jimerman Posted October 23, 2015 CID Share Posted October 23, 2015 Ever since the first week in October, TestMy.Net has shown a very large (slower) result than SpeedOf.Me and SpeedTest.net. For example, the other 2 are showing 85-90 Mbps Download, while TestMy.Net is showing 25 Mbps. I turned on Multithreading, and they are running from the same browsers. I tried Internet Explorer 10 and Firefox from Windows 7, plus Safari 9 and 8 from OS X 11.11 and 11.10. Are you aware of any system issues with your infrastructure that would cause that problem? I know, for example, that SpeedTest has local servers and discards a disproportionate portion of the test results, so their report is faster than reality. However, it should not be 300%+ discrepancy. And SpeedOf.Me is similar to TestMyNet in methodology and infrastructure. Quote Link to comment Share on other sites More sharing options...
CA3LE Posted October 23, 2015 CID Share Posted October 23, 2015 So check this out... https://testmy.net/db/-SMrztfuy See all of the locations you're testing against. uk, de, cn and au are bringing your speed down for sure because of the distance and international routing involved. Your multithread results before looked like... https://testmy.net/db/Aa1YNWoSI If it just says testmy.net then it gave you a mix of Dallas TX (4 servers), San Jose CA and Washington DC. (outside of the US the mix changes) I think that if you change to more favorable options your speed will pick back up. Try some tests with multithread disabled too. Quote Link to comment Share on other sites More sharing options...
CA3LE Posted October 23, 2015 CID Share Posted October 23, 2015 And SpeedOf.Me is similar to TestMyNet in methodology and infrastructure. Not really. Our tests work totally different. Quote Link to comment Share on other sites More sharing options...
jimerman Posted October 23, 2015 Author CID Share Posted October 23, 2015 (edited) Thanks, that was it! I didn't realize that all those locations were checked, and that meant it was testing them all. That explains it. I have to still disagree with you about single/multi thread, I think multi thread is the best way to test and tell what my pipe is allowing. I do understand there are technical differences in the different testing web sites, but at a gross level, at least you return stats on all the packets (rather than Ookla). And, you use PoP servers, instead of geographically-close servers. Edited October 23, 2015 by jimerman Additional similarity Quote Link to comment Share on other sites More sharing options...
CA3LE Posted October 23, 2015 CID Share Posted October 23, 2015 Sweet! Glad it's working for you. Curious, what do you get in single thread? Quote Link to comment Share on other sites More sharing options...
jimerman Posted October 25, 2015 Author CID Share Posted October 25, 2015 Single thread shows 77 down, multi shows 79.9 down from the same browser at similar moments in time. But apps and browsers are typically multithreaded, and that threading utilizes the bandwidth more fully, testing its peak performance. CA3LE 1 Quote Link to comment Share on other sites More sharing options...
CA3LE Posted October 25, 2015 CID Share Posted October 25, 2015 Single thread shows 77 down, multi shows 79.9 down from the same browser at similar moments in time. But apps and browsers are typically multithreaded, and that threading utilizes the bandwidth more fully, testing its peak performance. I just wanted to make sure your numbers were comparable between the two tests. Slightly higher one way or another is okay. It raises an alarm if one is much faster than the other. Yours looks great. Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.