nchw68
join:2003-03-19
Chula Vista, CA
2 edits | Cox vs. SBC
Right now I have both Cox HSI and SBC DSL(new service), both are dynamic accounts. I had SBC a few years ago in Anaheim and am considering them again since I want to share Mame roms via P2P. I only used WinMX and it worked good for awhile with Cox. Then suddenly all uploads failed, and I do mean all. Since then WinMX has shut down (it still survives) and now I only get a couple of upload connections all night, but some go through when connected via SBC. I'm using Frostwire now as well and uploads with it are working about 50% of the time with SBC. Tonight I will run Frostwire via Cox and see what happens. My guess is that no uploads will go through.
All I do to switch between providers is disable the local area connection in WinXP and switch ethernet cables going to the card, then enabling the connection again. Very quick.
I don't use a router with Cox, just the modem. With SBC I'm using a 2-Wire 1701HG modem/router with the firewall set to allow everything through since I'm using Sygate instead. So there's more hardware with the SBC connection.
As of now, it's not looking good for Cox. I will overlap the two services for a few months and if SBC proves to be reliable at my location then Cox is going in the crapper.
Yeah, I know. Cox couldn't care less if they lose my business. |
|
nchw68
join:2003-03-19
Chula Vista, CA
| I did a trace route to DSL Reports with Cox and SBC and here's what I got with SBC:
Tracing route to dslreports.com [209.123.109.175]
over a maximum of 30 hops:
1 8 ms 2 ms 1 ms home [192.168.1.254]
2 15 ms 15 ms 13 ms ppp-71-136-47-254.dsl.sndg02.pacbell.net [71.136
.47.254]
3 14 ms 13 ms 15 ms dist1-vlan50.sndg02.pbi.net [63.200.206.130]
4 14 ms 13 ms 15 ms bb2-g5-0-1.sndgca.sbcglobal.net [66.121.119.50]
5 13 ms 13 ms 15 ms 151.164.40.185
6 18 ms 17 ms 17 ms bb1-p10-1.crrvca.sbcglobal.net [151.164.43.26]
7 18 ms 17 ms 15 ms core2-p4-0.crrvca.sbcglobal.net [151.164.41.1]
8 44 ms 45 ms 44 ms core2-p10-0.crhstx.sbcglobal.net [151.164.42.4]
9 57 ms 56 ms 57 ms core1-p11-0.cratga.sbcglobal.net [151.164.240.11
4]
10 56 ms 57 ms 56 ms core2-p1-0.cratga.sbcglobal.net [151.164.241.82]
11 69 ms 68 ms 68 ms core2-p6-0.crhnva.sbcglobal.net [151.164.41.206]
12 69 ms 68 ms 68 ms bb2-p4-1.hrndva.sbcglobal.net [151.164.243.138]
13 68 ms 68 ms 69 ms ex2-p5-0.eqabva.sbcglobal.net [151.164.191.138]
14 69 ms 68 ms 68 ms eqix.ge-0-0-0.gbr1.ash.nac.net [206.223.115.69]
15 74 ms 75 ms 106 ms 0.so-2-2-0.gbr2.nwr.nac.net [209.123.11.29]
16 74 ms 74 ms 74 ms 0.so-0-3-0.gbr1.oct.nac.net [209.123.11.233]
17 75 ms * 74 ms www.dslreports.com [209.123.109.175]
Trace complete.
Here's what I got with Cox:
Tracing route to dslreports.com [209.123.109.175]
over a maximum of 30 hops:
1 14 ms 8 ms 11 ms 10.111.172.1
2 87 ms 26 ms 10 ms cv1xaggc01-gex0914.sd.sd.cox.net [68.6.10.186]
3 10 ms 24 ms 9 ms fed1sysc01-gex0910.sd.sd.cox.net [68.6.8.34]
4 16 ms 14 ms 30 ms fed1dsrj01-ge702.rd.sd.cox.net [68.6.8.194]
5 9 ms 20 ms 10 ms fed1bbrc01-pos0101.rd.sd.cox.net [68.1.0.204]
6 48 ms 44 ms 45 ms mtc3bbrc01-pos0103.rd.ok.cox.net [68.1.0.209]
7 66 ms 47 ms 46 ms mtc3bbrc02-pos0300.rd.ok.cox.net [68.1.0.121]
8 80 ms 80 ms 79 ms lkhnbbrc01-pos0202.rd.ok.cox.net [68.1.0.118]
9 80 ms 83 ms 82 ms lkhnbbrc02-pos0100.rd.at.cox.net [68.1.0.3]
10 95 ms 106 ms 95 ms 68.1.1.228
11 93 ms 95 ms 96 ms 68.105.30.94
12 96 ms 94 ms 95 ms 0.ge-0-0-0.gbr1.ash.nac.net [209.123.11.41]
13 101 ms 100 ms 100 ms 0.so-2-2-0.gbr2.nwr.nac.net [209.123.11.29]
14 102 ms 101 ms 102 ms 0.so-0-3-0.gbr1.oct.nac.net [209.123.11.233]
15 109 ms 104 ms 112 ms www.dslreports.com [209.123.109.175]
Trace complete.
The SBC trace looks faster. Would you call this a significant difference? |
|
coxengr
Premium,VIP
join:2002-03-09
Atlanta, GA
| The SBC trace isn't "faster" - it simply has a route that's got lower latency. To do a true apples-to-apples comparison, you'd need to perform the same test to a good set of random sample sites around the Internet. Then you can see if it's a factor or not.
Oh, and by the way, we (Cox) do care! 
--
Want the most out of BBR? Visit our help page: »members.cox.net/coxengr/dslr_help |
|
nchw68
join:2003-03-19
Chula Vista, CA
1 edit | said by coxengr  :The SBC trace isn't "faster" - it simply has a route that's got lower latency. I envision most people would translate "faster" into lower latency. I didn't mean SBC used faster electrons than Cox. Sheesh, ya gotta spell everything out for engineers! The two traces still favor SBC if you ask me, but I probably will do some other side by side traces to other destinations like you suggested to get a better picture.
said by coxengr :Oh, and by the way, we (Cox) do care! But not enough to allow P2P to fully function.
I see you made no comment on the P2P thing. Interesting. |
|
coxengr
Premium,VIP
join:2002-03-09
Atlanta, GA
| Well, there are a lot of people that don't understand the difference between speed and latency, and their relationship. So I point that out for others that may be reading this post.
As far as P2P, I really wasn't sure where you going with that. But there's no reason your P2P app shouldn't perform as well if not better than SBC.
--
Want the most out of BBR? Visit our help page: »members.cox.net/coxengr/dslr_help |
|
nchw68
join:2003-03-19
Chula Vista, CA
| P2P uploads in Frostwire are working with SBC and a router. Before going to bed tonight I'll switch to Cox using no router and run Frostwire all night and see what has transpired in the morning. I'll post what I find. If uploads all fail then I would say there is definitely a reason for it. P2P uploads are the only thing I find not functioning with Cox. That seems very specific, and therefore, suspicious to me. |
|
Tsume
join:2004-02-23
Johnson City, TN
 ·Embarq
 ·ViaTalk
 ·Comcast
| reply to coxengr
Only on COX do I get little to no gnutella uploads, and then the ones I do get only upload less than 16kB before magically canceling.
Granted using most p2p apps is like running a "server" and thus isn't allowed. I just hope in the future that COX realizes being able to run a "server" is something many consumers would like to see from cable providers.
I can't complain though. 9mbps/1mbps is great, and COX would have to somehow piss me off a ton to get me to leave within the forseeable future.
--
"True warriors do not follow paths, they make them. It is not just their desire, it is their nature." (Battletech) |
|
nchw68
join:2003-03-19
Chula Vista, CA | Makes you wonder just how useful 1mbps upload is outside of email and gaming. I guess if I host a game I would be violating the TOS, but I think Cox is afraid to start enforcing their AUP on hosting games. |
|
Tsume
join:2004-02-23
Johnson City, TN
·Embarq
·ViaTalk
·Comcast
| When they say no servers, they originally meant no HTTP or FTP servers, bandwidth whoring 16+ player game servers, anything that would cause you to consume way more bandwidth than you should. Now they're stretching the term so they have an excuse to do things the way they want to, such as blocking gnutella uploads.
Technically speaking, anything with a listening socket is pretty much a server. That's not what COX is trying to ban when they say "no servers" though.
COX isn't going to disconnect you for hosting a game on Starcraft or Warcraft or *insert other low-bandwidth game that doesn't require a dedicated server*
--
"True warriors do not follow paths, they make them. It is not just their desire, it is their nature." (Battletech) |
|
b5turbo
join:2005-01-28
Manhattan, KS
| reply to coxengr
The people at the corporate office are not going to care one bit if they lose a customer, since that ex customer will be replaced by someone else or by people who have no other choice in their market area. Me, Im lucky enough to either have SBC or Cox. Cox had the cheapest static IP package available so I went with them and I already had a cable modem laying around. |
|
nchw68
join:2003-03-19
Chula Vista, CA
1 edit | reply to coxengr
Doesn't seem to be a whole lot of people looking for Mame roms in the Gnutella community, but last night, using Frostwire with Cox, there were 14 transfers initiated from 12 different IPs and all of them were interrupted. Had I been connected with SBC some of those transfers would have been successful, probably not all, but some.
said by coxengr :But there's no reason your P2P app shouldn't perform as well if not better than SBC. You have got to be sh#*ting me with that statement. Either you're just playing dumb or you're being kept in the dark (I think it's the former). This is beyond traffic prioritizing. It's flat out preventing all P2P uploads. I couldn't work for a company that required me to lie or mislead the customer. If you're truly in the dark on this matter, then I apologize.
Edit:
I just ran Frostwire for about four hours with SBC and only got three uploads, but they all went through 100%. |
|
COXIT
join:2005-11-18
Santa Barbara, CA
| if they completed then it's certainly not a COX issue. try a different client (uTorrent, azurus...).
it's pretty much impossible to blame cox for a p2p problem when there a seriously dozens of factors involved, ie router not capable of so many connections, MPAA interference, remote hacking by tracker leeches, filtering at any one of the dozens of hops...
try to remember that most ISPs are federlaly mandated to manage certain threats to infrastructure. this would include spam, saturation, DoS attacks, viral broadcasting and books more of other stuff. Balancing the customer's needs and the gevernments demands is tough work. |
|
nchw68
join:2003-03-19
Chula Vista, CA
| said by COXIT :if they completed then it's certainly not a COX issue. You're not reading something right. All uploads fail with the Cox connection which i don't even use a router for. Uploads succeed with the SBC connection which I use a router for. This is with using the same computer and software for both services. I have both services right now and can switch from one to the other quickly.
It doesn't get much more black and white than that. You folks that can still upload via P2P with Cox should consider yourself fortunate, but it may not last.
However, I may just keep BOTH services so I can game online with one and my girlfriend could be sending out photos via email with the other without it affecting the game. |
|
2kmaro
Think
Premium,ExMod 1 BC
join:2000-07-11
ColossalCave
clubs:  
| Keeping both is certainly an option as long as the budget doesn't complain too loudly.
As far as that latency issue goes, nothing looks bothersome except possibly for gaming. While the difference between (last hop numbers) of 112 and 74 looks largish in absolute terms, remember we're talking in 1/1000's of a second here - so basically one message got through about 40/1000 of a second slower/faster than the other. That's .004 seconds difference, to put it in decimals.
You can also look at that another way: number of hops and thus number of chances for problems. 14 with Cox, 17 with SBC.
You can run DrPing from this page here at DSLR:
»/beta/doctorping and compare your results to mine (from Cox in OKC)
18.6ms 64.200.110.74
23ms ns1.dl.cox.net
28.6ms 151.164.67.134
29.6ms 151.164.70.132
30ms 66.185.133.154
34.6ms 0.so-7-0-0.xr1.dfw9.alter.net
35.6ms hsa1.orlando1.level3.net
35.6ms hsipaccess1.dallas1.level3.net
37.6ms hsipaccess1.houston1.level3.net
39.6ms hsa1.losangeles1.level3.net
40ms hsa1.raleigh1.level3.net
40.6ms dist1-vlan50.milwwi.ameritech.net
44ms hsipaccess1.denver1.level3.net
44.6ms hsa1.cincinnati1.level3.net
45.6ms 68.47.160.5
46.3ms 64.200.118.241
46.6ms p1-1-0-0.r03.mclnva01.us.bb.verio.net
47.6ms hsa1.sacramento1.level3.net
48ms ns1.no.cox.net
48.6ms ns.atl.bellsouth.net
50.3ms pos5-0.ur2.atl7.web.wcom.net
51ms hsipaccess1.atlanta1.level3.net
51.3ms ns1.oc.cox.net
51.6ms hsa1.miami1.level3.net
52.6ms 0.so-4-0-0.xl1.sat1.alter.net
52.6ms 0.so-6-0-0.xl2.atl5.alter.net
53.3ms hsa1.lasvegas1.level3.net
53.6ms ns1.at.cox.net
53.6ms 69.15.9.2
54ms 38.112.7.25
55ms sl-gw9-rly-9-0.sprintlink.net
55.6ms 0.so-3-0-0.cl2.iad5.alter.net
57.3ms 24.29.100.41
57.3ms 24.29.100.85
58ms 24.29.100.13
61ms 0.so-7-0-0.xl2.nyc1.alter.net
61.6ms 0.so-5-0-0.CL2.PHX2.ALTER.NET
65.3ms ns.msy.bellsouth.net
65.3ms 216.52.69.18
67ms 213.ATM6-0.GW1.VEG2.ALTER.NET
67.3ms hsipaccess1.chicago1.level3.net
68.6ms 66.28.4.93
72.3ms 196.ATM7-0.XR2.SF04.ALTER.NET
73ms hsipaccess1.boston1.level3.net
74.6ms ns1.mia.bellsouth.net
75ms so-0-2-0.dex1.Honolulu1.Level3.net
75.3ms 125.at-5-0-0.TR1.ATL5.ALTER.NET
76.3ms ge-0-2-0.a03.hstntx01.us.ra.verio.net
77.3ms hsipaccess1.detroit1.level3.net
77.3ms gige-g1-0-105.gsr12012.chi.he.net
78.3ms p2-0.lsanca1-br1.bbnplanet.net
78.6ms hsipaccess1.newyork1.level3.net
79.6ms sl-gw11-sea-8-0.sprintlink.net
80.6ms sl-w1-mae.sprintlink.net
84ms 69.25.0.14
84.3ms hsipaccess1.washington1.level3.net
85.6ms hevacis-04-pos51.conxion.net
86ms hsipaccess1.sanjose1.level3.net
87.3ms sl-bb11-nyc-4-0.sprintlink.net
88.6ms sl-bb23-chi-4-2.sprintlink.net
88.6ms noio-fe-0-0-0.lava.net
88.6ms hsipaccess1.seattle1.level3.net
89.6ms acr1-loopback.Chicagochd.cw.net
90.6ms 144.232.20.94
92ms as-0-0.mp1.phoenix1.level3.net
94.6ms core1.iad.above.net
95.6ms ns2.snet.net
97ms hsa1.stlouis1.level3.net
98.3ms 66.28.4.142
99.6ms acn13a.al-birmingh1.ne.earthlink.net
100ms sl-bb20-kc-11-2.sprintlink.net
100.6ms sl-gw7-nyc-0-0-0.sprintlink.net
102.6ms 216.109.118.69
103.3ms 0.so-2-0-0.xl2.nol1.alter.net
104ms r8-pao1.r3.sfo2.isc.org
106.6ms s3-1-gsr1.nwc.acsalaska.net
106.6ms 66.7.153.202
143ms 206.123.6.10
144ms iiwi-fe-0-0-0.lava.net
160ms pos1-0.gsr12012.sjc.he.net
227.6ms 192.104.242.82 |
|
inTulsa
Premium
join:2002-02-24
| said by 2kmaro :That's .004 seconds difference, to put it in decimals. .04 second difference. But it could also be stated as 50% slower transport response.
Latency in gaming seems to compound because of the back-and-forth data transfer during online play. The .04 can easily become 1.0+ when player movements and action clicks of the participants are combined through the sockets. That's what seems to happen in the few that I've tried. By the time the server can synchronize my aim & firing of the weapon, the target is long gone and my player may have already been hit a few times.
Latency in gaming matters quite a bit, and some DSL players seem to have a real edge. Or maybe it's my slow brain thinking that I should have aimed and meant to fire. |
|
needforspeed59
Cruise Ship Just Passing Through
join:2001-05-02
Glendale, AZ | reply to nchw68
I don't think the grass is all that greener for Qwest DSL customers. Read this. »Qwest DSL - terrible xbox live performance
--
Of all the people I know... you're one of them. |
|
reelbigfish
join:2002-06-06
Boston, MA
·Comcast
·Comcast Digital Vo..
| I have the 5/2 package and I play Battlefield 2, Battlefield 1942 and UT2004. So far I haven't notice too much of a slow down. Since I switched to cable from DSL, I haven't notice that much of a difference in my score. Occasionally it seems like it lags, but it is infrequent. I do play during the week at peak times, and my bandwidth may be diminished to about 2Mbps down, but I still get good latency. I have the same Linksys router I used with SBC DSL and a Motorola SB5100. I do normally play in servers located in the North East, Quebec and Virgina. I did try playing on a west coast server and didn't really notice too much difference. From my personaly expericence there isn't too much difference between Cox and SBC. SBC here routes from CT, to VA and then back to NYC because it is cheaper. I have better routes with Cox than SBC to NYC, upstate NY and Quebec. |
|
DarnellP
join:2004-10-12
Las Vegas, NV
| reply to needforspeed59
With all due respect, what in the world does Qwest have to do with this thread? The OP specifically states that he has SBC. |
|
needforspeed59
Cruise Ship Just Passing Through
join:2001-05-02
Glendale, AZ
| reply to nchw68
I just mentioned it due to a comment by InTulsa. "Latency in gaming matters quite a bit, and some DSL players seem to have a real edge. Or maybe it's my slow brain thinking that I should have aimed and meant to fire."
I wasn't trying to throw the thread off topic.
--
Of all the people I know... you're one of them. |
|
inTulsa
Premium
join:2002-02-24
| said by needforspeed59 :I wasn't trying to throw the thread off topic. I should have said SBC instead of DSL. I don't know about Quest.
Some geeky friends here have switched from Cox to DSL, and they're not coming back. At our house we don't have the SBC option. Valor Telecom isn't in the same league.
catseyenu posted some comparison after getting SBC, shown here: »[LA] Further Questions on Apparent Latency to Gateway
The graphs from those line quality tests is what I re-posted here, SBC shown above Cox for the same customer location and PC equipment. It's the same image posted in the other thread. It illustrates the kind great consistency that I've seen in other people's SBC service.
Distance matters for DSL, so the quality of service can be poor in a fringe locations. Also, maybe SBC's service varies greatly by geographical region like Cox's does.
Anyway I'm more apt to blame Cox than the 'slow brain' thing for my poor gaming  |
|
