This is still an ongoing issue just FYI if keeping up with this thread. I have a thought to why this is happening, but I think I am close to finding proof to my idea.
I believe this is solely on C.O.X to fix.
So I have both a Mac and Windows computer I can run some tracert/traceroutes from to the servers and the reason C.O.X will push this issue off as not their problem is because they say once the data leaves their server it is not their problem any more, but the only data you can really provide from Windows is a simply tracert which will use ICMP data. If you run these tracert using ICMP data there is no data loss from you to the servers and this is what C.O.X gets sent by most people experiencing this issue and they will just dismiss it. If you have a Mac or a Linux machine though you can run a traceroute with the specified data type as UDP. If ran in succession with the amount of data being 60 packets per second and with UDP after running the command about 4x you will see in one of those a dropped packet or no response for the latency of how long the data took to send. These drops are happening BEFORE they reach the C.O.X server that is sending the data off to Amazons servers. This is after testing this just a little bit ago and want to run some more tests. If this is the case it will be on C.O.X to fix this issue. A couple of thoughts to why this is happening now and did not happen before June of 2018. Cox is has probably changed the way they treat UDP data possibly, only problem is no other online games ex. COD BOPS4 seem to have this packet loss issue unless they are, but you can not actually tell or you do not get that information from the network information section of COD BOPS4. It could be possible that the game itself is set to only send out so much information going up that the data is under the capped rate C.O.X is setting on their networks to have going out. If you notice when you connect to a VPN service and run the game through this your packets per second going up change to a much lower number and you receive no loss, just higher ping. Also if you notice streamers pushing over 100 packets per second going up to the Fortnite servers have no problem with packet loss, so I do not believe this is an issue on Amazon's servers.
In conclusion, back in June of 2018 C.O.X probably changed something to how UDP data is being handled when going out.
If that is not the case then I believe around that time Fortnite went from 30mhz to 60mhz on their servers and I m not sure if that means at what rate data is being sent out, but I believe it does and that would explain why this is when this all started to happen. I also think around that time is when we received the ability to see the network debugging statistics on Fortnite and it could have been happening before this, but I doubt it since the blank shotgun hits were not always happening.
Anyways that's the ideas I have came up with after looking from all the different forums from C.O.X and Fortnite. I will be running some more tests to see if I can nail it down to only Amazon servers receiving the packet loss on upload or if it is any other online games or sites. If its any site receiving packet loss like this and the results still show the drops happening before C.O.X server it is on them and I will be giving their techs all this information which if they try denying that then it is to the FCC to file claims.
If anyone else has a mac or linux machine they can run some tests with also, and come up with the same findings please call C.O.X tech support and lay into them to have the issue fixed and if they do not file FCC complaints.
It is time this is fixed.
I noticed this began around the same time you mentioned. I moved in June and assumed it was my new address. I've moved again since then and its still an issue so i believe you're onto something with that June date. Game is horrible you can't even play. Cox is garbage for allowing this and dodging the problem.
