01-21-2012 06:16 AM
I purchased a Cisco E2500 router during the Christmas holidays (2011) to upgrade an existing Linksys WRT54GS (802.11g capable) router. The following notes express my candid experience with the E2500:
1. Setup using a Windows-Vista PC took me 2 hours because Cisco Connect would not finish installing despite following every step meticulously. After 3 attempts of uninstalling and re-installing, I switched to a newer Windows-7 laptop - success at the first attempt. My immediate reaction at that time was less of frustration and more of a bewilderement - there is NO reason for the setup to not finish installation on a Vista machine. Routers are seemingly straightforward products so, in theory, setup should be a 10-15 minute affair - anything more and the product's designers have got it wrong.
2. We live in an average-sized house in San Jose (2400 sq.feet). My prior experience with Linksys WRT54GS router had been positive w.r.t. signal strength and reliability. My bewilderement turned to surprise when I realized that the E2500 began to drop connections on a 802.11 b/g/n link everytime someone turned on the microwave oven in the kitchen. Huh? Having spent the past 18 years of my professional life with signal processing and modem stuff, I understand the noise levels emanating from the household microwave - but, would it be so bad to drop the connection?
3. Determined to resolve the microwave problem, I turned on the WLAN utility on my Windows-7 laptop; this utility (standard on most newer machines) measures the SNR of the different WiFi routers that it can reliably measure. I ensured that all the variables impacting wireless broadcast remained fixed - no change to the walls of the house, fixed spot for the E2500 router, no additional shielding for the microwave in the kitchen, fixed spot for my Windows-7 laptop, etc. Observations:
a) Signal strength of the E2500 in 802.11 b/g/n was approximately -68 dbm
b) Yes, the microwave did produce noise which impacted the SNR (signal to noise ratio) somewhat on the end-point but not substantially to completely blanket out the signal
c) Signal strength of the E2500 in 802.11 a/n (which is usually used in high-noise environments) was approximately -80dbm.
d) Ambient noise (without the microwave oven or any other household appliance turned on) was approximately -90 to -91 dbm.
e) I played around with the different channel settings on the router, determined to find a channel where the router would not drop the connection. After all, can you imagine if I have to instruct my family to use the microwave only at specific times of the day when my laptop was switched off? No luck in finding a channel that would hold the connection.
Multi-band routers like the E2500 are touted for their newer antennae (and I understand that antenna layout on the PCB is a very specialized art) but my measurements seemed to indicate that either the product was defective or that Cisco had done a very job of designing wireless performance of this product.
2 days back, the E2500 router began to slow down (browsers slowed down, Outlook performance became sluggish). And then it stopped working. My immediate reaction was that Comcast's Cable modem service had suffered an outage - after 10 minutes, I realized that it was the E2500.
We went back to the good ol' WRT54GS product. My technical curiosity got the better of me so I decided to run the same tests using my Windows-7 laptop. I ensured that the WRT54GS went back to the same spot in the house which was recently occupied by the E2500. I ensured that all the other parameters remained fixed. Observations:
1. Signal strength of the WRT54GS at the end-point was -57dbm. How can that be, I wondered. This router is so old (we've had it for about 4-5 years) so surely, there must be a mistake.
2. I turned on the microwave oven in the kitchen. Same spike in ambient noise. But the router held the connection steady.
The E2500 is being returned. I am hoping that Cisco will replace it with a good one. Until then, the WRT54GS will have to do.
I am disappointed with the quality of the newer products - lots of technical hype about performance and features. But, I wonder if corners are being cut to save that extra dime. I wonder if not enough is being done to test the performance of the product.
I am disappointed with the E2500. Sleek looks can never replace reliability of a product, even if it comes with an un-attractive pair of rabbit-ear antennae.
01-22-2012 02:19 PM
The performance of your new router depends on various factors apart from the router itself. The wireless devices in your network, are they wireless G or wireless N devices? If they're wireless G devices, you won't really see any difference, and in fact any N devices in your network would also suffer because of that. Apart from that, placement of the router also matters. Avoid keeping the router in a corner or under the table or something like that. Ideally you'd want to keep it at a higher location. The signal always travels better from a higher location to a lower location. Apart from this, you can also change channels on the router. Typically, wireless interference can cause poor signal strength and speed. Changing channels helps alleviate this problem to some extent. I'd suggest trying channels 1, 3, 6 or 9. You'll have to try each and observe which channel gives you the best performance. Every network environment is different, so what works for one network may not necessarily work for another. Please let me know if you need any further help with this.
If you are installing the router with the help of the Cisco Connect Software make sure of the followings steps:
• The router’s power adapter needs to be plugged to your router and into an available power outlet.
• If using a wired connection, the first Ethernet cable needs to be plugged to the computer’s LAN port into any of the router’s blue numbered Ethernet ports.
• The second Ethernet cable should be plugged to the modem’s Internet port and the other end into the router’s yellow Internet port.
• Ensure that you have no firewall enabled that may prevent from being detected.