Mods this may be IMHO thread, but I am really looking for factual information on how the system works and it’s flaws.
I have DSL service with Century Link. We had the 1.5mb service and our “special” of $29 expired. When calling to see what our options were we were told there was another “special” being run for $34 for 10mb service. Sounded good to me, so I signed up for a year contract.
The day came for the switch from 1.5 to the 10mb service. Quite honestly I really didn’t notice much speed difference, except that NetFlix streaming movies seemed to load faster and no buffering, so that is good. Wondering what speed I was getting I headed over to SpeakEasy to do a DSL speed test. I ran 4 or 5 test and the fastest speed I was able to show was 1.3mb on the download side. Hmmmm, no where near the 10mb advertised. I called Century Link and tell them what I have discovered. The rep asks me what speed test I used, then said to me “you need to test your speed using OUR test site”. He proceeded to give me the Century Link speed test address, I tested there and guess what? I am getting 8 to 9.75mb!
I hang up satisfied that I am getting the speed I am paying for, but then it hit me. BS, BS, BS … I go to an “independent” speed test site and am only getting 1.3mb, BUT if I go to their speed test site I am magically now pulling down close to the 10mb advertised. Something smells fishy.
So, the question is: Does anyone know if SpeakEasy is reliable for a true measurement of DSL speed? Does it sound fishy to anyone else that the provider says to get a “true” speed test one must use their site? Seems to me it wouldn’t be too difficult for them to fudge the numbers. Any thoughts would be appreciated.
Check Speedtest.net to get a reliable speed test. In fact, when I’ve been on the phone with AT&T, they’ve even had me use that site to verify my speed (after using a speed tester on their site). If you’re still getting slow speeds, I would call back again and tell them that you’re getting 1.3 (or whatever it comes back at) on multiple sites* and something still isn’t right.
If they won’t do anything for you, maybe you should drop back to their 1.5mb/s service. If you get 1.4 at, say 3 or 4 sites and 10 at their site, I’d say something is wrong with their speed test.
*Here are several more reliable testing sites you can use.
My first thought is that you might be confused by misleading units. Connection speed can be measured in megabits or megabytes (per second). Sometimes those are abbreviated in subtle ways: megabits/second might be shortened to “Mbit” or just “Mb”, while megabytes/second might be shortened to “MByte” or “MB”. Small b for bit, large B for Byte. One byte is eight bits (for these purposes), so if you’re getting 1.3 megabytes per second that’s entirely consistent with an advertised 10 megabit per second (1.3*8 = 10.4).
ISPs like to measure things in megabits per second, since that gives them a bigger number to flash around. But your computer measures things in bytes, so if you just measure how long it takes a file to download you’ll get megabytes per second.
Thanks for the info everyone. I am much more aware now of the bits/bites issue. I have re-run the test from several sites and am confident that I am getting the 10MB service promised. I can pull 8 to 9.5MB from several sites.
I hate to be so suspicious, but companies these days seem to have no problem being less than truthful with customers.
It’s good to be suspicious and test things independently.
I kept having to explain to the reps that I didn’t care that the ping test to my router from their equipment showed no issues - that the issues was with getting from that junction on our to the internet at large - their DNS routers, etc having issues. Raw speed means nothing if the routers/DNS/etc were always faulty - we ended up switching from their DNS to Google DNS and resolved much of our issues.
Then we switched back to Cable - where we have zero issues in comparison.
A big ISP corporation lie? Say it ain’t so, Joe! :eek:
Seriously, another trick they use is to quote the DOWNload bits/sec, and ignore the UPload rate. The ratio between down:up is typically 10:1, and they never mention the unimpressive low speed unless you ask for it (I’ve had salespersons who didn’t know what I was talking about).
It may not matter to the typical user, whose data is more down than up, but it matters to some – uploading to YouTube, for example.
I recall that when I called the phone company to tell them their name server wasn’t working, there was a long pause and then, “What’s a name server?” A few weeks later I switched to cable and have had no problem since.
The way I knew for sure it was a name server problem was that I know the actual IP address for my office machine and was able to sign on to that, no problem. Aside from that and one other I knew, I could not reach any site.
the speed test measured rates might indicate the data transfered though there is some amount of transfer overhead which is included it the total cited by the ISP.
Your DSL modem itself (or your router if you have one) may also keep its own statistics about data speeds, and there might be ways for you to look at that. You may be able to log in to your DSL modem just as if it were another host machine on your LAN, and look at its statistics.
Here’s one common way to do that: Go to your web browser’s address bar and type in: http://192.168.0.1 and see if you get anywhere with that. (But if that works, you might still have to type in a log-in name and password. Heaven forbid, you might have to RTFM.)
At least, this works with two different DSL modems I have here, from two different manufacturers, and I think this is common. The modem itself might have a sticker somewhere on it with instructions.