Actually I haven't mentioned this here, so I'll add it. One thing I really don't like about the 5a is the bandwidth between the WiFi and the Ethernet. Ethernet-Ethernet via the switch does seem good, although it shows up in the hub in an odd way which I think reflects the mix of on and off-chip Ports. It's possible there are some restrictions on or limitations with what you plug in where of 100 and 1G connections (or it may just be the way the user interface reports stuff).
When I moved my NAS onto a HH5a Ethernet port my read and write speeds via WiFi basically went to a third to a quarter of what I had previously. That's with a 975Mbps WiFi connection and full-duplex 1G Ethernet to the NAS. Writing to it at 12MBps isn't impressing me. However reads are twice that so it isn't a 100Mbps limit (reads via Ethernet were 70+MBps and still are via an Ethernet connected computer). It might just be a lack of buffering.
However £49 for a 802.11ac router with built-in modem and a gig ethernet switch is quite a deal. Enough that I'm willing to give it a good trial, especially as the HH3 wasn't too bad. I also like that it does Port Forwarding by host name rather than IP address and you can trivially attach names to IPs yourself. My next choice would have been the Draytek 2860ac as I had good experiences with them in the past (then recommended them to some colleagues who didn't, as they were reminding me at the Pub yesterday, such is life).
Someone at Draytek suggested the throughput on 11ac was a long way below the connection speed, but I'm pretty sure it's not the main cause. (I stopped designing this sort of stuff a while back so don't really know.) He suggested as low as a third, which for my 975Mbps would give 40MBps. That would still be a nice bump over my current write speed to the NAS. If the 5b is significantly faster it'll probably be done and dusted. Alas I can't find much on the Ethernet architecture inside the Broadcom chip.
Author
Topic: HH5a and HH5b firmware/general reliability comparison (Read 20340 times)