TPG Community

Get online support

FTTB super slow. TPG speedtest exaggerates actual download speed.

SOLVED Go to solution
ronichol
Level 3

Checking my speed to today on FTTB plan. I usually get 90-100 mbps dowload and 40 mbps up.

 

TPG custom speedtest says I am now getting 27.5 mbps download, 35 mbps up.15ms ping.

 

When I compare this to speedof.mew, I get 9.0 mbps down, 36 mbps up, 20ms latency. (Thi sis the BEST result I have got today, typically it is 5mbps down)

 

I have redone the test 3 times and TPG always exaggerates the download by 3 times.

 

These are the times I get plugged directly into the modem.

 

Obviously this sucks for fibre to the bulding.

 

Any suggestions?

1 ACCEPTED SOLUTION

Accepted Solutions
ronichol
Level 3
Hi Basil,

After about 4 hours of experimentation I worked out that the only way to get the Nest to work was to turn off both 2.4 and 5ghz wireless on the TP link router so that it would revert to being just a modem, then connect to Nest router via one of the 4 LAN ports. Then following instructions in Google home, including rebooting the TP Link when instructed.

It now works brilliantly. I have a much stronger signal throughout both floors of my townhouse.

Interestingly I am now getting close to full speed with speedof.me at aroung 80 mbps whuch before was hitting 18 mbps on a good day.

Very happy with the new setup.

Thanks for your assistance.

View solution in original post

29 REPLIES 29
BasilDV
Moderator

Hi @ronichol 

 

Thanks for raising this with us.

 

The speed test results that you've posted, were they taken via a WiFi or a Wired connection?

 

If via WiFi, can you try to run the test via Wired to compare?

 

I've tested the FTTB line and was not able to detect any fault. The modem is getting a good sync speed with 99.99 Mbps Downstream and 40.9 Mbps Upstream, so we need to check this further.

 

Regards,

BasilDV

ronichol
Level 3

They are from a wired connection. I get the same result whether I am plugged directly into the modem, wirelessly connected or on IOP (internet over power).

 

For the record, today I have again run the TPG custom speed test side by side against speedof.me.

 

Before I say the result, I am going to add that speedof.me is widely accepted as the BEST indepndent way of measuring actual internet performance. It works by cycling through larger and larger downloads/uploads.

 

I have now seen consistently, for several weeks, that the very small downloads speedof.me does initially are fast. The larger downloads get progressively slower.

 

I will now add some of my own editorialising - I have to say this looks like TPG has some kind of throttling of moderate to large downloads (but NOT uploads). it is VERY dramatic.

 

Now, to the latest results.

 

speedof.me:

9.12 mbps download (spiked to 38.9 mbps very early then rapidly dropped)

33.1 mbps upload (progressively got better as the test proceeded).

 

http://tpg.speedtestcustom.com:

84.7 mbpds dowonload (continued to ramp UP the longer the test proceeded)

38.0 mbps upload (hit early and stayed high)

 

Subjectively, the performance is even worse than this. Sometimes it is unusable and slower than dial up used to be (yes, I am old enough to have actually had a dial up connection over 56kb modem).

 

I think those results speak for themselves. I am not prepared to let this slide. There will be serious repercussions for TPG if you do not directly address this CHASM in result differences between your (bogus?) custom speed test and a widely accepted independent speed test. I have to say on the face of it, this looks like deliberate deception, but for now I will give you the opportunity to rectify this issue, preferabley once and for all.

 

 

Anonymous
Not applicable

Hi @ronichol,

 

Thanks for the additional details. If the connection speed you're getting using our speed test website is different to the speedof.me. we can do additional tests to determine the speed of the service.

 

Note: The host location of speed test website will affect the test result when you run a test.

That's why there are some speed test website made just like ours ( https://tpg.speedtest.net/ ) this is to ensure the accuracy of the test.

 

Also as mentioned above by @BasilDV the other way to determine the attainable speed of the service is by checking the actual 'Sync speed' of the modem/router.

Note: Sync speed, also known as the line speed, this is the maximum speed at which the router/modem can connect to the internet / the actual line.

 

For testing purposes and comparative testing.

 

Please post the following

 

 

ronichol
Level 3

Dear moderator,

 

speeof.me is widely accepted as the best and most reliable independent speed test.

 

As of right now, speedofme shows that I am gettign 2.1 mbps download. That's right. TWO. POINT. ONE.

 

Speedofme when using 4G on my mobile phone reports 118 MBPS. About *60* times faster!

 

Speedtest.net repprts 78.55 mbps download and your own http://tpg.speedtestcustom.com/ reports 86.5 mbps.

 

NOTE: speedofme works by downloading files of size which double in each attempt. Thi sis quite a different approach than other sites such as speedtest which do not test download of files of a realistic size. speedofme is vastly superior when attempting to test streaming speeds (for example).

 

I have been using google chromecast for streaming services. Images were buffering, breaking up and repeating, because the speed is soooo slooow, however now, even though my mobile phone says I am casting successfully, no casting happens and instead the streaming service just plays on my (or my partners) mobile phone.

 

I have a Huawei HG659 modem. I would happily bin this piece of junk and replace with something I have bought myself which works, even if it costs $500, however this is a TPG supplied and supported bargain basement rubbish modem, so I need to know whatever I replace it with is supported and configured properly. Also I have no way of knowing if this will even fix the problem.

 

All I know is, at the moment the most widely accepted independent speed test (which the NSW public service IT department uses) says I am getting 2.1 mbps while your speed test site says I am getting over 80 mbps. I also know from my own PERSONAL EXPERIENCE, I am simply *NOT* getting ANYTHING LIKE 10 mbps - LET ALONE 86.

 

I am very close to dumping TPG for good after having been a customer for more than 10 years.

 

Please come up with a legitimate (rather than canned) response to explain the difference. So far all you have done is respond the same way you do to every other customer that has problems like this - pretend there is nothing wrong simply because you can't detect any problems on your end.

 

This is simply not good enough!

ronichol
Level 3

Can you also please tell me how to check the synch speed of an HG659 router? I do not see this anywhere on the routers admin website.

Anonymous
Not applicable

Hi @ronichol,

 

Testing a different speed test website such as (www.speedtest.net) will help to determine the which test result is accurate. At the moment we do not recommend you to buy a different modem/router, we need to determine which test is accurate.

 

We'd like to arrange one of our Technicians to contact you for real-time testing and investigation. PM us your best contact number and preferred time.

 

In case you need a reference: How do I private message (PM) in the community

 

Cheers!

ronichol
Level 3

Shane I already explained that speedtest gives much better results than either speedofme or my own observation, which is that regular streaming services, pixelate, buffer, or do not even manage to connect.

 

The difference is that speedofme downloads files of size which progressively double in size using a single thread. The initial very small files give much better speed than the final ones (it starts at around 10-15mbps, which although still very slow, might actually be usable if it was maintained).

 

I will PM you but I don't appreciate the insinuation that speedof.me might be suspect. It isn't. I think your customers and potential customers need to know that.  Here is an explanation of why it is actually superior:

 

https://speedofme.zendesk.com/hc/en-us/articles/223659007-Why-are-SpeedOf-Me-test-results-different-...

Anonymous
Not applicable

Hi @ronichol,

 

I understand, but it is better for us to perform a real-time testing and investigation to understand what's causing slow connection rather than using different speed test result as a basis of the actual speed of the service.
 

I haven't received any private message from you, feel free to send me your best contact number and preferred time so we can arrange one of our Technicians to call you immediately.

 

Regards,

ronichol
Level 3

Just to provide an update, a tech did try to contact me but only when I was at work (I am only at home during business hours about one day per fortnight).

The next time I might have been able to call back, I did not have the problem. This problem is intermittent but frequent. I just can't guarantee that it will happen when it is possible for me to talk to a tech.

Haing said that, the last time I was able to test this really thoroughly I was getting 1MBPS wirelessly less than 50 cm away from the router. I plugged in directly by ethernet cable and got over 100 mbps. I then powered the modem off and on, and I started getting close to 60mbps wirelessly.

 

This problem happens far too often for that to be an adequate solution. I suspect something like a Google Nest might solve this problem (ie. using the Huawei as a modem only). That is probably what I will try next (after a couple more tests the next time the problem occurs), however I am still somewhat concerned that this might not be the whole problem.