3G in my area
Bob
Does this mean that 3G is available in my area? I can't find any
coverage maps on Cingular that show exactly where it would be the 3G
service vs. EDGE service.
Anyone have an idea? FYI, I'm in Hartford county, CT.
John Navas
In <1133876945....@g49g2000cwa.googlegroups.com> on 6 Dec 2005
05:49:05 -0800, "Bob" <shm...@msn.com> wrote:
>I noticed that for my zip code, I can buy the Sierra Wireless 860 card.
>Does this mean that 3G is available in my area?
Not necessarily.
>I can't find any
>coverage maps on Cingular that show exactly where it would be the 3G
>service vs. EDGE service.
>
>Anyone have an idea? FYI, I'm in Hartford county, CT.
See "Where are Cingular coverage maps?" in the FAQ below.
--
Best regards, SEE THE FAQ FOR CINGULAR WIRELESS AT
John Navas <http://en.wikibooks.org/wiki/Cingular_Wireless_FAQ>
GomJabbar
help the OP much.
Mark W
Mark W
seems all they list are laptop cards. Are there any phones coming out
soon that support it? As well, what is the speed of 3G?
John Navas
In <1133894420....@f14g2000cwb.googlegroups.com> on 6 Dec 2005
10:40:20 -0800, "GomJabbar" <dkba...@earthlink.net> wrote:
>I see no 3G HSDPA/UMTS maps in your link, or your links links. Doesn't
>help the OP much.
He's talking about "EDGE".
John Navas
In <11sbp1t5p7q615k4v...@4ax.com> on Tue, 06 Dec 2005 13:10:44
See the Cingular FAQ below.
GomJabbar
> He's talking about "EDGE".
Bob wrote:
> I can't find any coverage maps on Cingular that show exactly where
> it would be the 3G service vs. EDGE service.
Put on your glasses. The OP wants to know about 3G service. EDGE is
not 3G - it is considered 2.5G.
Hey, but you're in luck. A link I had found referenced, I couldn't
gain access to. But checking today, it works!
GomJabbar
additional 3G maps.
http://pcsintel.com/modules.php?name=News&file=article&sid=1095&mode=thread&order=0&thold=0
In addition the following link was pointed out to me on another forum.
It has 3G maps as well.
John Navas
In <1133962716.4...@z14g2000cwz.googlegroups.com> on 7 Dec 2005
05:38:36 -0800, "GomJabbar" <dkba...@earthlink.net> wrote:
>John Navas wrote:
>
>> He's talking about "EDGE".
>
>Bob wrote:
>
>> I can't find any coverage maps on Cingular that show exactly where
>> it would be the 3G service vs. EDGE service.
>
>Put on your glasses. The OP wants to know about 3G service.
I stand corrected.
>EDGE is
>not 3G - it is considered 2.5G.
I disagree.
GomJabbar
> EDGE is not 3G - it is considered 2.5G.
John Navas wrote:
> I disagree.
Take your pick..............
http://www.cnet.com/4520-7363_1-6361076-4.html
http://www.mobilein.com/2.5G.htm
http://www.devx.com/wireless/Door/11264
There are some that consider EDGE 3G. But mainly they are basing that
on the theoretical maximum thoroughput of EDGE which is 384 KBPS.
Cingular does not approach that kind of speed with their EDGE network.
John Navas
In <1133978844.4...@g14g2000cwa.googlegroups.com> on 7 Dec 2005
10:07:24 -0800, "GomJabbar" <dkba...@earthlink.net> wrote:
>I wrote:
>> EDGE is not 3G - it is considered 2.5G.
>
>John Navas wrote:
>> I disagree.
>
>Take your pick..............
>
>http://www.cnet.com/4520-7363_1-6361076-4.html
>http://www.mobilein.com/2.5G.htm
>http://www.devx.com/wireless/Door/11264
>
>There are some that consider EDGE 3G.
Indeed, including the major cellular players; e.g.,
<http://www.ericsson.com/products/hp/Mobile_Broadband_EDGE_bs.shtml>
<http://www.nokia.com/link?cid=EDITORIAL_1485>
>But mainly they are basing that
>on the theoretical maximum thoroughput of EDGE which is 384 KBPS.
>Cingular does not approach that kind of speed with their EDGE network.
It's actually based on real-world speeds.
Regardless, debating these arbitrary labels is silly and pointless, in part
because they are based more on politics than on technical realities.
SMS
> There are some that consider EDGE 3G. But mainly they are basing that
> on the theoretical maximum thoroughput of EDGE which is 384 KBPS.
> Cingular does not approach that kind of speed with their EDGE network.
Argh, the sub-dividing of 1G/2G/3G has gotten out of hand. Technically,
EDGE is not 3G, but some people refer to it as 2.5G, some as 2.75G.
GomJabbar
> Regardless, debating these arbitrary labels is silly and pointless, in part
> because they are based more on politics than on technical realities.
Well, on at least that we can agree.
John Navas
In <439758e1$0$38583$742e...@news.sonic.net> on Wed, 07 Dec 2005 13:49:21
-0800, SMS <scharf...@geemail.com> wrote:
>GomJabbar wrote:
>
>> There are some that consider EDGE 3G. But mainly they are basing that
>> on the theoretical maximum thoroughput of EDGE which is 384 KBPS.
>> Cingular does not approach that kind of speed with their EDGE network.
>
>Argh, the sub-dividing of 1G/2G/3G has gotten out of hand. Technically,
>EDGE is not 3G, ...
Actually it is.
GomJabbar
I am here in NY harbor right now. I have Cingular's EDGE service and
my coworker has Verizon's ED-VO service.
I did several speedtests. Following are my results. And no, I did not
confuse upload and download speeds. Both the EDGE and ED-VO speed
tests were done from the same location, to the same site, within 5
minutes of one another. I also tried disconnecting and reconnecting to
the service, but no real changes resulted. BTW, I had 5 bars of signal
strength when I did these tests. Admittedly, my speeds are usually
better, but near NYC, I'm just lucky to have a stable connection.
Note, I have seen EDGE download speeds of up to 200 KBPS in some areas.
Download Upload - EDGE
24 87
32 64
40 80
30 105
Download Upload - ED-VO
528 61
I have never seen any speeds on EDGE approaching 528 KBPS! You tell
me, which is the TRUE 3G network?! {I am not making any claims
regarding UTMS here}
This test also tends to confirm statements made elseware that Verizon's
service is superior to Cingular's in the NYC area. Cingular just can't
handle the traffic as well as Verizon can (at least in this area).
Otherwise, EDGE would be up to the task here in NYC and give at least
advertised download rates.
John Navas
In <1134022420.2...@o13g2000cwo.googlegroups.com> on 7 Dec 2005
22:13:40 -0800, "GomJabbar" <dkba...@earthlink.net> wrote:
>To JN: Balony!
Noted.
>I have never seen any speeds on EDGE approaching 528 KBPS! You tell
>me, which is the TRUE 3G network?! ...
Both (EDGE and EV-DO).
Is EV-DO sometimes faster than EDGE? Yes.
Is EV-DO sometimes slower than EDGE? Also yes.
What matters is the typical speed, not the peak speed.
I can't speak for your area, but here in the SF Bay Area,
EDGE (Class 10) is quite consistent at about 150 Kbps.
By comparison, EV-DO ranges from about 100 Kbps to about
300 Kbps, with a typical speed of about 170 Kbps.
Thus there isn't much difference between them.
Perhaps there is more excess EV-DO network capacity
in your area.
GomJabbar
speed test after we left NY harbor for LI sound. He got 775 KBPS down,
60 KBPS up. This is without the Venturi feature enabled.
I performed a large file download with my Cingular EDGE connection in
LI sound and saw download speeds for a couple of minutes hovering
around 220 KBPS, with one spike of 240 KBPS.
EV-DO knocks the pants off of EDGE. I just hope UTMS/HSDPA performs as
well as EV-DO when it is available in the areas I travel.
Cliff
"John Navas" <spamf...@navasgroup.com> wrote in message
news:OPulf.237841$zb5....@bgtnsc04-news.ops.worldnet.att.net...
3G and Edge are not the same thing.
John Navas
In <Rwjmf.30348$Au1....@tornado.texas.rr.com> on Fri, 09 Dec 2005 17:53:53
GMT, "Cliff" <missing...@yahoo.com> wrote:
>"John Navas" <spamf...@navasgroup.com> wrote in message
>news:OPulf.237841$zb5....@bgtnsc04-news.ops.worldnet.att.net...
>>
>> In <1133894420....@f14g2000cwb.googlegroups.com> on 6 Dec 2005
>> 10:40:20 -0800, "GomJabbar" <dkba...@earthlink.net> wrote:
>>
>> >I see no 3G HSDPA/UMTS maps in your link, or your links links. Doesn't
>> >help the OP much.
>>
>> He's talking about "EDGE".
>3G and Edge are not the same thing.
Nokia, Sony Ericsson and other major players disagree.
(See my prior posts for links.)
GomJabbar
> Nokia, Sony Ericsson and other major players disagree.
> (See my prior posts for links.)
Yeah, they'd sell you a bridge in Brooklyn if they could. LOL
Anytime you believe marketing hype, you're in trouble.
Or as someone once said: "There's a sucker born every minute".
What matters is real world performance. Doesn't 'almost' everyone
agree?
John Navas
In <1134064642.7...@o13g2000cwo.googlegroups.com> on 8 Dec 2005
09:57:22 -0800, "GomJabbar" <dkba...@earthlink.net> wrote:
>As a matter of fact my coworker on Verizon EV-DO performed another
>speed test after we left NY harbor for LI sound. He got 775 KBPS down,
>60 KBPS up. This is without the Venturi feature enabled.
>
>I performed a large file download with my Cingular EDGE connection in
>LI sound and saw download speeds for a couple of minutes hovering
>around 220 KBPS, with one spike of 240 KBPS.
Something is almost certainly wrong with your testing, because those speeds
are very unlikely in the real world. I suspect caching.
GomJabbar
> Something is almost certainly wrong with your testing, because those speeds
> are very unlikely in the real world. I suspect caching.
The first time he ran the test he got 528 KBPS. No chance for caching
there. The later tests, maybe. But I don't think so. Although he did
run the later tests when I wasn't around. As far as my own testing on
EDGE goes, I always clear the cache between tests - just in case.
When he got 528 KBPS, we were in NY harbor and I was getting GPRS
download speeds on my EDGE connection. When we were in LI Sound, he
got 775 KBPS down, and I got some of the best EDGE speeds I have seen.
The area you are located in makes a huge difference - at least with
EDGE.
>From Verizon's website regarding EV-DO:
http://www.verizonwireless.com/b2c/mobileoptions/broadband/serviceoverview.jsp
> Speed
> Quickly download complex files and view email attachments with average download
> speeds of 400-700 kbps capable of reaching up to 2.0Mbps.
The REAL world. It's where I live. ; = )
John Navas
In <1134432307....@o13g2000cwo.googlegroups.com> on 12 Dec 2005
16:05:08 -0800, "GomJabbar" <dkba...@earthlink.net> wrote:
>John Navas wrote:
>In <1134064642.7...@o13g2000cwo.googlegroups.com> on 8 Dec 2005
>09:57:22 -0800, "GomJabbar" <dkba...@earthlink.net> wrote:
>
>>As a matter of fact my coworker on Verizon EV-DO performed another
>>speed test after we left NY harbor for LI sound. He got 775 KBPS down,
>>60 KBPS up. This is without the Venturi feature enabled.
>>
>>I performed a large file download with my Cingular EDGE connection in
>>LI sound and saw download speeds for a couple of minutes hovering
>>around 220 KBPS, with one spike of 240 KBPS.
>> Something is almost certainly wrong with your testing, because those speeds
>> are very unlikely in the real world. I suspect caching.
>
>The first time he ran the test he got 528 KBPS. No chance for caching
>there. The later tests, maybe. But I don't think so. Although he did
>run the later tests when I wasn't around. As far as my own testing on
>EDGE goes, I always clear the cache between tests - just in case.
Here's why I'm very skeptical:
Current GPRS and EDGE implementations are limited to 4 time slots.
Theoretical maximum raw speeds (including overhead) are:
* 80 Kbps (4 x 20.0 Kbps) for GPRS with CS4 coding.
* 237 Kbps (4 x 59.2 Kbps) for EDGE with MCS9 coding.
But those speeds won't normally be seen because the those particular codings
are very limited in range and very sensitive to errors. Thus in practice the
best raw speeds are:
* 58 Kbps (4 x 14.4 Kbps) for GPRS with CS3 coding.
* 179 Kbps (4 x 44.8 Kbps) for EDGE with MCS7 coding.
(Cingular initially deployed GPRS limited to CD2 coding, and only enabled CS3
coding when the system has proven itself.)
Net data throughput (less protocol and error correction overheads) is
significantly less, on the order of:
* 52 Kbps for GPRS
* 160 Kbps for EDGE
Thus, as I wrote, the EDGE speeds you're reporting are very unlikely in the
real world.
The issues are somewhat different for EV-DO (cell load is a factor), but the
speeds reported by your coworker are likewise far above expected real world
speeds. I can't speak for your area, but here in the SF Bay Area, actual
EV-DO downlink ranges from as low as 100 Kbps to about 300 Kbps. (1xRTT
likewise delivers typical speeds far below theoretical maximums.)
That's why I suspect caching, data compression, or measurement error.
GomJabbar
> I performed a large file download with my Cingular EDGE connection
> in LI sound and saw download speeds for a couple of minutes hovering
> around 220 KBPS, with one spike of 240 KBPS.
I was downloaded two large programs that morning: OpenOffice.org 2.0
and Java 1.5. I don't recall which program I was downloading when I
saw the above speeds. These speeds are what the Sony Ericsson Wireless
Manager reported; realtime.
My coworkers speeds and my first speed post was done by going to
dslreports.com and running a speedtest there. These speeds are what
the speed test web page reported.
John Navas
In <1134452576.1...@g14g2000cwa.googlegroups.com> on 12 Dec 2005
Don't depend on what programs report.
dslreports is notoriously unreliable.
For dependable results, download uncompressible data by FTP, time how long it
takes, and calculate throughput from size and time. Do at least three trials.
Restart the computer between successive trials.
Here's a good test file:
<ftp://ftp.sonic.net/pub/testfile.compressed.10meg>
GomJabbar
connection for more than a minute or two. I am not sure when I will
see the above speeds again, so who knows when I could do the FTP.
Anyway, I believe the Sony Ericsson Wireless Manager to be accurate in
the speeds it reports, and it reports realtime, not average time. Your
test would report average time. I pretty sure I wouldn't see my above
posted speeds as an average for a 10 MB download.
BTW, when you download software like OpenOffice.org 2.0 or Java 1.5,
they are uncompressable files. They have been compressed already by
the software developer. Also of course, they are not in the cache.
Another piece of software that I use on occasion is the Opera browser.
When I do file downloads with it, it shows the tranfer rate in
kilobytes per second (as opposed to kilobits per second). In the past,
I've seen MPEG downloads in Opera at speeds over 25 kilobytes per
second - translated that's 200 kilobits per second. MPEG's are
basically uncompressable.
Ah, if only I could get those high speeds all the time. Yesterday, I
was transitting the Delaware river. Speeds were horrible all day. Web
page loading was often up to 5 minutes per page. The pages were
loading in fits and starts. It's the latency that's really the pits
with EDGE. I would see the Sony Ericsson Wireless Manager's reported
speed drop to zero often for some seconds, then it would start up
again, retrieving the requested page. This is with a 3-5 bar signal.
John Navas
In <1134475388.1...@o13g2000cwo.googlegroups.com> on 13 Dec 2005
04:03:08 -0800, "GomJabbar" <dkba...@earthlink.net> wrote:
>The problem is, I've never seen EDGE to maintain a maximum speed
>connection for more than a minute or two. I am not sure when I will
>see the above speeds again, so who knows when I could do the FTP.
That would seem to support what I'm suggesting -- variation due to factors
I've mentioned.
>Anyway, I believe the Sony Ericsson Wireless Manager to be accurate in
>the speeds it reports,
I've seen too much inconsistency (e.g., 1000 versus 1024 for kilo, different
assumptions on bits per byte, different assumptions on overhead) to trust the
reporting in any software, which is why I rely on my own timing to ensure
accuracy and consistency. That's why (the question of honesty aside) there's
just no way to meaningfully compare most reported speeds on Usenet.
>and it reports realtime, not average time. Your
>test would report average time.
Flash readings are definitely unreliable due to things like buffering in the
comm channel, error recovery, etc.
>I pretty sure I wouldn't see my above
>posted speeds as an average for a 10 MB download.
With all due respect, higher flash speeds aren't meaningful.
>BTW, when you download software like OpenOffice.org 2.0 or Java 1.5,
>they are uncompressable files. They have been compressed already by
>the software developer.
Some are; some aren't. You can't be sure unless you subject the data to
analysis (e.g., by trying to compress it further). That's why it's best to
use a reliable synthetic test file like the one I cited.
>Also of course, they are not in the cache.
On successive trials they may well be.
>Another piece of software that I use on occasion is the Opera browser.
>When I do file downloads with it, it shows the tranfer rate in
>kilobytes per second (as opposed to kilobits per second). In the past,
>I've seen MPEG downloads in Opera at speeds over 25 kilobytes per
>second - translated that's 200 kilobits per second. MPEG's are
>basically uncompressable.
Not necessarily. Again, it's best to use a reliable synthetic test file like
the one I cited.
>Ah, if only I could get those high speeds all the time. Yesterday, I
>was transitting the Delaware river. Speeds were horrible all day. Web
>page loading was often up to 5 minutes per page. The pages were
>loading in fits and starts.
You're not alone -- Cingular data has been erratic here in the Bay Area for
almost a week. I think the rollout of HSDPA may be a factor.
>It's the latency that's really the pits
>with EDGE.
Other than these recent network problems, EDGE latency has been surprisingly
low here in the Bay Area.
>I would see the Sony Ericsson Wireless Manager's reported
>speed drop to zero often for some seconds, then it would start up
>again, retrieving the requested page. This is with a 3-5 bar signal.
That's probably error correction, packet loss, or network congestion, not
latency per se. Except for the initial access latency, HTTP is capable of
uninterrupted transfers over high latency paths.
GomJabbar
obscure the facts. Although you spew a bunch of technical jargon, that
doesn't give validity to your arguements. Politicians spew out a lot
of double-talk too; expounding their 'truths'. Should we believe them?
I've not fit in with your stated opinion and limited user experience.
So you must contest.
If I have a different user experience than you, in your world I must be
wrong. I don't contest your personal findings. But I do contest your
extrapolating them to apply to all other users - including myself.
How does that song go by Carly Simon?
-----------------------------
The speeds I received occurred over a several minute time period. I
was moving at the time, so it's not surprising that my connection would
wane after a while. This particular area in LI Sound, in the wee hours
of the morning, happened to offer the best EDGE speeds I've seen. I've
had EDGE service for some 9-months now. I've seen speeds all over the
map. Some days in some areas usually in the wee hours of the morning,
I will average 200 - 220 KBPS for a time. This is NOT my average EDGE
service overall. The other day in the Deleware River, I couldn't even
average 50 KPBS.
The latest firmware update for the GC83 - R4D9 shows EDGE or GPRS in
the green triangle above the signal strength bars when using Sony
Ericsson's Wireless Manager. I never could tell before, but now I
occasionally see that I am getting GPRS not EDGE. Yet even when it
displays that I am on EDGE (literally and figuratively) there are still
times I can only get 25 - 50 KBPS service. : > (
Achieving 240 Kbps is possible under ideal conditions - but not likely
to be sustained over a period of time. The number of users accessing
the cell tower makes a hugh difference in data thoroughput. I saw a
chart that shows 130 Kbps is achievable with about 40 users per sector.
Up that number to 100 users per sector, and your data rate will likely
drop to around 50 Kbps.
You stated previously:
> Current GPRS and EDGE implementations are limited to 4 time slots.
> Theoretical maximum raw speeds (including overhead) are:
> * 80 Kbps (4 x 20.0 Kbps) for GPRS with CS4 coding.
> * 237 Kbps (4 x 59.2 Kbps) for EDGE with MCS9 coding.
I've seen the above stated in a white paper available from Cingular.
Interestingly the Sony Ericsson GC83 User Guide specifies the
following:
* 247.40 Kbps (4 X 61.85 Kbps) for EDGE with MCS-9 coding.
DecaturTxCowboy
> If I (GomJaber) have a different user experience than you (Navas), in your world I must be
> wrong.
"Now you get it!"
John Navas
In <1134505207.6...@g43g2000cwa.googlegroups.com> on 13 Dec 2005
12:20:07 -0800, "GomJabbar" <dkba...@earthlink.net> wrote:
>Sorry but I don't buy into your arguements. Just a lot of words to
>obscure the facts. Although you spew a bunch of technical jargon, that
>doesn't give validity to your arguements. ...
Those are facts, not arguments. But whatever -- you are free to believe
whatever you want, just I'm free not to waste any more time posting facts.
FWIW, this kind of response is why I often don't bother re-posting facts --
I know from experience that certain folks are just going to reject them out of
hand.
>Achieving 240 Kbps is possible under ideal conditions - but not likely
>to be sustained over a period of time. The number of users accessing
>the cell tower makes a hugh difference in data thoroughput. I saw a
>chart that shows 130 Kbps is achievable with about 40 users per sector.
> Up that number to 100 users per sector, and your data rate will likely
>drop to around 50 Kbps.
That's not how GPRS works. What matters is (a) the number of users and (b)
the number of available data timeslots (which can vary considerably).
>You stated previously:
>> Current GPRS and EDGE implementations are limited to 4 time slots.
>> Theoretical maximum raw speeds (including overhead) are:
>
>> * 80 Kbps (4 x 20.0 Kbps) for GPRS with CS4 coding.
>> * 237 Kbps (4 x 59.2 Kbps) for EDGE with MCS9 coding.
I also stated:
But those speeds won't normally be seen because the those particular codings
are very limited in range and very sensitive to errors. Thus in practice the
best raw speeds are:
* 58 Kbps (4 x 14.4 Kbps) for GPRS with CS3 coding.
* 179 Kbps (4 x 44.8 Kbps) for EDGE with MCS7 coding.
(Cingular initially deployed GPRS limited to CD2 coding, and only enabled CS3
coding when the system has proven itself.)
Net data throughput (less protocol and error correction overheads) is
significantly less, on the order of:
* 52 Kbps for GPRS
* 160 Kbps for EDGE
Thus, as I wrote, the EDGE speeds you're reporting are very unlikely in the
real world.
>I've seen the above stated in a white paper available from Cingular.
>
>Interestingly the Sony Ericsson GC83 User Guide specifies the
>following:
>* 247.40 Kbps (4 X 61.85 Kbps) for EDGE with MCS-9 coding.
The *data rate* of MCS-9 per timeslot is 59.2 Kbps per ETSI TS 145 001
<http://webapp.etsi.org/action/PU/20020129/ts_145001v040001p.pdf>.
That's the *raw* data rate -- throughput (exclusive of overhead) is less --
and MCS-9 isn't used in practice for EGPRS(EDGE), just as CS-4 (which would
give a raw speed of 80 Kbps) isn't used in practice for GPRS, for the reasons
I stated above.
I'm now done with this. Feel free to think whatever you want, and to have the
last word.
SMS
Cingular has never claimed that EDGE was 3G.
What they said was "EDGE is recognized as the most
economical way to overlay GSM/GPRS to achieve 3G
capabilities"
This statement is intentionally vague. 3G "capabilities"
which could mean a lot of things, i.e. services that
were expected to be available on 3G are now available
on EDGE, etc. This is in fact the same sort of claim the
Finnish carrier Sonera alluded to when they claimed that
they had deployed "3G."
Read "http://newsletter.nordicwirelesswatch.com/story.php?story_id=2133"
"There used be, before they were stolen, a couple of fairly
concrete criteria for 3G networks. One said that it must use
the 3GPP (3G Partnership Program) defined air interface
and the other that it must be capable of transmitting 2
megabits per second between the network and the 3G
handset."
SMS
> I have never seen any speeds on EDGE approaching 528 KBPS! You tell
> me, which is the TRUE 3G network?! {I am not making any claims
> regarding UTMS here}
Cingular does not claim that EDGE is 3G. Using the 3G air interface does
not make EDGE 3G. Navas has been claiming that EDGE is 3G for years, but
you will be hard-pressed to find anyone else in the world that believes
that.
> This test also tends to confirm statements made elseware that Verizon's
> service is superior to Cingular's in the NYC area. Cingular just can't
> handle the traffic as well as Verizon can (at least in this area).
> Otherwise, EDGE would be up to the task here in NYC and give at least
> advertised download rates.
AT&T Wireless has had severe capacity issues in NYC. Cingular was using
the T-Mobile GSM network prior to the acquisition.
John Navas
In <43a30d1c$0$95931$742e...@news.sonic.net> on Fri, 16 Dec 2005 10:53:29
-0800, SMS <scharf...@geemail.com> wrote:
>GomJabbar wrote:
>
>> I have never seen any speeds on EDGE approaching 528 KBPS! You tell
>> me, which is the TRUE 3G network?! {I am not making any claims
>> regarding UTMS here}
>
>Cingular does not claim that EDGE is 3G. Using the 3G air interface does
>not make EDGE 3G. Navas has been claiming that EDGE is 3G for years, but
>you will be hard-pressed to find anyone else in the world that believes
>that.
I guess you're speaking for yourself. I have no such problem; e.g.,
<http://www.ericsson.com/products/hp/Mobile_Broadband_EDGE_bs.shtml>
<http://www.nokia.com/link?cid=EDITORIAL_1485>
--
John Navas
In <43a30c26$0$95928$742e...@news.sonic.net> on Fri, 16 Dec 2005 10:49:22
-0800, SMS <scharf...@geemail.com> wrote:
>GomJabbar wrote:
>> I wrote:
>>> EDGE is not 3G - it is considered 2.5G.
>>
>> John Navas wrote:
>>> I disagree.
Here is what you snipped:
<http://www.ericsson.com/products/hp/Mobile_Broadband_EDGE_bs.shtml>
EDGE is 3G radio technology that triples the high-speed data communication
capacity of GPRS-enhanced GSM networks, enabling GSM operators to provide
high-speed mobile Internet services over existing infrastructure and with
an existing GSM license.
<http://www.nokia.com/link?cid=EDITORIAL_1485>
The world's first GSM/GPRS/EDGE 3G phone, the Nokia 6200, was launched in
November 2002 and has been in shops since May 2003.
GomJabbar
I have not had a decent EDGE connection in the last week to make any
further tests with EDGE worthwhile at this point. But my coworker did
perform some more EV-DO tests.
John Navas wrote:
> By comparison, EV-DO ranges from about 100 Kbps to about
> 300 Kbps, with a typical speed of about 170 Kbps.
> Thus there isn't much difference between them.
> Perhaps there is more excess EV-DO network capacity
> in your area.
John Navas also wrote:
> Don't depend on what programs report.
> dslreports is notoriously unreliable.
> For dependable results, download uncompressible data by FTP, time how long it
> takes, and calculate throughput from size and time. Do at least three trials.
> Restart the computer between successive trials.
> Here's a good test file:
> <ftp://ftp.sonic.net/pub/testfile.compressed.10meg>
OK, a couple of days ago my coworker on Verizon's EV-DO Broadband
Access performed the speed test from dslreports.com. The actual test
site we used for all speed tests was
http://middletown.speedtest.frontiernet.net/ . The cache was cleared
before and between all tests. He tested twice and got 1075 Kbps
download the first time and 966 Kbps download the second time.
Next he went to the site you posted and ran the test one time:
ftp://ftp.sonic.net/pub/testfile.compressed.10meg . I timed the
download from the time it started. It took 1minute, 18seconds. This
equals 78 seconds. Take 10 Mb X 1024 (1024 bytes to a megabyte) =
10,240 bytes. Divide 10,240 by 78 and you get 131.28 kilobytes/sec.
Multiply that by 8 and you get 1050 kilobits/sec. Sure a lot different
than the 'facts' you posted above. I see consistent speeds with EV-DO
under good conditions to be around 1000 Kbps. ' Smacks EDGE's behind.
LOL
I wrote:
> I performed a large file download with my Cingular EDGE connection in
> LI sound and saw download speeds for a couple of minutes hovering
> around 220 KBPS, with one spike of 240 KBPS.
Noticed I said a "one spike of 240 KBPS". That's exactly what it was -
a spike. However I did see the speeds average around 220 KBPS for a
couple of minutes. Note that I did upgrade the firmware to R4D9 in my
GC83 card on Dec. 6th, 2005. The upgrade was supposed to 'improve
performance'. I don't know if it had any real effect or not.
Notice what PC Magazine wrote:
http://www.pcmag.com/article2/0,1895,1770749,00.asp
> And though Class 10's theoretical maximum speed is 236 Kbps, we maxed
> out at 220 Kbps during testing. That's more than three times as fast as dial-up,
> but still much slower than cable, DSL, or EV-DO.
Hmmm.... Who to believe?