Linksys Dual-Band Wireless
A+G Notebook Adapter
Rating: 4 out of 5
Netgear Dual Band Wireless
Rating: 4 out of 5
What has lots of letters and helps people to communicate? If you answered "the
Post Office," you’re right, but it’s also an apt description of the two
Cardbus cards I tested here, the Linksys
Dual-Band Wireless A+G Notebook Adapter (model WPC55AG) and the Netgear Dual
Band Wireless PC Card (model WAG511). Each contain a veritable alphabet
soup of 802.11 letter designations.
Both of these products are based on the three-chip AR5001X chipset from Atheros,
which is capable of communicating across all three major WLAN technologies:
the 5GHz 802.11a and the 2.4GHz 802.11b and 802.11g.
Please note that both cards have rather tall antennae, and putting them in
the lower Cardbus slot is not advisable as it would certainly block access to
the remaining slot.
Product evaluations rarely go off without a hitch, and this one was no exception.
However, this particular evaluation exposed a very odd problem, the cause and
scope of which remain unknown as of this writing.
Both the Netgear and Linksys cards and their configuration utilities installed
on my test bed normally and without incident. Moreover, both cards were able
to locate and associate with wireless networks (all three flavors) via their
That’s when all the fun started.
When configured as DHCP clients, neither card was able to obtain an IP address
lease (from any of several routers). No matter, I specified a static IP address
When I did so, however, I found that I had no IP connectivity between the client
and the wireless access point or router, even though the address information
was valid and correctly configured.
Several un- and re-installations of the cards’ respective software and drivers,
as well as the TCP/IP protocol itself, did not rectify the problem. Neither
did a restoration of the system from a cloned virgin image unspoiled by potentially
Neither vendor was able to reproduce my problem or provide an explanation for
it. On the other hand, the fact that both cards exhibited the same behavior
strongly implies that the culprit is the chipset that they have in common or,
at least, some aspect of the firmware or drivers.
For the record, my original test system was an IBM ThinkPad I series Model
1400, running Windows 2000 Professional. At this point, there’s no evidence
that the problem afflicts anything more than this specific system, but it’s
likely that at least some other brands or models may be impacted as well.
In any event, to accomplish performance testing on the cards, I turned to another
notebook, a Compaq Presario 722 running Windows XP Home. Both cards installed
and, more importantly, operated normally on this system. For 802.11g testing,
I operated the both cards against a Netgear 54 Mbps
Wireless Access Point (model WG602), while for the 802.11a testing I used
HE102 access point from the same vendor.
The performance numbers produced by both the Linksys and the Netgear CardBus
products clearly indicate that neither suffers significantly from a "jack
of all trades and master of none" syndrome. In 802.11g mode, both turned
in performance numbers that compared favorably to dedicated 802.11g devices
I’ve tested in the past. In 802.11a mode there was more daylight between the
two products performance numbers at certain distances.
Neither was able to obtain a usable signal (in A or G mode) much beyond a 100
foot range from the access point in my environment.
When operating in 802.11g mode against the WG602 access point, the two cards
were in the same performance neighborhood at almost every interval, from between
22 and 23 Mbps at 10 feet to between 4.8 and 5.8 Mbps at 100 feet. The only
exception was at 75 feet, where the Linksys could only manage 7.4 Mbps to the
Netgear’s 13.3 Mbps. I attribute this to temporary interference, however.
In 802.11a mode, the two cards were less in lockstep. The Netgear exhibited
stronger performance at close range–16.7 and 14.6 Mbps at 10 and 25 feet respectively.
After that though, the throughput dropped to below 5 Mbps through 100 feet.
In comparison, the Linksys was less impressive out of the gate at (14.24 Mbps
at 10 feet), but it held its performance better throughout the range. Throughput
at 25 feet was 8.55 Mbps and 9.18 Mbps at 50 feet. At the longer ranges, though,
it turned in scores similar to that of the Netgear.
If the hardware and the performance of the two products were somewhat similar
given their common underpinnings, there was a more distinct difference between
the two products when it came to the supplied software.
For example, Netgear’s utility provides a reasonable level of control over
the card. In addition to the de rigueur stuff like site survey and WEP
configuration, it also lets you individually enable or disable any of the three
supported modes–802.11a, 802.11a Turbo, and 802.11b/g. The utility doesn’t
let you set a desired data rate for any of these modes, but it does offer drop-downs
for transmit power and power savings, features that are often relegated to the
bowels of the driver and accessible only via Device Manager.
If you install the WAG511 under Windows XP, you’ll have the option to use the
full utility or let XP’s "Wireless Zero Configuration" handle it.
If you chose the latter, you’ll only get informational tabs from the utility,
and the operating system will display whatever networks are in the vicinity.
Linksys’ configuration utility is much more minimalist. The WPC55AG’s software
offers basic functionality–site survey, profiles, and link information only.
Make no mistake–it will find available networks and let you connect to them,
but with a minimum of fanfare.
When it comes to Windows XP, Linksys abdicates the configuration chores to
the OS entirely– you can’t even install their utility under Windows XP. This
may be convenient for some, but it can be somewhat limiting. For example, I
could not enable the maximum level of WEP encryption, 152-bit, on the Linksys
card because Windows would only accept the standardized 40/64- or 104/128- bit
key length. Linksys says that an XP-compatible version of their utility is planned.
The Linksys WPC55AG and Netgear WAG511 are examples of the fact that you can
have your cake and eat it too; that performance and universal compatibility
need not be mutually exclusive. Though my initial experiences with the cards
on my test bed remain troubling, especially absent an explanation, my hope is
that my experience was an anomaly.
I give a slight edge to the Netgear WAG511 for its more comprehensive software
utility, but I would wholeheartedly recommend either of these cards to mobile
mavens who want to have the maximum flexibility to connect to whatever network
is present where their travels take them.