[Megathread] Star Citizen

[Xeenith]

This "will be" was sponsored by 2012.

it most likely will end like the movie of the same name huhuhu

[Proskill]

guys, why r u such buzz killers

[Calfredd]

guys, why r u such buzz killers

As opposed to blind sycophants?

[banmebaby]

Lol, SQ42 removed from the store, but the digital manual for it still selling for $12 ?

[Cynical Asshole]

Lol, SQ42 removed from the store, but the digital manual for it still selling for $12 ?

What? Are they pulling an Overwatch 2?

[Gorsameth]

Huh, this is the "Pledge" page.



Have to assume just some sort of error. Can't imagine why they'd pull it from sale.

*insert obligatory "Because it does not exist, it never has"*

[Cloverfield]

Huh, this is the "Pledge" page.



Have to assume just some sort of error. Can't imagine why they'd pull it from sale.

It is not an error. They have even updated the SQ42 page "pledge" button used to buy SQ42 into a "sign up" button instead now, just to sign up for the newsletter. It has been actively removed from sale. Why or whether that is temporary or not, we have no idea yet, but it has been already over a week with no word from the devs.

[banmebaby]

They polished the damned thing so hard that it went from Squadron42 to Squadron404. 2 years tops!

[Val the Moofia Boss]

I checked my account. SQ42 is still listed in the pledge package I bought nearly a decade ago.




... Though it has still been almost a decade and I haven't gotten the game I bought.

- - - Updated - - -

Why or whether that is temporary or not, we have no idea yet, but it has been already over a week with no word from the devs.

Perhaps they are trying to quietly memory hole the game. They stopped talking about SQ42 several years ago. As the oldtimers who backed for the singleplayer game aged out and lose interest and their voices are removed from the community, they will be replaced by a new generation of players who only care about the multiplayer component.

- - - Updated - - -

I've noticed that CIG is supposedly still shooting mo-cap with the SQ42 actors, so perhaps CIG has given up the game and is instead trying to make it into a 3D CGI animated movie.

[Gorsameth]

I've noticed that CIG is supposedly still shooting mo-cap with the SQ42 actors, so perhaps CIG has given up the game and is instead trying to make it into a 3D CGI animated movie.

God I now really want to know what those actors are thinking. Ofc its their job and they won't care much so long as they are getting payed but doing pick-ups 10 years later for a product that has never seen the light of day must be very weird.

[Yelmurc]

How small can those portions end up being? What if you have 1000 players congregating all within the exact same location, say one of the court yards in Orison? Will all those be managed by the same server or will that small area be sub partitioned by dynamic server meshing in even smaller areas and assigned to new servers? How small will those sub partitioned areas be?

Define "smoothly" please: Will players in contiguous portions of an area see each other across the boundaries of the portions? Will bullets or projectiles of any kind be able to cross those boundaries seamlessly and players be able to shoot each other from different portions of the area as if they were in the same portion? If the answer is yes, then all servers managing those portions will have to manage at the same time all that same information from other portions, so please remind me again, what is the point in having separate servers in the first place then?

This is what was said in December 2022 of the future of server meshing. This is taken from the Letter of the Chairman. So this seems to be what is left after PES and how it will function.

The next step will be to separate the Replication Layer (RL), the in-memory cache that remembers all dynamic object's state, from the Game Server and have scalable RL workers that communicate the state changes between the various game clients, servers, and the RL.

This is very important for Server Meshing, as universe state needs to be fully decoupled from server state. It will also make Star Citizen much more crash resilient on the client side as a server crash will no longer take down the RL as well, meaning that clients can stay connected while a new server takes over where the last one left off. It will also have the added benefit of having the view every client has of the universe around it be a little more fluid with less lag as the client’s update and refresh of object state will no longer be tied to the tick rate of the game server. This means players will see state changes from other remote clients and game servers as they happen at the rates the various clients and servers push them.

Once this happens, we’ll be able to have multiple game servers communicate with the RL (much like multiple game clients communicate right now with the RL) allowing Star Citizen to have multiple servers simulating the universe’s state, which we call Server Meshing (SM).

We will start with Static Server Meshing, where different servers are assigned to simulate different Entity Zones in a star system. Entity Zones, sometimes referred to as Local Grids, are separate simulation areas; the inside of a spaceship is a different Zone to the space around which the spaceship flies. A planet is a different Zone than the Star System Zone where all the planets and moons exist in, and a landing zone can also be a different Zone, nested inside the Planet Zone, nested in the Star System Zone.

At first, the servers will be bound to fixed Zones but we will quickly move to Dynamic Server Meshing V1, where we assign servers dynamically to Entity Zones, based on gameplay and simulation load. This will be a much more efficient use of the servers in the cloud, as you only need servers where the players are, whereas with Static SM you have servers assigned to Zones, even if there are no players there.

Dynamic Server Meshing V2 will take this dynamic assignment of servers one step further by subdividing Entity Zones into “simulation islands” (this is organizing the dynamic objects in one Zone / Local Grid into different groups, based on which objects can interact / collide with each other), allowing these islands to be distributed between servers to again help balance simulation load.

At this point we should be able to handle tens of thousands of players all playing in the same PES universe shard, bringing Star Citizen much closer to the ultimate goal of a huge living universe densely populated with players and AI.

[LanToaster]

I seriously have doubts Star Citizen will EVER get released in the state IT was promised.

I fully agree with you here. I dont doubt that at some point the devs will have to cut Corners/Features to get stuff in a Proper working condition.
However, personally i´m fine with it, if it never gets more lasting fun, it was a wasted Investment, if it does, it wasnt.

Its clear that even easier things are impossible for some developers, not to mention all the stuff that has piled up on SC by Chris Roberts.

I didnt really read much into the Blog post, but I dont think its an Appropriate comparison.
I dont think the "How" was the Problem ob OW2, more like the cost/return calculation. It would´ve just taken too much time/effort for what its worth.

While the Problems on CIGs end are not really a "Time" constraint, (even though understandibly more people might loose faith/patience) but rather a: "How the Fuck do we even Achieve this?".

Both are valid points to Cut corners/content. But entirely different, and as I said: I agree with you, at some point content/corners will most likely be cut.

But i wondered, what is even supposed to be the main gameplay loop of star citizen? Lirik said that he thinks the random generated missions are kind of boring and he rather flies around and tries to meet other people but that made me think... Is there more planned than grinding random generated missions and buy a new ship now and then?

From my Understanding its a Sandbox, and whatever yo do is meant to be fun. Its not like in WoW where you do content to increase your GearScore/Level.
You Buy different Ships to do different things, not neccessarily Better things.
Its weird for me to Explain this, but for example, I really Enjoy just Running Cargo, that means Buying somewhere Stuff, and selling it somewhere else for Profit. Other times I enjoy doing some Mining.

This is the thing that puzzles me about people claiming "SC has more content than other space games!!!"

"People claim", the thing is, Content is very Vague. You first need to Define what is actually Content.

You can easily Argue that SC has next to no Content, as its just a Star System with a handfull of Planets, some few Mission Types and prototype Gameplay loops.
On the Other side you can just aswell Argue that SC has a Shit ton of Content, with several Fully explorable Planets, Caves, and some very diverse Gameplay loops.

Both Claims are kinda true, yet are seen entirely Different, and thats OK. Not everything is for everyone.
Sometimes I log into SC and think, I can do so many things, and other times I´m there: "Meh, nothing to do".

I don't get it. I tried to get it. I spent more money than I should have trying to get it, but man... I feel like people need to raise their standards.

I´m sorry that you spend too much Money and regret it, hope you could´ve gotten a refund. But I wouldn´t agree with the latter statement, people have different Tastes in what is fun, enough people like it, and for them it offers what other games dont.

But isn't the game world instanced? How would that work in such an evironment? Do you own the property in all instances? Can you be attacked in all instances?

That is a good Question, I dont think CIG every fully explored how that works, I think its more a "We want to have it work kinda like that".
The overall Idea of the Universe is to have it work kinda like in WoW right now, where Each area in Space can be its own Server Instance, and you seamlessly move between the Servers/instances.

I would guess the Overall Idea is to only have one of Each Instance for each Area, and not Multiple. So every player that would meet up at these Property would actually be on the same server. Though, while I think they get the ServerMeshing tech at some point, in some form working. I cant see that happening. So, its basically wait and see what they do with that Feature.

[Cloverfield]

This is what was said in December 2022 of the future of server meshing. This is taken from the Letter of the Chairman. So this seems to be what is left after PES and how it will function.

Fully familiar. This does not seem to answer any of the specific questions I asked regarding "dynamic server meshing" either. I am pretty sure it helped tremendously selling ships though.

[LanToaster]

Fully familiar. This does not seem to answer any of the specific questions I asked regarding "dynamic server meshing" either. I am pretty sure it helped tremendously selling ships though.

Firstly, I think i´ve read you repeadetly (like several others, and if I´m wrong please forgive me, I tend to not pay much attention to the default "SC is a Scam and will never be" or "Of Course, SC will be whatever you want" posts, and might miss some actual Discussion) said Servermeshing or Magical Jesus Tech cant ever, and wont ever work/exist. So why Bother asking Questions about how its supposed to work?

Secondly you are Arguing with someone who only has access to the same Information as you, and expect specific answers to Questions the devs probably dont even know how to handle. What do you expect?

But if you insist, I´ll try to give you some more vague "It was said that way, so maybe its going to be that way" answers.

How small can those portions end up being? What if you have 1000 players congregating all within the exact same location, say one of the court yards in Orison? Will all those be managed by the same server or will that small area be sub partitioned by dynamic server meshing in even smaller areas and assigned to new servers? How small will those sub partitioned areas be?

Kinda irrelvant I think, technically if the Tech works how they want it to, they should be however small as they want or need them to be. I would guess its just a Value in the Code that says: "A Shard is anywhere between X and Y units in size".

From a gameplay/logical standpoint, I would say probably as small as they expect the Gameplay area to be. So everything relevant to the Player is inside your own Server.

As for what happens if X players go to a Y small Area, who knows? Maybe you get Sharding like in WoW. Thats something I think the devs never mentioned. But I think it would be wise to Shard the Players, but that also opens a whole new can of worms. But thats for the devs to figure out.

But I recall they said, if everything works how they want it to be that each Inside of a Building and Ship can potentially be managed by its own Server.

Define "smoothly" please: Will players in contiguous portions of an area see each other across the boundaries of the portions?

I think Smoothly means without the player noticing that he actually switched to a different Shard/Server, as for the second part, see my Previous answer to the size, and my other answer to Multiple servers beeing useless.

Will bullets or projectiles of any kind be able to cross those boundaries seamlessly and players be able to shoot each other from different portions of the area as if they were in the same portion?

From a QA I read/listened back in the day: Yes, Bullets are supposed to get from one Shard to another, though they probably wont need to if the Shard Area is properly managed.

If the answer is yes, then all servers managing those portions will have to manage at the same time all that same information from other portions, so please remind me again, what is the point in having separate servers in the first place then?

Not really, if you go back to the "Shard Size", if the Shard is the size of the Relevant Space, the Adjecant Shards dont need to feed as much and as accurate Information to the other Shards.
E.g. the Adjecant Shards dont need to know what each Person on a Ship is doing (or even if there are people), it only needs to know: "Theres a ship at that Location and Orientation", all the other data on the Shard where the ship actually is, is irrelevant to the adjecant Shards, this can be Adjusted by distance, so your Shard only knows about really big ships in Shards that are further away.

Obviously it takes some Serverpower for Managing that, but should still be less than having all on one Server.

(Note that I´m not a software engineer, so its only my Logic speaking here)

But in the end of the day, everything is only: "What CIG wants it to do" what it can actually do is only for sure when it gets done.
And as I said Previously, the notion shouldnt be: "It will never be done", or "It will be everybodys wet dream". But rather:
"What will it actually become". Because, if you really think about it, the entire Sharding/Phasing thing WoW does, is kinda ServerMeshing, so the Basic technology is not that Impossible, the Problem is how many of the things they want can be done.

If they at the end of the day settle on: "Each Planet and Orbit, POI in Space, and some Empty Space Regions are their own Server, and you switch them while Quantum Traveling", its still a solution to their Serverproblem, would it be good? If you ask this Thread probably not, if you ask any random SC Fan, probably yes. But in the end, one would only see if it gets done.

[Cloverfield]

So why Bother asking Questions about how its supposed to work?

I think the reason is perfectly clear, CIG has been touting how their tech is going to support the Best Damned Space Sim Ever, competing or exceeding with any of the best AAA out there, etc, for over a decade, with zero released products in gold to show for it. What they have done on the other hand, is to post extensive and frequent dreams.txt explaining us the details. Except that every time they have tried to implement the details, those the devil is usually in, it has either a) usually really never worked as intended, and often times breaking things in a spectacular fashion, or b) those details seem to have been lacking or not making much sense, as is the case that we discuss here.

Secondly you are Arguing with someone who only has access to the same Information as you, and expect specific answers to Questions the devs probably dont even know how to handle. What do you expect?

When some people just simply parrot those dream.txt from CIG I think it is just fair to confront them to the same issues and ask those questions.

But if you insist, I´ll try to give you some more vague "It was said that way, so maybe its going to be that way" answers.

That is precisely one of the issues highlighted in these discussions. CIG does not have a very solid track record in making things be the way they said it was going to be. If CIG can not make those things happen even after "saying it", why/how exactly do you think that just repeating what they said is going to help?

Kinda irrelvant I think, technically if the Tech works how they want it to, they should be however small as they want or need them to be.

See above, this does not make much sense as a response. That is precisely what is being put into question.

Man 1 - Hey, I have a spare Ford Fiesta engine block and a couple Honda Civic wheels in my garage, if I get the right oil I can build a car faster than the fastest Lamborghini!
Man 2 - Are you sure? I do not think that you can make that go as fast as you think, no?
Man 1 - Man, I told you I can do it if I get the right oil, don´t you understand?
Man 3 - Yeah! He said he will do it if he gets the right oil! Didn´t you hear what he said?
Man 2 - Yeah, I heard what he said, not sure that explains much though.

As for what happens if X players go to a Y small Area, who knows? Maybe you get Sharding like in WoW. Thats something I think the devs never mentioned. But I think it would be wise to Shard the Players, but that also opens a whole new can of worms. But thats for the devs to figure out.

Precisely. And fully agree there. That would be just classic instancing though, and which in my opinion is pretty much inevitable unless CIG finds a way to bend the laws of universal physics to their will. CIG has toyed in the past with the idea of simply denying access to players to certain areas if too crowded but that would be probably an even more serious headache game design wise and player experience wise. The thing is many SC apologists used to parrot the line that there would be no instancing, even CIG has for a very long time avoided the word "instancing" to sell the idea of "thousands of players in the same area" and the like. Many still believe that dynamic server meshing will somehow miraculously deliver that single instance experience, almost EVE style. Lots of millions have probably been pledged based on this alone.

As for you last comments "that also opens a whole new can of worms. But thats for the devs to figure out", that is indeed precisely the point. They have not, and they have not figured out yet even basic stuff such as physics, collisions, how to keep NPCs from standing on tables or T posing everywhere, and a myriad of other basic stuff that after over a decade should be more than nailed. And yet we expect them to "figure out" an almost miraculous dynamic server meshing that some still believe will allow players to "operate smoothly" across multiple partitioned areas as if it was pretty much a "de facto" single instanced universe?

I think Smoothly means without the player noticing that he actually switched to a different Shard/Server,

That does not help much I am afraid. There may be a ton of things going on at the moment a player decides to move from one partition to a contiguous one. Do you mean that everything that is relevant for that player will be synchronized across those two servers? What if we are talking hundreds of players instead of just 1, crossing that boundary? The amount of server information that needs to be replicated in all servers almost instantaneously is probably off the charts and would also probably defeat the purpose of multiple servers. Even someone without dev experience can probably see that is going to be everything but "smooth".

Again, back to my example of the 1000 players that decide to congregate in the same court yard in Orison. Imagine they decide to now initiate a free for all fight to the death in that spot. According to dynamic server meshing "ideas" that space will be subdivided in a ton of little sub partitions, managed by dedicated servers, that would also probably need to be adjusted on the fly real time, share tons of information across servers (positions, damage states, projectile trajectories etc) almost instantaneously as players move around. Now imagine that same scenario but in space with ships around an outpost... Do we really think a dev team incapable of implementing decent physics or getting NPC´s off the tables is going to really do this?

The closest we have for the idea of what dynamic server meshing can do out there is Dual Universe. A game that, unlike SC, was actually built from the ground up around the network elements to allow precisely what CIG is "saying" it wants to do. The result in Dual Universe so far is generally quite disappointing when players congregate in numbers, tons of lag, serious performance issues, disconnects etc etc etc.

Not really, if you go back to the "Shard Size", if the Shard is the size of the Relevant Space, the Adjecant Shards dont need to feed as much and as accurate Information to the other Shards.

This seems to be a bit more of the "they said it, so that is how is going to work" though.

The main point of server meshing is precisely to have servers manage smaller dynamic areas so to have to deal with smaller amounts of data which in turn can in theory allow for larger players density in that area etc etc. But if the data budget you manage to save due to smaller size area portions is eaten up by the need for that server to also manage information coming from other nearby servers (neighbouring asset locations, rendering, projectiles, damage states etc, anything in any contiguous area portions that is relevant to ALL players in this area portion) then you might end up in situations where the extra data required from contiguous partitions defeats the point of server meshing in the first place.

Unless you start instancing like basically pretty much every other mmo in history has done, of course. But given this is CIG we are talking about here, I have doubts they can even do proper regular instancing.

[Gorsameth]

Even EVE is heavy instanced if you look at the bigger picture. Every system is a completely separate instance.

But yes the problem with SC and CiG is that there is no reason to believe them when they say "x will fix this problem".

[LanToaster]

I think the reason is perfectly clear, CIG has been touting how their tech is going to support the Best Damned Space Sim Ever, competing or exceeding with any of the best AAA out there, etc, for over a decade, with zero released products in gold to show for it. What they have indeed done on the other hand, is to post extensive and frequent dreams.txt explaining us the details. Except that every time they have tried to implement the details, those the devil is usually in, it has either a) usually really never worked as intended, and often times breaking things in a spectacular fashion, or b) those details seem to have been lacking or not making much sense, as is the case that we discuss here.

When some people just simply parrot those dream.txt from CIG I think it is just fair to confront them to the same issues and ask those questions.

Yea, but I feel you are just trying to stir up conflict with that to be Honest. Though its still better than the Unending Parroting of, delayed deadlines and whatever old stuff gets brought up time after time.

I think one could have a much better Diskussion if the blatant Hostility would be avoided, especially from a few specific people. And I speak very generally, both sides of the Coin are very Annoying.

That is precisely one of the issues highlighted in these discussions. CIG does not have a very solid track record in making things be the way they said it was going to be. If CIG can not make those things happen even after "saying it", why/how exactly do you think that just repeating what they said is going to help?

True, but as I said from "Idea", or "How we want it to be" to actually doing it, stuff will most of the time Change. You can see it hower bad you want. That doesnt change that everyone outside of CIG doesnt have any more Information, and can either quote what CIG said, what they think it means, or what they Hope it will become.

But I can give the Question in that regard right back:
Why/How exactly do you think asking Questions no one can answer is going to help/change anything?
People who buy into the Idea of it beeing maybe possible will still hope and believe, and people who dont will not. And thats in my Opinion perfectly fine.
Problematic are only people who blindly throw cash at vague Ideas and then whine and complain that its not exaclty what they want, without even checking. And dont tell me its hard to see that SC might not be what people may first think.

While it wants to be the best damn Space sim ever, right now, its a niche product. For people who enjoy what it offers, how little or much it is.

See above, this does not make much sense as a response. That is precisely what is being put into question.

No, this is you changing the Question or Target of the Question, or beeing unable to ask the question in a way that gives people the ability to answer it.

You didnt ask anything Technical, you asked: "How Small/Large can Shards be", and its Irrelvant.
The Goal of the Tech is to have Server/Shards of varying sizes, that worky dynamicly together.

No Matter how Large or Small the Areas are, the Technology will need some form of knowledge what Adjecant Shards contain, and a way for the Players and Objects to transition between shards.

From a Players standpoint though its Irrelevant how Large a Shard is Area wise, because it should be as large/small as it needs to be, and the player shouldnt have to worry about that.

Precisely. And fully agree there. That would be just classic instancing though, and which in my opinion is pretty much inevitable unless CIG finds a way to bend the laws of universal physics to their will.

The general Idea of ServerMeshing is a form of Instancing. And in the end its just a "Switch" to allow Multiple Instances in the same Space, or just a single one.

The thing is many SC apologists used to parrot the line that there would be no instancing, even CIG has for a very long time avoided the word "instancing" to sell the idea of "thousands of players in the same area" and the like. Many still believe that dynamic server meshing will somehow miraculously deliver that single instance, almost EVE style.

I dont know the Mechanics of Eve, but I would wager Eve also has each StarSystem/Region in its own Instance, they just dont allow multiple Instances in the same place and let the game slow down to a crawl.

As for you last comments "that also opens a whole new can of worms. But thats for the devs to figure out", that is indeed precisely the point. They have not, and they have not figured out yet even basic stuff such as physics, collisions and a myriad of other basic stuff that after over a decade should be more than nailed.

We already Established that the Exact Idea might be Impossible, though Just because it seems impossible and were not done doesnt neccessarily mean it is, I mean the Entire Nanite Thing from UnrealEngine 5 kinda feels like Black Magic for me, and if you would´ve told anyone beforehand they want to give you basically Infinite amount of Polygons in a Scene they probably would´ve been laughed at and said its Impossible. So who knows.

But from my Experience the Basic Stuff that Breaks, and doesnt work Properly is always tied to the Overall Performance. On the Times the Server runs Good, next to no things Broke for me. (Be carefull, this is a Personal Experience and might not be true for others).

And I specifically meant Solutions for Instancing, not Technical but Designwise. If you take WoW as Example, if two People cross from one zone to the next, they might end up in different Shards in the next zone. This is something CIG would need take into Account. (I Think) It would for most people be "Good Enough" if the Instancing would not be as obvious as in WoW. And they could easily Achieve this, if they have Regions, these Regions have Layers or Several Instances however the fuck you want to call it. (CIG arent the Only one who Invent Clever words for Mundane Technologies, "Oh no, we dont do Phases on Classic, we do Layers", ... its the fucking same.)
E.g. The Entirety of Crusader and its Orbit would be a Region, and have several Layers, so if 1000 People visit Crusader, 500 go to Layer A and 500 go to Layer B, these Layers could again be Split into these Dynamic Shards. Remember A dynamic Shard only needs progressivly less Information from Adjecant Shards the further they are away. These Dynamic Shards are however Unique in its Layer.

And yet we expect them to "figure out" an almost miraculous dynamic server meshing that some still believe will allow players to "operate smoothly" across multiple partitioned areas as if it was pretty much a "de facto" single instanced universe?

As I said: You apparently expect them to get noting working, I expect them to get "Something" to work. There are a shit ton of Solutions possible, the question is just how much of the stuff, they would like to do, are they going to sacrifice.

That does not help much I am afraid. There may be a ton of things going on at the moment a player decides to move from one partition to a contiguous one. Do you mean that everything that is relevant for that player will be synchronized across those two servers? What if we are talking hundreds of players instead of just 1, crossing that boundary? The amount of server information that needs to be replicated in all servers almost instantaneously is probably off the charts and would also probably defeat the purpose of multiple servers. Even someone without dev experience can probably see that.

To actually make any relevant statement on the amount of Data that needs to be Synchronized between Shards you would need to have actually knowledge how it works. I dont, I can only speak Logicly, it if you switch from one server to another, its probably just like Logging out of one and Logging back into another. Other servers can Handle a few hundred players Logging in and Out, why shouldnt dynamic SC servers do that?

Again, back to my example of the 1000 players that decide to congregate in the same court yard in Orison. Imagine they decide to now initiate a free far all fight to the death in that spot. According to dynamic server meshing "ideas" that space will be subdivided in a ton of many little sub partitions that would also probably need to be adjusted on the fly real time and almost instantaneously as players move around. Now imagine that same scenario but in space with ships around an outpost... Do we really think a dev team incapable of implementing decent physics is going to really do this?

This is you again Nitpicking, you can still interpret whatever you want into the (you yourself said) vague not really telling much Explanations of ServerMeshing, but the reality still stands:
It doesnt Exist, and whatever it will be, it will be. The only thing reasonable is, to actually use some Logic.

A ServerShard should not be smaller than the Relevant gameplay Area. So, if 1000 Players rush to one Spot, they should be either:
A: Be in the same Shard or B: Split up into different Shards in the same space. The Latter part cold be creativly implemented without much Immersion breaking.

The closest we have for dynamic server meshing out there is Dual Universe. A game that, unlike SC, was actually built from the ground up around the network elements to allow precisely what CIG is "saying" it wants to do. The result in Dual Universe so far is generally quite disappointing when players congregate in numbers, tons of lag, serious performance issues, pop in everywhere etc etc etc.

I cant comment on the Servermeshing part, because I dont know how Dual Universe works technically, but yea. I´m with you there.
But I think CIG at some (fairly Recent) point actually talked about Challenges with their persistance on multiple Shards in the same Area. So I would guess they already settled (at least internally) on having multiple Instances of Areas.

This seems to be more of "they said it, so that is how is going to work" though.

No, its: "Thats what I would do"

I would limit the Area/Size of Shards to the Relevant Gameplay. As the FlightSpeed in Atmosphere is much smaller than in Space, you would have Planetside smaller Shards, (Just a Couple KM² maybe? [Beware Random Number]). But only as small as the Gameplay is Relevant.

That way Adjecant Shards would only need limited Information. And in regards your: "Suddenly alot of Information replicated" you can have an Overlap/Buffer zone where the Data gets slowly updated and more complete for Adjecant Servers.

So. E.g. Server A to the North of Server B, gets all the Information about stuff that is on the North Side of B, but next to no Information on the SouthSide of B.

[Surfd]

I dont know the Mechanics of Eve, but I would wager Eve also has each StarSystem/Region in its own Instance, they just dont allow multiple Instances in the same place and let the game slow down to a crawl.

Actually, you would be half right: they don't do multiple instances of zones or systems, but they do occasionally let the game slow to a crawl. Deliberately in fact.

Their solution to "giant space battles slowing the systems to a crawl" was to actually develop an in house a method, known as Time Dilation, to intentionally slow things down to allow the servers to cope instead of simply imploding into a mess of unplayable lag: https://www.eveonline.com/news/view/...-dilation-tidi

Eve is "one universe", but it manages that universe by dynamically load balancing the various segments of the universe across the available server power. Major Hub segments would get entire servers, or even multiple servers dedicated to running just those areas. Meanwhile, fringe segments out in the boonies which maybe see one player at a time on a good day may have dozens of segments being handled by a single server.

It's actually been noted by the EvE devs that they are capable of manually tasking additional server power to an area if they know there is going to be a need for it as well, if I recall right. I am pretty sure I recall them saying that they have done it before for special events and some big corp vs corp fights.

[Gorsameth]

We already Established that the Exact Idea might be Impossible, though Just because it seems impossible and were not done doesnt neccessarily mean it is, I mean the Entire Nanite Thing from UnrealEngine 5 kinda feels like Black Magic for me, and if you would´ve told anyone beforehand they want to give you basically Infinite amount of Polygons in a Scene they probably would´ve been laughed at and said its Impossible. So who knows.

The important thing is that Unreal didn't come out saying they were going to handle an infinite amount of polygons and then not deliver on that for 10 years.

Especially when the thing they are failing to deliver is a critical requirement to delivering the product they set out to make 10 years ago.

And Unreal isn't doing it with money their customers payed in advance for a product they will likely never get.

[banmebaby]

Unending Parroting of, delayed deadlines and whatever old stuff gets brought up time after time.

Last time people forgot that Chris Roberts over promises and delays his projects till the money runs dry, people gave him over half a billion dollars for another one, and after spewing multiple of those parroted deadlines, there is still no sign of a single player that just needed a "little more polish" years ago.