Download speed

Ron hath wroth:

Sure. First, you're mixing bits and bytes. If you absolutely must abrev. please use "b = bits" and "B= Bytes". Your test speeds are probably 195KBytes/sec and 320KBytes/sec respectively. That's equal to 1.56Mbits/sec and 2.56Mbits/sec.

Looking at your NNTP, I notice you're using Thunderbird. That implies that you're also using Firefox for downloads. Firefox has a nasty bad habit. It starts downloading immediately when you click on a file to download. However, the indicated Bytes/sec starts when you click on where to save the file. This results in unusually high indicated download speeds, depending on how quickly you decide where to save the file.

Also, I would not expect downloads from the MS site to run at the full speed of your broadband connection. If you're going to use downloading as a benchmark, use a file located on your ISP's servers, with the fewest number of hops and delays between the server and your computah.

If this is not what's happening, kindly disclose how you're measuring these speeds. Also, try one of the benchmarking sites to see if there are any problems:

Also, verify that you have the latest Proset drivers and utilities for your Intel 2200BG card. |

Thanks for your reply, Jeff. Sorry for the (bits and bytes) mix-up. Yes, you're correct, the downstream speed is equal to 1.56Mbps and 2.56Mbps which is right about my broadband advertised speed (I know I would never get 3Mbps downstream speed - 2.5 is about right).

Thanks again for that explanation. I do notice difference between IE and firefox transfer rate. Yes, firefox does indicate higher transfer rate than IE. I use both Firefox and IE (more firefox than IE). The download speed posted earlier is taken from IE "Transfer Rate" (IIRC) statistics, not from Firefox.

I do get greater than 300Kbps transfer rate with IE on my desktop PC downloading WinXP SP2. My notebook PC, however, maxes out to somewhere in the 190Kbps neighborhood. No, I don't download on both (desktop and laptop) at the same time. I test each one at different time, a few minutes apart. Tested it multiple times, I still get the same result... fast DL on desktop and slow DL on laptop.

I use

Some of my customers have have 3000/384 Kbit/sec DSL lines. I typically see speeds of about 2500 Kbit/sec using the ISP's speed test programs. I vaguely recall seeing 2700 once, but I don't recall what I was doing to get it.

If you need some entertainment, start a download with Firefox 1.5.0.6 and wait a while before hitting the "save" button. You'll get some really impressive (and impossible) initial download speeds. Eventually, if the file is large enough, it will produce realistic download speeds.

Is there anyone else using your wireless? (i.e neighbors, freeloaders, hackers, etc).

Dunno. I assume you have a nice strong signal to the laptop and that you're not rediculously close to the wireless router.

Try downloading a file from the desktop, through the wireless, and to the laptop. Do it in both directions. That will take the internet out of the picture and perhaps offer a few clues. Also, try dragging the laptop to another wireless router and see if the problem persists.

Is Speakeasy your DSL ISP? If yes, then it's probably the best test. If not, you might be going through too many hops. Try running traceroute to whatever machine is running their speedtest.

Well, if you were downloading the file from the same server as the speed test server, I would expect the download speeds to be identical. However, if your downloading is being done from a random server on the internet, there are many opertunities for either the server or the intermediate routers to slow you down. Most download sites limit the download speeds to some maximum. Same with the number of download connections per user IP address. With such a server, you would never even approach your maximum possible speed.

Try IPerf between your desktop and laptop: