Scenario 3 - 11b associated and simultaneously active
Once 11b clients connect to your AP, sooner or later they're going to start using it! Figure 11 shows one example of what you might expect.
Figure 11: Linksys WPC54G & NETGEAR MA401 Stations
The screen shot shows a test done with Linksys WPC54G and NETGEAR MA401 clients. The Linksys draft-11g card starts first, and is joined by the NETGEAR 802.11b card at the 20-second mark. Both cards run until the Linksys card completes its programmed number of transfers and stops, letting the NETGEAR card finish by itself.
The plots clearly show that the Linksys draft-11g client's throughput really takes it on the chin, dropping well below the NETGEAR 11b client's. The effect, however, is once again client dependent, as shown in Figure 12.
Figure 12: Two pair test - NETGEAR WAB501
This second screen shot shows the same test, but this time with the second card being a NETGEAR WAB501 Dual-Band client forced to 802.11b mode. This test shows behavior more like what you should expect in all cases, once Linksys gets its firmware released that has draft 6.1 improvements, which mandate that 11g clients get more air time than 11b clients in mixed WLANs.
802.11g Wireless Performance Test Results
|Test Conditions - WEP encryption: DISABLED |
- Tx Rate: Automatic
- Power Save: disabled
- Test Partner: Linksys WPC54G Wireless-G Notebook Adapter
|Firmware/Driver Versions AP f/w: |
1.01.4 Jan 27 2003
Wireless client driver:
Wireless client f/w:
|Test Description||Signal Quality (%)||Transfer Rate (Mbps)||Response Time (msec)||UDP stream|
|Throughput (kbps)||Lost data (%)|
|Client to AP - Condition 1||0||21.2 |
|1 (avg) |
|Client to AP - Condition 2||0||20.5||1 (avg) |
|Client to AP - Condition 3||0||11.6||60 (avg) |
|Client to AP - Condition 4||0||4.8||8 (avg) |
See details of how we test.