How bad are wireless speeds?

I just got a di-524 dlink router, certainly not the cream of the crop but if it says 54mbs on the box I expect it to work that way. Anyways everything is setup in my room I have good connectivety, strong 54mbs connection to the wireless router. I have WEP enabled on the router too, but I run some file transfer tests.

On wired-wired my speeds where 80mbits solid, On wireless-wired speeds dropped to 11mbits (this would be 802.11b speeds right?) even though my connection was at 54mbs. I understand WEP can reduce speeds.. but this much? On wireless-wireless it got even worse. 5mbits!!

It would help if you told us what kind of client device you are using.

I am going to guess and say a USB dongle of some sort?

I ask because USB 1.1 is rated at 12mbps. So if you had some sort of wifi device plugged into a USB 1.1 port, you couldn't expect any more speed than 12mbps at most.

What are you using for client radios? How far away from the DI-524 in feet?

How are you measuring this? I suggest you try IPerf:

First a bit of explanation. Under ideal conditions, you'll get and maintain a 54Mbit/sec association up to about 8ft from the access point. Put a few reflectors and add some interference from other radios in the room, and the speed will drop rapidly. I scribbled a long and detailed explanation on speed adjustments effects yesterday. The discussion is on diversity reception, but the explanation involves speed reduction to deal with error rate reduction: |

The following was stolen from: | don't consider the numbers very accurate and are very optimistic. If you want speed, you loose range. Period. Incidentally, the above article is excellent reading on how Wi-Fi really works.

Rate Approximate Max Indoor Range

1 Mbps 350 ft. 2 Mbps 250 ft. 5.5 Mbps 180 ft. 6 Mbps 300 ft. 9 Mbps 250 ft. 11 Mbps 150 ft. 12 Mbps 200 ft. 18 Mbps 170 ft. 24 Mbps 140 ft. 36 Mbps 100 ft. 48 Mbps 95 ft. 54 Mbps 90 ft.

Also, 54MBits/sec is NOT the transfer speed you'll see with a benchmark program. It's the wireless connection or association speed. Transfer speed is roughly 33 to 50% of the connection speed. The best you can do with a TCP file transfer benchmark under Windoze at

54Mbits/sec is about 24Mbits/sec.

This is stolen from an Atheros PDF at:

ros_range_whitepaper.pdf with some additions and corrections by me.

Non-overlapping Modulation Max Max Max Channels ------- | Link TCP UDP | | | | |

802.11b 3 CCK 11 5.9 7.1 802.11g (with 802.11b) 3 OFDM/CCK 54 14.4 19.5 802.11g only 3 OFDM 54 24.4 30.5 802.11g turbo 1 OFDM 108 42.9 54.8 802.11a 13 OFDM 54 24.4 30.5 802.11a turbo 6 OFDM 108 42.9 54.8

Note that if you have 802.11b compatibilty mode enabled, you get to slow down even more. The receiver has to listen for 802.11b stations and simply stops everything while this is done. If it finds an

802.11b signal, it deals with it before returning to 802.11g file transfers. If you even have an 802.11b device anywhere in range, your 802.11g file transfer performance will suffer.

Finally, all the above is between a wired LAN computah, plugged into the wireless router, to a wireless computah. If you go wireless to wireless, divide your thruput by at least two. That's because the access point is simplex and can only transmit or receive one at a time. The access point is doing a store and forward on each packet between wireless clients.

I don't wanna speculate on the validity of your benchmark test without first having a clue as to your setup and testing methodology.

The "strong" refers to the signal strength

yes.. I thought that was what I was implying :)

various reasons. I'm aware I'm not going to get the maximum speed out of the network but when it's less than half the established speed the question needs to be asked. If I was only getting 20mbits/sec on file transfers on a wired connection I would try to find out why too.

Yes of course. The wireless network properties says it established a 54mbps, but my file transfers (throughput) only achieve 11mbps. How did I determine it? I transferred a file from one of my computers to another over the wireless. The server is on a wired link and the client is on wireless. I regularly transfer files through ftp from one computer to another.

right but it's only confirming to me my thoughput is still 11mbps when my connection to the router says 54mbps

fair enough, I was just wondering if it was WEP, but maybe I probably need to get an even faster wireless router :)

Johann Beretta wrote in news: snipped-for-privacy@4ax.com:

AND, any other USB devices plugged in will also affect the speed of a wireless USB since most PC's I've seen use one single interupt for all of the on-board USB devices. CPU time is then needed to process those other devices as well.

Most all in one motherboard combine the IRQ lines from video, ethernet, USB, firewire, and the kitchen sink, into one IRQ line. They can do this because NONE of these devices use the IRQ line for anything more than indicating an error condition or exception. The bad old daze of serial I/O and hard disk programmed I/O, where the IRQ line was thrashed for every character or block, are long gone. Same with the parallel port that can easily run without ever thrashing it's IRQ line. I've tested about 12 simultaneous USB devices on a single USB 2.0 hub without the slighted indication of any IRQ related performance hits. There was a problem when I attached a DVD burner and three wireless devices. However, it was the DVD burner that took the performance hit, not the wireless devices.

Incidentally, the maximum number of devices that can be attached are

127 USB devices.

Have you actually tried plugging in a mess of USB devices and seen any deterioration in wireless performance? I haven't.

There is a difference between clock rate and data rate. In a shared Ethernet system wireless or a HUB wired system you get around 30% throughout compared to clock speed. So 30% of 54M is around 15M.

When you go from client to client via wireless this will be halved since the first client sends the data to the Access point and then the access point sends the data to the second client.

Tony