Wireless Performance
Although I have a USB2-equipped desktop machine, the notebook that I normally use for client testing remains at USB 1.1. But not to be kept from my appointed throughput-testing rounds, I just connected a Belkin F5D7130 AP [reviewed here] to my notebook and left the USB-G54 lashed to my 2.4GHz Pentium4 desktop, sitting where the test AP normally is. Figure 2 shows the test results.
Figure 2: Four location throughput
(click on the image for a full-sized view)
I should note that you can't directly compare these results - or any results going forward - with those of previous reviews before October 1, 2003. My recent move meant setting up new test locations, which are described here.
The new Location 3 test is the only one done from a point one floor below the stationary AP location. The poorer performance for Location 3 might be due to the unusual antenna design of the USB-G54, but it could just be the way the radio waves crumble in the new test location. Only time - and more product testing - will tell for sure.
I also ran tests in WEP128, WPA-PSK / TKIP, WPA-PSK / AES modes and found a significant throughput loss (24%) only in WPA-PSK / TKIP. All other modes had variation within the accuracy of my test methods. This is consistent with what I've found on other Broadcom-based products, and does put you at a throughput disadvantage if you mix the USB-G54 (or any other Broadcom-based client supporting WPA) with WPA-enabled APs or wireless routers based on GlobespanVirata (Intersil) or Atheros chipsets.
The best-case average throughput of about 20Mbps is consistent with Broadcom-based 11g products based on Ethernet or CardBus interfaces. So it appears that - unlike 802.11b USB1.1 adapters - you won't pay a throughput penalty for connecting via USB 2.
You will lose throughput, though, if you use the adapter with a computer supporting USB 1.1, as Figure 3 shows.
Figure 3: Throughput connected to USB 1.1
(click on the image for a full-sized view)
Smooths things right out, doesn't it? Although you wouldn't buy the USB-G54 specifically to attach to a USB 1.1-equipped machine, it's nice to know that you at least have the option.
802.11g Wireless Performance Test Results
| Test Conditions
| Firmware/Driver Versions AP f/w: | ||||
|---|---|---|---|---|---|
| Test Description | SNR (dB) | Transfer Rate (Mbps) | Response Time (msec) | UDP stream | |
| Throughput (kbps) | Lost data (%) | ||||
| Client to AP - Condition 1 | 0 | 20.4 [No WEP] 19.9 [w/ WEP] | 1 (avg) 2 (max) | 499 | 0 |
| Client to AP - Condition 2 | 0 | 13.2 | 1 (avg) 3 (max) | 499 | 0 |
| Client to AP - Condition 3 | 0 | 2.8 | 3 (avg) 6 (max) | 347 | 29 |
| Client to AP - Condition 4 | 0 | 11.7 | 2 (avg) 3 (max) | 499 | 0 |
Gn5jbP Kudos! What a neat way of thinking about it.