Jump to content


Sophist Member
  • Content count

  • Joined

  • Last visited

  • Speed Test

    My Results
  • Days Won


Sean last won the day on February 12

Sean had the most liked content!

1 Follower

About Sean

  • Rank
    TMN Friend

Profile Information

  • Gender
  • Location

Recent Profile Visitors

6,412 profile views
  1. Host Graph

    To me, that seems like either high packet loss or intermittent drop-outs, both which you can check by running an extended ping test while streaming. On a Windows PC, open a command prompt (Start -> Windows System -> Command Prompt) and type the following command: ping -t You should see a continuous run of "Reply from". Leave that window open and start streaming a programme. As soon as the streaming stalls, check the window for any "Request timed out" lines. If you see three or more in a row, your connection had a brief outage. Press CTRL + C on the keyboard to stop the ping utility. Look at '% loss' figure. If that is 1% or higher, scroll up by holding the mouse down on the top-right up-arrow and look out for lines that say 'Request timed out'. If you see three or more grouped together, that is another brief outage. If you come across five or more 'Request timed out' lines in one screen-full, this is a high packet loss issue, which can also interrupt streaming. I've had an issue in the past with at least 3 D-Link routers failing with a high packet loss with this exact symptom, i.e. no issue with browsing or speed tests, but could not reliably stream YouTube.
  2. Going by the ATTN DR figure which seems to indicate attainable downrate, there is something seriously wrong with the line itself between your modem and the exchange. This is not your ISP limiting the speed or a contention issue. Based on those figures, it looks like you have severe noise on the line. Usually water ingress or a DSL filter fault would push the line attenuation figure over 60dB. An attenuation of 50dB should still be able to deliver around 4 to 5Mbps. Do you have any type of signal amplifier running in your building such as a mobile signal booster or TV antenna amplifier? If you do, switch them off and reboot your router to see if the speed improves. The next step would be to try a process of elimination unplugging electrical appliances in case you have a faulty appliance generating noise. For example, an electric heater with an arcing contact can generate severe noise. I don't think it's a bridge tap being added back in. Based on the info of what I could find about bridge taps, a bridge tap is used where the telecom provider runs phone lines down the full length of a street / circuit and each customer is connected somewhere along those cable runs. With DSL, they cause havoc as the signal will reflect up/down the unused sections of cable beyond the bridge tap. When a bridge tap is removed, the telecom provider physically disconnects the wire beyond the where the customer's line connects to the circuit. There would be no reason for them to ever rejoin that unused section of cable. The exception would be if you terminate your service altogether and the telecom provider would like to reuse that line to serve a new customer further down the street.
  3. Going by that Downstream rate, your modem appears to be facing a lot of noise or attenuation on the line. Check the downstream attenuation and noise margin (or S/N) figrues. If the S/N value is higher than 6dB, try rebooting the modem and see if the Downstream figure increases. If the S/N figure remains quite high such as over 12dB, then the Internet service provider is likely restricting your maximum speed. For example, on my DSL line, the maximum I can get is 5120Kbps as that is what the ISP deems is the maximum my line can support. If the downstream attenuation figure is very high such as over 50dB (and you're not 2+ miles from the exchange), try another phone lead between your modem and the socket and check if the figure improves after the DSL syncs again. If the figure does not improve, then there is likely some other fault with the line such as degradation, water ingress, corroded contacts (including at the exchange end), etc. If you have a DSL filter/splitter, try removing it.
  4. Where are all these tests coming from?

    This is common with dynamic IP addresses. When your router restarts or the IP lease time expires, the server provider will likely assign a different IP address from its pool, after which the IP you add will become available in the pool to assign to another customer. A side effect is that you will see test results from subscribers that were previously assigned the IP address you currently have, highlighted in yellow. You will also see these type of results if you run speed tests on another device on your network that is logged out on TestMy. You can hide those test results by other users by clicking the following button at the bottom of the test results page: Your speed is very good compared to others on the Hughesnet network.
  5. Extremely low speed on European servers

    I had a quick look at the database and interestingly every speed test on a French ISP from about 1pm on the 31st January to the German TestMy server was under 2.5Mbps (except a few tests run on hosting providers), so there may have been a peering issue between TestMy's German server and ISPs in France. With ISPs in Germany, the majority of test results were well over 10mbps with the German TestMy server, with some test results over 100Mbps. However, I did not see a similar pattern with the UK TestMy server from France, even with tests carried out on Free SAS. For example, the test on the left was run at the same second coincidentally as your test on the right:
  6. So you want to stream up all the data?

    Ah, was wondering. I think there's just one ISP in Ireland that throttles to 1Mbps. The rest either cut the connection (e.g. LTE and wireless ISPs) or charge for excessive usage. I checked the Windows Resource Monitor while playing a video forced in 4K mode: I see that YouTube now streams over two simultaneous connections, as those two connections were going steady until I stopped the video. The following is a trace route to this IP: According to the TCPIPUTILS website, the last 4 ISPs shown belong to Google. is a non Internet routable IP, so is within the ISP. This indeed means that either Google has servers within the ISP or has peering with Google that is given high traffic priority.
  7. So you want to stream up all the data?

    Is that 500kbps limit you set or ISP enforced? My ISP (LTE based) surprisingly prioritises most streaming traffic such as YouTube (and Netflix I think) during peak time. My monthly limit is 750GB, however, that cap is strictly enforced. Go over it and it is a eye popping €50 per GB, which I've heard people getting hit by. To give an idea of traffic prioritisation, the following is a YouTube "Stats for Nerds" screen playing 4K, which should be impossible at the speed TestMy reports on the right:
  8. latency issues

    Going by your trace route, the server is both far away (Virginia, US) and requires a lot of hops to reach it. Your ISP's latency appears to be fine going by the first few hops. The two largest latency issues are between hop 10 & 11 (Sydney to Los Angeles) and 12 to 19 (West to East US, i.e. Los Angeles to Washington DC). See if they offer a server much closer to you, in Australia if possible. If the people you are playing with are located outside of Australia, try using a server somewhere centralised between you and them, such as in Europe if they are in European countries. Even if they are in the US, try choosing a server towards the south west of the US. This will reduce the latency by about 50ms and the number of hops, i.e. hops 11-19.
  9. From some experimenting and research - The culprit is HTTP2! I created a simple HTML file and used the image TestMy was fetching. I inserted it multiple times, changing the '?=' text on each line, so it looked like this: <!DOCTYPE html> <http><head><title>test</title></head> <body> <img src="https://cloud.testmy.net/tmn/img/tmn-sti_2.jpg?q=9ijehwae1" class="test" height="20" width="20" /> <img src="https://cloud.testmy.net/tmn/img/tmn-sti_2.jpg?q=389gehae1" class="test" height="20" width="20" /> <img src="https://cloud.testmy.net/tmn/img/tmn-sti_2.jpg?q=3h89uw2g1" class="test" height="20" width="20" /> <img src="https://cloud.testmy.net/tmn/img/tmn-sti_2.jpg?q=zdpq2jga1" class="test" height="20" width="20" /> <img src="https://cloud.testmy.net/tmn/img/tmn-sti_2.jpg?q=18saegaw1" class="test" height="20" width="20" /> <img src="https://cloud.testmy.net/tmn/img/tmn-sti_2.jpg?q=aghh23871" class="test" height="20" width="20" /> <img src="https://cloud.testmy.net/tmn/img/tmn-sti_2.jpg?q=h3o8aw891" class="test" height="20" width="20" /> <img src="https://cloud.testmy.net/tmn/img/tmn-sti_2.jpg?q=jhh2983h1" class="test" height="20" width="20" /> <img src="https://cloud.testmy.net/tmn/img/tmn-sti_2.jpg?q=zxchoiwq1" class="test" height="20" width="20" /> <img src="https://cloud.testmy.net/tmn/img/tmn-sti_2.jpg?q=2ho8zs871" class="test" height="20" width="20" /> </body> </html> Generally when multiple images are fetched, I expect to see multiple TCP connections. However, this was not the case as with running TestMy's multithread test with the Cloudflare CDN. One thing that came to mind is HTTP2, which I was recently reading about. I remember it used a different method of connecting to servers and sure enough the FAQ has the answer: To check if this is the culprit, I disabled HTTP2 in Firefox's about:config page: I reran a multithread speed test using the Cloudflare CDN and sure enough it makes multiple TCP connections: I'm not sure if you're able to disable HTTP2 on Cloudflare, otherwise it looks like the only way one can run a multithread test with the Cloudflare CDN is by disabling HTTP2 in the web browser.
  10. I found one issue with the Cloudflare server, but not sure if it's what you encountered before. At a first glance, the test seems to work fine. However, with a few test results lower than the UK for the multithread test, I had a quick look in the Windows Resource Meter. When I run a Multithread test using the Cloudflare server, it only downloads over a single connection: For comparison, this is how it appears when I retest using the UK server: One thing I've observed already is that the Three 4G network prioritises Cloudflare traffic. I certainly don't get 2MB+/s with any of the linear speed tests.
  11. A few years ago, TestMy provided Google CDN as a speed test server. This made a pretty good test for wireless and cellular data ISPs to test the bandwidth between the user and the ISP, especially for making antenna adjustments since most ISPs have good peering with Google. That came to an end when they discontinued that CDN service. Now that TestMy uses Cloudflare, I wonder if it could be set up as a multithread test server, much like what was done with the Google CDN.
  12. That's great - These seem to be working fine now. I also noticed that after you switched to HTTPS, the multithread tests now only test with a single domain per server. This threw up an interesting surprise with my 4G LTE connection on the Three network - It throttles per host. Initially I wondered why my multi-threaded test results were much lower than recently, until I tried selecting more than one server. For example, I tried a handful of multithread tests in a row, selecting London, then Germany, then London+Germany, all with 25MB blocks. I repeated this sequence three times to rule out a coincidence. In the test results below, the red-highlighted lines indicate the tests I ran with London+Germany. In the top result, I selected London+Germany+Miami+New York. In the past, I wouldn't have seen this as selecting a single server such as London resulted in testing across multiple domains such as oldtestmy.net, tmnstatic.com, testmercury.net, etc.
  13. When a multithread test is in progress, the [STOP] button does not work properly, bringing up a 'Secure Connection Failed' error screen. When I hover my mouse over the [STOP] button... ... the address shown in the status bar has a ':80' after the domain name: It affects the mobile website also, again bringing up an error screen when I touch [STOP], with the ':80' clearly in the URL it tried loading: This issue also affects the upload test on the mobile website when Multithread mode is enabled (left image). The Express upload test also fails with Multithread mode enabled (right image):
  14. If ISPs decide to throttle traffic, they would need to do this all services to be effective, in which case it would probably be easier for them to charge by maximum speed like many ISPs did in the post, e.g. one price for 10Mbps, a higher price for 50Mbps and so on. Otherwise, it would be relatively straight forward to overcome with the use of a VPN. Let's say an ISP prioritises port 8080 to deliver fast speed tests while throttling everything else, just use something like OpenVPN over port 8080 with a VPN privacy service. To the ISP, all your traffic would be seen and treated as speed test traffic. For example, about two years ago the Irish cellular networks Three and Vodafone were doing something similar, i.e. throttling most traffic over the standard web ports (e.g. HTTP port 80), while letting port 8080 run at full speed to deliver fast speed tests, at least with the well-known Ookla Speedtest App. At the time, it meant one could get 4G speed tests over 20Mbps, yet faced slow browsing speeds similar to a 1Mbps connection. All I had to do was make a VPN connection over port 8080 (same port # as Ookla uses for its speed tests) and everything performed a heck of a lot better. A few months later, Three changed their tactic by prioritising certain services such as YouTube when the network is congested. This means YouTube can potentially play 4K fine even when the speed tests (including Ookla) deliver low test result figures.
  15. Now working fine Thanks for fixing this, it kept catching me out earlier.
Speed Test Version 15.9
© 2018 TestMy Net LLC - TestMy.net - Terms & Privacy
© 2018 TestMy Net LLC - TestMy.net - Forum - Terms & Privacy