why do i speedtest 5x faster when hooked up via cable, than w/ 54mpbs wireless?

[theNEOone]

i don't get it....

my d/l speeds are near 6mb/s when directly connected via cable to the router (not directly to the modem), but only 1mb/s over wireless. i've got a "very good" connection, and i'm on wireless g.

what's going on?

comcast provider. netgear wgr614 v6. wpa-psk encryption.


=|

 

[jlazzaro]

even on wireless g, one can only expect 2 - 3 MB/sec. are you getting much interference? what is your signal ratio? try switching your AP to channel 1, 6, or 11.

 

[theNEOone]

AP?


=|

 

[jlazzaro]

Access Point aka your wireless router.

 

[JackMDS]

Link to: "Speed" (Bandwidth) expectation of Ethernet Home Networks using Windows 98/2000/XP.

 

[kevnich2]

That's not totally true. I have a buffalo WHR-54GS, on a wired PC, I get around 6.7mb through speedtest, on my wireless connection about 3 rooms down (still excellent connection through three walls BTW), I get between 4-5mb, but it has NEVER hit the same speed as my wired PC, but it's pretty close. I'm satisfied with it anyway. I'd say yes it's with your wireless router/AP, but not ALL wireless AP's will give that low of speed.

 

[RedWolf]

I have a netgear wgr614 v4 and I have the same exact problem. My speed degrades over time and I have to reboot the router to get it back up. It has been worse the last few weeks and was bad enough that I did some testing. My best speeds from wireless are 2mbps while the wired computer gets 6mbps. Upload speeds are around 300kbps for both wired and wireless.

I would reboot your router, update the firmware, clear the settings, reenter your information manually, and then reboot the router again. I would also look at the positioning of the router. If it is near other electrical wires you could be getting interference. Try putting it on it's own power (separate from other things).

 

[Madwand1]

Don't confuse "b" and "B". "B" = byte = 8 bits. Typical wireless can do 15-22 Mb/s max, which is much higher than typical home internet, so should not in itself be a bottleneck with a good connection, etc. You'd need to run diagnostics to see the actual local speed of your wireless connection.

E.g. using iperf 1.7:

server: iperf -s
client: iperf -c server -l 64k -t 12 -i 3 -r

Here are some of my sample results:

F:\tools\bench\iperf>iperf -c 192.168.0.144 -l 64k -t 12 -i 3 -r
------------------------------------------------------------
Server listening on TCP port 5001
TCP window size: 8.00 KByte (default)
------------------------------------------------------------
------------------------------------------------------------
Client connecting to 192.168.0.144, TCP port 5001
TCP window size: 8.00 KByte (default)
------------------------------------------------------------
[812] local 192.168.0.141 port 31062 connected with 192.168.0.144 port 5001
[ ID] Interval Transfer Bandwidth
[812] 0.0- 3.0 sec 5.00 MBytes 14.0 Mbits/sec
[812] 3.0- 6.0 sec 6.06 MBytes 17.0 Mbits/sec
[812] 6.0- 9.0 sec 6.06 MBytes 17.0 Mbits/sec
[812] 9.0-12.0 sec 6.06 MBytes 17.0 Mbits/sec
[812] 0.0-12.1 sec 23.3 MBytes 16.1 Mbits/sec
[788] local 192.168.0.141 port 5001 connected with 192.168.0.144 port 1732
[ ID] Interval Transfer Bandwidth
[788] 0.0- 3.0 sec 5.88 MBytes 16.4 Mbits/sec
[788] 3.0- 6.0 sec 5.75 MBytes 16.1 Mbits/sec
[788] 6.0- 9.0 sec 5.87 MBytes 16.4 Mbits/sec
[788] 9.0-12.0 sec 5.69 MBytes 15.9 Mbits/sec
[788] 0.0-12.1 sec 23.3 MBytes 16.2 Mbits/sec

This is across a wireless 802.11g bridge; short to moderate distance; a couple of obstructions... a pretty good result for the equipment and environment.

I can remote desktop from a wired connection into a computer which is connected using this wireless 802.11g; run a web benchmark going back through the wireless bridge to the internet, and get essentially the same internet benchmark as I get using a wired connection to the internet. In my case, the service is stated 5 Mb/s, and effective around 4.3 Mb/s max using those benchmarks.

In fact, at this time, the fastest benchmark result I've gotten was through the wireless bridge in this manner. But this just shows that such benchmark results can vary significantly due to external factors, so I'll just say that the speeds are essentially the same, not that my wireless is faster than wired for internet benchmarks.

There are some tweak that can be used to improve wireless Internet results (before Vista, which does such tuning automatically). E.g. TCPWindowSize and TCP1323Opts as in the following guide:

http://www.speedguide.net/read_articles.php?id=157

A reboot is required for such settings to take effect.