Optus, iiNet crush Telstra in speed challenge

Customers of Optus and iiNet, along with users of TPG and Internode, have a much faster Internet connection than their Telstra-using counterparts, according to results generated by ZDNet Australia's Broadband Speedtest.

Thousands of Web users across the globe have taken the ZDNet Australia broadband speedtest, which determines a user's broadband speeds, since it was launched in September.

Since its introduction, more than 270,000 tests have been completed, the majority of them from Australian users.

Battle of the Australian ISPs
Though the results will change as more people take the test, Exetel is currently in top place -- with an average connection speed of 11,258 kbps. However, it should be noted that only two tests have been completed using the ISP's network at the time of writing. Second place was claimed by iiNet, with an average speed of 8,688 kbps after more than 38,000 tests.

OptusNet, TGP Internet and Internode currently stand at fifth, sixth and seventh, with average speeds of 6,942 kbps, 6,924 kbps and 6,183 kbps respectively.

After more than 55,000 tests, Telstra has notched up both the fastest (169,054 kbps) and slowest (1 kbps) recorded speeds. BigPond currently records an average of 5,646 kbps and sits in ninth place overall.

On the slower end of the table, AAPT only managed an average seed of 1,568 kbps after more than 5,000 tests while Unwired, with 605 tests, managed an average performance of 364 kbps. At the foot of the table is satellite ISP Gobush Broadband, which after 31 tests managed a top connection speed of just 701 kbps and an average of 349 kbps.

For the complete ISP table with more than 40 providers' results click here.

State of origin -- Web style
NSW took the top spot with an average speed of 6,746 kbps, with Queensland and Victoria a close behind on 5,842 kbps and 5,682 kbps respectively. Despite the resources boom, WA is second-last with an average of 3,960 kbps. Tasmania lags behind with a sluggish 1,107 kbps average.

The Australia-based rankings are available here.

The World Cup
Quiet achiever Switzerland took the top spot on with a blistering average speed of 20,978 kbps, followed by neighbour Germany at 19,852 kbps. Surprisingly, high-tech pioneer Japan is only ranked sixth with an average connection of 12,187 kbps.

An average speed of 6,381 kbps took Australia to 22nd place, just pipping Singapore.

Check the country rankings for yourself.

ZDNet Australia's Broadband Speed Test measures the data throughput between a computer and a geographic distributed network of servers to calculate how fast broadband speeds are in the real world. The results are not an indication of the actual line speed between the user and the ISP. There are many external factors such as network congestion and routing between multiple networks that can affect individual results, which means the broadband speed test does not necessarily warrant 100 percent accuracy.

Like this article? Click below to send it to your mobile for free!

Advertisement

Talkback 11 comments

  1. Stupid survey Anonymous -- 06/11/07

    What an idiotic survey. Since when did BigPond oiffer a 169Mbps service? Or 104Mbps from TPG. And 1kbps services? Someone is pulling your chain. Putting Exetel on top from just two 'tests' is a joke. Seems like the Melbourne Cup party at zdnet started too early

    1. maybe you should re read the story? Anonymous -- 06/11/07

      where does it say Mbps?

    2. Maybe you should read the story! Anonymous -- 06/11/07

      169Mbps is the same as 169,054Kbps

    3. I wonder how many people out there have a Gigabit connection? Anonymous -- 07/11/07

      The only way someone will get over a 100Mbps (or 100,000kbps or 100,000,000bps) is by having a GE connection on their computer, a GE switch, a GE router, a fibre connection to their ISP and their ISP having a GE link from the local exchange to their POP.

      Highly unlikely which just goes to prove what a flawed test this is.

      It is also important to note that someone testing a dial up is lumped in with someone testing an ADSL2+ service. Why don't they actually breakdown the results by the actual end user connection type / speed so wireless connections can be compared with wireless, 1.5M ADSL with 1.5M ADSL, ADSL2+ with ADSL2+ etc.

      Once again ZDiot prove that they are uneducated, a second rate publication as well as clambering to ride on the bandwagon of bagging Telstra and promoting it's competitors.

    4. Agreed Anonymous -- 09/11/07

      It should also be noted that people with faster internet connections (ie. ADSL2+) are more likely to be the kind testing the speed of their connection than those on 1500.

      People on dialup, 256, 512, and 1500 connections are more likely to know what kind of speeds they are obtaining without needing a test to tell them.

  2. Speed test. Anonymous -- 06/11/07

    I've used the meter multiple times with multiple readings. Does ZDnet count every attempt or does it count only one attempt per person?

    1. Attempts at Speedtest Suzanne Tindal -- 08/11/07

      Each test done was recorded. If a person did multiple tests, all of their tests were recorded.

    2. Mutant Speedtest Malcolm Moore -- 27/11/07

      Good 1

      Which ISP was connected to what Major Switch/Router and which Major Router/Switch was each caller connected to via their local Exchange Router/DSLAM? (Was that recorded too?)

      What was the available connection speed from the ISP into the network at that time? (Was that recorded too?)

      We know with IP transmisison, as the path is used the connection speed usually increases. With changes in speed during connection did you include this data? (Was that recorded too?)

      Did all the calls so far connect via ISPs that are local to the caller - or were they deliberately cross continent connected? (Was that recorded too?)

      How many test ISPs are located internationally so that international traffic can be included? (Was that recorded too?)

      Praise be to thee..... Amen

  3. Broadband speedtest Anonymous -- 07/11/07

    It would be interesting to see a cost comparison to see what $ per mb those participants pay for their speediness. I am sure the likes of iiNet , Internode etc would move further to the top, and Telstra would move further to the bottom.

    1. Day or night Anonymous -- 07/11/07

      Are you talking about the free traffic they offer between 3am and 3:15am or the higher priced traffic when people actually use the internet service?

      You know the saying about peanuts and monkeys. Want a banana?

  4. We desperately need some good figures Greg Alexander -- 23/11/07

    Nice stuff. I would be VERY interested in seeing results which exclude all the results over 22Mbps, and exclude all the results that are under 1.5Mbps.

    That removes internal ISP results (and fibre connections), and removes the skew to the results from people purchasing 1.5, 512K or 256K connections.

    It leaves us with what speeds are really possible, rather than speeds which include artificial restrictions.

    While your current info is great, any chance of seeing results like that?

Add your opinion


Sponsored content

Power Centre - Content from our premier sponsors

Blogs

  • Alex Serpo Is green IT a marketing fad?
    It seems that green IT has dropped off the radar, with other technology issues moving to the fore. But was green IT ever a real technology movement, or was it just a marketing fad?
  • Array Gutless studios have the wrong target
    I have one word for the Australian Federation Against Copyright Theft (AFACT). Gutless.
  • Array NBN needs workers on board
    Without consensus on labour issues, the eventual winner of the NBN may end up as little more than a lame duck and a cashed-up symbol of the conflict between the desire for progress and the lack of mechanisms to deliver it.
  • More blogs »

Tags

Back to top

Featured