SpeedTest Beta works on LTE

[flipteg]

i'm still getting very high uploads with this beta. with the old application, i was getting around 20-40mbps upload... now, i'm getting 10-14mbps.

 

[VideoEngineerAJS]

Its off on my home wifi, the router is right under me and its reading about 1/4 of the right speed on download, upload is fine.

I guess VZ has already throttled central Ohio 4.5 down and 2 up

Try a different server. I'd say see what the Chicago Comcast server gives you (it's the first of the 2 Chicago servers). That's the one I use and my fastest result has been...

41.79Mbps down
9.87Mbps up

And I seem to get over 30 just about every single time.

 

[robdec17]

Test Date: Apr 18, 2011 9:03:13 am
Connection Type: Lte
Server: Atlanta, GA
Download: 10.41 Mbps
Upload: 5.27 Mbps
Ping: 99 ms

 

[KarbonKopy]

Does anyone else to tries to hit he Los Angeles server get time outs and odd speed results? I swear this server is on the fritz.

 

[Dcpmark#AC]

Same here in LA. The app doesn't work well for me....either times out or gives really low results.

I just changed to the San Diego server....works perfectly.

 

[KarbonKopy]

Same here in LA. The app doesn't work well for me....either times out or gives really low results.

I just changed to the San Diego server....works perfectly.

Yeah thats what I did too, wasn't sure if it was just me for some reason. I sent an email to them hopefully they can get it figured out.

 

[lippstuh]

AGREED LA server doesn't work, SD server works. I'm in OC

 

[robrecht]

Are you using the same server on both? Servers do make a difference. Just because one server is closer doesn't mean it's going to be the fastest. I always get the best speeds using Chicago servers even though there is one in St. Louis.

Sent from my Thunderbolt

I've seen differences of nearly 10-fold between servers using the Speakeasy website a couple of weeks ago. Using the older SpeedTest app, I was also seeing differences of nearly the same magnitude from one moment to the next on the same server.

Is the differences between servers affected by volume of tests that they are running? Someone else mentioned that some servers may be defective.

Was I wrong to dismiss these tests as little more than a novelty?

 

[shaddix]

I would stick to desktop versions of speedtest on the tb still... Possibly the servers they use for phones are not as robust or do not have as phat a pipe going to them.

 

[FD2]

Does anyone else to tries to hit he Los Angeles server get time outs and odd speed results? I swear this server is on the fritz.

Same here in LA. The app doesn't work well for me....either times out or gives really low results.

I just changed to the San Diego server....works perfectly.

AGREED LA server doesn't work, SD server works. I'm in OC

Word. That LA server is on life support. I emailed the dev.

 

[robrecht]

For what it's worth, I just used the new app and got 89ms, 4424 kbps down and 7028 kbps up from the Sayerville, NJ server (I'm in central Jersey). Moments later, from the same server, I got 5756 down and 3196 up. Those are at least in the same order of magnitude.

But does it make sense to get a 30% faster download speed on the second test while the first test was 219% faster in upload speed?

For the 3rd test, I switched to a server in Washington, DC, and I got 74 ms, 24,479 kbps up, and 7,119 kbps down.

Greater than 7-fold faster upload speed between the 1st server and the 2nd server.

Should anyone be paying much attention to these tests?

 

[asianrage]

When I'm at work, I'm in the "center" of 4G @ 23-24Mbps Down & ~4Mbps Up
When I'm at home, I'm in the outskirt of 4G @ 6-8Mbps Down & 2-4Mbps Up. But since my FIOS connection at home via Wifi is about 20Mbps, that's what I end up using.

 

[FrankXS]

Should anyone be paying much attention to these tests?

All speed test websites have the potential to vary, test to test. Generally, when using any speedtest, you should believe whatever the highest reading is. Because, this shows what your phone/system is capable of under the correct Internet conditions, which can vary day to day and location to location across routers. After all, you don't want to test the overall Internet conditions, you want to test your own system.

Now I am in no way saying that some speedtest programs might not have programing errors that would result in inaccurate readings. Certainly, that is what was going on with speedtest.net mobile. I understand that the upload packet size in the mobile version was being swamped by the TBolts faster/larger packet size, causing erroneous upload speeds to be reported. But apparently that is fixed now. Looks like it to me anyway.

-Frank

 

[robrecht]

All speed test websites have the potential to vary, test to test. Generally, when using any speedtest, you should believe whatever the highest reading is. Because, this shows what your phone/system is capable of under the correct Internet conditions, which can vary day to day and location to location across routers. After all, you don't want to test the overall Internet conditions, you want to test your own system.

Now I am in no way saying that some speedtest programs might not have programing errors that would result in inaccurate readings. Certainly, that is what was going on with speedtest.net mobile. I understand that the upload packet size in the mobile version was being swamped by the TBolts faster/larger packet size, causing erroneous upload speeds to be reported. But apparently that is fixed now. Looks like it to me anyway.

-Frank

Fair enough, but keep in mind, I'm not talking about variations from place to place and day to day, but rather a tests done on the same server within moments of each other varying by more than two-fold, for example, just now:

5:18 pm: 8854 kbps down
5:18 pm: 4291 kbps down

Other times, I've seen even greater differences, nearly 10-fold, on tests done one right after the other on the same server.

 

[shaddix]

i'm still getting very high uploads with this beta. with the old application, i was getting around 20-40mbps upload... now, i'm getting 10-14mbps.

I'm envious