|
Viodi View Menu
Interested in Sponsoring
the Viodi View? Send an email to: sponsor@viodi.com
Please forward this free publication
to anyone you know who is involved in some way with independent telephone
companies.
Mission of the Viodi View:
In this on-line publication, we share our analysis, opinions
and direction on the interactive television news and views that we believe
will be of interest and use to our friends associated directly or indirectly
with independent telephone companies. For more information as to the various
ways Viodi works with independent telephone companies, please go to http://www.viodi.com/alliance/
Disclaimer:
The Viodi View [Viodi, LLC] and its associates used their
best efforts in collecting and preparing the information published herein.
However, the Viodi View [Viodi, LLC] does not assume, and hereby disclaims,
any and all liability for any loss or damage caused by errors or omissions,
whether such errors or omissions resulted from negligence, accident, or
other causes.
All displayed trademarks, logos and service marks are
the property of their respective owners. © 2005Viodi,
LLC. All Rights Reserved.
5255 Stevens Creek, #127 Santa Clara, CA 95051 |
Viodi
View Newsletter - September 6th, 2006

Click here to learn more about the Technology of Local Content Workshop
AT&T vs. Comcast Testing
Last Updated, 11:00 P.M.
P.S.T., 9/6/06
By Ken Pyle, ken.pyle at viodi.com, Managing
Editor, Viodi View
The following videos record some of the tests we have
been performing on the Comcast cable modem connection versus the AT&T
DSL connection. This page is a work in progress and an update date will
be provided above to give provide a reference (as well an archived page
for historical purposes).
We are open to commentary and ideas for tests. Probably
the best way to add this commentary is to go to http://viodi.com/board/index.php?topic=11.0.
You don't need to be a member to comment or read the commentary.
Some of these videos have been triple compressed, so
they may not provide the best results (first through the Packet8 compression,
then through Windows Media and then through Revver’s Quicktime compression).
We may put up the Windows Media format at some point, eliminating the
Revver compression effect.
Some of the other things that we may want to add to this
page are included at the bottom of the page.
|
AT&T Speed Test
This test was peformed using the San Francisco
server at the web site http://www.speedtest.net/.
The test was performed September 2nd. These results are at the high
end of what has been acheived with the AT&T DSL service. Typically,
the speed is 300kb/s up and down, although speeds as low as 200kb/s
have been measured..
It is important to note that this location is approximately
11 to 12 kilofeet from the central office and is on home-run copper
(no Service Area Interface), which is rare these days. The location
is in the middle of the Silicon Valley. |
|
|
Comcast Speed Test
This test was peformed using the San Francisco
server at the web site http://www.speedtest.net/.
The test was performed September 5th. These results are pretty typically
of what has been acheived with the Comcast Cable Modem service.
It is consistently well above 6 Mb/s downstream and clocks in at
350 kb/s on the upstream.
The location is in the middle of the Silicon Valley.
|
|
|
Packet8 Video call over a Comcast connection.
In general, the performance of this connection
was pretty good. The lips were fairly well synchronized to the voice.
Pixelization was minimal and only appeared during rapid movement.
One end of this call used a DSL connection that
is typically 1.5 Mb/s downstream, 400 kb/s upstream. The far end
(Roger Bindl's end) used a Packet8 softphone on a Dell Notebook
computer, while the near-end used a Packet8 DV326 Video Phone. Unfortunately,
there were some recording issues, so it was filmed instead of being
direct recorded from the output of the Packet8 Video Phone. The
router used was the NetGear WGR 614 v6.
|
|
|
Packet8 Video call over an AT&T Connection
The performance in the video will not look much
different than what was seen as compared to the Packet8 call over
the Comcast connection. There is more pixelization, especially with
fast movement. The lips are a bit out of sync and the delay might
be a bit more noticable. Based on a few other calls we made, this
bandwidth seems to be marginal. In one of those other calls we experienced,
a few seconds of delay, which was unacceptable.
One end of this call used a DSL connection that
is typically 1.5 Mb/s downstream, 400 kb/s upstream. The far end
(Roger Bindl's end) used a Packet8 softphone on a Dell Notebook
computer, while the near-end used a Packet8 DV326 Video Phone. Unfortunately,
there were some recording issues, so it was filmed instead of being
direct recorded from the output of the Packet8 Video Phone. The
router used was the same NetGear WGR 614 v6 described above. .
|
|
|
Video Download from YouTube - 16 second
This is a side-by-side comparison of the download
performance of Comcast and AT&T using a short video from the
YouTube video site.
The AT&T is using a relatively new (January
2006) Toshiba lap-top (1.4 GHz , 512 MByte), while Comcast uses
an older Toshiba (Sept 2001, 800 MHz, 384 Mbyte). The AT&T connection
uses a Linksys router, while the Comcast connection uses the NetGear
WGR 614 v6. |
|
|
Video Download from YouTube - 1 minute
This is a side-by-side comparison of the download
performance of Comcast and AT&T using a medium-length video
from the YouTube video site. You may be surprised as to which connection
delivers the video first.
The AT&T is using a relatively new (January
2006) Toshiba lap-top (1.4 GHz , 512 MByte), while Comcast uses
an older Toshiba (Sept 2001, 800 MHz, 384 Mbyte). The AT&T connection
uses a Linksys router, while the Comcast connection uses the NetGear
WGR 614 v6. |
|
|
Video Download from YouTube - 2 minute+
This is a side-by-side comparison of the download
performance of Comcast and AT&T using a bit longer video from
the YouTube video site. This longer length video, titled Relay for
Life, really shows the difference in speed between the two broadband
offerings.
The AT&T is using a relatively new (January
2006) Toshiba lap-top (1.4 GHz , 512 MByte), while Comcast uses
an older Toshiba (Sept 2001, 800 MHz, 384 Mbyte). The AT&T connection
uses a Linksys router, while the Comcast connection uses the NetGear
WGR 614 v6. |
|
|
Video Download from Google - 1 minute
This is a side-by-side comparison of the download
performance of Comcast and AT&T using a one minute from the
Google video site. This is the a shortened version of the Relay
for Life video that had played from the YouTube site.
The AT&T is using a relatively new (January
2006) Toshiba lap-top (1.4 GHz , 512 MByte), while Comcast uses
an older Toshiba (Sept 2001, 800 MHz, 384 Mbyte). The AT&T connection
uses a Linksys router, while the Comcast connection uses the NetGear
WGR 614 v6. |
|
Things that could be added and will be if enough interest.
- Post results from Yahoo! video site. Some very interesting
results that need to be reviewd with an independent third party.
- Add block diagrams of system configuration with equipment
models, etc.
Back to Top
Next
Article
Previous
Issue
|
Viodi
View Subscribe

Order before September 18th, 2006
& Get an Extra Issue at No Charge
Click here to subscribe to 12 months of the Viodi View (Does
not include Club Viodi membership)
|