View Full Version : Leics consistently low speeds anyone else?
Chrysalis
10-01-2006, 17:40
Hi I noticed browsing was crawling and a ftp download going of a previously alwats fast ftp go very slow and got these speed test results.
Tue, 10 Jan 2006 17:32:34 UTC
1st 512K took 4094 ms = 125.1 KB/sec, approx 1031 Kbps, 1.01 Mbps
2nd 512K took 3359 ms = 152.4 KB/sec, approx 1256 Kbps, 1.23 Mbps
3rd 512K took 2516 ms = 203.5 KB/sec, approx 1677 Kbps, 1.64 Mbps
4th 512K took 3297 ms = 155.3 KB/sec, approx 1280 Kbps, 1.25 Mbps
Overall Average Speed = approx 1311 Kbps, 1.28 Mbps
1st 512K took 2641 ms = 193.9 KB/sec, approx 1598 Kbps, 1.56 Mbps
2nd 512K took 6047 ms = 84.7 KB/sec, approx 698 Kbps, 0.68 Mbps
3rd 512K took 5187 ms = 98.7 KB/sec, approx 813 Kbps, 0.79 Mbps
4th 512K took 3719 ms = 137.7 KB/sec, approx 1135 Kbps, 1.11 Mbps
Overall Average Speed = approx 1061 Kbps, 1.04 Mbps
1st 512K took 6718 ms = 76.2 KB/sec, approx 628 Kbps, 0.61 Mbps
2nd 512K took 4047 ms = 126.5 KB/sec, approx 1042 Kbps, 1.02 Mbps
3rd 512K took 3860 ms = 132.6 KB/sec, approx 1093 Kbps, 1.07 Mbps
4th 512K took 3625 ms = 141.2 KB/sec, approx 1163 Kbps, 1.14 Mbps
Overall Average Speed = approx 982 Kbps, 0.96 Mbps
Tue, 10 Jan 2006 17:39:29 UTC
1st 512K took 2891 ms = 177.1 KB/sec, approx 1459 Kbps, 1.42 Mbps
2nd 512K took 2172 ms = 235.7 KB/sec, approx 1942 Kbps, 1.9 Mbps
3rd 512K took 2234 ms = 229.2 KB/sec, approx 1889 Kbps, 1.84 Mbps
4th 512K took 2297 ms = 222.9 KB/sec, approx 1837 Kbps, 1.79 Mbps
Overall Average Speed = approx 1782 Kbps, 1.74 Mbps
last one slight improvement but still below 2meg all the others nearer 1meg.
Upload seems fine, pings are only 1 in 10 at proper latency rest at least 10ms higher then they should be although massive spikes are not a lot. Have rebooted both pc and modem.
---------- Post added at 17:40 ---------- Previous post was at 17:40 ----------
forgot to add I noticed leics down for upgrades this week I wonder if this is to do with the problem maybe it went wrong or something.
I am also from Leicester and my speeds are all over the shop.
Sometimes the speed is down to a crawl. Not good? :td:
same problem here, in Oxford - on 10MB, getting constant 3 meg.
Checked modem config, im on the correct tier still.
Tried releasing IP and renewing on WAN interface on router, no difference.
Tried connecting modem directly to PC, no difference.
If i trace route back from easynews. i get 10% packet loss from oxfd-t2core-b-pos31.inet.ntl.com and oxfd-cache-4.server.ntli.net but i think that generally this is normal.
Downloads are constant slow from everywhere, not just easynews - microsoft, gamespot, file planet.
Not happy at the moment. Only hope its sommet to do with these planned works.
LemonyBrainAid
10-01-2006, 20:47
same problem here, in Oxford - on 10MB, getting constant 3 meg.
Checked modem config, im on the correct tier still.
Tried releasing IP and renewing on WAN interface on router, no difference.
Tried connecting modem directly to PC, no difference.
If i trace route back from easynews. i get 10% packet loss from oxfd-t2core-b-pos31.inet.ntl.com and oxfd-cache-4.server.ntli.net but i think that generally this is normal.
Downloads are constant slow from everywhere, not just easynews - microsoft, gamespot, file planet.
Not happy at the moment. Only hope its sommet to do with these planned works.
Over here in surrey it seems I'm only getting 100kbps on easynews which usually gives me max speed...
Really annoying -
Tue, 10 Jan 2006 20:47:28 GMT
1st 512K took 2937 ms = 174.3 KB/sec, approx 1394 Kbps, 1.36 Mbps
2nd 512K took 3062 ms = 167.2 KB/sec, approx 1338 Kbps, 1.31 Mbps
3rd 512K took 3531 ms = 145 KB/sec, approx 1160 Kbps, 1.13 Mbps
4th 512K took 2734 ms = 187.3 KB/sec, approx 1498 Kbps, 1.46 Mbps
Overall Average Speed = approx 1348 Kbps, 1.32 Mbps
Over here in surrey it seems I'm only getting 100kbps on easynews which usually gives me max speed...
should note the only way i was able to get 350k off easynews was to try several different routes (from the route selector on site) and segment the download 4x with getright.
Each segment was roughly 80-100kpsec in speed.
Im another user in the Leicester area that is having speed problem issues im hoping all the planned work this week will help us all get a better service.
Anybody know what work is going on ? Is it more new UBR's or just rebalancing and segmentation work going on . The amount of work we have had in the Leicester area in the last 3months indicates to me that technically it was very poor before they started it now it seems to be getting worse rather than better.
Chrysalis
11-01-2006, 16:43
the status page says "upgrades" this time instead of "optimisation" which I would guess means more bandwidth, but I think something obviously didnt goto plan because I have a defenite downgrade in performance, my latency is way down from what it was a few days ago and my traceroutes look even more messy.
Pinging www.ntlworld.com [212.250.162.47] with 32 bytes of
Reply from 212.250.162.47: bytes=32 time=19ms TTL=119
Reply from 212.250.162.47: bytes=32 time=31ms TTL=119
Reply from 212.250.162.47: bytes=32 time=25ms TTL=119
Reply from 212.250.162.47: bytes=32 time=15ms TTL=119
Reply from 212.250.162.47: bytes=32 time=35ms TTL=119
Reply from 212.250.162.47: bytes=32 time=19ms TTL=119
Reply from 212.250.162.47: bytes=32 time=22ms TTL=119
Reply from 212.250.162.47: bytes=32 time=44ms TTL=119
Reply from 212.250.162.47: bytes=32 time=18ms TTL=119
Reply from 212.250.162.47: bytes=32 time=68ms TTL=119
Ping statistics for 212.250.162.47:
Packets: Sent = 10, Received = 10, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 15ms, Maximum = 68ms, Average = 29ms
as you can see 15-20ms is what I should be expecting, and 6 out of 10 packets is delayed due to some sort of congestion between me and ntlworld.com.
Yea i also noticed the "upgrades" listed instead of optimisation. Upgrades could just mean physical re-segmentation (distributing users more evenly across UBRs). UBR's cost a lot so i dont think installation of more UBR's is likely but it could mean better hardware of some sort. NTL have plently of bandwidth.
One things for sure, NTL have changed all their speed tiers and introduced 10Mbit, they will definatly be monitoring areas more closely and providing an action plan for the problematic areas.
Not sure what the upgrades are though this week.
Fingers crossed.
Chrysalis
11-01-2006, 22:18
things have gone from bad to worse I was doing a traceroute to diagnose a website that is down and look at this. Not just spikes but some very big spikes, I am going to see if bbking got time to have another look at my ubr stats.
1 22 ms 56 ms 8 ms UBR [10.8.143.254]
2 130 ms 37 ms 21 ms leic-t2cam1-b-ge914.inet.ntl.com [82.3.35.213]
3 32 ms 19 ms 18 ms leic-t2core-b-ge-220-0.inet.ntl.com [82.3.33.138
]
4 48 ms 64 ms 78 ms nth-bb-b-so-230-0.inet.ntl.com [62.253.188.114]
5 70 ms 26 ms 102 ms nth-bb-a-ae0-0.inet.ntl.com [62.253.185.117]
6 43 ms 31 ms 69 ms so-0-3-2-0.loncr4.london.opentransit.net [193.251.254.141]
7 20 ms 44 ms 38 ms po2-0.loncr3.London.opentransit.net [193.251.242.21]
8 * * * Request timed out.
9 46 ms 42 ms 26 ms so-5-3-0.cr1.lhr3.uk.above.net [208.184.231.178]
10 162 ms 126 ms 125 ms so-7-0-0.cr1.dca2.us.above.net [64.125.31.186]
11 165 ms 199 ms 166 ms so-6-1-0.mpr2.atl6.us.above.net [64.125.27.66]
12 109 ms 134 ms 121 ms so-0-0-0.mpr1.atl6.us.above.net [64.125.27.49]
13 133 ms 125 ms 195 ms so-4-0-0.mpr2.iah1.us.above.net [64.125.29.70]
14 150 ms 146 ms 131 ms so-0-0-0.mpr1.iah1.us.above.net [64.125.31.61]
15 169 ms 167 ms 215 ms so-5-1-0.mpr2.lax9.us.above.net [64.125.29.97]
16 169 ms 180 ms 197 ms 216.200.249.141.available.ipowerweb.com [216.200.249.141]
and to ntlworld.com
Tracing route to www.ntlworld.com [212.250.162.47]
over a maximum of 30 hops:
1 12 ms 47 ms 72 ms UBR [10.8.143.254]
2 47 ms 46 ms 102 ms leic-t2cam1-a-ge914.inet.ntl.com [82.3.35.149]
3 13 ms 67 ms 34 ms leic-t2core-a-ge-220-0.inet.ntl.com [82.3.33.10]
4 92 ms 105 ms 37 ms lee-bb-a-so-220-0.inet.ntl.com [62.253.188.110]
5 33 ms 38 ms 16 ms lee-bb-b-ae0-0.inet.ntl.com [62.253.187.186]
6 58 ms 28 ms 23 ms man-bb-a-so-700-0.inet.ntl.com [62.253.185.193]
7 34 ms 29 ms 61 ms man-bb-b-ae0-0.inet.ntl.com [62.253.187.178]
8 70 ms 21 ms 16 ms win-bb-a-so-300-0.inet.ntl.com [62.253.185.138]
9 46 ms 70 ms 122 ms win-dc-a-v902.inet.ntl.com [62.253.187.222]
10 44 ms 22 ms 17 ms www.ntlworld.com [212.250.162.47]
---------- Post added at 22:18 ---------- Previous post was at 21:53 ----------
hmm I read ignition's sticky thread after seeing his post on manchester thread and am now wondering if ntl did something to affect my modem stats.
These quotes are from ignition in his thread, and I will put my stats below them.
Downstream Receive Power: Ideally +/- 3dBmV, cause for concern at around +/- 8dBmV, different modems handle the higher and lower levels differently however I'd expect to see service degredation at that point.
Downstream Receive Power Level : -11.3 dBmV
Downstream SNR: Minimums are 24.5dB on the Langley platform and 26dB on the Bromley platform. Again these aren't the minimums required to function but when you're down this low you're teetering on the brink of losing your service.
Downstream SNR : 32.2 dB
Upstream Transmit Power: Anything between 28dB and 53dB is fine. Up to 55dB is also fine so long as it's fairly stable. Above 55dB suggests issues. Depending on your modem connectivity is jeopardised when transmit power hits 58dBmV or in the case of Pace set tops 61dBmV.
Upstream transmit Power Level : 50.0 dBmV
So if I am right my downstream power is way out of whack, my upstream power is in spec but close to borderline, my snr is fine, so I should ring tech support for an engineer concerning my downstream power level?
I was hoping as we are now over half way through the planned work for Leicester that we may start to see some improvement but none as yet, I've had the modem reboot couple times last few nights in fact was off for over 3 hours last night.:erm:
Chrys:
1) You can't read all your modem parameters, I can (not your fault, the other two are read at the UBR - upstream SNR and upstream receive power). Downstream receive at -11 isn't going to cause problems in itself, particularly if the SNR is fine. Ideally it would be higher, but only so it's more tolerant to fluctuations - losing -4 when you're at -3 is ok, when you're at -11 it puts you out of spec.
2) Your problem, even without checking (which I have now) is upstream congestion. The wildly fluctuating RTTs in the traceroute are the giveaway.
Chrysalis
12-01-2006, 22:42
ok thanks, I dont need to call ntl for an engineer then at least, but dissapointed about the congenstion it is so frustrating an isp cant get this right, I am just hoping at the end of the week they have this fixed.
BB king do you know if all the work in Leicester this week is going to ease or solve the problems we have been having recently ? I spoke to an engineer i saw in the street today and he was saying how bad the network has been coping since the upgrade especially those served by Northfield just dont seem to be able to get the balance right and may need some new fibre laying or replaced to solve it whatever that means ?
Ignition
13-01-2006, 07:41
*cough* students *cough*
---------- Post added at 07:35 ---------- Previous post was at 07:32 ----------
Yea i also noticed the "upgrades" listed instead of optimisation. Upgrades could just mean physical re-segmentation (distributing users more evenly across UBRs). UBR's cost a lot so i dont think installation of more UBR's is likely but it could mean better hardware of some sort. NTL have plently of bandwidth.
Actually the kit needed to support a reseg where a new uBR is installed tends to be more expensive than the actual uBR. The new optics, etc, needed, and in the case of Leicester the new fibre were far more expensive than the additional routers.
Resegmentation nearly always involves new uBR kit being used.
---------- Post added at 07:41 ---------- Previous post was at 07:35 ----------
Considering the amount of work that's already gone into breaking bits of Leicester down and increasing the amount of available bandwidth per home here's hoping it doesn't become another Coventry of a while back with upstream bandwidth just being anhialated and there being absolutely no reason for ntl to spend the loads of money required to break the network down even further when every time they give more bandwidth it just gets eaten.
I don't know the exact figures but know that in some areas worldwide cable operators do just decide they've had enough. The argument being that x amount of cash has been spent giving x bandwidth per xx homes passed which is more than enough for most but here as soon as extra bandwidth is made available sometimes at quite large cost it gets eaten without any increase in revenue.
Ig thats understandable but customers all over the country are getting a bad service. From recent threads we are seeing poor service in
- Manchester
- Leics
- Derby
- Stafford
- Luton
- Dorset
Thats a rather large disection of the UK and that excludes areas where people didn't post their location.
I'm not ranting and raving about this because it gets you no-where. But what springs to mind is maybe NTL rushed into pushing higher speeds simply to be more competitive in the UK broadband market. As a result some areas just can't cope with the speeds and congestion is a real threat.
I would say this, when the students come back the service is consistently poor. Its blatently obvious that they are on BT / P2P all day as the fluctuations in ping, packet loss, connection stall and inability to sustain a smooth throughput is just the result of upload congestion; esp now the upload speeds have increased on 10Mbit specifically. I still think NTL have plenty of bandwidth to play with but the limitations of upload speed on their hardware (UBR's) and standards is the achilles heel of the service.
10Mbit through a Samsung and WRT54G router:
Normally get 4-7MBps, today getting between 2-3 tops.
As they say in Yorkshire "summat's up".... :dozey:
Chrysalis
13-01-2006, 23:06
In response to ignition it isnt the customer's fault tho is it.
If NTL were to decide they will leave the situation as it is because upgrades just get swallowed up then they are selling what they cant provide and need to change their product line (higher prices, lower speeds etc.).
I decided today to downgrade, I dont see the point in paying for a 10meg product with a 75gig cap that fails to burst higher then 2meg, when I can pay for 2meg unlimited for £13 less. It was okish up until monday and then took a sharp turn downwards. I am still seeing some 100+ms spikes on traceroutes and pings.
Ignition
13-01-2006, 23:14
These problems were expected and inevitable I'm afraid in such a huge shift from 3Mbit to 10Mbit and a near doubling of upstreams.
OKish until Monday? Hrm anyone wanna place a bet that a college started back or the students started coming back around that time? ;)
Sounds like a simplification but it's true.
As far as the customer goes no it isn't their fault, they should have their hands slapped if their taking the michael.
Chrysalis
13-01-2006, 23:32
I am thinking along the same line. University may have started after xmas break and I am assuming if this problem is still here monday it will last until the next uni break.
sharpygreen
14-01-2006, 09:54
getting the same problem in northampton annoying thing is when you ring support they tell you there is nothing wrong there end and must be your pc
total crap i had engineers out yesterday and they couldnt find anything wrong and said it must be a bandwith problem so ntl's network cant support 10 meg ?
I have had a friend do some reasearch for me regarding the students in Leicester and it makes interesting reading he confirmed they went back Mon but more related to the long term problems is that in the last 12months Leicester Uni have added nearly 1200 new rooms and De Montfort uni have added over 2000 new rooms all of which im told all have Ntl connections and remember thats just new accomondation plus all the ones in private accomondation who are probaly the worst offenders and 14% of population of the city is a student.
It appears this could be where the draining of the bandwith is coming from I have a friend in halls who tells me everybody leaves computer on 24/7 downloading/uploading he knows of people running mini servers just connected p2p plus loads of people using voip services as cheaper than Ntl option and some providers allow unlimited calls for yearly fee. Also a trend developing of people watching tv over the net especially the foreign students some of the stuff is at over 1500k i mean when i watch some motor cycling programme it used 1.6 gig in a hour so it soon mounts up.
I think some of these are in for a shock when they start enforcing the download cap.
sharpygreen
15-01-2006, 07:45
I really wish they would enforce it now I have had hardly any service since Jan 1 works for a couple of hours mid afternoon then the rest of the time im stuck with less than 56k performance.
on the verge of telling them to stick it paying 37.99 for this is a joke .
even been told if I drop to 2 meg no guarantee it will improve
come on ntl sort ya network out
yup the students here in Leic even get special 9 month contracts!
Its the same issue in stafford. When the students are in term time the service is un-usable, with constant packet loss, spiking and un-even throughput. Now that 10Meg is onboard add slow download speed to that list too. Try 11Kb SEC from Microsoft.com
Chrysalis
15-01-2006, 18:31
umm the cap is removed and only in place on the top tier. Interesting reading that although I hear from ntl staff we have had tons of upgrading the extra 3200 rooms sort of wipes out that upgrading. A landlord I know own's over 200 properties in my area all letted out to students and all fitted with ntl, he doesnt allow sattelite/bt to be fitted so I would guess most of his tenants are actively using ntl.
Chrysalis
16-01-2006, 14:40
new development, my sister's pc is vastly improved over mine her upstream is fairly good. I am hoping I can get my upstream channel changed. Mine is still very dire.
Just been reading a document off the press side of Ntl on how the network goes forward and wondered if somebody could clarify whether in Leicester Oswin road,Northfield and Shepshed are headends or hubs as that will allow me to understand the diagrams better as has info on how future could go and how is at present, its a pdf file maybe worth some people having a look at to understand the network better is quite technical in places especially the telephone side.
Ignition
17-01-2006, 08:16
Oswin Road is a headend, Northfields and Shepshed are both hub sites.
bugheadcjd
17-01-2006, 10:26
Just been reading a document off the press side of Ntl on how the network goes forward and wondered if somebody could clarify whether in Leicester Oswin road,Northfield and Shepshed are headends or hubs as that will allow me to understand the diagrams better as has info on how future could go and how is at present, its a pdf file maybe worth some people having a look at to understand the network better is quite technical in places especially the telephone side.
Hi uno,
Is there any chance you could post a link to the pdf as I am interested in reading it.
Cheers,
bughead
This is the link that i used to see the information i found it really interesting reading http://www.ntl.com/locales/gb/en/investors/presentations/network-overview-final-vr.pdf It was published in 2004 but gives info on how network works and may do in future.
Chrysalis
24-01-2006, 01:01
I managed to get my upstream chan changed by using the reset to defaults option on the modem, thanks to the people in particular who messaged me with help.
vBulletin® v3.8.11, Copyright ©2000-2024, vBulletin Solutions Inc.