Network extenders and performance....

Discussion in 'Mac Accessories' started by davidra, Dec 8, 2014.

  1. davidra macrumors 6502

    Joined:
    Oct 12, 2011
    #1
    OK, maybe someone can clarify this. I have my Mac in a guest house located only about 100 feet from my cable modem/router combo. However there's lots of glass in the way so I have a Cisco/Linksys extender located in the guest house that can pick up the signal just fine. It's a dual band router but I just use the 2.4 GHz band because the 5 band won't get that far. Looking at the utilities section of Wireless Diagnostics, when I scan for wifi I see three bands; the two 2.4 bands and the one 5 band. Ordinarily I stay connected to the extender's 2.4 GHz band. Sometimes, though, that extender will crap out and the Mac will connect to the other 2.4 band. Here's the question. When I'm connected to the extender band, Utilities shows excellent quality with a transmit rate over 140. When I'm connected to the 2.4 band not coming from the extender, Utilities shows poor quality with tx rates in the 30 to 50 Mbps range. That makes sense. Here's what doesn't: when I use Speedtest I get download speeds of around 7 to10 Mb/s using the extender band with excellent quality and high transfer speeds; and download speeds of 26 to 32 Mb/s using the house band with the poor quality and low transfer speeds. Now I realize that using an extender affects the bandwidth in some way, but I guess I expected that it would affect both connections equally, while it appears to only affect the extender band. Can someone make sense of this? My inclination is to just not use the extender as long as I can connect, regardless of what the utilities show me.
     
  2. Altemose macrumors G3

    Altemose

    Joined:
    Mar 26, 2013
    Location:
    Elkton, Maryland
    #2
    Wireless extension often uses more than 50% of the original bandwidth.
     
  3. davidra thread starter macrumors 6502

    Joined:
    Oct 12, 2011
    #3
    That is certainly said frequently. I had always assumed that if the entire 2.4 GHz band were halved by using the extender, then wouldn't you expect two devices, one attached to the extender and one to the remainder of the band, be running at similar speed? Or the obvious explanation: only the part of the band running through the extender suffers from the decreased bandwidth?
     
  4. Alrescha macrumors 68020

    Joined:
    Jan 1, 2008
    #4
    The traffic from your computer to the extender is retransmitted from the extender to the access point in the same band so it takes up twice the available bandwidth.

    A.
     
  5. Altemose macrumors G3

    Altemose

    Joined:
    Mar 26, 2013
    Location:
    Elkton, Maryland
    #5
    Alrescha's explanation is spot on!
     
  6. davidra thread starter macrumors 6502

    Joined:
    Oct 12, 2011
    #6
    Got it. Thanks a bunch. I find it interesting that I have such poor quality and low tx rate but I'm still downloading at the same speed as my wife's Mac which is much closer with higher quality signal. But this makes sense. Think I'll stick to the lower quality faster signal and I start to have connectivity problems I'll go back to the extender.
     
  7. Altemose macrumors G3

    Altemose

    Joined:
    Mar 26, 2013
    Location:
    Elkton, Maryland
    #7

    Perhaps it would be possible to install an outdoor AP on the main house and connect to that. Normally, I would suggest Ubiquiti NanoStation Loco M5s but they are overkill for this situation.
     
  8. ragetty, Jan 10, 2015
    Last edited: Jan 10, 2015

    ragetty macrumors newbie

    Joined:
    Oct 26, 2009
    #8
    i ran into a very simlar problem.

    a cable modem was etherneted to a 3G AExtreme (simul. dual band) to provide wifi via separately named 2.4GHz & 5GHz networks. a 2G AExtreme (either/or dual band) was connected over wifi to *either* band in order to extend that band. all of this is downstairs in the extended cellar under the house.

    upstairs on the ground floor, wifi tools indicated good reception, regardless of whether the MBA was connected to (i.e. directly above) the 3G or the 2G AE, with 5GHz data rates better than 2.4GHz - although this obviously only indicated the maximum data rate with the respective AE.

    true internet speeds were good when connected to the 3G AE regardless of whether via 2.4GHz or 5GHz. however, if i roamed so that the MBA connected to the 2G AE, 1). true internet speeds for an extended 2.4GHz network were roughly half that of when connected to the 3G AE via 2.4GHz, and 2). the extended 5GHz performed much worse still.

    obviously we have some range issues in the house (through the cellar walls - not surprising given the vintage), in which case 2.4GHz is the better choice than 5GHz. but the disadvantages of extending over wifi (rather than ethernet) are also evident.

    then it ocurred to me to swap the AEs. the 2G is now the base station (connected to the cable modem), providing (longer range) wifi at 2.4GHz, with the 3G AE now remotely extending over both bands. if the MBA connects to the 3G via 2.4GHz, there is still the same bandwidth drop (compared to conecting via the 2G AE) - i.e. performance is unchanged compared to the reverse configuration. but, if the MBA connects to the 3G AE via 5GHz, there is little bandwidth loss compared to the 2G AE via 2.4GHz - i.e. the trick is to separate the communication bands (device to 3G AE, 3G AE to 2G AE)!

    this works quite well for me in that the aTV can connect over 5GHz (fixed location directly above the 3G AP), while all other (roaming) devices can make use of 2.4GHz. upstairs on the 2nd floor (office), the MBA connects to the 2G AP, so no roaming bandwidth issues there.

    there is one thing i am not clear on - if i do *not* separately name the 5GHz network, does the 3G AE automatically put 2.4GHz+5GHz capable devices on 5GHz (signal permitting)? or do the devices choose themselves? either way, can i assume that apple has somehow built this intelligence into the 'system'?

    thx for any help ...
     

Share This Page