A Linksys WPC600N Cardbus card was inserted into a Fujitsu P7120 Lifebook (1.2 GHz Intel Pentium M, 504 MB) notebook running WinXP Pro SP2 with all the latest updates. I used the 22.214.171.124 driver with the Linksys client application running.
The router had 1.01.33 "Draft 2.0" firmware and I left all factory default settings in place, except as noted.
Maximum Throughput - 2.4 GHz
To start, I ran some close range open-air IxChariot test to look at maximum performance and throughput variation. This also provides baselines to check the Azimuth results against. Testing was done with the router and notebook about 10 feet apart in open air sitting in my lab with no other networks in range.
Figure 14 shows simultaneous up and downlink results in the 2.4 GHz band with the mode forced from its default of Auto to its Standard - 20 MHz Channel mode, which yields around 82 Mbps of total throughput. Running up and downlink separately yielded results of 64 and 69 Mbps respectively.
Figure 14: Up and downlink throughput - 2.4 GHz band, 20 MHz bandwidth
There is an anomaly in the 600N's throughput characteristic that is worth noting, since it has the effect of lowering average throughput. Figure 15 shows 2.4 GHz downlink throughput where you can see a large throughput "dropout" around the 16 second mark. This same effect is seen in varying degrees in both bands and in both 20 and 40 MHz channel modes.
Figure 15: Downlink throughput - 2.4 GHz band, 20 MHz bandwidth
Switching back to the default Auto mode resulted in the 600N's using a 40 MHz channel bandwidth, with a subsequent improvement of total up and downlink throughput to 110 Mbps (Figure 16). Unfortunately, you can't exceed the 100 Mbps mark running up or downlink alone, even using bonded channels. Running up and downlink separately yielded results of 73 and 85 Mbps respectively.
Figure 16: Up and downlink throughput - 2.4 GHz band, 40 MHz bandwidth
Maximum Throughput - 5 GHz
I next checked throughput in the 5 GHz band. Like the Apple [reviewed] and Buffalo [reviewed] dual-band routers, the 600N defaults to using 40 MHz channel bandwidth (actually, its Auto mode), so that's what I tested first. The total throughput of 112 Mbps is essentially the same as achieved with channel-bonded 2.4 GHz operation, which is nice to see (Figure 17).
But, as in 2.4 GHz, you can't exceed 100 Mbps running up or downlink only, which clocked in at 80 and 83 Mbps respectively. Note that the average throughput would have been closer to 120 Mbps, if not for the multi-second glitch around 15 seconds into the test.
Figure 17: Up and downlink throughput - 5 GHz band, 40 MHz bandwidth
Shifting down to 20 MHz bandwidth mode, total up/down throughput drops to around 82 Mbps (Figure 18), which is virtually identical to the 2.4 GHz, 20 MHz channel results. Throughput is a little choppier, which appears to come from the uplink side. Running up and downlink only turned in 1 minute average throughputs of 66 and 72 Mbps, respectively.
Figure 18: Up and downlink throughput - 5 GHz band, 20 MHz bandwidth
Security mode throughput
The 600N did a bit better than most Draft 11n routers in terms of throughput loss when using WEP or WPA-TKIP wireless security. Figure 19, taken with the 2.4 GHz radio running uplink in 20 MHz channel mode, shows 60% and 53% losses respectively in downlink tests.
While this is better than the 90% WEP throughput loss seen in Atheros XSPAN-based products such as the Apple Airport Extreme, it's not as good as the 30% WEP loss I found in the Marvell TopDog-based Netgear RangeMax NEXT Wireless-N Router - Gigabit Edition [reviewed].
Figure 19: Security mode throughput comparison - uplink
Downlink loss is a bit better for WEP 128, which comes in closer to the 50% WPA-TKIP loss (Figure 20).
Figure 20: Security mode throughput comparison - downlink
The good news is that WPA2/AES showed no significant throughput loss in both up and downlink modes. So, once again, it's the wireless security mode you should be running if maximum throughput matters.