James Clark Posted December 11, 2012 CID Share Posted December 11, 2012 Hello testmy.net members.. I used to use testmy in the 90s and I decided to come back after my provider informed me that I was getting an upgrade to my DL speed. I tested before and after the "upgrade" but there was no change (about 7mb/s) However, if I go to microsoft or Adobe and download a large file (50MB+) I am getting around 16mb/s download... It seems as though the testmy.net site is not getting my full bandwith, or their servers are overstrained? Not sure why the tests are so far away from what my actual speeds are... Any ideas? Pekexolve 1 Quote Link to comment Share on other sites More sharing options...
CA3LE Posted December 11, 2012 CID Share Posted December 11, 2012 The file that you're getting from Microsoft and Adobe is no doubt being served via CDN (content delivery network) so the file is served from an optimal location. TestMy.net is also served from an optimal location, where a lot of the websites you visit are actually hosted. Unless you specify a location you're tested from Dallas, TX. But if you're anywhere in the US this is a great, centralized testing location. The amount of bandwidth going through Texas is the reason TMN has been hosted there since the early days. Where are you located? I see that you haven't tried any of the other testing locations. Go to the TestMy.net homepage and try changing the server at the top of the page. Due to distances your speed will be different to the various servers. A significantly slower speed to any of the servers usually points to a problem with the route to that server. If you have certain configuration issues you'll see the same slow speeds to all servers. Here's my last tests https://testmy.net/st...q=6436733977554 This is a pretty typical result from my Cox Communications 50/5 connection. Sometimes I'm faster, sometimes I'm slower. These results were pretty expected and didn't send off any alarms that I had any problems. Will I be able to download faster or slower than these speeds... yes and yes. Depends on the server I'm taking the file from... also depends on if that download is multithreaded. (meaning it's split and opens more than one connection to download the file, TMN is single thread.... a good connection on a properly configured computer will be able to max on a single thread and shouldn't need multiple threads to max the speed... if you only achieve your advertised speed by opening multiple connections, you may have a configuration issue. [most likely MTU] This could also point to limitations of the provider or the provider limiting the consumer. Multithreaded speed testing is another way I think that many of the provider hosted speed tests make customers connections look better than they are. They often open 4 to 8 connections, then add the results together. Yeah, you may be maxing out your line... but why do you have to open so many connections to do that?) First tests were taken with SmarTest sizing. Then I did 50MB download and 6MB upload to each of the servers. As you can see, my speeds to Europe are slower. Which is to be expected given the distance. Below you can see a result taken around the same time from Seattle WA, which is a greater distance than I am from Amsterdam. So it doesn't always have to do with distance. It's the quality of the connections that you're routed through along those distances. Testing your connection to a further location is actually more of a test than close by. It's tells a lot about who your provider peers through and who those peers in turn peer with. It's my belief that the further out you can test and maintain quality connection... the better the connection and ISP. The Internet is everywhere, not just near by... you want a connection that performs at distance. ... Depending on where you are, you also may be effected by the aftermath of Super Storm Sandy. A lot of lines were cut and a lot of traffic was rerouted and in turn congested some routes on the east coast USA. Test to the other servers and you can get a better idea if you just have a bad route to the main server. mudmanc4 1 Quote Link to comment Share on other sites More sharing options...
James Clark Posted April 5, 2013 Author CID Share Posted April 5, 2013 Sure I'll give it a try, I am in Syracuse NY... I guess I just assumed it would choose the server that had the best connection to and available bandwidth Quote Link to comment Share on other sites More sharing options...
James Clark Posted April 5, 2013 Author CID Share Posted April 5, 2013 I tried DC, but the results gave me a much lower speed. On the plus side DALLAS is now giving me 14Mbit, which is a lot closer to my actual speed Quote Link to comment Share on other sites More sharing options...
CA3LE Posted April 6, 2013 CID Share Posted April 6, 2013 Sure I'll give it a try, I am in Syracuse NY... I guess I just assumed it would choose the server that had the best connection to and available bandwidth It does choose the best server... my server is Texas is the most powerful. Many people assume that closer is better when in fact the opposite is true if you want a real test of your connection. You could be hitting bottlenecks or issues further in your route that will only be apparent if you test at distance. The best connections peer well across the Internet regardless of distance. I've seen 200+ Mbps across international waters and over distances of many thousands of miles on TestMy.net. You want a connection that performs well across the Internet not just near by. ... side track, in regards to international routes. People always blame poor performance on this factor. But it doesn't have to be the case if your provider is delivering. The reason many connections slow down over these long routes, especially transatlantic or transpacific routes, is because the lines are oversold. They don't leave enough available bandwidth to meet peek demand and some use packet shaping to limit throughput to any one user. Good providers use routes that have room available. If I can get these speeds at distance... anyone can. Your provider just has to give you what you're paying for. Comercial bandwidth in my experience performs much better over distance than residential. One of the points to this website is that I'm trying to help bring awareness to this and hopefully bring that same level of service to every consumer. I tried DC, but the results gave me a much lower speed. On the plus side DALLAS is now giving me 14Mbit, which is a lot closer to my actual speed How are your speeds downloading files across the Internet? Pretty inconsistant? Is your speed of browsing inconsistant too? Adobe and Microsoft downloads are delivered with CDN which means that it's pulled from a server near you. Most things you download aren't like that. 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.