Well, my computers all access my 5N via a gigabit Ethernet switch, which I believe is how it was designed to be used and doesn’t defeat its purpose at all. WiFi is a very poor substitute for a cabled connection, being intrinsically slower and subject to interference that slows it further. The only thing it has going for it is the fact that it’s sometimes a bit more convenient than having cables trailing across the floor.
Here’s my suggestion. Try it with cable and see what it’s capable of. Then, if you insist of going the WiFi route, get some appropriate software (I use WiFi Explorer from the Mac App Store) and do a survey. Choose the least congested 40 MHz of contiguous spectrum (that’s double the usual 20 MHz channel bandwidth and remember that each channel overlaps its nearest four neighbours) and use that. If you choose channels 1 and 6, for example, you’d better hope that everyone else is using channel 11, because there isn’t a lot else in the 2.4 GHz band. Use the 5 GHz band if your equipment supports it as it’s generally less crowded but has more difficulty penetrating obstacles, such as walls. The long and short of it is that WiFi is nowhere near as good as cabled Ethernet.
Are you sure you’re not confusing Mb/s (megabits per second) with MB/s (megabytes per second) in your comparisons?
Right. The consensus is that “N” wireless generaly sucks. Go for an “AC” router, that can deliver gigabit speeds (that’s if your computer supports AC). In my case, I get same speed over wireless or wired, using an AC1750 router. The D-Link Wireless AC1750 router (DIR-868L) is a good option.
First I would test the Drobo via a wired cable, and see if the speeds is reasonable for you. Drobo’s are generally ‘slow’ so you need to see if it works for you before investing in additional hardware.
Nah - my Drobo 5n isn’t slow at all. I have the slowest possible ethernet connection from my systems to the Drobo (I’m talking using an $8 10/100 switch to connect everything). I easily do 80MB/sec transfers to it. Heck, I can simultaneously decode two 1080p/6chan streams from my Plex Media Server to it and feed them to two different wireless Rokus without a hiccup.
The “suckiness” isn’t to do with the IEEE 802.11n specification. It’s to do with the fact that the 2.4 GHz band is overcrowded and full of noise, plus the fact that a wireless access point behaves like an Ethernet hub, not an Ethernet switch, so all clients of that hub share its bandwidth. IEEE 802.11n specifies two chunks of spectrum. The other is in the 5 GHz band but not all 802.11n equipment supports both bands. The OP is lucky. His does and he’s happy with the solution.
IEEE 802.11ac is still inferior to a cabled connection because of that fact that cabled gigabit Ethernet is invariably switched and any pair of hosts connected to the switch have a full duplex 1 Gb/s path through that switch. So a decent (i.e. non-blocking) 16-port switch can support eight pairs of hosts each communicating at 1 Gb/s full duplex, simultaneously. That’s a total bandwidth of 16 Gb/s. Whatever throughput an 802.11ac access point can provide is shared by its clients.[hr]
You’re confusing bits (b) and bytes (B). You can’t get 80 MB/s through a 100 Mb/s switch because 80 MB/s = 640 Mb/s.
And? There’s no denying the laws of physics, my friend. I suggest you look at your cabling. Maybe you have a direct connection or your switch is actually capable of gigabit/s speeds but you simply can’t get 71.1 megabytes per second through a 100 megabits per second (= 12.5 megabytes per second) switch.