New ISP, All the Difference - Page 2

Basic

Canada288 Posts

December 23 2014 21:07 GMT

#21

Here is the result I get if I use one of the more commonly used speed test websites. I do not have an advanced understanding of networking but I did it have it explained to me that most sites "game" the uploader giving you inaccurate results. I trusted that explanation as the person knows much more about networking than I do.

Regardless here is a speed test from speedtest.net, a site I was informed games results.

icydergosu

528 Posts

December 23 2014 21:31 GMT

#22

Since i do not know what the person actually said i will refrain from passing judgement. The only thing i can think of is that some shady websites might just fake a speedtest and not transfer any data at all to make the people feel good and make them return to the website but that logic seems pretty fucked up.

Regarding the speedtest.net testing method: They use several Download and Upload threads concurrently for a test to saturate your connection. Using multiple threads concurrently is not unusual since that is what is usually needed when using tcp to be able to make sure that your connection is fully saturated. Using a low latency connection like fiber and a server within your vicinity in canada should not require multiple download threads to max out your connection.

However testing the way speedtest.net currently does it masks providers which throttle download/upload threads to certain Mbps values. You should have no troubles maxing out your connection with only one download thread from a canadian server.

To test this you could download one of the files i linked and time how long it took to download completely. Chrome for example does not save that information(total time or average speed), i am not sure if there is an addon for chrome or other browsers. Otherwise you would probably need a download manager for windows which saves the average speed or total time.

It is important that you download the whole file since some ISPs seem to be masking their shady behaviour by bursting(letting your download thread run without a leash) from time to time. In this case the larger the file the better since even if they apply short periods of bursting the average download value will still show because we looked at a larger set of data.

Basic

Canada288 Posts

December 23 2014 22:06 GMT

#23

It took roughly thirty seconds, which if my math does not fail me equals out to 35 Mbps.

For the records there are two high speed internet providers in Saskatchewan so options are slim and I simply happy this provider tries to provide me with as optimal results as possible.

Fuchsteufelswild

Australia2028 Posts

December 24 2014 02:46 GMT

#24

Errm, unless I'm missing something, you didn't download the full 1GB file.
Maybe my brain's gone silly but 30 seconds at 35Mbps (Megabits) would be 1050 Mb (Megabits) downloaded, which is 131.25 MB (Megabytes).
To get 1000 MB (1GB) in thirty seconds would be 1000÷30=33.33 MB/s or 266.66Mb/s.
266.66 ≠ 35 and I doubt you have 266.66Mb/s if they tell you you're getting (up to) "50Mb/s".