted in terrible connections
(very long stalls), even when your tweaks were applied. (See Note 1.) Worse,
both caused Windows DUN to exhibit the same behavior, suggesting that a lot of
reported data problems are due to running or having run Cingular Communication
Manager. To restore normal Windows DUN performance, I had to run Motorola
Phone Tools (once).

>Of course, this improvement would not apply to the San Francisco Bay
>area as users there never have any problems.

Windows DUN works well in any location *when* configured properly *and* not
screwed up by Cingular Connection Manager. The usual problem is the
configuration, not the location. My advice, particularly with MEdia (rather
than Data Connect), is *not* to install Cingular Communication Manager. If
you have, it may be difficult to undo the damage if you don't have a Motorola
phone and Motorola Phone Tools software. (See Note 2.)

Notes:

1. Cingular Connection Manager (the version I tested at least) works by
creating a Windows DUN connection called "GSM", but the settings are weird,
which may be part of the problem.

2. Motorola Phone Tools is obviously a good choice for those with Motorola
phones *except* that the Windows DUN connection it creates (called "Cingular
GPRS") has a port speed of only 115200 bps. For better results this should
probably be increased (in Control Panel -> Network Connections) to 460000 bps
(or at least 230000 bps).

--
Best regards, SEE THE FAQ FOR CINGULAR WIRELESS AT
John Navas <http://en.wikibooks.org/wiki/Cingular_Wireless_FAQ>

John Navas wrote:
> As I've noted above, these tweaks shouldn't really matter, much less help.
> If they do, then either something must have really been screwed up (e.g.,
> improper configuration settings, software not working properly)

Configurations hadn't been changed and software was reinstalled. So that
negates that guess.

> or Cingular data service must be really bad in your area (e.g., very high latency).

Problem was consistent from first of March for three weeks, suddenly got
better after the tweeks. So that negates that guess.

>> Of course, this improvement would not apply to the San Francisco Bay
>> area as users there never have any problems.
>
> Windows DUN works well in any location

I must have missed my sarcasm.

[POSTED TO alt.cellular.cingular - REPLY ON USENET PLEASE]

In <[email protected]> on Tue, 28 Mar 2006
18:34:31 GMT, DecaturTxCowboy <[email protected]> wrote:

>John Navas wrote:
>> As I've noted above, these tweaks shouldn't really matter, much less help.
>> If they do, then either something must have really been screwed up (e.g.,
>> improper configuration settings, software not working properly)
>
>Configurations hadn't been changed and software was reinstalled. So that
>negates that guess.

Nope. Read what I wrote more carefully.

>> or Cingular data service must be really bad in your area (e.g., very high latency).
>
>Problem was consistent from first of March for three weeks, suddenly got
>better after the tweeks. So that negates that guess.

Nope. Could have been coincidence. Only way to tell for sure is with
controlled tests.

>>> Of course, this improvement would not apply to the San Francisco Bay
>>> area as users there never have any problems.
>>
>> Windows DUN works well in any location
>
>I must have missed my sarcasm.

You must have missed the rest of what I posted (that you snipped), which
explains how to avoid the kind of problems you've been experiencing. "You can
lead a horse to water..."

--
Best regards,
John Navas <http://navasgrp.home.att.net/>

"A little learning is a dangerous thing." [Alexander Pope]
"It is better to sit in silence and appear ignorant,
than to open your mouth and remove all doubt." [Mark Twain]

John Navas wrote:
>>> or Cingular data service must be really bad in your area (e.g., very high latency).
>> Problem was consistent from first of March for three weeks, suddenly got
>> better after the tweeks. So that negates that guess.
>
> Nope. Could have been coincidence. Only way to tell for sure is with
> controlled tests.

Three weeks of poor service. Change settings and 30 seconds later better
service for the past week. Yeah, that's coincidence.

[POSTED TO alt.cellular.cingular - REPLY ON USENET PLEASE]

In <[email protected]> on Tue, 28 Mar 2006
20:39:59 GMT, DecaturTxCowboy <[email protected]> wrote:

>John Navas wrote:
>>>> or Cingular data service must be really bad in your area (e.g., very high latency).
>>> Problem was consistent from first of March for three weeks, suddenly got
>>> better after the tweeks. So that negates that guess.
>>
>> Nope. Could have been coincidence. Only way to tell for sure is with
>> controlled tests.
>
>Three weeks of poor service. Change settings and 30 seconds later better
>service for the past week. Yeah, that's coincidence.

With you anything is possible.

--
Best regards,
John Navas <http://navasgrp.home.att.net/>

"A little learning is a dangerous thing." [Alexander Pope]
"It is better to sit in silence and appear ignorant,
than to open your mouth and remove all doubt." [Mark Twain]

DecaturTxCowboy wrote:
> John Navas wrote:
>
>>>> or Cingular data service must be really bad in your area (e.g., very
>>>> high latency).
>>>
>>> Problem was consistent from first of March for three weeks, suddenly
>>> got better after the tweeks. So that negates that guess.
>>
>>
>> Nope. Could have been coincidence. Only way to tell for sure is with
>> controlled tests.
>
>
> Three weeks of poor service. Change settings and 30 seconds later better
> service for the past week. Yeah, that's coincidence.

I'll just enable masochism mode, jump into the middle of this and say:
Though I don't know you or the details, my (fairly long) experience with
troubleshooting says it could be a coincidence. I have seen it happen
many, many times over the years. A fairly quick (and too often skipped)
technique to gain a lot more confidence is to restore the old settings
and see if the original symptoms return. Use an objective measurement.
You can't trust your own subjective impression on a good day, certainly
not when you are engaged in a pissing contest.

You're welcome!

--
Phil Nelson

John Navas wrote:
> With you anything is possible.

Of course, *I* am a professional and you are???

[insert your last word here]

Phil Nelson wrote:
> A fairly quick (and too often skipped)
> technique to gain a lot more confidence is to restore the old settings
> and see if the original symptoms return.

That is so funny as I just indeed set my Cingular Connection Manager
settings back a little bit ago. Good connections before, then all to
hell again after the settings back to the original. Put them back the
way I had, and am good to go.

The coincidence factor is rapidly evaporating.

[POSTED TO alt.cellular.cingular - REPLY ON USENET PLEASE]

In <[email protected]> on Wed, 29 Mar 2006
01:05:02 GMT, DecaturTxCowboy <[email protected]> wrote:

>Phil Nelson wrote:
>> A fairly quick (and too often skipped)
>> technique to gain a lot more confidence is to restore the old settings
>> and see if the original symptoms return.
>
>That is so funny as I just indeed set my Cingular Connection Manager
>settings back a little bit ago. Good connections before, then all to
>hell again after the settings back to the original. Put them back the
>way I had, and am good to go.
>
>The coincidence factor is rapidly evaporating.

You have problems.
Lots of others have no problems.
I show that CCM causes problems.
No coincidence?

--
Best regards, SEE THE FAQ FOR CINGULAR WIRELESS AT
John Navas <http://en.wikibooks.org/wiki/Cingular_Wireless_FAQ>

Somewhere around Tue, 28 Mar 2006 16:01:09 -0800, while reading
alt.cellular.cingular, I think I thought I saw this post from Phil Nelson
<pCUT_THiS_TEXTdn@soAND_THiS_TEXTnic.net>:

>I'll just enable masochism mode, jump into the middle of this and say:
>Though I don't know you or the details, my (fairly long) experience with
>troubleshooting says it could be a coincidence. I have seen it happen
>many, many times over the years. A fairly quick (and too often skipped)
>technique to gain a lot more confidence is to restore the old settings
>and see if the original symptoms return. Use an objective measurement.
>You can't trust your own subjective impression on a good day, certainly
>not when you are engaged in a pissing contest.

I think that's a good idea. In my long experience troubleshooting
electronic, then computer problems, I've found that theories are great first
approximations, but sound troubleshooting and process of elimination are
always good if you want answers. I check the impossible, and when I find
that the impossible has happened, I've learned something new. And the
impossible becomes the usually impossible, or even the unlikely.

Most people, though, don't care, and when it works, that's all they care
about. No problem with that (but what I hate are people who "fix"
something, and when you ask them what they did, they just fiddled with it,
and it started working).


--
Marty - public.forums (at) gmail (dot) com
"Those are my principles, and if you don't like them...
well, I have others." - Groucho Marx

Marty wrote:
> Most people, though, don't care, and when it works, that's all they care
> about. No problem with that

Its disturbing to say the least when you don't know the process what
fixed it, If the new card works, I put the "bad" one back in and see if
the problem reoccurs, but then...a simple reseating the card or whatever
fixed it.

> (but what I hate are people who "fix"
> something, and when you ask them what they did, they just fiddled with it,
> and it started working).

I'll tell a customer:

1) I couldn't find what was wrong, because I couldn't duplicate the
problem, but made some changes that might prevent it in the future.
2) I couldn't duplicate the problem, but I know there must be a problem,
otherwise you wouldn't have called me.
3) I could find a problem, but moved that phone to another desk. Let me
know if that person begins to have problems or if you still have problems.

Admitting what you could or couldn't do usually gives confidence to the
customer if presented the right way.

Disclaimer - This does not apply to people living in the San Francisco
Bay area as nothing ever goes wrong for them.

[POSTED TO alt.cellular.cingular - REPLY ON USENET PLEASE]

In <x%[email protected]> on Wed, 29 Mar 2006
23:46:37 GMT, DecaturTxCowboy <[email protected]> wrote:

>Disclaimer - This does not apply to people living in the San Francisco
>Bay area as nothing ever goes wrong for them.

On the contrary. The real difference is that when that when something isn't
working properly I (a) carefully determine the real problem and (b) fix it
correctly.

--
Best regards, SEE THE FAQ FOR CINGULAR WIRELESS AT
John Navas <http://en.wikibooks.org/wiki/Cingular_Wireless_FAQ>

[POSTED TO alt.cellular.cingular - REPLY ON USENET PLEASE]

In <[email protected]> on Wed, 29 Mar 2006 14:38:42
-0800, Marty <[email protected]> wrote:

>Somewhere around Tue, 28 Mar 2006 16:01:09 -0800, while reading
>alt.cellular.cingular, I think I thought I saw this post from Phil Nelson
><pCUT_THiS_TEXTdn@soAND_THiS_TEXTnic.net>:
>
>>I'll just enable masochism mode, jump into the middle of this and say:
>>Though I don't know you or the details, my (fairly long) experience with
>>troubleshooting says it could be a coincidence. I have seen it happen
>>many, many times over the years. A fairly quick (and too often skipped)
>>technique to gain a lot more confidence is to restore the old settings
>>and see if the original symptoms return. Use an objective measurement.
>>You can't trust your own subjective impression on a good day, certainly
>>not when you are engaged in a pissing contest.
>
>I think that's a good idea. In my long experience troubleshooting
>electronic, then computer problems, I've found that theories are great first
>approximations, but sound troubleshooting and process of elimination are
>always good if you want answers. I check the impossible, and when I find
>that the impossible has happened, I've learned something new. And the
>impossible becomes the usually impossible, or even the unlikely.
>
>Most people, though, don't care, and when it works, that's all they care
>about. No problem with that (but what I hate are people who "fix"
>something, and when you ask them what they did, they just fiddled with it,
>and it started working).

There can be a problem if they start spreading their "fix" as gospel, thereby
starting yet another Internet Myth that does a disservice to those looking for
real help.

--
Best regards, SEE THE FAQ FOR CINGULAR WIRELESS AT
John Navas <http://en.wikibooks.org/wiki/Cingular_Wireless_FAQ>

Somewhere around Wed, 29 Mar 2006 23:46:37 GMT, while reading
alt.cellular.cingular, I think I thought I saw this post from
DecaturTxCowboy <[email protected]>:

>Marty wrote:
>> Most people, though, don't care, and when it works, that's all they care
>> about. No problem with that
>
>Its disturbing to say the least when you don't know the process what
>fixed it, If the new card works, I put the "bad" one back in and see if
>the problem reoccurs, but then...a simple reseating the card or whatever
>fixed it.
>
>> (but what I hate are people who "fix"
>> something, and when you ask them what they did, they just fiddled with it,
>> and it started working).
>
>I'll tell a customer:
>
>1) I couldn't find what was wrong, because I couldn't duplicate the
>problem, but made some changes that might prevent it in the future.
>2) I couldn't duplicate the problem, but I know there must be a problem,
>otherwise you wouldn't have called me.
>3) I could find a problem, but moved that phone to another desk. Let me
>know if that person begins to have problems or if you still have problems.
>
>Admitting what you could or couldn't do usually gives confidence to the
>customer if presented the right way.

I agree that honesty is the best policy - and it's usually easier. I was
thinking more of non-technical people who say they fixed something when they
know nothing about what they're doing, and don't really know the difference
between "fixed" and "problem went away". :-)

>Disclaimer - This does not apply to people living in the San Francisco
>Bay area as nothing ever goes wrong for them.

I live in the Bay Area. We have the same experiences, but my definition of
right and wrong is just different. Wrong can just be an opportunity to
expand my knowledge about something; a bump in the road. But it can be a
drag when there are too many bumps at one time.

--
Marty - public.forums (at) gmail (dot) com
"Those are my principles, and if you don't like them...
well, I have others." - Groucho Marx