Linksys Wireless-N

Any opinions on the Linkysys wireless-N router? It got lukewarm reviews last summer from PcMag and Toms Hardware. There have been firmware upgrades since then. I've had good luck with Linksys equipment for years, so despite the negative reviews, I took the plunge last week and bought the WRT300N router and WPC300N notebook adapter. My cable modem is the Motorola SB5120. My ISP is Comcast and I routinely connect at an astounding 270 Mbps. The servers at Speedtest.net produce download speed results over 15Mbps. I am happy so far. My only regret is not getting the WRT350N with the wired gigabit switch.

Reply to
shoreke
Loading thread data ...

shoreke hath wroth:

Yep. Basically what they said about draft 802.11n routers is that they don't really offer any improvement over more conventional 802.11g wireless devices and are anything but interoperable with other draft

802.11n devices.

Holdit. That's your wireless connection speed, not your thruput. My laptop connects at 54Mbits/sec but only transfers files at perhaps

22Mbits/sec. What you need to do is run a local speed test, without the cable modem creating a bottleneck.

Run: iperf -h for help and syntax examples. Find a wired ethernet connected PC and run IPerf 1.70 in server mode as just: iperf -s

On your wireless laptop, run IPerf as a client as in: iperf -c ip_addr_of_server

You should get a real file transfer speed, instead of just the connection speed.

Duz your cable ISP offer "burstable" speeds, where the nominal download speed is 6Mbits/sec, but which might go to several times that for short periods of time?

Maximum wire speed for 100baseT-FDX is typically about 88Mbits/sec. I'll believe you need more than that when I see the results of IPerf.

This is what I get between my junk desktop and my Panasonic CF-M34 laptop with a Netgear WG511 card going through a WRT54GS v3.? router with a connection speed of 54Mbits/sec, but where there's all kinds of traffic moving (SNMP, syslog, ping probes, etc) with the 802.11b compatibility mode enabled and light 802.11b traffic. It's usually about 20Mbits/sec but I think I've got some water in my outdoor coax and antenna. I get roughly the same lousy 16.6Mbits/sec results in either direction. I'm very interested in what you get with your new Linksys setup.

C:\\IPerf> iperf -c 192.168.1.116

------------------------------------------------------------ Client connecting to 192.168.1.116, TCP port 5001 TCP window size: 8.00 KByte (default)

------------------------------------------------------------ [840] local 192.168.1.11 port 2212 connected with 192.168.1.116 port

5001 [ ID] Interval Transfer Bandwidth [840] 0.0-10.0 sec 19.9 MBytes 16.6 Mbits/sec
Reply to
Jeff Liebermann

Jeff Liebermann hath wroth:

Turning off 802.11b compatibility didn't help very much: C:\\IPerf> iperf -c 192.168.1.116

------------------------------------------------------------ Client connecting to 192.168.1.116, TCP port 5001 TCP window size: 8.00 KByte (default)

------------------------------------------------------------ [840] local 192.168.1.11 port 2342 connected with 192.168.1.116 port

5001 [ ID] Interval Transfer Bandwidth [840] 0.0-10.0 sec 21.3 MBytes 17.9 Mbits/sec

I also have frame burst and Afterburner turned off. I've never benchmarked this particular laptop. Something is wrong... argh!

Reply to
Jeff Liebermann

Jeff Liebermann hath wroth:

I found the problem. The neighbor was using their wireless and I was picking up the interference. They finally turned off their computer and went to bed. Now, I get a more normal:

C:\\IPerf> iperf -c 192.168.1.116

------------------------------------------------------------ Client connecting to 192.168.1.116, TCP port 5001 TCP window size: 8.00 KByte (default)

------------------------------------------------------------ [840] local 192.168.1.11 port 5463 connected with 192.168.1.116 port

5001 [ ID] Interval Transfer Bandwidth [840] 0.0-10.0 sec 28.4 MBytes 23.8 Mbits/sec
Reply to
Jeff Liebermann

Hi Jeff,

I have the basic Comcast broadband package. I dont pay for extra for burstable lines or speedbooster gimmicks to increase extra bandwidth.

There is an easy- as- pie IPERF already configured to run under Windows XP at

formatting link
I ran the test and repeated it. These are my results:

Client connecting to 192.168.1.102, TCP port 5001 TCP window size: 8.00 KByte (default)

------------------------------------------------------------ [1868] local 192.168.1.101 port 1443 connected with 192.168.1.102 port

5001 [ ID] Interval Transfer Bandwidth [1868] 0.0-10.0 sec 32.1 MBytes 26.9 Mbits/sec

Client connecting to 192.168.1.102, TCP port 5001 TCP window size: 8.00 KByte (default)

------------------------------------------------------------ [1868] local 192.168.1.101 port 1441 connected with 192.168.1.102 port

5001 [ ID] Interval Transfer Bandwidth [1868] 0.0-10.0 sec 32.0 MBytes 26.9 Mbits/sec
Reply to
shoreke

shoreke hath wroth:

The last time I checked Comcast locally (Santa Cruz County, Calif), there was a hard limit at 6Mbits/sec. I'll check with my customers and see if anything has changed.

Thanks. That's the same binary as I was using. Looks like you ran yours with the default parameters.

At 26.9 Mbits/sec thruput, you're getting only a bit better than what I would expect from a good solid 54Mbit/sec connection. You could probably do better by tinkering with the IPerf parameters as suggested in the above URL. Also, UDP will go faster than TCP. However, it's a long way from 270Mbits/sec connection speed and hardly any better from my stock 802.11g system.

There may be some benifits to MIMO in tolerating internal reflections and multipath, which will show up as additional range. Comparing your range with mine is a waste of time because there are just too many differences. What would be interesting is how far away you can get between the two MIMO devices and still maintain something close to the

26.9Mbits/sec speed. You'll probably do somewhat better than conventional 802.11g which is what Tom's Hardware found in their simulations.

Hmmm... I'm getting consistent failures testing with UDP for some reason. I'll play with it later.

iperf -c 192.168.1.116 -b 30M WARNING: option -b implies udp testing

------------------------------------------------------------ Client connecting to 192.168.1.116, UDP port 5001 Sending 1470 byte datagrams UDP buffer size: 8.00 KByte (default)

------------------------------------------------------------ [840] local 192.168.1.11 port 1131 connected with 192.168.1.116 port

5001 [ ID] Interval Transfer Bandwidth [840] 0.0-10.0 sec 35.7 MBytes 29.9 Mbits/sec read failed: Connection reset by peer
Reply to
Jeff Liebermann

Cabling-Design.com Forums website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.