Jonkman Microblog
  • Login
Show Navigation
  • Public

    • Public
    • Network
    • Groups
    • Popular
    • People

Conversation

Notices

  1. 🇳🇴 Thor — backup account (thorthenorseman@octodon.social)'s status on Friday, 05-Oct-2018 16:11:29 EDT 🇳🇴 Thor — backup account 🇳🇴 Thor — backup account

    Just performed an interesting test: I put an Asus RT-AC66U tri-band (AC1700) router up against a Linksys LAPAC1200 dual-band (AC1200) access point.

    I ran SpeedTest on them both from my iPhone 7 Plus at a 1 meter distance parallel to the antennas, connected to 5 GHz.

    I tested a 80 MHz channel on the ASUS and 80 and 40 MHz on the Linksys.

    Results (rounded):
    - ASUS 80 MHz: 400/400
    - Linksys 80 MHz: 400/400
    - Linksys 40 MHz: 200/200

    In conversation Friday, 05-Oct-2018 16:11:29 EDT from octodon.social permalink
    1. 🇳🇴 Thor — backup account (thorthenorseman@octodon.social)'s status on Friday, 05-Oct-2018 16:14:12 EDT 🇳🇴 Thor — backup account 🇳🇴 Thor — backup account
      in reply to

      The reason for the 40 MHz channel width test on the Linksys is that, since it's only dual-band, it doesn't have many 5 GHz channels, and only a single possible extension channel at 80 MHz, and thus only one non-overlapping channel, whereas in the 40 MHz mode, there's two overlapping channels, which is more helpful to me, since I'm doing a multi-floor multi-building WiFi setup for my landlord, and I don't want interference.

      In conversation Friday, 05-Oct-2018 16:14:12 EDT from octodon.social permalink
      1. 🇳🇴 Thor — backup account (thorthenorseman@octodon.social)'s status on Friday, 05-Oct-2018 16:17:34 EDT 🇳🇴 Thor — backup account 🇳🇴 Thor — backup account
        in reply to

        It may seem a bit of a bummer to cap a 500/500 connection to 200/200 per floor on WiFi, but that's better than having an unreliable 400/400 connection per floor. I might still go for 80 MHz if there's sufficient signal attenuation between the floor. Just gotta figure out what "sufficient" is in terms of dBm, so off to Google I go.

        In conversation Friday, 05-Oct-2018 16:17:34 EDT from octodon.social permalink
      2. 🇳🇴 Thor — backup account (thorthenorseman@octodon.social)'s status on Friday, 05-Oct-2018 16:17:56 EDT 🇳🇴 Thor — backup account 🇳🇴 Thor — backup account
        in reply to

        It may seem a bit of a bummer to cap a 500/500 connection to 200/200 per floor on WiFi, but that's better than having an unreliable 400/400 connection per floor. I might still go for 80 MHz if there's sufficient signal attenuation between the floors. Just gotta figure out what "sufficient" is in terms of dBm, so off to Google I go.

        In conversation Friday, 05-Oct-2018 16:17:56 EDT from octodon.social permalink
        1. 🇳🇴 Thor — backup account (thorthenorseman@octodon.social)'s status on Friday, 05-Oct-2018 16:25:45 EDT 🇳🇴 Thor — backup account 🇳🇴 Thor — backup account
          in reply to

          "A ratio of 10-15dB is the accepted minimum to establish an unreliable connection; 16-24dB (decibels) is usually considered poor; 25-40dB is good and a ratio of 41dB or higher is considered excellent."

          Okay, so I need a minimum of 25 dB attenuation between floors before I can consider any channel overlap.

          In this concrete building, the attenuation looks to be about 32 dB between this floor and the one above it. I can say this because I have identical access points on each floor.

          In conversation Friday, 05-Oct-2018 16:25:45 EDT from octodon.social permalink
          1. 🇳🇴 Thor — backup account (thorthenorseman@octodon.social)'s status on Friday, 05-Oct-2018 16:30:05 EDT 🇳🇴 Thor — backup account 🇳🇴 Thor — backup account
            in reply to

            I think the setup in this building is about as close to ideal as you can get. I have zero channel overlap on both 2.4 GHz and 5 GHz, the 5 GHz is running at a 80 MHz channel width, and there's 25 dB of signal attenuation between the floors. I don't think it can get much better than that.

            I've been fine-tuning the setup in this building after I discovered that not all 5 GHz channels are created equal: To get a 80 MHz channel width, you have to pick the right channels.

            In conversation Friday, 05-Oct-2018 16:30:05 EDT from octodon.social permalink
            1. 🇳🇴 Thor — backup account (thorthenorseman@octodon.social)'s status on Friday, 05-Oct-2018 16:39:31 EDT 🇳🇴 Thor — backup account 🇳🇴 Thor — backup account
              in reply to

              Specifically, for 5 GHz, to get an 80 MHz channel width, you have to use channel 42, 58, 106, 122, 138 or 155 for the extension channel. If your radio chipset supports 160 MHz channels, you have to use channel 50 or 114. I'm not sure if an 80 MHz radio chipset will use the 160 MHz channels. ASUS RT-AC66U routers operate at 80 MHz if you use channel 36, 52 or 100 as the main channel, respectively auto-picking channel 42, 58 or 106 as the extension channel.

              In conversation Friday, 05-Oct-2018 16:39:31 EDT from octodon.social permalink
              1. 🇳🇴 Thor — backup account (thorthenorseman@octodon.social)'s status on Friday, 05-Oct-2018 16:44:08 EDT 🇳🇴 Thor — backup account 🇳🇴 Thor — backup account
                in reply to

                Linksys LAPAC1200 access points only have a single 5 GHz band, operating on channel 36, 40, 44 or 48. In 80 MHz mode, they will use channel 42 as the extension channel no matter which main channel you pick. In 40 MHz mode, they will pick extension channel 38 for main channel 36 or 40, and extension channel 46 for main channel 44 or 48. If you need non-overlapping 80 MHz channels, get a LAPAC1700 instead.

                In conversation Friday, 05-Oct-2018 16:44:08 EDT from octodon.social permalink
  • Help
  • About
  • FAQ
  • TOS
  • Privacy
  • Source
  • Version
  • Contact

Jonkman Microblog is a social network, courtesy of SOBAC Microcomputer Services. It runs on GNU social, version 1.2.0-beta5, available under the GNU Affero General Public License.

Creative Commons Attribution 3.0 All Jonkman Microblog content and data are available under the Creative Commons Attribution 3.0 license.

Switch to desktop site layout.