Yup. You guessed it. We Want Cross play.

Yesterday in Shinrarta,four human players in instance,lag/stutters,FPS dropping,imagine on foot combat in Odyssey with even more players (some on the ground some in the air) 13 fps,then add to that crossplay...good luck!

That's not a valid argument against crossplay. Crossplay itself has no inherent additional performance cost. Any current performance issues should be sorted out regardless, but that's a completely orthogonal topic.
 
That's not a valid argument against crossplay. Crossplay itself has no inherent additional performance cost. Any current performance issues should be sorted out regardless, but that's a completely orthogonal topic.
Of course it is valid because with crossplay you'll have way more people from other platforms in the same instance,now the servers cannot handle a few people "in space",imagine what will happen on the ground with NPCs too,3d buildings and other stuff,good luck again.It's not othogonal topic is just sheer numbers wich seems you don't realize.
Any current performance issues should be sorted out regardless,
Do you have proof of that? Because in all these years multiplay has always been ty as it is now.
 
Last edited:
I hope that goes for ODY too, then and they can crossplay as much as they want.

By the Far Gods I hope not, it certainly helps with my shoddy aim but when the sights ‘snap’ to targets as I’m trying to manually adjust my aim it throws me right off. I much prefer if the game leaves me to miss my target of my own volition.

In fact, a slight tangent, the first thing I do on any FPS if I can is disable any sort of aim assist and crosshairs even on multiplayer twitch shooters, it’s all about the immershun innit. I’ll forgive the futuristic projected HUD though where it fits with the sci-fi setting as long as it disappears when someone smashes my helmet with a rifle butt. I managed to get almost all the way through Metro 2033 on the Survival or Ranger difficulty until I got stuck on an autosave with about ten seconds left on my last air filter with no hope within reach...
 
I'd like to see it, but at the same time, there are a lot of other things I'd rather see first. And crossplay is one of those things that you can reasonably expect to take a lot of dev time. Unless you're No Man's Sky, and you can apparently just flip a switch and it just works in one of your five yearly updates... sigh.
 
Of course it is valid because with crossplay you'll have way more people from other platforms in the same instance,

That's not how instancing works.

The load on the back end is based on total playerbase numbers alone -- how those numbers are distributed across platforms has no bearing on performance. A web request is a web request no matter what type of device it came from.

The real time sessions are based on p2p hosting so one player acts as a server for 1..n players, n being some max number of players per session set by the developers (edit to add: or n is determined algorithmically at runtime based on measurements on how the host and the clients are coping with their current load). There are 1..m such sessions for each game location that's currently occupied by players. These are called instances.

Instances do not directly cross-communicate any real-time data that needs to be updated frequently like position or orientation, so the total number of instances doesn't impact performance either. Player actions may modify persistent data stored in the back end and those effects propagate across instances that way, but that rate of updates is small enough to be negligible.

The number of instances per game location depends on instance occupancy and matchmaking rules such as Open/PG/Solo, blocking, platform, etc.

Poor performance when there are other CMDRs around are an indication of some host or client having performance issues in general. It is an orthogonal problem to matchmaking rules which is what crossplay essentially is.


I have designed, developed and am now managing the development of a rather similar system and we match Oculus Quest users with MacOS and high end Windows PC users on a daily basis. I just got my first US patent on multiplayer networking technology in 2020 so even if I may not be exactly on point for 100% of the implementation details, I know a little bit about multiplayer networking and none of ED's networking tech appears to be exceptional in any way.

Therefore I argue that stupid implementations aside, crossplay has no inherent performance implications.

Do you have proof of that? Because in all these years multiplay has always been poopooty as it is now.

I didn't say they are going to be sorted out, I said they should be sorted out. As they obviously should.
 
Last edited:
That's not how instancing works.

The load on the back end is based on total playerbase numbers alone -- how those numbers are distributed across platforms has no bearing on performance. A web request is a web request no matter what type of device it came from.

The real time sessions are based on p2p hosting so one player acts as a server for 1..n players, n being some max number of players per session set by the developers (edit to add: or n is determined algorithmically at runtime based on measurements on how the host and the clients are coping with their current load). There are 1..m such sessions for each game location that's currently occupied by players. These are called instances.

Instances do not directly cross-communicate any real-time data that needs to be updated frequently like position or orientation, so the total number of instances doesn't impact performance either. Player actions may modify persistent data stored in the back end and those effects propagate across instances that way, but that rate of updates is small enough to be negligible.

The number of instances per game location depends on instance occupancy and matchmaking rules such as Open/PG/Solo, blocking, platform, etc.

Poor performance when there are other CMDRs around are an indication of some host or client having performance issues in general. It is an orthogonal problem to matchmaking rules which is what crossplay essentially is.


I have designed, developed and am now managing the development of a rather similar system and we match Oculus Quest users with MacOS and high end Windows PC users on a daily basis. I just got my first US patent on multiplayer networking technology in 2020 so even if I may not be exactly on point for 100% of the implementation details, I know a little bit about multiplayer networking and none of ED's networking tech appears to be exceptional in any way.

Therefore I argue that stupid implementations aside, crossplay has no inherent performance implications.



I didn't say they are going to be sorted out, I said they should be sorted out. As they obviously should.
Nice try, a wall of text discharging on players connection the responsibility for a terrible open play experience in ED,keep your convinctions,I'll keep mine ;)
 
Nice try, a wall of text discharging on players connection the responsibility for a terrible open play experience in ED,keep your convinctions,I'll keep mine ;)

What an awful, awful response to read.

You have no counter argument and prefer pointing out the fact that his response is a "wall of text" ?

Well yeah, he just explained to you how instancing can work, as well as what his background is to base his information, he probably had the intention to make you learn something, after noticing he knew more than you did on the subject, you don't even have the decency to read and potentially learn, let alone questioning your knowledge on the matter ?

"keep your convinctions" I must admit though, that made me laugh, for how sad it was to read.
 
What an awful, awful response to read.

You have no counter argument and prefer pointing out the fact that his response is a "wall of text" ?

Well yeah, he just explained to you how instancing can work, as well as what his background is to base his information, he probably had the intention to make you learn something, after noticing he knew more than you did on the subject, you don't even have the decency to read and potentially learn, let alone questioning your knowledge on the matter ?

"keep your convinctions" I must admit though, that made me laugh, for how sad it was to read.
Yes yes as you please. ;)
 
What an awful, awful response to read.

You have no counter argument and prefer pointing out the fact that his response is a "wall of text" ?

Well yeah, he just explained to you how instancing can work, as well as what his background is to base his information, he probably had the intention to make you learn something, after noticing he knew more than you did on the subject, you don't even have the decency to read and potentially learn, let alone questioning your knowledge on the matter ?

"keep your convinctions" I must admit though, that made me laugh, for how sad it was to read.
Its pretty much the default response for people that get proven wrong since.... ever.

If facts and logic dont match the position they hold then they are dumb and for suckers, the all mighty unsubstantiated opinion clearly knows whats best.

This why we still have people swearing up and down that clean drives run cooler in supercruise over dirty drives, for instance.
 
Back
Top Bottom