Wireless channel advice

[bodean]

Why is it frowned upon to set Channel to AUTO and let the router decide the channel?

 

[stevech]

Why is it frowned upon to set Channel to AUTO and let the router decide the channel?

Because often they choose other than 1, 6 or 11, the three non-overlapping channels. Also, if you live in a dense urban area, it is likely to choose a channel on which there's a neighbor with intermittent high usage (e.g., evening video streaming).

 

[bodean]

Because often they choose other than 1, 6 or 11, the three non-overlapping channels. Also, if you live in a dense urban area, it is likely to choose a channel on which there's a neighbor with intermittent high usage (e.g., evening video streaming).

I live in a Condo builiding, have about 28 units in the building. 90% of the time I am the only person running 5.0GHZ, sometimes I see 2 others. There are about 15-20 2.4GHZ SSIDS listed. So for the 2.4GHZ, I see why manual channel selection is important. But as for 5.0GHZ, only having me, or maybe 2 others at times, would AUTO be ok, or still stick with manual?

 

[sfx2000]

Try 1 or 11 first in 2.4GHz, and then try 6 - most SOHO routers default to 6 in auto-select mode, and when running wide-channels, they tend towards 1 or 11 for the primary channels... channel 6 seems to be the noisy one in the 'burbs...

Don't consider this a rule of thumb however... every case is different, so a good site survey with Innsider or similar tools is important - and do several locations inside the house or apartment (or office)

Oddly enough - tend to see a lot of ATT-DSL/Uverse routers camp on channel 8 on the 2.4Ghz band - which is downright un-neighborly... same with HP all-in-one printers with WiFi which default to 10, esp the older ones that only do 802.11b...

sfx

 

[bodean]

Thanks. What about the 5.0 channel given what I posted above.

 

[sfx2000]

Thanks. What about the 5.0 channel given what I posted above.

149 is a good place to start...

 

[azazel1024]

149 is a good place to start...

Good place to start and don't use auto on 5GHz either. Even if there isn't a lot of congestion, 5GHz is pretty darned wide. From around 5.2GHz up to close to 5.9GHz. Absorbtion is different, as is reflection, which can all have impacts on speed and distance. Plus, 5GHz is fun in that some of the channels are allowed much higher radio power than others. The lower channels tend to be limited to 50mw or some on 250mw of radio power. The upper channels are generally allowed to be 250mw or 1w.

1w doesn't matter because there are no consumer routers (that I am aware of) that'll do 1w of radio power. However, plenty will do over 50mw and most "high power" consumer routers do 250mw or so. So some of the channels might limit radio power by 1-2dBm over what higher channels might (on a consumer router).

 

[stevech]

Good place to start and don't use auto on 5GHz either. Even if there isn't a lot of congestion, 5GHz is pretty darned wide. From around 5.2GHz up to close to 5.9GHz. Absorbtion is different, as is reflection, which can all have impacts on speed and distance. Plus, 5GHz is fun in that some of the channels are allowed much higher radio power than others. The lower channels tend to be limited to 50mw or some on 250mw of radio power. The upper channels are generally allowed to be 250mw or 1w.

1w doesn't matter because there are no consumer routers (that I am aware of) that'll do 1w of radio power. However, plenty will do over 50mw and most "high power" consumer routers do 250mw or so. So some of the channels might limit radio power by 1-2dBm over what higher channels might (on a consumer router).

Consumer routers with HPAs (high power amps) claiming 100+ mw, tend to market themselves using only the highest power spec which occurs at the lowest modulation order (WiFi bit rate). As the OFDM mode rates increase up the stepped scale, the power declines by about 6dB. This is due to the nature of OFDM's high peak-to-average signal ratio. Conversely, the slower speeds in WiFi are non-OFDM and the ratio is smaller. If the 6dB "backoff" is not done at high bit rates, the transmitted signal is distorted. This leads to higher frame error rates at the receiver. This becomes lower throughput due to retransmission. It gets worse at lower signal to noise condition (weaker signals).

Transmitted (radiated) power gains best come from antenna gain rather than HPAs, as antennas do not introduce distortion. The root cause of the HPA backoff issue is economics: it's too costly to sell consumers a radio with an HPA that can meet the rho standards (highly linear gain) at high modulation orders. So the backoff is a compromise for cost. Also, radio firmware knows the rho demand and usually ignores calls for more power than can be achieved without excessive distortion, e.g., the UIs in 3rd party firmware.

The WiFi alliance has standards for transmitted waveform quality, often called rho. But many vendors don't truly comply.