802.11g Demonstrates White Lies of Wi-Fi

Wireless Supersite Editor Ross Rubin calls on the wireless industry to start labeling gear with actual capabilities the way monitor and hard disk companies do.

When is a 17-inch monitor not a 17-inch monitor or a 30GB hard disk not a 30GB hard disk? When you actually use them. Following the outcry over misleading vendor statistics, monitor and hard disk manufacturers began modifying their measurement of screen size and hard disk capacity to include such qualifiers as "usable" screen space and "formatted" capacity. The move was a positive one because it allows users to compare apples to apples.

However, this truth in advertising doesnt exist in the land of wireless networks, where a consumer or hapless small business owner would be left in a quandary over which networks can travel at which speeds—that is, once theyve figured out the other pressing issues of compatibility and range.

Take 802.11b, which is touted as providing 11 Mbits per second. However, its typical maximum throughput is closer to 4 or 5 Mbps.

Like other white lies, ones about bandwidth tend to spiral and grow. Even larger gaps exist between the theoretical and stated throughputs of 802.11a and 802.11g. While both are rated at exactly the same speed of 54 Mbps, youll be lucky to see even half of that, and only with 802.11a. The real-world throughput of 802.11g is even slower. However, vendors are reporting that, while neither high-speed Wi-Fi variant can match the range of 802.11b, 802.11g can do more with fewer access points. This is likely due to its better penetration of walls.

The stat raised some eyebrows when allegations leaked out that 802.11g had been throttled from 24 Mbps to 20 Mbps in response to performance in the presence of 802.11b. It turned out that nothing had actually changed, but someone had the courage to plow through at least some of the marketing malarkey and note the more typical lower speed. However, those efforts will be largely ignored anyway because 802.11g is still widely marketed as being capable of 54 Mbps.

Ultimately, the gap between the theoretical and actual speeds of networks is an issue neither new nor endemic to wireless. But the ease of use of installing wireless networks has put them in the hands of businesspeople and consumers who are not network experts. Lay people need better information on what they can realistically expect from their networks.

The disparity may remain less of an issue in the consumer market, where the most popular application of shared broadband has requirements far below the threshold of even 802.11b. As more demanding tasks such as networked video begin to overwhelm the real limits of todays Wi-Fi gear, however, members of the Wi-Fi Alliance may find things less appealing things than packets flying through the air toward them.

Should Wi-Fi vendors report the actual throughput of their wireless network products? E-mail me.

Wireless Supersite Editor Ross Rubin is a senior analyst at eMarketer. He has researched wireless communications since 1994 and has been covering technology since 1989.

More from Ross Rubin: