Since my primary router lost programming, I reset the following features as suggested.
- 2.4 & 5GHz set to "n" only (no "a", "g", or combo).
- WPA-2 with AES only (ok for a test but not acceptable in real life since cell phones and tablets will not work with this encryption scheme).
- Within 10ft of the router (again, ok for a test but impractical in real life).
- Use 5GHZ band only.
- Only one device on the 5GHz wireless band (again, ok for the test but impractical in real life).
With this setup, I did note an improvement in the transfer rate of my test iso file. Where previously I saw a maximum transfer speed of 7.2MB/sec, I now measured a maximum speed of 9.8MB/sec. Where I previously measured an average sustained transfer speed of 4.5MB/sec, it increased to about 7.0MB/sec. This IS as significant increase but at the cost of loss of network functionality and practicality.
I have read other articles and posts about different "n" routers with much higher transfer wireless speeds. While taking them with a grain of salt, the sheer number of posts claiming speeds 30% -50% higher than my measured speeds has to be acknowledged.
I am also curious as to whether anyone has done a calculation of the realistic expected transfer speeds possible based on bandwidth characteristics of the router, signal strength, distance, and antenna configuration. It would seem that DLink engineers would have done these calculations and real life measurements. DLink makes all kinds of routers, AP's, signal boosters, switches, and accoutrements for wireless networks. They needed to do testing when changing their designs. Is that information available somewhere in a white paper to help me benchmark my stuff? All the literature on the DLink website is marketing hype with no real data to support their claims. Not that DLink is different from any other WiFi router mfg. This is an industry wide problem.