Unencrypted Throughput - Client to Server:

Our throughput test compares the performance of our D-Link DIR-625 against the three Draft N routers and the NetGear RangeMax 240 based upon Pre-N MIMO technology. This router features the proprietary third generation Airgo chipset and to date this technology has held the lead in wireless throughput and distance. Belkin has chosen not to support 802.11n or 802.11g only transmission modes so we tested the N1 router with the wireless option set to mixed n/g/b mode. In the test results we are reporting today there are not any 802.11b or 802.11g compliant devices active so the scores should be comparable. In our next series of articles we will look at true mixed mode results along with compatibility between our 802.11n Draft 1.0 devices and other wireless standards.




The NetGear WPNT834 Pre-N router has the highest throughput overall and continues to show a very consistent throughput pattern up to 120 feet. The Belkin N1 and D-Link DIR-625 excel at 40 and 80 feet in maximum throughput compared to our other Draft N routers. At 120 feet the throughput of the Atheros XSpan chipset equipped Belkin and D-Link routers fall off considerably. The NetGear WNR834B still provides the most consistent throughput across the full test spectrum of the Draft N routers.

Encrypted Throughput (WPA2-PSK) - Client to Server:




We spent an inordinate amount of time testing our routers and clients with WPA2-PSK security enabled as the test results were not always consistent or complete. After the final firmware and driver updates we were able to consistently achieve the results shown. However, at first glance you will notice that our encrypted results in the average throughput test are higher than the unencrypted results at 10 feet. We already know that 802.11n Draft 1.0 specifies WPA2 security protocol as the preferred method of encrypted transmission and it appears the hardware is optimized for it at close distances. In order to reach the throughput results we reported your network will have to be free of legacy devices and utilizing the recommended wireless adaptor for the router.

Unencrypted versus Encrypted Throughput
Maximum Mbps - Percentage Difference
Client to Server 10 Feet 40 Feet 80 Feet 120 Feet
D-Link DIR-625 -18.24% -9.94% -6.25% -27.03%
Belkin N1 -18.13% -15.41% -6.18% -16.28%
NetGear WNR834B -9.91% 8.29% -14.19% -6.58%
Linksys WRT-300N -9.25% -7.80% -41.61% -18.34%
NetGear WPNT834 (Pre-N) -9.55% -14.19% -11.56% -13.70%
.

Unencrypted versus Encrypted Throughput
Average Mbps - Percentage Difference
Client to Server 10 Feet 40 Feet 80 Feet 120 Feet
D-Link DIR-625 4.04% -53.85% -51.04% -58.11%
Belkin N1 4.23% -62.80% -54.64% -25.00%
NetGear WNR834B 4.92% -6.74% -25.82% -11.81%
Linksys WRT-300N 13.17% -18.18% -47.35% -10.84%
NetGear WPNT834 (Pre-N) -1.79% -5.36% -5.48% -9.15%
.

While the 10 foot distance shows very little loss and actually gains for most of the routers we see an issue with the Belkin N1 and D-Link suffering an 18% loss with encryption enabled in the maximum throughput results. This is not surprising as both units are based on the same chipset. On our 40 foot test we see the D-Link DIR-625 experiencing a 53% performance hit in the average throughput numbers and 51% at 80 feet. Our Linksys router after the last firmware update still has a performance issue with WPA2-PSK enabled at the 80 foot range. Once again the NetGear WPNT834 offers the most consistent performance.

Test Setup Server to Client Performance
Comments Locked

15 Comments

View All Comments

  • yyrkoon - Friday, September 22, 2006 - link

    How you got 'overheating NICs' out of what I said, is beyond me. As for routing overhead. the PCI version of the intel Pro GbE NIC, was second ONLY to its onboard counterpart last I checked (which WAS about a year ago).

    Anyhow, thats only the LAN portion of said router anyhow, it seems its the wireles portion of these devices that is severly lacking.
  • Beenthere - Wednesday, September 20, 2006 - link

    Based on D-Link's history of shoddy products and I have personal experience that supports this, D-Link's Non-existent Customer and Tech Support, (sic) and I have proof of this, and based on class action lawsuits against their defective products, I couldn't in good conscience recommend any D-Link products to anyone. D-Link may not even be around when the dust settles on law suits?

    It's no surprise their product didn't perform as advertised. I wouldn't be at all surprised if the FTC files a class action lawsuit for fraud on this product too.
  • seabat - Wednesday, September 20, 2006 - link

    I couldn't agree more.

    I have had the misfortune of using D-Link products at home and work, in both situations their products have let me down. My company recently spent 1000's changing over 80 D-Link switches because of reliability/stability issues.
  • drebo - Wednesday, September 20, 2006 - link

    We've had good success with D-Link's print servers, but I don't really like their routers all that well.

    Just personal preference, though, of course.
  • TowerShield - Wednesday, September 20, 2006 - link

    Of course, all draft-N routers from every manufacturer so far have basically been crap, so might as well file a class lawsuit against everyone if you think that is justification.
    (Can only think of one decent D-Link, that's the DGL-4300).

Log in

Don't have an account? Sign up now