Results 1 to 7 of 7

Thread: How do I interpret this in terms of expected wifi performance?

  1. #1

    Default How do I interpret this in terms of expected wifi performance?

    First of all, thanks for the free inSSIDer utility. It is best software of this kind out there!

    Now there is my confusion.

    I just purchased a Lenovo Yoga 13. It is beautiful ultrabook but I have heard about its poor wifi performance. So I checked it out with inSSIDer V3, and compared the results I got from the older HP laptop I was planning to retire.

    The simple test I did was starting at 3ft from the AP to establish the best possible wifi condition for the test subject, and then slowly walk away from the AP to my family room, which is about 24ft away with a couple layers of walls in between.

    The HP starts out with a good RSSI (-25dbm) but a somewhat poor link score (56) and ends up with still decent RSSI (-50dbm) with link score of 28 (http://www.flickr.com/photos/28467392@N08/9078602467/)

    The Lenovo Yoga 13 starts out with a poor RSSI at the same place (-45dbm) but a good link score of 85. At the far end, the RSSI drops to a weak -71dbm with a decent link score of 56 (http://www.flickr.com/photos/28467392@N08/9078600473/)

    The real world bandwidth test shows that they are pretty close when they are not far from the AP. But the old HP laptop outperforms the new Yoga 13 by large margin when they are at the fringe of the coverage.

    My question is, how do you make sense out RSSI and link score to determine relative performance of the wifi client? Thanks.

  2. #2

    Default

    Hello

    You must be careful when comparing 802.11n Wi-Fi clients. The reason is that not all 802.11n Wi-Fi clients are the same and consequently will not perform to the same level. Can you post the tech specs for each card and be sure to include the number of Rx and Tx antenna each one has along with the amount of spatial streams each is capable of e.g 3x2:2.

  3. #3

    Default

    Yoga 13 uses a Realtek wifi/bluetooth combo card. HP dm-4 uses Intel Centrino Wireless N-1000 adapter. I am not sure about their antenna designs.

    RSSI seems easy to understand. I believe that it is the AP transmitter power seen by the wireless card. But what about the quality score? Is it reported by the adapter or calculated by inSSider? How is it derived?

  4. #4

    Default

    Quote Originally Posted by litewavve View Post
    Yoga 13 uses a Realtek wifi/bluetooth combo card. HP dm-4 uses Intel Centrino Wireless N-1000 adapter. I am not sure about their antenna designs.

    RSSI seems easy to understand. I believe that it is the AP transmitter power seen by the wireless card. But what about the quality score? Is it reported by the adapter or calculated by inSSider? How is it derived?
    Correct, but 8011n mimo techniques can improve the signal strength seen by an client by sending multiple streams of the same data to multiple antenna on the client. In the specs for any 802.11n client you will see a number like 2x2:2 or 1x1:1 or 2x3:2. To explain these numbers, a 2x3:2 would be a client with 2 Tx antennas, 3 Rx antenna and capable of 2 spatial streams. A client with 1 Rx antenna may not see the signal strength as high as a client with 3 Rx antenna.

    This is one of the ways in which 802.11n improves range and throughput.

  5. #5

    Default

    Keep in mind that RSSI units are arbitrary and vary among chipset makers.
    For example a reading of -50 does not represent the same signal strength
    for every wireless nic.

    Atheros reports power on a scale of -0 to -127 while the Cisco scale is -0 to -100.
    Zero is the highest power level.
    Old Mod by the Sea

  6. Default

    The Link score is a grade calculated for each network based on signal strength, channel power (with a Wi-Spy), and the amount of networks competing for airtime.

    The points made about MIMO are good ones. A 1x1:1 card only has one radio chain, where a 3x3:3 card has three radio chains, all of which can be used simultaneously. That's roughly 3x the throughput! On mobile devices, you typically see less radio chains and antennas to conserve space and power. My Nexus 4 (cell phone) probably only has a 1x1:1 card, but my MacBook Air has a 2x2, whereas MacBook Pros have 3x3 cards. When evaluating a card against other cards, using inSSIDer to watch the signal strength of an AP is fine, but I'd probably use a throughput test with iPerf. I wouldn't pay attention to the Link Score.

  7. #7

    Default

    I never see better than 50db on my Toshiba (Realtek) laptop, even when I'm within 3 ft of the router.

    Russ

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •