Datalink wrote:
Absolutely agree with VivienM. I think we are going to see more issues like this in this forum, which is definitely ok, as there are people here who can help. Prior to posting for help however, have a very close look at every piece of equipment in your network and try to determine if in fact they will support the data rate you believe you should be seeing. That entails everything from cable specs, to switch specs to WiFi modes of operation and also computer and/ or laptop ethernet or WiFi specs. I suspect that carefully examining each network segment and it's spec might in fact answer the question of why the data rate that you are seeing isn't as fast as it should be. Also think about the possibility of a bad cable or bad connection, and look for ways to test each segment, starting with the CGN3, and working outwards, towards the other end of the circuit, adding one segment at a time. When all else fails, send up the white flag and post for help, including as much info about the network circuit as possible so that others can help troubleshoot.
I agree with all of this, but I also want to add one other point: real estate.
For example, at my parents' condo, there is a bathroom between the router and my mom's equipment. And I believe those walls to be concrete. So she gets an awful wifi signal - with 2.4GHz equipment, she was maybe lucky to get about 100KB/sec (so 800 kilobits) Internet download speeds. With a CGN2 it was more like zero. I tried weird powerline access points too, that improved things for a while, but then that gear got D-Link disease and became flakey. An additional part of the problem was that the bathroom weakened their 2.4GHz signal enough that the neighbours' network actually had comparable or higher signal strength.
I finally 'fixed' the problem with brute force - an Asus RT-AC68U on one end and her new retina MacBook Pro on the other. The laptop reports a 351 megabit/sec link rate. Since that gear, if it was next to each other, could do 1300 megabits/sec, that means the walls are eating over 70% of the theoretical performance. But luckily, the 27% she does get is more than enough for networked backups and their 30 megabit/sec Rogers service...
27% with a lot of cheaper gear, though, would turn you (or Rogers) into an embarassing Netflix performance statistic.