Elite Dangerous Unplayable with 4G and 5G connections

IP Address Changed. Disconnected. When I connect with port redirection (with mac address fixed to a dynamic address allocation) I get the error resource could not be allocated.
So all combat logging on my account is your fault Frontier. Fix your code.
 
My connection details. Always 75%
 

Attachments

  • net.png
    net.png
    36.6 KB · Views: 375
  • net2.png
    net2.png
    15.6 KB · Views: 354
This is the error
The whole point of port redirection is there to overcome this problem

The final piece of the puzzle,...my port redirection

Expecting to use phone level internet seems extravagant to me.
Expecting to get fibre in Wales when most cannot even get copper wire is wishful thinking.
 

Attachments

  • net3.png
    net3.png
    383.6 KB · Views: 387
  • net4.png
    net4.png
    12.6 KB · Views: 314
Last edited:
4G based connections often suffer from CGNAT issues.

Your router is essentially part of a hive and there is a specific server that sorts your connection to whichever IP adress is required, this however means as you don't have your own public IP, your router wont show up as a single connection to another, resulting in an inability to host or play with large groups of people.

I'm unsure as to what the limitations of this is as some people can connect consistently and seemingly on opposite sides of the earth, but sometimes people comparitivley just down the road cannot.

Mobile communications infastructure at its finest.
 
Maybe you should get in touch with tech support instead of posting it as a suggestion. They can either help you out setting things up or maybe someone is willing to put this on their to-do list. Chances that this discussion will draw their attention is almost zero.
 
Maybe you should get in touch with tech support instead of posting it as a suggestion. They can either help you out setting things up or maybe someone is willing to put this on their to-do list. Chances that this discussion will draw their attention is almost zero.
If a thing is not supported in the first place, how can it be considered to be a bug?
 
Research port redirection. Frontier attempted to implement it, but failed.
This problem occurs every hour on the hour. It is as regular as clockwork.

AFAIK thats still a router based issue but that could go a long way to explain why instancing goes kaput fairly often, but that, in a lot of cases is more of a ten minute thing, then it picks up again.

I'll have a look now, nice one.
 
AFAIK thats still a router based issue but that could go a long way to explain why instancing goes kaput fairly often, but that, in a lot of cases is more of a ten minute thing, then it picks up again.

I'll have a look now, nice one.
I have over 35 years in real time financial services software development experience using c/c++. I've written Inmarsat LES / MES test harnesses. I've written real time servers for the city. I understand communications and real time. I worked for Arcontech (15 years) and Thomson/Reuters (10 years) most of my life. This problem exists in Frontier's TCP/IP library module and is most likely to do with retry attempts or bad timing code. Could be a low retry count. Their development department needs to be more aware of the variable timing of going from cell to cell, especially during busy times. Also they may need to consider packet bit-loss repair to help improve reliability, as most satellite tech uses these days. A data packet windowed approach would help allowing for flaws and faults to be automatically repaired without the need to re-transmit data already received. UDP is not reliable, but you can program around that.
They may also need to implement throttling which is useful in high volume times. In real time if new data arrives while old data did not it's ok to say the new data can replace the missing data re-request. This means anything queued to go out can be deleted from the out queue as it is no longer needed.
Finally, and most significantly, they did not implement a distribution layer on the client, which is needed for P2P.
 
Last edited:
I have over 35 years in real time financial services software development experience using c/c++. I've written Inmarsat LES / MES test harnesses. I've written real time servers for the city. I understand communications and real time. I worked for Arcontech (15 years) and Thomson/Reuters (10 years) most of my life. This problem exists in Frontier's TCP/IP library module and is most likely to do with retry attempts or bad timing code. Could be a low retry count. Their development department needs to be more aware of the variable timing of going from cell to cell, especially during busy times. Also they may need to consider packet bit-loss repair to help improve reliability, as most satellite tech uses these days. A data packet windowed approach would help allowing for flaws and faults to be automatically repaired without the need to re-transmit data already received.

EDIT. 35 years of IT experience - Buys 4G router to play online games. For real.


So this beggars the question, if such a thing can be diagnosed so easily then why have they not employed anyone capable of dealing with it? For goodness sake they were looking for a network engineer a year or so back.
 
Last edited:
Again, get in touch with their tech support. It's not about this being a bug, it's about finding someone who is willing to work on this.

If you want something like this to change, you need to get it on their radar. If it's logged in their system that tracks tech issues, it's more likely to get attention.

It's clearly a technical issue that prevents you from enjoying the game - so let them treat it as one.

You won't get anywhere with this post. Unless you just want to vent, in which case this is the perfect place.
 
Back
Top Bottom