Throughput vs. Sync Rate for Different ISPs

TalkTalk staff tell their fibre customers that: throughput speeds are usually 80% of the sync speed

Is that a standard fibre stat or is it dependent on provider?

Does it mean for TT customers with a perfect line delivering max synch on the 80/20 product that throughput won't be above approximately 64meg (80% of 80)?

More rubbish ... they say it to fob people off when their speeds are slow and there is congestion. Train them at day one and it will perpetuate!

The max throughput is over 95% of Sync (cannot remember the exact figure - 96.8% I think) and with no congestion you should see that figure or very close. Even with a small amount of errors and resends you will still be up at 95%.

For a "perfect" line on 80/20 you should be seeing 76Mbps - the figure BT now uses.

Ummm?

Your 96.8% sounds like the BT Wholesale IP Profile of 96.79% on non-G.INP, and your 76Mbps the Ofcom imposed figure for the sync 10% of customers can achieve. Neither of which refer to throughput.

Throughput may be being quoted from speed tests, many run wirelessly connected to the router.

We often see speed tests on these forums above 74mbps, but 76Mbps?

Edited by RobertoS (Fri 03-Jun-16 09:23:58)

They're telling you porkies. On TT I sync at the full 80/20 and speedtests show 75/18 meg day and night...with no sign of any congestion whatsoever

http://www.speedtest.net/my-result/5355634956

http://www.thinkbroadband.com/ping/share/07b4d36b37d...

Edited by deleted (Fri 03-Jun-16 09:30:56)

with full sync

BT - 74.82Mbps
Talktalk - 75.21Mbps

Its never 80%

closer to 95%

Forgive me if I've misunderstood this, but throughput will be partly dependent on the equipment in the home? So when TT say throughput is usually 80% (which they say repeatedly to many customers on their forum), could this be TT tacitly acknowledging that their router is poor and is therefore increasing the spread between sync and throughput? Such that 80% really is the average for most of their customers, thus making the statement truthful? Otherwise it seems strange they would constantly repeat a downright lie.

Baby F, Do you use the TalkTalk provided modem/router or have you bought your equipment and are you on a normal TT domestic product or business fibre? (Also your 75meg output is great! What's your max line rate stat taken from the router?)

Edited by deleted (Fri 03-Jun-16 10:53:36)

It's mostly just to stop people ringing them up complaining about throughput.

To be honest there are so many things that can affect throughput, some down to the user and some to the ISP (and many that are wider Internet related) that giving any figure is a nonsense. Any figure chosen is made up and not relevant to most people.

Talktalk routers that support VDSL2 can support full sync rate throughput.

If you where lucky to get a HG635 "super router" they're in my opinion one of the best packaged routers ever from an isp.

I have a HG635 by talktalk and it's the best router I have ever used.
The HG633 is also able to pull the full speeds as well.
So in essence, the router is able to support the speeds & is not rubbish.

In reply to a post by RobertoS:

Ummm?

Your 96.8% sounds like the BT Wholesale IP Profile of 96.79% on non-G.INP, and your 76Mbps the Ofcom imposed figure for the sync 10% of customers can achieve. Neither of which refer to throughput.

The 10% is nothing to do with Ofcom but CAP/ASA, and relates to throughput not sync.

https://www.cap.org.uk/~/media/Files/CAP/Help%20note...

Edited by deleted (Fri 03-Jun-16 11:01:35)

In reply to a post by Alexir:

Baby F, Do you use the TalkTalk provided modem/router or have you bought your equipment and are you on a normal TT domestic product or business fibre? (Also your 75meg output is great! What's your max line rate stat taken from the router?)

I'm with TalkTalk Business (direct), but my neighbour is with TalkTalk residential after getting a poor connection with Zen and our VDSL connection stats are 99.99% the same (both on 80/20). Of course i get better support (uk based) and also up to 4 static ip addresses and higher priority at the exchange at busy times...allegedly. Other than for testing purposes, I never ever use ISP supplied routers as there are far better routers out there, though i've heard many good things about TT's latest HG633/5 router. I'm using an Openreach HG612 VDSL2 modem connected to a Linksys EA8500 Mu-Mimo 802.11ac 4 stream router and the wifi speeds are almost as good as wired speeds in our house.

Screenshot of my VDSL2 stats:
http://s33.postimg.org/ov52mwc7z/Capture_jpg.jpg

Edited by deleted (Fri 03-Jun-16 11:12:47)

In reply to a post by Ignitionnet:

The 10% is nothing to do with Ofcom but CAP/ASA,

I think this MOU sets out how it is plenty to do with ofcom AND asa http://stakeholders.ofcom.org.uk/binaries/consultati...

and relates to throughput not sync.

It clearly relates to Sync Speed and not throughput as you assert. It emphasises the effect of attenuation which impacts Sync Speed not throughput.

Edited by deleted (Fri 03-Jun-16 11:16:55)

In reply to a post by BatBoy:

In reply to a post by Ignitionnet:

The 10% is nothing to do with Ofcom but CAP/ASA,

I think this MOU sets out how it is plenty to do with ofcom AND asa http://stakeholders.ofcom.org.uk/binaries/consultati...

and relates to throughput not sync.

It clearly relates to Sync Speed and not throughput as you assert. It emphasises the effect of attenuation which impacts Sync Speed not throughput.

Find me where Ofcom mention broadband speeds or 10% in that MOU if you would, given it's specifically Ofcom imposed. My browser can't seem to find any match for 'broadband' 'speed' or '10%'.

I'm aware the ASA and CAP are connected to Ofcom but as far as I'm aware this restriction on broadband advertising is entirely within the ASA's remit and per my comment is not Ofcom imposed.

As far as the CAP note goes perhaps consider reading the document again. It specifically mentions speed claims should be measured in ways that represent user activity and says nothing about sync speed being an acceptable measurement.

Measurement

There is no established standard for testing broadband speeds. Marketers should therefore use
methods based on relevant industry standards.

Measurements used should be appropriate to the claims made, for instance:

� Specific claims, such as �50Mb Download speed�, should be based on tests of protocols
relevant to downloading large files.
� Speed claims for a service in general should be based on speed testing that is
representative of the activities that users generally perform.

Attenuation is emphasised because the majority of broadband subscribers are on services impacted by attenuation, however a bunch of other factors that have no impact on sync speed are also discussed, the relevant sections to fixed lines being as follows, I'm pretty sure the ones I've put in bold do not impact on sync speed:

Relevant factors
Marketers should account for all the relevant factors that cause a reduction or variations in speeds. The principal factors are:
� Signal attenuation
� Congestion/contention
� Traffic/network management
� Protocol overheads

All ISP's that resell BTW or use BT's back haul throughput is restricted by the use of BT's IP profile Both SKY and TT use their own backhaul so don't have BT's IP profile,
So theoretically the max throughput should be slightly higher ,

One exception is Plusnet who also use their own additional IP profile as well as BT's which combined with their prioritization of traffic (traffic shaping ) reduces throughput a little

Syncing at the max 80/20 rates:
Plusnet 73-74mbps
Zen 74.50 75.40mbps

Edited by tommy45 (Fri 03-Jun-16 12:52:26)

As you can see in that document should you care to read it

The legal foundation for the self-regulatory system envisioned by this MoU is the DCOA. Where a Minister or other public authority (such as Ofcom) has been given specific functions or duties by legislation, DCOA allows the Minister or public authority to delegate or �contract out� those functions to another person or organisation.

Maybe you should change your browser?

As ignition rightly says, advertised speeds represent throughput. The big ISPs easily meet 10% of users at 80 sync, and 76 is what they reckon (rightly or wrongly) to be the calculated corresponding throughput.

The 10% 'rule' was decided by CAP and BCAP. It wasn't Ofcom's choice, who were consulted and didn't think much of it, instead preferring a "typical speed range" to be advertised.

You obviously don't understand DCOA

In reply to a post by Ignitionnet:

In reply to a post by RobertoS:

Ummm?

Your 96.8% sounds like the BT Wholesale IP Profile of 96.79% on non-G.INP, and your 76Mbps the Ofcom imposed figure for the sync 10% of customers can achieve. Neither of which refer to throughput.

The 10% is nothing to do with Ofcom but CAP/ASA, and relates to throughput not sync.

https://www.cap.org.uk/~/media/Files/CAP/Help%20note...

You are right it is an ASA requirement, not Ofcom. What amuses me is that I have often corrected people on this in just the way you have corrected me . Just brain fade.

As for the 10% being throughput not sync, frankly I don't believe that. BT may approach it by the simple means of not allowing anyone onto the product with estimates below 15Mbps, particularly if the 14.9Mbps is the lower end of the B range. It certainly isn't the top figure of the A range.

I'd be surprised if many other ISPs restrict sales that way, though unless they've changed Plusnet don't let people with 40'ish estimates have 80/20, and of course Sky and TT plug the 40Mbps services more strongly than the 80Mbps ones.

Are there any 80/20 suppliers that advertise it below 76/19? To get 10% of customers on 80/20 via BT Wholesale getting 76Mbps throughput when the highest IP Profile is 77.432Mbps would be remarkable.

Sky and TT don't have that IP Profile but AIUI they do restrict data feed speed in some way as per SIN 498 requirements.

The 2011/12 article you linked to is a real gobblegook piece of work, considering "This Help Note is intended to help the industry interpret the Misleading Advertising sections of the UK Code of Non-broadcast Advertising, Sales Promotion and Direct Marketing (the CAP Code) and the UK Code of Broadcast Advertising (the BCAP Code)". I'm sure a year or so ago I read what the ASA said, specifically looking for this point, and concluded it could only mean sync speed.

As I just replied to IgnitionNet, yes I know it is ASA. I have frequently said so myself . But if it applies to throughput, not sync, then an awful lot of ISPs must be lying and not been told off.

The guidance is clear that advertised speeds should refer to throughput not sync. The ASA maintains standards reactively by complaint on a case-by-case basis, TV and radio ads are vetted to some degree by clearance companies.

The main ISPs all genuinely have 10% of users on 80 sync (for the appropriate product). I think it was BT who came up with a corresponding throughput calculation for advertising of 76 (or >76) Mb/s and everyone else copied. They reckon this calculation is sufficient for substantiation, but this can only be known if it is tested by a complaint.

The BTw IP profile is not restrictive, it reflects the max throughput on the Openreach product (PTM, VLAN, ethernet overhead).

In reality throughput is 75 for Windows users, a bit less for Mac, and I believe this matches the true calculated max.

Your detailed rebuttal, most of which you had already said , and an overlapping "most" you are very well aware I certainly know , ends by confirming what I say. "In reality ...." There is no way that 10% get 76Mbps throughput. Particularly if using wireless connection to the router.

How about you post a link or three to user speed test results showing 76Mbps on FTTC.

As I said, the CAP document is a loudly written document considering it is supposed to be specifically intended to clarify things. It's a complete mess. As such I place no faith in what it says. IIRC it doesn't even mention the effect of using wireless and that throughput needs to be measured wired.

As for BT having done these "sophisticated calculations", (my eye!), the adverts are supposed to relate to each individual ISP's performance. It is not permissible for them to do as you say, and just use the BT figure. Is the BT Consumer anyway? Or is it a BT Openreach theoretical one?

Lies, damn lies and statistics.

If and when I am in the mood I shall go to the ASA documents.

I may be wrong. Nothing you have said proves I am. But I am not an idiot and find your tone extremely offensive.

We see people nailing the 76 Mbps so have never seen the point of making a big noise over the percentage
http://www.thinkbroadband.com/speedtest/results.html...

Whether its 10% is hard to say for sure, but if one was to sift through the results and reject those which are Wi-Fi/phone/tablet based and take just PC that appear Ethernet connected and on the up to 76 Mbps service then believe even our observations are going to put close to 10% in the 75 to 77 Mbps region. To do this exercise properly needs a good many hours sifting the many results manually and thus needs a good reason (or someone mad enough to pay for time) to do it.

We are seeing speed tests hitting the figure (or to within 1 Mbps) and this is not FTTP ones either.

With the current debate about changing the rules, the up to 76 Mbps is an irrelevancy and fingers crossed they might be sensible enough to talk to a cross section of industry people to try and understand some of the science involved.

There is the problem - the providers will insist it does not include wireless connections...and I believe you can probably show 10% within 1 Mbps of 76 Mbps. If ASA accepted SamKnows testing in the way Ofcom do then I have absolutely no doubt it is possible.

Perhaps focus here is better spent on not the current rules, but debating what they should change to, or whether they should stay as they are.

It is 76. Within 1Mbps of that is a fail. 10% need to reach or exceed 76. 75.5 does not. And it is not valid for every ISP to quote BT's figure as vimto-girl suggests. It has to be their own customer base. Not BT Consumer or Openreach theoretical calculations.

But I accept your point this discussion has become pointless as things are likely to change. The problem is that the regulators once again don't seem to fully understand the technologies they are regulating. This has seemed obvious to me for years, and you imply the same wrt this aspect in one of your other posts just now.

As I know that NO speed test can be absolutely accurate when I say within 1 Mbps I am telling you that one has to have some degree of tolerance.

Nothing wrong with every ISP using BT up to 76 Mbps figure if they can show this is the case when asked to prove it by ASA. A real risk if everyone wants to be behave in a black and white fashion is that any order for a 80/20 GEA-FTTC product would be rejected by providers unless they are pretty sure sync will be full tilt.

Remember some advertisers have presented what looks like reasonable data on various aspects of a service, but these get rejected, i.e. ASA rulings are very hard to predict even if someone has a good set of data. If presented poorly then good data is very likely to be rejected, but if done slickly you stand a better chance.

I'll go as further and say Ofcom may have some who understand the technology inside and out but this understanding is often diluted or ignored too.

Huh? I said I do not agree with the advertised figure! And that I do not believe 76 download can be achieved even in calculation! And I was far from endorsing the way I think ISPs copied BT consumer's figure and took safety in numbers as is possible!

I'm not really sure what your point is. I was just telling the facts, not really rebutting anything. It's not even up for discussion about throughput vs sync, it's in black and white what should be advertised.

In reply to a post by MrSaffron:

fingers crossed they might be sensible enough to talk to a cross section of industry people to try and understand some of the science involved.

They did a thorough consultation process actually last time, I guess any new approach should follow suit. A taster of what it involved:
- Consultation Proposals
- Consultation Responses
- Regulatory Statement

In reply to a post by MrSaffron:

Perhaps focus here is better spent on not the current rules, but debating what they should change to, or whether they should stay as they are.

Yet if people don't understand (or refuse to accept) the basics of the current rules, how can they debate whether or how they should change...? Same goes for how the ISPs interpret and attempt to comply.

In reply to a post by vimto_girl:

Yet if people don't understand (or refuse to accept) the basics of the current rules, how can they debate whether or how they should change...? Same goes for how the ISPs interpret and attempt to comply.

The people making the rules don't understand them, so what chance does anyone else have?!

With my Fibre with Plusnet since 2014 my line throughput always around 74-75 Meg. Never get above that. The Plusnet current line speed was set at 77.3Mb while BTw IP Profile set at 77.35Mb

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

Edited by adslmax (Sun 05-Jun-16 00:18:14)

Wireless is becoming less of an issue with wireless AC. I certainly see people getting 200Mbps over wireless on virginmedia regularly on iPads etc.

In reply to a post by MrSaffron:

We see people nailing the 76 Mbps so have never seen the point of making a big noise over the percentage
http://www.thinkbroadband.com/speedtest/results.html...

Maybe it's a linux/seamonkey thing, but I've always got saner results from your flash test (though there is one issue with it).

I guess testing data throughput with tcp is always tricky, but the new test is clearly capable of giving impossible results - just look at this freak, both avg and burst upload being impossible for a 20 meg sync ptm.

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

On the 76 thing, I have no idea what it is, but with adsl bt iprates were actually defined by bt as aal5 payload (so ip +2 ppp). On adsl the (so called) ip profile was even less related to ip, though IIRC bt docs did call that one bras rate = atm payload which did historically reduce speed below syc rate. Not sure about 21cn % rates - also not sure whether on fttc the profile is slightly limiting (being on plusnet with a locked down ECI modem it's hard to test that).

If the profile doesn't limit then on 80 sync it seems to calculate (accepting my sums may be wrong) to ip payload. Of course if bras limits then it it could be something else.

If talk talk test better maybe it means BT do limit - but then do TT use pppoe? I don't know but if they don't that could account for it. I came across a mail searching once that said sky don't use pppoe and it also said something about them not wanting to send fcs - if LLU providers avoid that + pppoe that's 12 bytes per packet saved which will easily show @ 80meg.

Upload yes, and its pretty clear from the graph that upload is spiking up/down, if I added smoothing issue would go away, but prefer to avoid things like smoothing or interpolation as these can hide useful bits of data.

Fair enough, though given the speed I bet the upload leaving my line was rock solid max rate at ip level, with the bursting being a tcp artifact/feature due to "me" dropping packets off the end of a buffer somewhere.

Out of all speed tests for upload I really rate your flash test, it pretty much gives consistent results, is single threaded and so lets me "play" like turning off wscale. It seems to be quite accurate eg. flipping timestamps on/off the difference seems correct from what I expect.

Saying that I may just be biased because it agrees with my calculated rates - it may be totally wrong and just re-enforcing my misconceptions

The flash had the smoothing turned on some years ago which explains a lot.