How much does distance affect speedtest.net results?

Forums › New Zealand Broadband › How much does distance affect speedtest.net results?

NonprayingMantis

6434 posts

Uber Geek
+1 received by user: 1528

#152239 21-Sep-2014 08:24

Testing speedtest.net to local servers gets me basically my line speed which is cool. (Via Ethernet of course)
Testing to overseas servers and the speed drops off, which I would expect give the distances involved

But my question is:
What sort of drop of should I expect when testing to say, Sydney, LA, London.
Would it be a percentage or a fixed number?

I've seen some speedtests to LA that seem to get 80% of their line speed, but others that only get 20-30%, which seems like a huge difference.

How much of this is due to the way speedtest.net works vs actual throughput available or lack of capacity from my ISP?

1 | 2

6028 posts

Uber Geek
+1 received by user: 461

Trusted

Geekzone

Lifetime subscriber

#1133024 21-Sep-2014 08:24

Hello... Our robot found some keywords in your post, so here is an automated reply with some important things to note regarding broadband speeds.

If you are posting regarding DSL speeds please check that

- you have reset your modem and router

- your PC (or other PCs in your LAN) is not downloading large files when you are testing

- you are not being throttled by your ISP due to going over the monthly cap

- your tests are always done on an ethernet connection to the router - do not use wireless for testing

- you read this topic and follow the instructions there.

Make sure you provide information for other users to help you. If you have not already done it, please EDIT your post and add this now:

- Your ISP and plan

- Type of connection (ADSL, ADSL2, VDSL)

- Your modem DSL stats (do not worry about posting Speedtest, we need sync rate, attenuation and noise margin)

- Your general location (or street)

- If you are rural or urban

- If you know your connection is to an exchange, cabinet or conklin

- If your connection is to a ULL or wholesale service

- If you have done an isolation test as per the link above

Most of the problems with speed are likely to be related to internal wiring issues. Read this discussion to find out more about this. Your ISP is not intentionally slowing you down today (unless you are on a managed plan). Also if this is the school holidays it's likely you will notice slower than usual speed due to more users online.

A master splitter is required for VDSL2 and in most cases will improve speeds on DSL connections. Regular disconnections can be a monitored alarm or a set top box trying to connect. If there's an alarm connected to your line even if you don't have an alarm contract it may still try to connect so it's worth checking.

I recommend you read these two blog posts:

- Is your premises phone wiring impacting your broadband performance? (very technical)

- Are you receiving a substandard ULL ADSL2+ connection from your ISP?

I am the Geekzone Robot and I am here to help. I am from the Internet. I do not interact. Do not expect other replies from me.

Referral links: Quic Broadband (free setup code: R587125ERQ6VE) | Samsung | AliExpress | Wise | Sharesies

Support Geekzone by subscribing (browse ads-free), or making a one-off or recurring donation through PressPatron.

sbiddle

30853 posts

Uber Geek
+1 received by user: 9996

Retired Mod

Trusted

Biddle Corp

Lifetime subscriber

#1133053 21-Sep-2014 09:26

It's a combination of various things, including TCP window issues that possibly still occur, and transit beyond Australia and the US gateways that your ISP has very little control over. Because the internet in the US is also such a shambles there are many congested interconnection points which can also affect speeds within the US.

NonprayingMantis

6434 posts

Uber Geek
+1 received by user: 1528

#1136973 24-Sep-2014 22:04

sbiddle: It's a combination of various things, including TCP window issues that possibly still occur, and transit beyond Australia and the US gateways that your ISP has very little control over. Because the internet in the US is also such a shambles there are many congested interconnection points which can also affect speeds within the US.

Cool. I'm wondering about the impact of SXS capacity bought by an ISP - and how much that really determines experience in NZ.
To quote Rod Drury:
http://www.nzherald.co.nz/business/news/article.cfm?c_id=3&objectid=11284849

"The Southern Cross cable monopoly has shown its adeptness at keeping competition out of the market, paying hundreds of millions in dividends while business calls from New Zealand on Skype stutter along."

I don't have any problem with skyping people even in the UK - about as far away as it is possible to be, so I don't really believe what he is saying, but it got me thinking about how much the international cable plays a role in our experience with the internet today

Obviously my choice of last mile technology doesn't impact how much international bandwidth my ISP purchases, but presumably I get a bigger 'share' if I have more capacity to my house to suck it down with.
i.e. better plan in NZ delivers a better speed result overall, so a ADSL speedtest to the USA will be worse than a VDSL speedtest to the USA, which is worse than a 100Mbps UFB speedtest, which is in turn worse than a 1Gbps speedtest, right?

Is there any sort of speed I should be 'expecting' with an overseas speedtest to 'test' sxs, and therefore test how much capacity my ISP is purchasing, (if I pick the sxs landing point to avoid as much USA congestion as possible)?
I have VDSL which gets around 50Mbps, and my USA speedtest is around 15-20Mbps typically. Is that about right? When I had ADSL, at around 20Mbps, my USA speedtest was around 6-8Mbps. Does that sound about 'right'? If (eventually) I get UFB, should I get better speeds again?

Inphinity

2780 posts

Uber Geek
+1 received by user: 1184

#1136994 24-Sep-2014 22:18

There are a lot more factors in it, though, which make it hard to predict. As a comparison, I just ran a speedtest to CA... connection is 100Mbps UFB with Spark.

And here's another one, over fibre with another major ISP, to the same location. Far better upload, but significantly worse download.

chevrolux

4962 posts

Uber Geek
+1 received by user: 2638
Inactive user

#1137033 24-Sep-2014 23:19

The Southern Cross cable monopoly has shown its adeptness at keeping competition out of the market, paying hundreds of millions in dividends while business calls from New Zealand on Skype stutter along.

That's just being so overly dramatic. No real business uses freken Skype for a business call. Ffs. I don't care how successful Xero is and how clever he actually is that is just purely attention seeking.

Is he mates with Morgan? Did he say that when they announced Pacific Fibre.

sbiddle

30853 posts

Uber Geek
+1 received by user: 9996

Retired Mod

Trusted

Biddle Corp

Lifetime subscriber

#1137051 25-Sep-2014 05:54

There are no issues with capacity on SXC, none at all.

If an ISP chooses to buy less capacity than needed, then that's a business decision they've made. Transit beyond intermediate points (such as the US to the UK) is also typically completely beyond the control of your ISP.

As for Skype, remember it's a best effort service.

Samsung

Shop now on Samsung phones, tablets, TVs and more (affiliate link).

NonprayingMantis

6434 posts

Uber Geek
+1 received by user: 1528

#1137081 25-Sep-2014 08:33

sbiddle: There are no issues with capacity on SXC, none at all.

If an ISP chooses to buy less capacity than needed, then that's a business decision they've made. Transit beyond intermediate points (such as the US to the UK) is also typically completely beyond the control of your ISP.

As for Skype, remember it's a best effort service.

That's pretty much what I thought.

I am wondering if there is any way to tell whether my ISP is under-purchasing on capacity by running speedtests and comparing them with local, but allowing for the impact of latency.

If I get 50Mbps locally with VDSL, what's the maximum speedtest result I should be able to get the USA if my ISP has heaps of capacity?

e.g. on a 50Mbps VDSL connection, obviously I would never expect to get 50Mbps to the states on speedtest.net (right?), but is there any way to estimate what I could get - by looking at the latency and calculating the maximum theoretical speedtest result. e.g. latency is 200ms, therefore my 50Mbps VDSL cannot do better than XMbps to LA.

Then by comparing that to the actual result I get , I can see (roughly) if I am getting a good international connection or not.
if Maximum is, say, 30Mbps and I am getting 25Mbps, that's pretty decent. But if I am only getting 5Mbps then that's not so good.

NonprayingMantis

6434 posts

Uber Geek
+1 received by user: 1528

#1137082 25-Sep-2014 08:34

chevrolux:
The Southern Cross cable monopoly has shown its adeptness at keeping competition out of the market, paying hundreds of millions in dividends while business calls from New Zealand on Skype stutter along.

That's just being so overly dramatic. No real business uses freken Skype for a business call. Ffs. I don't care how successful Xero is and how clever he actually is that is just purely attention seeking.

Is he mates with Morgan? Did he say that when they announced Pacific Fibre.

loads of startups do - even biggish ones.

Aredwood

3885 posts

Uber Geek
+1 received by user: 1749

#1137899 25-Sep-2014 22:56

Try doing some speedtests to Fiji and Hawaii. As the Southern cross cable runs through both of those countries on it's way to the USA. And lots of other cables stop in at 1 or both of those countries.
https://www.speedtest.net/my-result/3786148265

sorceror

163 posts

Master Geek
+1 received by user: 47

#1143101 28-Sep-2014 20:02

e.g. on a 50Mbps VDSL connection, obviously I would never expect to get 50Mbps to the states on speedtest.net (right?), but is there any way to estimate what I could get - by looking at the latency and calculating the maximum theoretical speedtest result. e.g. latency is 200ms, therefore my 50Mbps VDSL cannot do better than XMbps to LA.

Then by comparing that to the actual result I get , I can see (roughly) if I am getting a good international connection or not.
if Maximum is, say, 30Mbps and I am getting 25Mbps, that's pretty decent. But if I am only getting 5Mbps then that's not so good.

using latency to determine max speed is a bit simplistic - sure you can figure it out for a single TCP connection, but most speedtest.net sites use multiple connections so the 'expected maximum' calculation becomes complex quite quickly (overheads start becoming more of a consideration)

the number of allowed connections is also dependent on what version of the Ookla speedtest software the server is running (and there's no easy way to check this from what I've found)

Coil

6614 posts

Uber Geek
+1 received by user: 2153
Inactive user

#1143102 28-Sep-2014 20:04

AliExpress

Shop now on AliExpress (affiliate link).

wired

191 posts

Master Geek
+1 received by user: 72

#1143252 29-Sep-2014 08:38

Have a look at https://www.sandvine.com/downloads/general/global-internet-phenomena/2013/2h-2013-global-internet-phenomena-report.pdf there is a section on "does speedtest.net deserve a failing grade" (page 7) It has a good example about how distance affects the speed test results.

timmmay

20857 posts

Uber Geek
+1 received by user: 5349

Trusted

Lifetime subscriber

#1143258 29-Sep-2014 08:46

Aredwood: Try doing some speedtests to Fiji and Hawaii. As the Southern cross cable runs through both of those countries on it's way to the USA. And lots of other cables stop in at 1 or both of those countries.
https://www.speedtest.net/my-result/3786148265

I was in Fiji recently, the only worse internet I've had was Samoa where it effectively didn't work. In Fiji I got 0.38Mbps down, 0.81Mpbs up, and 1539ms latency - that was between an Island and their capital, Suva. Unsure what kind of connections they use to the islands, but they're not great.

sorceror

163 posts

Master Geek
+1 received by user: 47

#1143736 29-Sep-2014 18:51

here are some real world results (done just now from my UFB connection):

this made 16 TCP connections to reach that speed - clearly latency isn't an insurmountable issue anymore for speedtest.net

NonprayingMantis

6434 posts

Uber Geek
+1 received by user: 1528

#1143748 29-Sep-2014 19:32

sorceror: here are some real world results (done just now from my UFB connection):

this made 16 TCP connections to reach that speed - clearly latency isn't an insurmountable issue anymore for speedtest.net

assume you are on a 200Mbps plan?

based on TimA's chart above, 16 TCP connections seems too low to get that result.
Ping of 172ms gives less than 5Mbps per connection (based on that chart), so 16 connections would, presumably, max out at less than 80Mbps, no?

(also, how do you know how many connections the speedtest.net test has done, I can't see it anywhere on speedtest.net? - the speedtest website claims to use up to 4 HTTP threads https://support.speedtest.net/entries/20862782-How-does-the-test-itself-work-How-is-the-result-calculated- although that 'article' is a couple of years old so maybe they use more now IDK

1 | 2