ksweis
join:2008-01-01
Spring, TX
| AT&T U-Verse Elite performing sluggish! Please help.
I upgraded from the sbc yahoo dsl elite package to the AT&T U-Verse Elite 6mbps package. Now with the old package in ANY centralized server I would get an average latency or "ping" of 18-20. After upgrading to the U-Verse which should of been a pretty decent change I now receive a latency of 29-32. Now that guy that installed the service said the internet would take awhile to perform at its max being DSL. How long is this suppose to take? It has been about a week. |
|
Kylan
join:2000-07-31
Houston, TX
| Welcome to the interleaved profile my friend. 
I'm pretty sure U-verse does not have that ramp up period. You can either get it or you can't.
With SBC/AT&T's ADSL, you're on what's called the fastpath profile. You have lower pings, but a higher chance of having errors with your line, etc. With U-verse, everyone is on the interleaved profile for better stability, as much more than just internet relies on your connection to be perfect. |
|
MarkyD
Premium
join:2002-08-20
Oklahoma City, OK
clubs: | reply to ksweis
FWIW, I noticed much slower page loads on U-Verse elite than I did on my old PPPoE (FTTP) elite. The interleaved profile did make a difference in latency for me, a noticeable one at that.
--
MCSE, ACSA, and a lot more |
|
ksweis
join:2008-01-01
Spring, TX | reply to ksweis
I find myself asking why did I upgrade for higher latency. |
|
removed
I performed my own circumcision
Premium,VIP
join:2002-02-08
Houston, TX
clubs:
| reply to Kylan
said by Kylan  :Welcome to the interleaved profile my friend. I really don't think that the interleaved profile makes any difference -- at least it doesn't make any difference whatsoever in my case. With that said, how about some traceroutes:
$ traceroute 66.7.208.63
traceroute to 66.7.208.63 (66.7.208.63), 30 hops max, 40 byte packets
1 192.168.1.254 (192.168.1.254) 6.967 ms 6.840 ms 6.804 ms
2 70-140-188-2.lightspeed.hstntx.sbcglobal.net (70.140.188.2) 25.133 ms 25.536 ms 25.556 ms
3 * * *
4 * * *
5 * * *
6 151.164.43.136 (151.164.43.136) 28.936 ms 21.035 ms 21.581 ms
7 (151.164.40.29) 28.152 ms 27.921 ms 28.471 ms
8 asn336-level3.eqdltx.sbcglobal.net (151.164.248.202) 29.157 ms 31.034 ms 30.893 ms
9 vlan69.csw1.Dallas1.Level3.net (4.68.19.62) 30.855 ms 30.839 ms 35.315 ms
10 ae-61-61.ebr1.Dallas1.Level3.net (4.69.136.121) 31.955 ms 31.248 ms 29.401 ms
11 ae-9-9.car2.Orlando1.Level3.net (4.69.133.69) 66.329 ms 66.671 ms 67.353 ms
12 HOSTDIME.car2.Orlando1.Level3.net (4.79.118.38) 67.397 ms 67.331 ms 67.290 ms
13 brazos.root--servers.net (66.7.208.63) 66.929 ms 67.068 ms 67.275 ms
100 pings to the same IP:
--- 66.7.208.63 ping statistics ---
100 packets transmitted, 100 received, 0% packet loss, time 99001ms
rtt min/avg/max/mdev = 65.229/66.678/69.512/0.772 ms
67ms from Houston to Orlando. Interleaved profile with Elite U-Verse package. I'm really curious to see how much of a difference other users see.
--
irc.removed.us - #dslr | DSLR Phishtracker | Email: removed@dslr.net |
|
Enlightener
join:2006-01-28
Cedar Park, TX
 ·AT&T U-Verse
edit:
January 1st, @08:55PM
| reply to ksweis
Tracert from Austin, TX with FTTP Elite. Note that my first hop is 20ms faster then yours and that my hop 6 is your hop 8 and again I've got you beet by 20ms. At the destination it's about 16ms. That's about a 25% reduction in latency.
1 1 ms 1 ms 1 ms home [192.168.1.254]
2 7 ms 5 ms 5 ms bras1-l0.austtx.sbcglobal.net [151.164.182.136]
3 4 ms 4 ms 6 ms dist1-vlan60.austtx.sbcglobal.net [151.164.20.16
2]
4 4 ms 4 ms 4 ms bb1-g8-3-0.austtx.sbcglobal.net [151.164.94.14]
5 10 ms 10 ms 10 ms 151.164.95.186
6 10 ms 11 ms 10 ms asn336-level3.eqdltx.sbcglobal.net [151.164.248.
202]
7 19 ms 16 ms 17 ms vlan69.csw1.Dallas1.Level3.net [4.68.19.62]
8 20 ms 17 ms 18 ms ae-61-61.ebr1.Dallas1.Level3.net [4.69.136.121]
9 52 ms 51 ms 51 ms ae-9-9.car2.Orlando1.Level3.net [4.69.133.69]
10 51 ms 51 ms 51 ms HOSTDIME.car2.Orlando1.Level3.net [4.79.118.38]
11 51 ms 52 ms 51 ms brazos.root--servers.net [66.7.208.63] |
|
koolkid1563
Premium,MVM
join:2005-11-06
Powell, WY
clubs: | reply to MarkyD
You wouldn't be on the interleaved or fastpath channel on either simply because you aren't using DSL anywhere. |
|
daveshome
@sbcglobal.net | Just switched from Elite DSL to Elite U-Verse.
Found games to get jerky. Checked and found the same.
Latency is higher by about 50%. Hope it gets fixed.
Dave |
|
koma3504
Advocate
Premium
join:2004-06-22
North Richland Hills, TX | reply to ksweis
consider your self lucky it only rose that much.
i have a buddy that is luckey to get 200ms on his ping time. |
|
MarkyD
Premium
join:2002-08-20
Oklahoma City, OK
clubs:
 ·Cox HSI
·AT&T FTTP
 ·AT&T DSL Service
| reply to koolkid1563
said by koolkid1563 :You wouldn't be on the interleaved or fastpath channel on either simply because you aren't using DSL anywhere. true, I didn't think of that. I wonder why my page loads on the lightspeed network were so much slower than on the legacy network. Also, my pings and tracerts were much better on the legacy network. strange. |
|
Enlightener
join:2006-01-28
Cedar Park, TX | reply to koolkid1563
I'm running over the same ATM back end that DSL is. I hit the same redbacks, get IP's out of the same address block. My physical network is a different implementation, but the logical backend network is all the same. |
|
Enlightener
join:2006-01-28
Cedar Park, TX | reply to MarkyD
Probably because the IP network was opitmized for a different purpose and round trip latency wasn't it. |
|
koolkid1563
Premium,MVM
join:2005-11-06
Powell, WY
clubs: | reply to Enlightener
Same backend yes, but you aren't using DSL anywhere therefore you can't be using the interleaved or fastpath channels as that is for DSL. |
|
Enlightener
join:2006-01-28
Cedar Park, TX | From what I've read on the subject, cell interleaving is an ATM virtual circuit concept and it really doesn't matter what physical transport ( dsl vs fiber) connects me to the ATM cloud. I suggest you do a bit more reading, kid. |
|
en102
Canadian, eh?
join:2001-01-26
Valencia, CA
·RoadRunner Cable
·DSL EXTREME
·DSL EXTREME
edit:
January 3rd, @08:08PM
| I was kind of wondering that myself... if the interleaving was at a different layer. I suspect that it could be (and possibly is) as this would make deploying consistent regardless of physical layer.
Trace from L.A.
--
Canada = Hollywood North |
|
arealdsltech
Premium
join:2007-02-27
Oklahoma City, OK
·AT&T U-Verse
| reply to MarkyD
FTTP does not have interleaved as there is no DSL signal involved. The Uverse network does use different routes to the Internet than PPPoE FTTP Internet. I am not sure why the performance seemed different. Maybe poor DNS performance or the load on the 2wire router from IPTV traffic?? |
|
Enlightener
join:2006-01-28
Cedar Park, TX
·AT&T U-Verse
| reply to ksweis
Other then latency, how is the thoroughput?
I'm on the phone with ATT FTTP and they are offering me to get Internet Elite only for $40. That's $10/mo cheaper since I get to dump my unused metered line.
This is assuming I can even get it. They have to call me back since the first system said I'm qualified and the second system said I'm not. The wierd thing is she said all new FTTP orders are being forced to go U-Verse and this gets `tricky` in Cedar Park, TX. |
|
ksweis
join:2008-01-01
Spring, TX
| reply to ksweis
I experienced a 10ms increase in latency solid. Never drops below 29 to servers that are maybe 4 hours away from me. I dont actually FEEL the lag difference but it just bugs me that it's still there. DSL doesn't randomly crash anymore but the fact that I'm upgrading and getting higher latency bugs the hell out of me. |
|
en102
Canadian, eh?
join:2001-01-26
Valencia, CA | To me, an extra 10ms isn't a big deal, as long as the line is stable. I don't want to try running Vonage over a line that's high/low on latency.
--
Canada = Hollywood North |
|
arealdsltech
Premium
join:2007-02-27
Oklahoma City, OK | reply to Enlightener
The forcing to Uverse on new installs for Internet Only I can confirm in Oklahoma City. Seems to have started on January 1. I did a speed test on one and got 5800 down and 980 up. I did not test latency and this was on FTTP. |
|
