«
»

iPhone 3G Tethering Speed vs. Roger’s Portable Internet Stick

The latest download 3.0 software from Apple gives you the ability for Tethering your internet connect through your iphone to your computer. Here is the summary from the Apple web site, “If you travel with a PC laptop or Mac notebook, you know it can be difficult to find a Wi-Fi hotspot. With Internet tethering on iPhone, you don’t need one. Now you can share the 3G connection on your iPhone with your computer and connect to the Internet anywhere. When your iPhone is tethered, you can still send and receive data and make phone calls.” Link to Tethering on Apple website.

Since I have both a portable Internet stick from Rogers (Novatel Wireless Ovation MC9500D) I wanted to conduct a test to compare the upload and download speeds using my iPhone 3G and my Internet stick. Since both devices are using the 3G network from Rogers, my assumption that the speed tests should be fairly close. For the speedtest I used the Shaw Speed Test tool, available here.

I started by conducting a speed test using my wireless connection running Shaw’s High-Speed Internet. Here are the results of that speed test.

Shaw Internet Speed Test


Here are the results from the Portable Internet Stick.

Roger's High Speed Portable Internet Stick


Here are the results from using my iPhone 3G connected with a USB cable.

iPhone 3G Tethering USB Cable


Here are the results from using my iPhone 3G connection via Bluetooth.

iPhone 3G Tethering Bluetooth

As you can see from the results above, I was quite surprised at the lack of performance when comparing the iPhone 3G with the Portable Internet Stick. In both the download and the upload speeds, the portable internet stick clearly out performs tethering with the iPhone. This is interesting considering both where using the Roger’s 3G network.

Based on this test my conclusion; in a pinch tethering using your phone for quick connections would be completely fine, however, if you are a power portable user I would not rely solely on tethering as the speed is simply not there.

NOTE: this test was performed in the order listed above on July 3, 2009 between 11:30-11:50am MST.

Popularity: 22% [?]

Tags: , , ,

This entry was posted on Saturday, July 4th, 2009 at 12:22 pm and is filed under iphone. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.

  • I've always thought of it as low end DSL, like the DSL that might of appeared in rural areas 10 years ago. They really hype up the 3G speeds in advertising but really it's only good enough for email, weather updates, and some browsing of small web pages on a mobile phone. I've had 3G with both AT&T and Verizon national networks in the US, and it's just as slow as you say it is.
  • Ross
    I just did a small test using tethering over USB and Bluetooth via the Fido/Rogers 3G and an iPhone 3GS, and I derived some better throughput than your Shaw Speed test, but still not quite as good as the Stick.

    The speed estimates were derived from SpeedTest.net. According to their servers, they estimated that the closest test server was in Toronto, ON. I am physically situated in Saskatoon, SK. Each test was a one-off, althought doing multiple tests per server for an average would have been better.

    iPhone 3GS tethered to Fido 3G via USB

    Toronto, ON Server: 214ms ping, 2.06Mb/s down, 280Kb/s up
    Calgary, AB Server: 261ms ping, 750Kb/s down, 270Kb/s up
    Brandon, MB Server: 464ms ping, 3.96Mb/s down, 190Kb/s up
    Montreal, QC Server: 214ms ping, 3.38Mb/s down, 320Kb/s up

    iPhone 3GS tethered to Fido 3G via Bluetooth

    Toronto, ON Server: 260ms ping, 1.32Mb/s down, 170Kb/s up
    Calgary, AB Server: 292ms ping, 1.24Mb/s down, 300Kb/s up
    Brandon, MB Server: 526ms ping, 1.59Mb/s down, 230Kb/s up
    Montreal, QC Server: 245ms ping, 1.59Mb/s down, 100Kb/s up
  • salman
    thanks for this useful analysis...how is ur 3g speed on the iphone itself?
  • "Salman" - I actually find the speed on my iPhone 3Gs not to bad overall. What I have noticed is that occasionally it takes quite a long time to activate itself, but once it gets going seems okay.
  • "Ross" - thanks for posting your numbers
  • Similar results from downtown Toronto, generally about 1.2Mb/s - far from the advertised speed.

    iPhone 3GS tethered to Fido 3G via USB

    Toronto, ON Server: 288ms ping, 1.09Mb/s down, 300Kb/s up
    Calgary, AB Server: 180ms ping, .980Kb/s down, 310Kb/s up
    Brandon, MB Server: 212ms ping, .880Mb/s down, 270Kb/s up
    Montreal, QC Server: 142ms ping, 1.79Mb/s down, 310Kb/s up
  • On device from downtown Toronto, generally about 2.00Mb/s - better than tethered

    iPhone 3GS (Fido) untethered . . .

    Toronto, ON Server: 288ms ping, 2.32Mb/s down, 134Kb/s up
    Montreal, QC Server: 180ms ping, 3.83Kb/s down, 85Kb/s up
    Institute, WV Server: 2674ms ping, 0.525Mb/s down, 54Kb/s up
    Kitchener, ON Server: 239ms ping, 0.581Mb/s down, 158Kb/s up
    Kalamazoo, MI Server: 313ms ping, 2.78Mb/s down, 90Kb/s up
  • "Idealpragmatist" thanks for the numbers
  • I think the main problem with Tethering or USB Stick is latency. Local within Rogers network the latency is in 100ms+ , to outside servers, the latency goes as high as 500ms+ sometimes reaching over 1000ms to servers in Texas/California.

    My information:
    iPhone 3GS v3.1 USB Tethering
    iTunes 9.0 latest
    Location: Ottawa, ON

    I did two tests:

    1) SpeedTets.Net

    Closest was Montreal with results:
    Download: 3.55Mbps
    Upload: 300Kbps
    Ping: 146ms

    2nd Location Toronto with results:
    Download: 3.53Mbps
    Upload: 230Kbps
    Ping: 302ms

    2) Second test I did using the Shaw test site with results:
    Download: 3.3Mbps
    Upload: 256Kbps
    Ping: 544ms


    I don't have USB stick to try, but had tested a Bell Stick in the past, their upload speed was a bit higher, but the main difference was Latency (Ping time).

    I think Tethering is good, I use it often for remote connections, SSH, VPN. But the latency kills you when you're in a hurry. I also noticed 3G network is slow in downtown between high rise offices.

    I also tried Roaming Tethering on AT&T in the US (Boston, MA). Speed seemed to be better than Rogers even though I heard lots of people complain about it.
  • Tamouh,

    Thanks for sending over the numbers. I agree with you - in a pinch Tethering can help you get online if required, but I would not rely on it as the sole traveling service for connecting to the web - just to slow. For me I use extensively the Portable Internet Stick when traveling in Canada and try to find some good WIFI locations when down in the US. I have not tried Tethering using Roaming in the US - will give that a shot next and record the speeds when I am down there.
  • @idealpragmatist - in Calgary, I can only dream of those speeds!

    Since Rogers introduced HSPA+ in Calgary (or at least started marketing it) I have seen a rapid degradation of speeds/reliability on my 3GS.

    Downtown, I often find myself in blackspots with inexplicable fluctuations between full strength 3g and patchy EVDO. There seems to be no obvious environmental cause - obviously you can't expect much in an elevator or basement - but even in open areas, the service is often stretched to breaking point.

    Tethering is great in principle and at certain times of day, if you are very lucky, speeds are similar to 1st generation broadband from 1999. Whenever I try to tether for a client meeting or informal demo, the tethering has been next to useless.

    Just for info, my last 3 speed tests as follows (untethered):

    9/18 - Calgary, AB server - PIng 2403 ms Down 516 kbps Up 212 kbps
    9/21 - Calgary, AB server - Ping 2409 ms Down 393 kbps Up 122 kbps
    9/24 - Calgary, AB server - Ping 416 ms Down 415 kbps Up 91 kbps

    So these speeds from a supposedly 7.2 Mbps network suggest that the new HSPA+ will yield maybe 1.5 Mbps on a good day :P There is always a gap between advertised and actual real world speeds but Rogers' 3G, in Calgary, seems to be monumentally poor. Perhaps by the end of my ridonculous 3 year term, the speed might be close to what was promoted when I signed it.
  • paul
    you guys should conduct tests using PDANet on an ad-hoc wifi connection.

    Ive heard USB tethering introduces much bottle-neck; i can reach speeds on speednet.net of up to 2.75mbps on my computer using this method. ive never tried the 'standard' usb tethering method.

    PDANet requires jailbreaking.
blog comments powered by Disqus