Extremely Slow

[Steve]

Certainly my evenings Think Broadband,Namesco broadband test etc show no relationship at all to my BT speedtest from earlier. Its difficult to complain about poor speeds when the evidence required is a BT speedtest ,which certainly this evening and others is contradictory to other tests and downloads.Are we certain that a BT speedtest bypasses idnet as part of the chain.

[Pistol]

Are we certain that a BT speedtest bypasses idnet as part of the chain.

AFAIK, yes, only if ye log in with their username in your router, not your own.

http://217.35.209.142:50302/cgi-bin/home.page.pl

instructions here http://217.35.209.142:50302/instructions.html

[Dopamine]

Are we certain that a BT speedtest bypasses idnet as part of the chain.

I don't know, but was told by Pipex's support that that was the case.

All I want is a connection that's fairly reliable...... I don't want to have to become an internet expert, which seems to be becoming necessary since I've been an IDNet customer. Worse, I (with hindsight, foolishly) recommended IDNet to two non-tech friends who, now they have a nice, expensive and slow line, think I'm an idiot!

[vitriol]

Test1 comprises of Best Effort Test:  -provides background information.
    Your DSL connection rate: 8128 kbps(DOWN-STREAM),  448 kbps(UP-STREAM)
    IP profile for your line is - 7150 kbps
    Actual IP throughput achieved during the test was - 3834 kbps

Not what I'd expect at 23:14

[Pistol]

I have to say actually there does seem to be a problem, yes. I just ran the test: and it has dropped.

Test1 comprises of Best Effort Test:  -provides background information.
    Your DSL connection rate: 8128 kbps(DOWN-STREAM),  832 kbps(UP-STREAM)
    IP profile for your line is - 6500 kbps
    Actual IP throughput achieved during the test was - 4467 kbps



Thinkbroadbands results
http://www.thinkbroadband.com/speedtest/results/id/122039403812267930546.html
Speed Test Results
Date    02/09/08 23:21:58
Speed Down    2126.23 Kbps ( 2.1 Mbps )
Speed Up    572.17 Kbps ( 0.6 Mbps )
Port    8095
Server    speedtest2.adslguide.org.uk


edit- btw, i just tried running the bt test logged in as bt... and it wont work :s  i can connect ok but i can not load the bt test page at all.. my router doesnt let me not imput a password (it says ye dont need one for bt, but i read elsewhere to put "testing", though i doubt it matters) but i cant think why else i cant get it to run?

edit: its getting worse again :s

http://www.thinkbroadband.com/speedtest/results/id/122039498881019322242.html
Speed Test Results
Date    02/09/08 23:37:12
Speed Down    1357.39 Kbps ( 1.3 Mbps )
Speed Up    573.76 Kbps ( 0.6 Mbps )
Port    80
Server    speedtest1.adslguide.org.uk

[jaydub]

My results have got better as the vening has gone on, but are significantly lower than I've seen since Friday night.

My results also show a pattern of TBB (and Zen's test files) giving lower results than the BT Speedtester.

My understanding is that since BT changed to the new speedtester, it doesn't bypass your ISP.  However, it is possible it is routed in a different manner, which could explain the difference in results. (Just musing - horribly out of my depth here. )

Anybody any ideas?

[Inactive]

TB Results;

http://www.thinkbroadband.com/speedtest/results/id/122039612398419822383.html

Not sure if that helps anyone.

[jaydub]

You are doing far better than me:

http://www.thinkbroadband.com/speedtest/results/id/122039665897728422441.html

Off to bed.  Hoping for a better tomorrow.

[Inactive]

Ouch, 

[Pistol]

I managed to "hotwire" routerstats to work with my d-link router, (a nice app and his instructions to get it working on routers that it doesnt by default are good) so ill run that for a couple of days and see whats happening and let ye know.

ps just noticed i have lost 1000Kbps on my synch this evening.

[Ann]

I think thinkbroadband's tester is inaccurate.  It doesn't agree with any of the other testers out there.

This one is way better  http://myspeed.visualware.com/servers/lhr.html

[David]

As a relative novice,to me all test sites are correct in the speeds surely,although they may differ in actual reported speeds they are all pretty much consistant.

For me the confusion arises when I start to test using different sites and assuming they  will all show the same speedsin theory they should but wont.

I do the usual stuff and test in the morning,maybe in the early afternoon and then in the evening,I know most of you know all this and may tell me were I am going wrong,the other thing for me at least I found that speed tests are addictive and I found myself testing for no real reason,so although most of the people here know thier onions there are others who may be at my level and should be aware of falling into this trap.

I do prefer BT test to all to give me a relable test but as I have said they are no more consistant at times than TB.in fact the other night TB was showing speeds much slower than all,including BT..

I dont speed test unless I notice a problem...............thhis way I keep my sanity and can focus on more productive things.......................wish I knew what I was talking about but it gets me through and after a year with Tisc I think I learned a littel about speed tests..in my case they turned into speed traps

[Pistol]

pianoplayer. Yep, when i test i always run www.speedtest.net as well as others, simply because it gives me a nice history of all my tests.

Potentially, differences between test sites could be due to the number of people testing at one time? ie. perhaps you may not get full bandwidth from the site itself..?  Potentially also the routing between ye and that site could have issues while not quite the same to others? Generally speaking though, especially at quieter times of the day/night, most tests should return similar results if all is well.

Thanks fer that one Ann, that is quite a nice test for the results it shows. I think someone mentioned above they prefer TB due to the size of the dl.. there is something to that aye, less burst speed relevant. However perhaps it also depends on ones normal downloading habits. eg for me i tend to not dl huge files, so in some ways a couple of meg file dl test is accurate for me?

Things picking up again now.. again i wonder if midnight is a relevant changing point as before?

[Dopamine]

I think thinkbroadband's tester is inaccurate.  It doesn't agree with any of the other testers out there.

This one is way better  http://myspeed.visualware.com/servers/lhr.html

That tester downloads a very small amount of data, around only 20% of that used by Thinkbroadband and BT's testers. Consequently, I'd be very reluctant to rely on its results.

The best way to monitor speeds is in normal use over a longish period of time. Thinkbroadband's test files allow you to do that, and if you have access to a server somewhere, it's worth putting a large file - 100mb or so - on it so that you can download it and replicate normal web usage.

Many of the small-download speedtesters are unlikely to be fantastically accurate, for various reasons - for example, can you possibly imagine ISPs that use traffic shaping not programming their shaping software to allow unthrottled downloads from speed test sites?

[Simon]

I have to admit, I did notice a brief slowdown myself earlier this evening, although I wasn't affected by the recent capacity issues.  All seems OK now though.

[jaydub]

I've woken up to a disconnected session this morning.  A swift reboot of the router quickly resolved it.  Speeds back to 6000kbps, which is no great surprise.

In terms of validating TBB speedtest results, I tend to use Zen's test download files at:

http://download.zensupport.co.uk/

It has the advantage of TBB's test files in that if the TBB speedtest is overloaded, then the TBB test files may also be.

Will keep a track on speeds and log a call if I have issues again tonight.

[MO]

An update with my speed problems.

I have had contact with support who believed, as did members on here, that my speed issue was a BT one. I tried plugging my router directly into the test socket but no improvement was noticed. BT have looked at my line and made inprovements, so now I am receiving better speeds although not as good as I have been used to.

On another note, I did buy a new wireless router as my Belkin was 3-4 years old and maybe sported older technology, this has also made an improvement. I now have stats and as I guessed, as being sooooo far from the exchange, have a lot of noise. What do you think?

Thanks again for everyone's input.



[attachment deleted by admin]

[David]

Mine were down a little last night,did a Bt test and saved results but I really think it was a capacity issue,everything is fast as it is usually today.

Sorry if I went on a bit last night.........

[Rik]

Hi MO

Those stats look OK for a line of that length (actual attenuation may be higher than the report, routers can only got yo 63/63.5db for some reason). Your noise margin may reduce if you can maintain a stable connection for a couple of weeks or more, so it's possible you could gain another 5-700k in time.

[Simon_idnet]

We are still working at re-balancing our Centrals. The BT systems are supposed to do this automatically but they don't seem to be very good at it.

The root of the problem is that we are nearing the limits of our capacity. The iPlayer traffic during the Olympics brought this to a head. We would normally have had a new BT Central installed by now. Back in April we placed an order with BT for a 1Gbps 21CN (21st Century Network) WBC Host Link (the next generation of BT Central). The official lead time is 65 working days. We have now received an install date of 4th November = 180 working days! This time last year BT said that IP Stream traffic (i.e. all traffic on the current ("old-style") BT Centrals) would be migrated to the Host Link "at launch". When the product was "launched" (I put that in quote marks due to the install delay shambles mentioned above) in April they then said that the migration could take place by Christmas. The latest timing that we have been given is "March". We will though be able to upgrade customers to ADSL2+ (those who are connected to upgraded Exchanges) as soon as our Host Link is installed.

In the meantime we will be working hard to get the balancing right.
Simon

[Rik]

It's a tough juggling act for you, Simon. You couldn't migrate me, even in November, as my exchange activation has been put back six months, and others around here have moved to a 'no date' situation.

It's a shame that BT can get away with messing ISPs around like this, yet seem to have to answer to no-one.

[Pistol]

Yes, cheers for keeping us in the loop Simon, greatly appreciated and does much to keep our confidence in IDnet as the best ISP.

[Inactive]

Thanks for the information Simon, it is appreciated, I don't mind any ISP having a blip in the service as long as I am kept informed and something is being done to fix it.

[Steve]

I find it rather worrying for myself: I see no potential capacity increases until March next year unless a mass migration to ADSL2+ occurs in November. If rebalancing fails to cure my current poor evening download speeds and browsing experience I have only one option left but to look round for an ISP that has capacity (if they exist) and I find that rather sad. However if you are paying a premium price one has the right to expect a premium service.I praise their honesty, the quality of their support,their desire to solve the problem within their current capacity constraints but ultimately I have a line which has the ability to support a fast connection and initially with idnet it did and I want that performance back.

[Rik]

Though it's not limited to people moving when their exchange is activated, Steve. BT are due to release IPStream Connect, which will allow IPStream customers to be transferred to the new HostLink, regardless of the state of their exchange. It's just a matter of when BT fulfil their already broken promises on that.