Arcadyan KVD21 ethernet

I have the same issue.  when I am on band n41, I can get 170-200mbps download when doing a speedtest over wifi with my laptop.  If I run the same test over ethernet with the same laptop, I get 90-110.  

I’m seeing the same issue. Any updates?

jallain wrote:

I have the Arcadyan gateway whic h seems to work ok for the most part. My issue is the ethernet speed compared to the wireless speed. With wireless I usually get over 100MBS. With ethernet usually the best I can do is rougly 20MBS. No way should wired be slower than wireless. I know it is not my computer as I have not yet cancelled my Xfinity service and I get around 920MBS with Xfinity. Is it the hardware in the Arcadyan or the firmware that T-Mobile is using? And what would be the advantage of slowing down the ethernet ports? I, like a lot of people, only want to plug my existing router into the gateway so I don't have to reconfigure my home network. T-Mobile support says the only option at this time is a replacement with another Arcadyan. I would like to try the Sagemcom but right now that doesn't seem to be an option. Thanks for any anelp.

You say that "T-Mobile support says the only option at this time is a replacement with another Arcadyan". How would that fix the issue would they be replacing it because your unit is faulty? Or would a new unit have some unlocked or different software? Upon speaking to them and telling them what your end goal which was. And correct me if I'm wrong,  to use your own router essentially making the gateway the modem (only Nokia can turn off Wi-Fi therefore modem in the sense of the word that's what one wishes to achieve with Arcadyan device although technically isn't purely possible). I'm just trying to understand what a new u it would do that your current wouldn't. And is Sagemcom a different gateway T-Mobile offers is it a newer device or was it among the first or is that the Nokia gateway?

Just a thought. I am with you guys on the Ethernet port performance being poor. Just to have another data point. If you run a local client to client performance test say with iperf it might be insightful to know if the transfer between two wired clients behaves the same or better/worse than running a speed test with the client conducting the speedtest out across the internet to a server. 

https://sourceforge.net/projects/iperf/

T-Mobile claims the three 5G gateways are intended to be equivalent but well in a twisted sort of way maybe but maybe not. I run a Nokia and have since the BETA period and it performs very well with wired or wireless testing. I have not used the Arcadyan or the Sagemcon so I don't know how much different they actually perform vs the Nokia.

You could run the command netstat -s before and after the test to see if there are errors or odd statistics that stand out from the before and after results. Just a thought. I am not sure how much is possible to get from the Arcadyan gateway in terms of statistics. 

Be sure when you are running the speedtest that you have the preferred server set so that the target server does not change from test to test. Try to reduce the possible variables.

I have the Arcadyan and using my own Asus mesh router connected to the gigabit port. No issues with wired bandwidth. The only difference in my case is that I connect to the WAN port on my mesh router.

I have 2 gigabit switches, one on each port on my Nokia and I see no bandwidth issues. It works just fine. 

For those seeing performance on ethernet equivalent or better than wifi as you would expect what firmware are you on? Per https://s.tmocache.com/support/devices/firmware-updates--t-mobile-home-internet there is currently 1.00.14 and 1.00.16.