close
MENU
14 mins to read

RAW DATA: TrueNet May 2016 Broadband Report - Incredible Copper Speeds - (ADSL & VDSL)

Another month of excellent copper connection performance by all ISPs that we measure, the best on record.

Staff Reporter
Tue, 28 Jun 2016

(TrueNet is contracted to the Commerce Commission to monitor broadband performance. It uses a network of probes attached to 400 volunteers' internet connections to rate ISPs — Editor.)

Another month of excellent copper connection performance by all ISPs that we measure, the best on record.  This means that all major ISPs are delivering ADSL and VDSL evening speeds that are nearly equal to the speed achieved when there is no traffic - see Charts 8 & 9.  Thats the equivalent of having all cars on a motorway doing 95-100km/h during the rush hour.  Fantastic!

Trustpower performance improvement is impressive.  After reporting their fibre performance for only 3 months, TrueNet's test results now show that Trustpower has removed all congestion.  We report the combination of 50 and 100Mb/s products showing the variation by time of day, which this month is almost zero.  See hart 7A

We also see a major improvement in Vodafone speeds for 100Mb/s fibre, increasing from just 60Mb/s to a more regular 95Mb/s for most TrueNet panelists on Vodafone connections.

 

Please complete TrueNet's consumer perception survey to help us gather more information about end-users experience of ISP broadband performance. This Survey  is a permanent feature on TrueNet's website, and helps us monitor the importance our readers assign to a range of measurement factors.

Summary of Performance Measures

After objections raised by ISPs when TrueNet changed to reporting ISPs and not individual brands, we have returned to reporting all brands for each ISP this month.  However, as BigPipe panelist numbers have dropped due to migration to other providers, TrueNet is no longer confident of the accuracy of our test results for BigPipe so they are not included in the May 2016 report.

May test results for ADSL and VDSL are extraordinary, with all New Zealand ISPs achieving better than 95%.  

Vodafone fibre speed, actual as a percentage of advertised speed, improved in May. This was based on a non-specified change that Vodafone initiated on the 10th May.   Previous speed results at 60% to 70% of advertised rate (e.g. 100Mb/s service) increased to over 95% in May.  This improvement is also seen in the Sydney speed results, up from around 40Mb/s to over 60Mb/s.

Improved speed performance for ordinary copper connections, (ADSL and VDSL) is continuing to improve, so much so that we thought it worthwhile showing our ADSL performance trend chart this month, although this is always available on the "ISP Performance" dropdown above.  This shows how much effect publicly available information has on the ISPs who, over time, have responded with better quality broadband services.  Take a look at the reduction in performance when movie services were started - Lightbox in July 2014 and Netflix in March 2015.  In May 2015 only Auckland Servers were available unfortunately, meaning southern panelists' connections traversed all the way up the country potentially affecting perfomance results.

(Click to zoom)

Speed test results in the Summary Table 1 are based on tests from the Auckland server only.  A problem with our Wellington Test Server occurred during the month, followed by a cable failure off the same server, and therefore Wellington results were not used in the "Best of" calculation.   Because of this change, it is inappropriate to compare directly with previous months, and panelists in the deep South may be impacted worse than Auckland results.  This is particularly noticable in the 100Mb/s fibre speed results, but should have had little impact on the Max/Min results.

 

Table 1: Summary of Performance Measures

(Click to zoom)

*Notes: Cable and Fibre speed comparisons show minimum hour (slowest) speed to Advertised Service Speed; ADSL & VDSL compare minimum hour (slowest) speed to the hour with maximum (As fast as your line will allow) speed. Due to problems with the Wellington server during May 2016, NZ Speed results are from the Auckland server to each test probe.

 

Webpage Download Time

Methodology

Responsive website browsing is valued by most Internet users, and conversely, slow-loading sites can be extremely frustrating.

TrueNet tests Internet browsing by downloading a selection of Live Webpages from NZ, Australia, and the USA, measuring the time to fully download all files on the page.  These pages are changed from time-to-time. This month there are 7 NZ pages, 4 Australian, and 4 USA pages.

The charts in this section have three points for each ISP entry showing the Min. Hour (minimum time to download a webpage); the Max. Hour (maximum time to download a webpage); and the Median of all hours.  The best ISP has a low median, and a short arrow showing the spread between peak and off-peak performance.

We download 7 popular NZ websites from each connection, and compare the average time taken to download all websites.  This test replicates daily activity for many people, and we group the websites into regions, NZ, Australia and USA, so that readers can compare ISPs based on their own preferences.

 

NZ Webpages

The May results of NZ webpage tests are in line with April results, with a few minor differences.  Overall webpage download times are down a very small amount, while min / max hour variation is up in a few cases. Some ADSL, VDSL and Fibre results have greater variation than April. Peak period performance was under pressure for MyRepublic fibre, Vodafone fibre 50Mb/s, Vodafone VDSL/ADSL, and 2DegreesADSL panelists . The grouping of "Other ISPs" in 100Mb/s service was much closer to other 100Mb/s services compared with previous months.

 May is the second month with 200Mb/s fibre included, and the results are again very similar to other fibre service speeds for each of the 3 country destinations of the live webpage tests.

Chart 1: NZ Webpage Download Time - in seconds

US Webpages

We download 4 popular US websites from each connection, and compare the average time taken to download all websites.   This test replicates daily activity for many people, but to provide the ability for readers to compare between ISPs based on their own preferences, we group the websites into regions, NZ, Australia and USA.

 

The average download time of US Webpages is fairly equal across Cable and Fibre ISP results, with no obvious advantage to speed, or technology. This implies that any constraints on Webpage download times are outside of NZ, and thus likely outside ISP control.

Chart 2: USA Webpage Download Time - in seconds

Australian Webpages

We download 4 popular Australian websites from each connection, and compare the average time taken to download all websites.  This test replicates daily activity for many people, but to provide the ability for readers to compare between ISPs based on their own preferences, we group the websites into regions, NZ, Australia and USA.

 

Vodafone 100Mb/s Cable continues to have a poor response for Australian pages, potentially reflecting congestion showing in Chart 4 below.

For our Australian Live Webpage test, there was an increase in variation of ADSL and VDSL results, as well as Vodafone 50Mb/s Cable. There was a slight increase in webpage download times, despite there being no apparent reduction in the measured Australia file download speed tests.

Chart 3: Australian Webpage Download Time - in seconds

Speed (File Download Performance)

For TrueNet's speed tests each panelist's probe regularly downloads a 1MB or 2MB file from Auckland, Wellington, Dallas and Sydney.  The faster the connection, the larger the file we need to download to ensure that the maximum speed is reached during our test. 100Mb/s connections easily reach full speed before 2MB of data is downloaded.  Slower connections can test accurately with much smaller files. Our tests on Satellite, Wireless and ADSL can tolerate a file size of just 300k.

New Zealand tests usually take the best of each test run from Auckland or Wellington. However, due to problems with the Wellington server, only Auckland results are used this month.

 International tests take the result from each test run from Dallas, or Sydney. 

Our reported measure is the average hourly speed as a percentage of the best hourly speed in the month, this measures the ISP performance rather than the technology performance.

 

International File Download Speed

TrueNet tests File Download Speeds from our servers in Dallas & Sydney to measure performance to our international neighbours.

We ensure the download file is not held in New Zealand (cached), so that the test truly measures international performance.  The results are selected from the 8pm to 10pm evening busy period, when any congestion is most likely to be observed.

 

Fibre & Cable

We have observed an unusual increase in the Dallas test speed from some probes this month.  While we are investigating whether results are correct, they are not included in this month's reporting.

Results from the Sydney speed tests show a big improvement in Vodafone 100Mb/s fibre, while 2Degrees fibre is again doing well, averaging almost line speed for the 30Mb/s service.   2DegreesVodafoneSpark & Voyager 100Mb/s services all had similar results,  MyRepublic and Orcon are close but slower.   As with previous months, the aggregate of Other ISPs had significantly slower speeds than the named ISPs on the chart.  

Chart 4: Fibre & Cable File Download Speed - Sydney, Australia

ADSL & VDSL (Copper Connections)

Copper connections (ADSL & VDSL) have a speed that is dependent on the distance between the home modem, and the exchange equipment which means that ISPs do not have any influence on the peak speed of each connection.  To overcome this limitation, the Australia, and USA speeds are referenced to the average NZ download speed of the respective ISPs.  The speed reported is from the 8pm-10pm evening period.

Our Dallas results look fine at the slower ADSL and VDSL speeds, so we are continuing to use them while we investigate the results of faster connections.

ADSL: Spark had the highest results from Sydney but Vodafone did not maintain the great results from April.

VDSL: 2Degrees and Spark improved in relative speed from the US, the other ISPs results are similar to April. For the Sydney test, Voyager maintained April's results, but the other ISPs were down comparatively.

Chart 5: ADSL, VDSL File Download Speed - Dallas & Sydney

New Zealand File Download Speed

Comparing performance by time-of-day is important as it shows the service degradation when everyone is using the Internet during the evening hours of 8pm to 10pm.  TrueNet uses the best of a pair of Auckland and Wellington download tests to calculate the median NZ results by hour over the month for each monitored connection.  We take the average of all median results with each ISP for each hour.

A poor result typically shows the line drop below 90%, which usually occurs in the busy period between 7pm and 10pm, i.e. if this is true, the average user for that ISP is getting less than 90% of their line capability.

 

Technology Comparison

Copper services (ADSL, VDSL) have improved to an exceptional level this month with almost no contention (where broadband users compete for access on the same network resulting in slow connection speeds) evident in any ISP. 

Overall there was a noticable increase in contention in Cable and Fibre

However, bear in mind these results only include Auckland speed tests (Wellington not used), making absolute comparisons between months difficult for the faster speeds.  Normally we take the best of two tests each hour to compare results, this means if contention is absent from one site, and the connection is faster then no contention will show, even if it exists.   Having said that, contention to the Auckland peering exchange on Fibre and Cable is clearly an issue.

Chart 6: Fibre, Cable, and Copper (DSL) File Download Speed

Fibre & Cable ISP Comparison indexed to Advertised Speed

There are several notable changes compared to last month's Time of Day chart for Fibre and Cable.

  • Trustpower seems to have solved the dip in evening performance of the last couple months, to have an almost flat graph in May.  
  • Vodafone's Fibre performance has improved significantly, from between 60% - 70% of advertised speed, to nearly 100%. This change occurred at a specific time and date for many of the Vodafone probes, indicating a sudden change in the network.
  • There was also signs of busy hour contention returning to Orcon & Spark fibre
  • A further decline in evening performance for MyRepublic
  • Other ISPs have improved from last month, but their contention is worse, although not dipping to the flat 60% levels of last month. 
Chart 7: Fibre & Cable Performance as Percentage of Advertised Speed

Trustpower performance has improved a lot over the last six months, but especially so since TrueNet first started publishing comparisons of speed by time of day.  We recalled that earlier performances looked poor for all Trustpower panelists, so we checked to see how much their performance had changed since January 2016.  Note that January and February data is not reliable due to the limited number of panelists (3 & 4 respectively), hence the dotted lines.  However earlier data fits the picture of change, most likely due to increasing network capacity.  Chart 7A, shows the extent of the change, and shows that Trustpower now has one of the best fibre performances TrueNet measures by time of day.  Note, that our Wellington issue did not affect Trustpower because all these panelists are in the upper North Island, so all tests since January are from the Auckland server.

Chart 7A: Trustpower Fibre Improvement by Month, March to May 2016

DSL Performance by Time of Day (ToD)

High Speed Copper (VDSL) File Download Speed

VDSL Time of Day performance was excellent in May, with only marginal differences between ISPs, and each staying above 95% of best hour speed throughout the day. The small April variations with Vocus (Orcon included with Slingshot for April publication), Voyager and Vodafone are essentially gone (when only testing to Auckland).

Chart 8: VDSL File Download Speed by Time of Day

Low Speed Copper (ADSL) File Download Speed

ADSL consistency is very high this month, with all NZ ISPs achieving better than 95% of maximum speed throughout the day.

In contrast, our sample of panelists on TPG Australia show severe contention in the Australian evening period, and hence a sharp dip at 10PM (Sydney time).

Chart 9: ADSL File Download Speed by Time of Day

Upload Speed - Performance

Upload speed is important to users sending large amounts of data through the Internet, or loading files to the Cloud.   TrueNet's upload test sends a 1MB file to our Wellington server, and records the results using a similar method to the download tests, but measured from Wellington only.

 

By Technology

Average upload speed by technology remains steady again.

Chart 10: Upload Performance by Technology

 

Upload Time of Day Performance - by ISP

 Uploads are conducted to our Wellington server, which showed no issues with upload speeds so we report all results.

In Chart 11, the only significant Time of Day effect is in the Orcon 50Mb/s Upload service, echoing the evening dip in download speed results, see Chart 7 above.

We continue to observe that "Other ISPs" deliver below advertised 20Mb/s service, while the named ISPs deliver about advertised.

Chart 11: Fibre & Cable ToD Upload Speed by ISP

Volunteers are needed in the following Technologies: 

ADSL - 2Degrees, Spark, Vodafone;
VDSL - Slingshot, Vodafone, Voyager;

Staff Reporter
Tue, 28 Jun 2016
© All content copyright NBR. Do not reproduce in any form without permission, even if you have a paid subscription.
RAW DATA: TrueNet May 2016 Broadband Report - Incredible Copper Speeds - (ADSL & VDSL)
59441
false