No Thargoids, No Rebalance. How about...

Fixing player instancing? It's been a problem for over a year.

Fix what is broke before breaking things to fix. How bout it?
 
Networking issues in a p2p are not easy to fix unless issues are reported well. If submitting a ticket you should send logs from all affected players wherever possible and I would bet money this almost never happens.

In a standard server + client model you slap logging on the server and on the client. Somebody reports a problem and submits logs you immediately grab the server logs for the same time and you have a complete picture of what happened and what went wrong. You can then (usually) implement a fix.

In p2p environments much of the matchmaking is invisible to the server so you get a DC and submit your logs expecting FD to go "oh, I see what went wrong" and implement a fix. The problem is that FD only have half of the networking "conversation" in your logs and all they know is a connection failed with no visible cause, one end simply didn't answer. To give them both ends of the conversation they need logs from 2 people at least. If you are in a wing and someone DCs get all members of the wing to submit their logs on the one ticket. That way FD can see that CMDR A tried to connect to CMDR B using X details but CMDR B was actually listening for connections using Y details.
 
They need to make it quick. 12+ months on a bug affecting multiplayer in a multiplayer game is.... sad.

That's a classic:

"Hey FDEV, stop working on everything else and fix the network! Even if the people who are working on something else don't know anything about network!"
"They are currently working on it, we'll see some fixes soon."
"They need to make it quick. 12+ months on a bug affecting multiplayer in a multiplayer game is.... sad."

How about: "Thanks, this is good news!" or "Looks like this thread is unnecessary, please close"?
 
So how would you advise they jump to the conclusion of an iterative process? I'm thinking of a number between 100 and 0, work out where it is and I just give higher or lower values. So you go 50, I say lower -> you go 25, I say higher -> You say 37 I say lower etc till you arrive at 28.
What you are asking is Frontier to do the first step or two then immediately arrive at the conclusion 28.

Make fix
Test fix as much as possible in house. Try to ensure it won't cause more issues than it resolves.
Rollout fix
Gather data
Go back to step 1

It's impossible to do quickly given the nature. Lots of the networking issues are hardware related. I suffered greatly at the hands of a student router where I had no access to the settings. Getting my own router now I have a job and such It increased from maybe 1-2 people best case at CG's to 10-15 on average! As a relatively active fuel rat I can confirm things have changed massively in the past 3 months making the game a ton more stable. No more do you crash out to menu when hyperspacing in a wing. I could go on.

One key thing to remember is whilst FDev have an array of hardware to test fixes and patches they do not have anything that can accurately simulate users with hundreds of thousands of different networking hardware and settings configurations talking to literally anybody in the world like Australia etc.

The standard port forwarding and UPNP advice applies. Make sure it's not your end. Submit the data to help FDev and watch it improve. It isn't an instant fix.
[video=youtube;EvJPyjmfdz0]https://www.youtube.com/watch?v=EvJPyjmfdz0[/video]

From the Devs themselves, I understand bits (I'm an engineer not an network specialist) but that doesn't even talk about what I was on about with routers, adapters and settings and stuff but you can see the complexity just by watching 10 minutes in. The standard bug may be 3 or 4 lines of code that are bad in the 3 million lines of your game. This right here is like having millions of different games each with their own very different potential issues.
 
Last edited:
Fixing player instancing? It's been a problem for over a year.

Fix what is broke before breaking things to fix. How bout it?
+ 1 ^

my thoughts exactly when 'the news broke'. some dude gets pulled out by a flying sunflower, everyone freaks out.

yet we are still dealing with tons of broken things ruining the game.
 
Last edited by a moderator:
+ 1 ^

my thoughts exactly when 'the news broke'. some dude gets pulled out by a flying sunflower, everyone freaks out.

yet we are still dealing with tons of broken things ruining the game.

it wont happen, this is why new content is created, to hide the stuff they cannot fix.
 
Last edited by a moderator:
8 CMDRs, same minor faction, in open. No Instance. Working as intended?

Of course not. And while like most people, we all wish FD could have fixed these issues a long time ago, without knowing the inside details and a whole lot of technical knowledge, we don't know why it has taken so long. It might not just be a matter of priorities.

Its one thing to report an issue, its another thing to recreate those issues on a dev box in a standardized environment, or figure out why it works ok on your system but not on another.
 
8 CMDRs, same minor faction, in open. No Instance. Working as intended?

All 8 of you in different instances?
And the minor faction bit is irrelevant. Your router model, ISP and geographical location might help though!
 
Last edited:
it shouldnt be a problem in a game where multiplayer is such an important part. would love to be in the meeting at fdev when they decided ED should have p2p.
 
it shouldnt be a problem in a game where multiplayer is such an important part. would love to be in the meeting at fdev when they decided ED should have p2p.

ED is a game where everyone online having their own unique instance is common. You cannot have dedicated servers run 20,000 unique instances concurrently without a subscription fee. P2P is a requirement for any game this size. There wasn't even a meeting because the P2P requirement is quite obvious from the start.
 
Last edited:
so we should just accept an bad gameplay experience? all the problems with instancing are typical for p2p gameplay, so I highly doubt it they can fix it without gutting the complete p2p infrastructure.
 
Not necessarily fixed but understood and addressed. The thread on the beta forums suggest that they do know what they are doing and as anxious as we are to avoid bad gameplay experience
 
so we should just accept an bad gameplay experience? all the problems with instancing are typical for p2p gameplay, so I highly doubt it they can fix it without gutting the complete p2p infrastructure.

Wrong all instancing problems are typical of bad routers. If you want to play modern games, get a modern router as well. I never have a problem. I see everyone in my wing 100% of the time and my supercruise transitions are near instant.
 
Last edited:
Keep aliens, keep updating....just follow star citizen and change to "Lumberyard" reportedly 1000 player instances!
No matter what engine ED uses, you will never see 1000 player instances because the game is P2P-based, not client/server. SC uses servers to handle all that, so the client doesn't have to.

- - - Updated - - -

Oddly enough, no instance issues, before or after, the lumberyard change in SC.

Because SC uses client/server, not P2P.

- - - Updated - - -

Wrong all instancing problems are typical of bad routers. If you want to play modern games, get a modern router as well. I never have a problem. I see everyone in my wing 100% of the time and my supercruise transitions are near instant.

And/or because of a poor internet connection (speed, jitter, latency etc).
 
Last edited:
Keep aliens, keep updating....just follow star citizen and change to "Lumberyard" reportedly 1000 player instances!

Lumberyard would require the use of CryEngine. It was easy for SC to do so because it was already using CryEngine.

ED isn't using CryEngine. They'd have to in many ways make the whole game over again.
 
Back
Top Bottom