All of the Internet Service Providers (ISP) offer a wide range of different services in an attempt to gain and then keep customers. One service they provide that everyone uses on a daily basis is “fast” download speeds. I personally know of several people who changed services because they thought that it was taking too long to download something, anything really.
If you listen to all of the commercials, while some ISPs offer different cable channels and DVR options, the one thing that they all seem to focus on is Internet speed. The problem is that none seem to provide the type of download speeds they guarantee. You may actually hit that speed once or twice over the course of a week, but it is rarely if ever sustained.
Individually, the best way to find out your Internet speed is to benchmark it using an Internet speed test. Generally, this will give you a fairly accurate indication of how much bandwidth is available to you at the current time. Notice the phrase, “at the current time.” Internet speed tests are great for proving that you are, or are not, getting the bandwidth from your ISP that you're paying for.
According a recent article in the Wall Street Journal, some ISPs are delivering download speeds up to 41 percent slower than they advertise. Of course, if you listen to all the commercials discussed above, this is something that is to be expected. So while you could use a speed test to determine if your ISP is leading you on, the problem is finding a network speed test that is reliable.
Last week, CNET released some information it found by running several speed tests on a variety of different speed calculators. The problem was that each program delivered different results with some of the variations being as great as a factor of 10. The range of speeds was anywhere between the turtle’s pace of 783Kbps to the rabbit speed of 11.237Mbps. This discrepancy came about using DSLReport.com’s Flash-based test.
The highest download speed test result came in at 13.06Mbps. This was accomplished using the HTML5-based test at SpeedOf.me. To show you the reliability of each of these programs, the same test generated a download speed of 4.87Mbps on the same connection two days later. Obviously day and time become a factor when trying to figure out download speeds.
Dennis O’Reilly, at CNET conducted the speed test and he states that his ISP promises download speeds up to 12Mbps. The tests were run in both Firefox and Google (News - Alert) Chrome on a Windows 8.1 laptop. As you can see, consistency in the testing method was a factor, however when DSLReports.com was used over 10 times, there were almost as many results.
This does bring up the question of whether or not the type of speed test makes a difference. It seems that a lot of the experts are claiming HTML5-based speed tests are more accurate than tests that use Java and Adobe Flash. Other experts point out multithread tests, such as those used by Speedtest.net’s Ookla (News - Alert) don't represent real-world network traffic as well as single-thread tests.
From all of the results, it appears that the most consistent test results were recorded at Speakeasy (News - Alert)'s Flash-based Speed Test and at TestMy.net's HTML5-based tester. Each test was conducted over several days at various times. All of the test results came in at slightly over 11Mbps.
There's been quite a bit of debate recently about Speedtest.net (News - Alert), Speedof.me, as well as other testing sites. Questions are raised when speeds are slow on one but not the other. Is this because the server itself is overloaded? Is multithreading masking slow speeds? These are just some of the questions that the test results cause.
As you can see, the results of the tests conducted by CNET offer varying results based on a lot of factors. One of the biggest complaints I hear is Internet speeds are much faster during the day and slower at night. Is this due to the fact that more people of downloading data and movies when they get home from work, or is it that ISP are slowing the downloads themselves?
HTML5-based speed tests such as those offered by SpeedOf.me and TestMy.net seem to have an advantage because they do not require any additional software. However, regardless of the installation process, if you think that your download speeds are slower than you expect, it would probably be a good idea to download a network speed test application.
While it would be nice to tell you which one to download and use, most of the tests conducted come down to whether or not any of the speed tests give a true representation of and ISP’s download speed. There are a lot of factors that come into play and I’m not sure there is a speed calculator that can take everything into consideration.
Edited by Maurice Nagle