Why would one of my hardwired PCs has faster intenet speed than wireless PC on same network?

Ok, here is more information. I just unplugged the desktop,a nd plugged in my laptop. When plugged in, the laptop also gets 6mbs, while only 2mbs on the wireless.   I tend to agree with your comment, except it seems like such a dramatic difference on teh download speed.  
Upload speeds on hardwired (desktop or laptop) are about the same at 1.6-1.7 mbs.
Download speed is what is really 1/3 speed on laptop when in wireless mode.  

I also just tested on both machines with FF3 and IE7, all speed results the same as before on either browser.

I am using this speed tester: http://www.internetfrog.com/mypc/speedtest/

What wireless connection protocol are you using?

For example on .11g your absolute best rate is 54mbps while the minute you connect a wire you instantly have 100mbps.

To achive the best wireless speed you need to have an unobstructed access to the router's reciever. If you want to see how you are doing look at the connection speed in either Network Connections - Properties or if you are using a third party wireless manager using their utliity. You will see quite quickly the speed will alter simply with proximity to the access point.

Add to this potential interference with the signal so that data has to be sent twice and encryption and you can see that there simply is no contest between the two.

With Gigabit Ethernet arrivng the difference even with Wireless -N is about to become more apparent.

http://www.cnet.com.au/wireless/routers/0,239028894,339273383,00.htm

Wireless also has more overhead.

Encryption methods, mac filtering, additional bindings in the wifi properties etc....

WIFI is at 54mb/s, and your hard wire is at 100mb/s......

I wouldnt think that you would get 6mb/s on the wifi....

What kind of router? In Linksys routers, there is an option in the Advanced Wireless for Connection types, being Automatic, or you can force 54mb/s.

Try stripping out the encryption, and mac filtering if present.

Also...

http://www.netstumbler.com

Check your environment for other wifi networks, and see if you can isolate a good channel for your use....

Hello jjltpc,

Aren't you making the mistake of comparing Mbits v.s. Mbytes/sec?

As far as I could find out, FiOS gives you 50 Mbits/second (which is around 6 Mbytes/sec) which is about the speed you get when you connect through ethernet.
A wireless G connection is limited at around 20 Mbits/second (54 Mbits/sec - 50% overhead - interference) which is around 2.5 Mbytes/sec in an optimum situation. So the 2 Mbytes/sec you're getting is actually pretty good.

Regards,

LucF

LucF said:
>> A wireless G connection is limited at around 20 Mbits/second

Close... its theoretical limit is about 27Mbps, and WPA2 adds zero payload overhead.
http://www.oreillynet.com/lpt/a/4085

I think you may be on the right track if the connection speed is actually 50Mbps, though.

jjltpc,
Try the speed tests from a couple different sites at speedguide.net - http://www.speedguide.net/speedtest/
After the test is done that site gives you a URL below the results that lets you go back and review the results, or you could just copy/paste some of the results to a message here with your IP xxx'd oiut if you prefer.

The image files the URLs reference don't have the IP in them anyway.

Darr247,

>>Close... its theoretical limit is about 27Mbps, and WPA2 adds zero payload overhead.

I'd appreciate it if you'd take my full comment instead of breaking a line in the middle...
A wireless G connection is limited at around 20 Mbits/second (54 Mbits/sec - 50% overhead - interference)

54 Mbits/sec - 50% overhead = 27 Mbits/sec. Then minus interference will leave in terrific conditions 20Mbits/sec effectively.

WPA doesn't add any "extra" overhead if the hardware can handle it (especially with TKIP this can cause issues on slower hardware/computers), the 802.11x overhead however still applies.

Thanks for your addition though.

LucF

Hi. I am adding some more info and I appreciate the ideas so far. They made me think about what is theoretical vs actual, but I still feel that a difference of 4 mbps between a wired connection and wireless seems excessive.

My plan is the basic FIOS plan, with following specs:
Download up to 10 Mbps / Upload up to 2 Mbps

I am reading that as mega bits per second.... and since Verizon says "up to" on both, they are obviously theoretical.

So my actual test results for hardwired are close enough on both to say I am getting a reasonable return on my investment in FIOS (actually it has been extremely reliable for past several years )

Wireless, let's say 6 Mbps download - 50% overhead = about 3 Mbps, then I am not that far off right?

More on the wireless:
Authentication Open
Encryption WEP
Key Index 1
AdHoc channel 802.11bg    11

I see a setting for accepting only 802.11g, would that make sense to set?  Wouldn't the router automatically set to highest speed?


I am attaching screen shots of two tests - one is internetfrog.com and the other is speedguide.net (that Darr247 suggested above). They are both similar in results.

Hi jjltpc,

>>Download up to 10 Mbps / Upload up to 2 Mbps<<
This is a speed which should easely be managable over 802.11G and ethernet. However you mention you only get 6 over ethernet which already seems rather slow to me considering that fiber shouldn't fluctuate that much.

Anyways, as you already mentioned, changing your wireless settings to G-only will help as the access point won't have to keep compatibility with slower 802.11B clients. This will improve your wireless speed.

Which model router do you have exactly?

Regards,

LucF

You should easily get faster than 5Mbps (let alone 2Mbps) even if it WERE an 802.11b connection.

Does your wireless router/AP have Quality of Service settings in it?

The router from Verizon is:
Platform:       Westell UltraLine Series3
Model Number:       A90-9100EM15-10

Current Wireless Status       
Wireless:       ON
SSID:       XXXXXXXXXXX
64-BIT WEP:       ON
64-BIT WEP KEY:       XXXXXXXXXXX
Channel:       2
SSID Broadcast:       Enabled
MAC Authentication:       Disabled
Wireless Mode:       Mixed - accepts 802.11b and 802.11g connections
Packets Sent Total       410395
Packets Received       197826



Note that I did download one of those war driving programs, and there was a much weaker wireless router that was using channel 1, the rest were on 11, so I hardcoded channel 2.  No gain or anything from that.

I am not sure what the quality of service setting would be.  I am attaching the advanced settings:

You might try channel 6 instead.  The 'channels' in the 2.4GHz ISM band are 22MHz wide, but their centers are separated by only 5MHz.  See first attachment.

The channels are really wider, but beyond 11MHz from center the signal is supposed to be attenuated at least -30dB from peak.

The second attachment shows what the signal on channel 6 looks like on a spectrum analyzer. Note how the signal overlaps channels 2 through 10.  Channel 2's signal would overlap on channel 1 (and vice-versa) the same way channel 6 does on channel 5 (and vice-versa... and on channel 7, too).

When the frequencies overlap they drown each other out like people talking 2 different languages trying to talk over each other... often it's so confusing the only choice is to stop talking and then repeat what was just said.

I hope this demonstrates why it's preferrably to use only channels 1, 6 and 11 for WiFi in the 2.4GHz ISM band. (and channel 14 in Japan... that just barely overlaps 11 thanks to a larger offset from 13 than any other channels have.)

Wow.... Nice charts...

:^)

Agreed - my guess is the people on 1 and 11 are already near to each other and aware of this - you are literally "stuck in the middle" so "hardcoding" 6 would give you least interference (until someone else near you starts a wireless system ;))