

Now we finish where we started, with Linksys stealing the show and ZyXEL putting in a strong bid for runner-up. Netgear’s WNDR3700 practically drops off the map here. And what happened to D-Link? Well, we ran into a bizarre little software glitch. For some reason, PerformanceTest reported back D-Link UDP results well above 2,000 Mb/s, which is obviously impossible. Along with these stratospheric scores, CPU utilization rocketed above 80 percent. No other router exhibited this oddity, and we reran the test several times, always getting similar numbers. D-Link claims it has been unable to replicate our results in its own lab. Anyway, we threw the numbers out as obviously being meaningless. In the 5 GHz scores, only Netgear’s repeated client adapter lethargy comes as a bit unexpected after some of its other excellent scores under Zap.


Linksys again shows how to do it up right with its terrific CPU efficiency, although Ruckus just might deliver more bang for the processor cycle buck. Either way, you can’t go wrong.
- Router Reignition
- Asus RT-N13U And RT-N16
- Belkin N150 And N1 Vision
- D-Link DIR-685
- Linksys WRT610N
- Netgear WRN2000v2 And WNDR3700
- Ruckus Wireless 7811
- TP-Link WR741ND
- TRENDnet TEW-654TR And TEW-671BR
- ZyXEL X550N
- How We Tested
- Benchmark Results: 1GB Transfer, Many Files
- Benchmark Results: 1GB Transfer, Single File
- Benchmark Results: IxChariot Throughput
- Benchmark Results: IxChariot Response Time
- Benchmark Results: Zap TCP
- Benchmark Results: Zap UDP
- Benchmark Results: PerformanceTest TCP
- Benchmark Results: PerformanceTest UDP
- Conclusion
Good point.
Which firmware was installed on it?
I have one (V1), but am very unhappy about the signal range! I have it replaced with a WNDR3700 and have now a twice as strong signal as before!
Bit the bullet with the $$ and opted for the Linksys and am very pleased.
pato, my WRT600N was the v1 variant. I forget the release version of the firmware, but it was the latest version, as Linksys has not released any updates for it in roughly a year (I've had the router since a few months after it was first released). I liked it due to the dual radios, however, but it would drop wireless clients randomly (which was aggravating and required me to reset the router about once every other month) and it would not retain my port forwarding settings for my home server. And I agree with you, signal range was marginal with that router.
2) Should have tested N + G concurrency on 2.4GHz as well as N only on 2.4 + 5GHz concurrency (for devices that had dual radio). This data is important for most people who will run a Wireless N device or two, but likely also have a few smart phones or a game console that only supports 2.4GHz... I know the Airport Extreme currently has a bug making this dog slow, do some of the others?
3) onyl 2 concurrent devices? how about 5 or 6? I regularly have 7 or 8. I notice performance drops off consistently just based on the number of connected devices, even if only one is "in use" actively downloading, and want to know if some routers hold out better with that.
4) no feature comparison chart?
How much did you guys test the shareport function? (Not much from what it looks like). The shareport function hooked up to an external hard drive only works if you are transferring a file or two using windows. It totally fails when you try to us it with a 3rd party backup program (such as acrea). I personally haven't tried connecting a printer to shareport. I also couldn't get it to work using eraser (a disk erasing utility. I concluded that it just doesn't work with third party apps. So far, none of the driver, firmware or shareport software updates have fixed this problem.
D-link does have a nice forum on their site where people can post their problems. For fixable problems, other users will helpfully solve your problems. For unsolvable problems (shareport being one of them) the user complaints just keep piling up. Rarely do d-link's own tech support grace the forums. Apparently, D-link is currently collecting all of the shareport grips and cataloging them. Ostensibly, this will result in a fix at some point in the future. Still waiting.