As they still haven't figured out how to show members how much their referral funds are yet, I remain skeptical about any "Dynamic Gridding" or any other "forward-looking technology" they might want to discuss.
All time classic MY NEW FAVORITE POST! (Keep laying those bricks)
"I should point out that no other company has shipped out a beta on a disc before this." - Official Mortal Online Lead Community Moderator
Proudly wearing the Harbinger badge since Dec 23, 2017.
Coined the phrase "Role-Playing a Development Team" January 2018
"Oddly Slap is the main reason I stay in these forums." - Mystichaze April 9th 2018
As they still haven't figured out how to show members how much their referral funds are yet, I remain skeptical about any "Dynamic Gridding" or any other "forward-looking technology" they might want to discuss.
You wont have to be skeptical for long considering its supposed to launch with alpha 2 which releases in a few months unless the plans have changed. We'll know if it works or not soon.
Just trying to live long enough to play a new, released MMORPG, playing New Worlds atm
Fools find no pleasure in understanding but delight in airing their own opinions. Pvbs 18:2, NIV
Don't just play games, inhabit virtual worlds™
"This is the most intelligent, well qualified and articulate response to a post I have ever seen on these forums. It's a shame most people here won't have the attention span to read past the second line." - Anon
As they still haven't figured out how to show members how much their referral funds are yet, I remain skeptical about any "Dynamic Gridding" or any other "forward-looking technology" they might want to discuss.
You wont have to be skeptical for long considering its supposed to launch with alpha 2 which releases in a few months unless the plans have changed. We'll know if it works or not soon.
Wouldn't call a few months very "forward looking"
The referral amounts were supposed to be launched a few months away for the last 7 years.
Any update on that? Or any thoughts on why this is beyond their technical ability?
All time classic MY NEW FAVORITE POST! (Keep laying those bricks)
"I should point out that no other company has shipped out a beta on a disc before this." - Official Mortal Online Lead Community Moderator
Proudly wearing the Harbinger badge since Dec 23, 2017.
Coined the phrase "Role-Playing a Development Team" January 2018
"Oddly Slap is the main reason I stay in these forums." - Mystichaze April 9th 2018
I take discussions about "server meshing" as a bad sign. I'm not sure why, but I do.
Probably because Star Citizen has been throwing around this terminology for years now and have yet to really deliver.
That's it.
I actually watched a bit of a video on this concept, and I hate to admit it but it was interesting. So, they're actually multi-threading the Unreal 5 engine for this - seems to me they've the expertise and drive to do this thing right.
I wasn't going to give the game a second thought but damn...
Can't wait to play this mmorpg in 2035......or maybe 2045 if UE6 comes out and then they say they have to keep it in further 'development' so it can be 'next-gen'
As they still haven't figured out how to show members how much their referral funds are yet, I remain skeptical about any "Dynamic Gridding" or any other "forward-looking technology" they might want to discuss.
You wont have to be skeptical for long considering its supposed to launch with alpha 2 which releases in a few months unless the plans have changed. We'll know if it works or not soon.
Wouldn't call a few months very "forward looking"
The referral amounts were supposed to be launched a few months away for the last 7 years.
Any update on that? Or any thoughts on why this is beyond their technical ability?
I'm gonna be honest. I couldn't care less about your referral money.
I'm gonna be honest. I couldn't care less about your referral money.
Its not MINE. It's what is owed to literally tens of thousands of people. Promised by Stephen and used as a tool to push the game on social media and raise funds.
But noted that you do not care.
All time classic MY NEW FAVORITE POST! (Keep laying those bricks)
"I should point out that no other company has shipped out a beta on a disc before this." - Official Mortal Online Lead Community Moderator
Proudly wearing the Harbinger badge since Dec 23, 2017.
Coined the phrase "Role-Playing a Development Team" January 2018
"Oddly Slap is the main reason I stay in these forums." - Mystichaze April 9th 2018
As they still haven't figured out how to show members how much their referral funds are yet, I remain skeptical about any "Dynamic Gridding" or any other "forward-looking technology" they might want to discuss.
You wont have to be skeptical for long considering its supposed to launch with alpha 2 which releases in a few months unless the plans have changed. We'll know if it works or not soon.
Wouldn't call a few months very "forward looking"
The referral amounts were supposed to be launched a few months away for the last 7 years.
Any update on that? Or any thoughts on why this is beyond their technical ability?
I'm gonna be honest. I couldn't care less about your referral money.
While I don't really care about the referral money either, it's important that the devs try to run their game at least somewhat honestly.
No-one is perfectly honest, but Ashes of Creation devs seem to be pulling a lot of dirty tricks when it comes to money and advertising.
I take discussions about "server meshing" as a bad sign. I'm not sure why, but I do.
Probably because Star Citizen has been working on delivering its version of server meshing for several years now, apparently more challenging than they first anticipated.
I'm sure the "glorious" developers at Intrepid will make it all look rather easy much to Chris Robert's embarrassment.
Just trying to live long enough to play a new, released MMORPG, playing New Worlds atm
Fools find no pleasure in understanding but delight in airing their own opinions. Pvbs 18:2, NIV
Don't just play games, inhabit virtual worlds™
"This is the most intelligent, well qualified and articulate response to a post I have ever seen on these forums. It's a shame most people here won't have the attention span to read past the second line." - Anon
I take discussions about "server meshing" as a bad sign. I'm not sure why, but I do.
Probably because Star Citizen has been working on delivering its version of server meshing for several years now, apparently more challenging than they first anticipated.
I'm sure the "glorious" developers at Intrepid will make it all look rather easy much to Chris Robert's embarrassment.
When they talk about making Unreal 5 multi-threaded.... dang... That sounds really technical.
I take discussions about "server meshing" as a bad sign. I'm not sure why, but I do.
Probably because Star Citizen has been working on delivering its version of server meshing for several years now, apparently more challenging than they first anticipated.
I'm sure the "glorious" developers at Intrepid will make it all look rather easy much to Chris Robert's embarrassment.
When they talk about making Unreal 5 multi-threaded.... dang... That sounds really technical.
So I watched about half of it and skipped around and skimmed some of the rest to understand what they are doing at a high level and I never heard them talk about making Unreal 5 multi-threaded. Matter of fact I would be shocked if Unreal 5 wasn't multi-threaded out of the box. It's difficult to imagine a high performance client engine being single threaded. Now that may just be a misunderstanding on your part. I don't know how technical you are. Not big deal if you're not. However multi-threading is simply the ability of a program to execute multiple threads (or commands) against the CPU simultaneously. It's how a program can benefit from a multi-core CPU which every CPU these days is multi-core. I don't know that you could even buy a single core CPU anymore. Even your phone has multiple cores.
Anyway...
What I did hear that they had to go tear the guts out of Unreal 5 was that they had to make it multi-server capable. That is, the client has to interact with 1-n servers, not just one server which is how Unreal 5 is coded out of the box. That was the complicated bit from the client side coding.
What they're doing server side is VERY interesting. I'm pretty sure the Star Citizen team is taking notes because basically they are solving the same problem Start Citizen has. The difference being that where Star Citizen has some slides and lots of talks about what needs to happen, AoC has videos of it working in real time with debugging enabled to show you where the server boundaries are. This corroborates my thought on the Star Citizen team that they are basically just not up to the technical task before them. I don't want to use the word incompetent because it's not some easy task to solve, and I'm sure individually they have a lot of smart people on their team, but it seems like maybe they didn't have the right smart people on the team (or poor leadership) because here we have the AoC coming along and solving the exact same problem with less budget and fewer developers (that I am aware of).
My quick short version of how this works so you don't have to watch an hour-long video follows. Feel free to correct me if I mess anything up.
First of all, everything is server authoritative meaning they will have much tighter controls on hacking. "Actors" is the terminology they user for anything in the game world that can have an interaction. The only thing that wouldn't be an actor would be like the terrain itself. Even the trees are actors because the wind will blow in a direction and the trees will react to the wind. Trees can be cut down, etc. If it can act, it's an actor.
Now understanding that, each server is responsible for a geographic area and all the actors in it. As an actor starts to approach a server boundary, the server starts negotiating with it's neighbor to spawn what they call a "proxy" of the actor which looks and acts just like the actor itself. The server that owns the actor starts replicating its data for the actor to the neighboring server which is how the proxy is indistinguishable to the actual actor because it will be in the same location, play the same animations, take damage, get healed, etc just like the actual actor does. This way you can have a real-time pvp battle with someone across the server boundary in real time. The video shows no perceptible lag in this scenario. When the proxy crosses the server boundary, the actor is basically handed off to the neighboring server and it now becomes the owner of the actor and the process is reversed until the actor moves out of visual range an no longer needs to be replicated across servers. This can happen in a corner situation as well where there are 4 servers in a corner boundary situation. The proxy/replication will happen between all 4. Finally the server boundaries are not fixed. They can be dynamically reallocated to apply more server processing where it's needed in real time. For example if there's a large pvp battle going on, the servers will reallocate such that the battle itself is split up among several servers to divide the load while relatively empty areas will be consolidated into a single server since it would be plenty to manage a large, empty area.
Pretty cool and sophisticated stuff. I'm not a big pvp guy, so I'm not sure the game is going to be my cup of tea, but the technology behind it is really cool and interesting for sure.
I take discussions about "server meshing" as a bad sign. I'm not sure why, but I do.
Probably because Star Citizen has been working on delivering its version of server meshing for several years now, apparently more challenging than they first anticipated.
I'm sure the "glorious" developers at Intrepid will make it all look rather easy much to Chris Robert's embarrassment.
When they talk about making Unreal 5 multi-threaded.... dang... That sounds really technical.
So I watched about half of it and skipped around and skimmed some of the rest to understand what they are doing at a high level and I never heard them talk about making Unreal 5 multi-threaded. Matter of fact I would be shocked if Unreal 5 wasn't multi-threaded out of the box. It's difficult to imagine a high performance client engine being single threaded. Now that may just be a misunderstanding on your part. I don't know how technical you are. Not big deal if you're not. However multi-threading is simply the ability of a program to execute multiple threads (or commands) against the CPU simultaneously. It's how a program can benefit from a multi-core CPU which every CPU these days is multi-core. I don't know that you could even buy a single core CPU anymore. Even your phone has multiple cores.
Anyway...
What I did hear that they had to go tear the guts out of Unreal 5 was that they had to make it multi-server capable. That is, the client has to interact with 1-n servers, not just one server which is how Unreal 5 is coded out of the box. That was the complicated bit from the client side coding.
What they're doing server side is VERY interesting. I'm pretty sure the Star Citizen team is taking notes because basically they are solving the same problem Start Citizen has. The difference being that where Star Citizen has some slides and lots of talks about what needs to happen, AoC has videos of it working in real time with debugging enabled to show you where the server boundaries are. This corroborates my thought on the Star Citizen team that they are basically just not up to the technical task before them. I don't want to use the word incompetent because it's not some easy task to solve, and I'm sure individually they have a lot of smart people on their team, but it seems like maybe they didn't have the right smart people on the team (or poor leadership) because here we have the AoC coming along and solving the exact same problem with less budget and fewer developers (that I am aware of).
My quick short version of how this works so you don't have to watch an hour-long video follows. Feel free to correct me if I mess anything up.
First of all, everything is server authoritative meaning they will have much tighter controls on hacking. "Actors" is the terminology they user for anything in the game world that can have an interaction. The only thing that wouldn't be an actor would be like the terrain itself. Even the trees are actors because the wind will blow in a direction and the trees will react to the wind. Trees can be cut down, etc. If it can act, it's an actor.
Now understanding that, each server is responsible for a geographic area and all the actors in it. As an actor starts to approach a server boundary, the server starts negotiating with it's neighbor to spawn what they call a "proxy" of the actor which looks and acts just like the actor itself. The server that owns the actor starts replicating its data for the actor to the neighboring server which is how the proxy is indistinguishable to the actual actor because it will be in the same location, play the same animations, take damage, get healed, etc just like the actual actor does. This way you can have a real-time pvp battle with someone across the server boundary in real time. The video shows no perceptible lag in this scenario. When the proxy crosses the server boundary, the actor is basically handed off to the neighboring server and it now becomes the owner of the actor and the process is reversed until the actor moves out of visual range an no longer needs to be replicated across servers. This can happen in a corner situation as well where there are 4 servers in a corner boundary situation. The proxy/replication will happen between all 4. Finally the server boundaries are not fixed. They can be dynamically reallocated to apply more server processing where it's needed in real time. For example if there's a large pvp battle going on, the servers will reallocate such that the battle itself is split up among several servers to divide the load while relatively empty areas will be consolidated into a single server since it would be plenty to manage a large, empty area.
Pretty cool and sophisticated stuff. I'm not a big pvp guy, so I'm not sure the game is going to be my cup of tea, but the technology behind it is really cool and interesting for sure.
I think they meant the multithreading part at 1:08:20 where they talk about multithreading the Unreal replication graph to improve server performance and how Unreal replication is one of the most single threaded systems. I've also heard about how Unreal Engine is very single threaded.
As for comparing to Star Citizen, my understanding is they want individual servers down to room scale or individual ships. All while the planets and ships themselves are moving because they are so stubborn in wanting every tiny object to be persistent. Meanwhile Ashes of Creation seems to work on grids since they call it dynamic gridding and from what their slides have demonstrated. I don't have the technical knowledge for how much of a difference that is but the amount of physics simulation in Star Citizen does seem to make things more difficult. Both games are also at the stage where they can be demonstrated and tested with Star Citizen having already done an 800 player test, so it will be interesting to see how AoC does.
What they're doing server side is VERY interesting. I'm pretty sure the Star Citizen team is taking notes because basically they are solving the same problem Start Citizen has.
AoC has a 72km² static map while just the current single Star Citizen system has 750 millions km across of simulated space of 17 full planets and moons with cities rotating in real time + thousands of npc's, complexity of ships with thousands of moving parts (entities), the density of physic calculations, speeds etc etc.
So the challenges and needs for Star Citizen and their Server Meshing tech aren't remotely close to AoC, as in they aren't even in the same ballpark.
AoC is creating a SM for static zones and small amount of simple entities duplicated from one server to another.
SC is creating a SM for dynamic zones and huge amount of nested entities managed by a replication layer (no duplicates).
It's like comparing a car to a plane, both have wheels, but only one can fly. The technique used by AoC cannot be used to dynamically assign a server to a moving object (a capital ship), which is the ultimate goal of CIG
What they're doing server side is VERY interesting. I'm pretty sure the Star Citizen team is taking notes because basically they are solving the same problem Start Citizen has.
AoC has a 72km² static map while just the current single Star Citizen system has 750 millions km across of simulated space of 17 full planets and moons with cities rotating in real time + thousands of npc's, complexity of ships with thousands of moving parts (entities), the density of physic calculations, speeds etc etc.
So the challenges and needs for Star Citizen and their Server Meshing tech aren't remotely close to AoC, as in they aren't even in the same ballpark.
AoC is creating a SM for static zones and small amount of simple entities duplicated from one server to another.
SC is creating a SM for dynamic zones and huge amount of nested entities managed by a replication layer (no duplicates).
It's like comparing a car to a plane, both have wheels, but only one can fly. The technique used by AoC cannot be used to dynamically assign a server to a moving object (a capital ship), which is the ultimate goal of CIG
Yeah sure Star Citizen is big but 99.9% of that is empty space or procedurally generated planets with no wildlife or anything. Even if it wasn't, I think that would just be a matter of scaling up. I don't see why the same tech wouldn't apply here.
Capital ships being its own moving server definitely sounds unique though but honestly... I am skeptical on that one ever working.
Every object being persistent is also probably unique to SC but whats going to stop trolls from piling tens of thousands of objects in random spots on a planet? I have doubts on that one too.
I wonder if this is going to be another Dual Universe. They claim to have new server technology to allow thousands of players in 1 area, etc, etc; yet, their game was a failure.
I take discussions about "server meshing" as a bad sign. I'm not sure why, but I do.
Probably because Star Citizen has been working on delivering its version of server meshing for several years now, apparently more challenging than they first anticipated.
I'm sure the "glorious" developers at Intrepid will make it all look rather easy much to Chris Robert's embarrassment.
When they talk about making Unreal 5 multi-threaded.... dang... That sounds really technical.
So I watched about half of it and skipped around and skimmed some of the rest to understand what they are doing at a high level and I never heard them talk about making Unreal 5 multi-threaded. Matter of fact I would be shocked if Unreal 5 wasn't multi-threaded out of the box. It's difficult to imagine a high performance client engine being single threaded. Now that may just be a misunderstanding on your part. I don't know how technical you are. Not big deal if you're not. However multi-threading is simply the ability of a program to execute multiple threads (or commands) against the CPU simultaneously. It's how a program can benefit from a multi-core CPU which every CPU these days is multi-core. I don't know that you could even buy a single core CPU anymore. Even your phone has multiple cores.
Anyway...
What I did hear that they had to go tear the guts out of Unreal 5 was that they had to make it multi-server capable. That is, the client has to interact with 1-n servers, not just one server which is how Unreal 5 is coded out of the box. That was the complicated bit from the client side coding.
What they're doing server side is VERY interesting. I'm pretty sure the Star Citizen team is taking notes because basically they are solving the same problem Start Citizen has. The difference being that where Star Citizen has some slides and lots of talks about what needs to happen, AoC has videos of it working in real time with debugging enabled to show you where the server boundaries are. This corroborates my thought on the Star Citizen team that they are basically just not up to the technical task before them. I don't want to use the word incompetent because it's not some easy task to solve, and I'm sure individually they have a lot of smart people on their team, but it seems like maybe they didn't have the right smart people on the team (or poor leadership) because here we have the AoC coming along and solving the exact same problem with less budget and fewer developers (that I am aware of).
My quick short version of how this works so you don't have to watch an hour-long video follows. Feel free to correct me if I mess anything up.
First of all, everything is server authoritative meaning they will have much tighter controls on hacking. "Actors" is the terminology they user for anything in the game world that can have an interaction. The only thing that wouldn't be an actor would be like the terrain itself. Even the trees are actors because the wind will blow in a direction and the trees will react to the wind. Trees can be cut down, etc. If it can act, it's an actor.
Now understanding that, each server is responsible for a geographic area and all the actors in it. As an actor starts to approach a server boundary, the server starts negotiating with it's neighbor to spawn what they call a "proxy" of the actor which looks and acts just like the actor itself. The server that owns the actor starts replicating its data for the actor to the neighboring server which is how the proxy is indistinguishable to the actual actor because it will be in the same location, play the same animations, take damage, get healed, etc just like the actual actor does. This way you can have a real-time pvp battle with someone across the server boundary in real time. The video shows no perceptible lag in this scenario. When the proxy crosses the server boundary, the actor is basically handed off to the neighboring server and it now becomes the owner of the actor and the process is reversed until the actor moves out of visual range an no longer needs to be replicated across servers. This can happen in a corner situation as well where there are 4 servers in a corner boundary situation. The proxy/replication will happen between all 4. Finally the server boundaries are not fixed. They can be dynamically reallocated to apply more server processing where it's needed in real time. For example if there's a large pvp battle going on, the servers will reallocate such that the battle itself is split up among several servers to divide the load while relatively empty areas will be consolidated into a single server since it would be plenty to manage a large, empty area.
Pretty cool and sophisticated stuff. I'm not a big pvp guy, so I'm not sure the game is going to be my cup of tea, but the technology behind it is really cool and interesting for sure.
I am not fully defending Star Citizen and their development fiasco, but even so in the recent few months they have been ever so slightly justifying the prolonged development cycle.
At their "CitizenCon" event in October they have shown off an interesting amount of content, which as far as I understood, nearly half of it has already or is on it's way to be implemented and during one of their presentation they have shown off a functional demo of the "Replication Layer" and "Server Meshing" technologies, which does seem quite intriguing and showcases that even while on different servers, entities and players are able to interact with one another seamlessly (or that is at the very least the intent). Furthermore they have conducted multiple test of the early meshing technology this year and judging by the many posts on reddit, the players have been pleasantly surprised by it's functionality.
While the Star Citizen developers are notorious for letting their fanbase down and we are used to judge Star Citizen as a money laundering scheme, the recent few months have been interesting in observing as they are slowly progressing in development. On the other hand AoC seems to be tackling a similar task and while i couldn't agree more that their progress is much faster, thus proving the lacking organization of Star Citizen dev team.
Either way i am looking forward to this competition between the two. It will be interesting to see who will be able to bring this technology into the hands of the players first and how it will affect them. So far it seems that Scam Citizen has a bit of an advantage with the already performed tests.
Comments
All time classic MY NEW FAVORITE POST! (Keep laying those bricks)
"I should point out that no other company has shipped out a beta on a disc before this." - Official Mortal Online Lead Community Moderator
Proudly wearing the Harbinger badge since Dec 23, 2017.
Coined the phrase "Role-Playing a Development Team" January 2018
"Oddly Slap is the main reason I stay in these forums." - Mystichaze April 9th 2018
You wont have to be skeptical for long considering its supposed to launch with alpha 2 which releases in a few months unless the plans have changed. We'll know if it works or not soon.
Wouldn't call a few months very "forward looking"
"True friends stab you in the front." | Oscar Wilde
"I need to finish" - Christian Wolff: The Accountant
Just trying to live long enough to play a new, released MMORPG, playing New Worlds atm
Fools find no pleasure in understanding but delight in airing their own opinions. Pvbs 18:2, NIV
Don't just play games, inhabit virtual worlds™
"This is the most intelligent, well qualified and articulate response to a post I have ever seen on these forums. It's a shame most people here won't have the attention span to read past the second line." - Anon
I'm not sure why, but I do.
Any update on that? Or any thoughts on why this is beyond their technical ability?
All time classic MY NEW FAVORITE POST! (Keep laying those bricks)
"I should point out that no other company has shipped out a beta on a disc before this." - Official Mortal Online Lead Community Moderator
Proudly wearing the Harbinger badge since Dec 23, 2017.
Coined the phrase "Role-Playing a Development Team" January 2018
"Oddly Slap is the main reason I stay in these forums." - Mystichaze April 9th 2018
The referral amounts were supposed to be launched a few months away for the last 7 years.
Any update on that? Or any thoughts on why this is beyond their technical ability?
It's a weird hill to die on, but I respect your diligence.
Mess with the best, Die like the rest
Probably because Star Citizen has been throwing around this terminology for years now and have yet to really deliver.
I actually watched a bit of a video on this concept, and I hate to admit it but it was interesting. So, they're actually multi-threading the Unreal 5 engine for this - seems to me they've the expertise and drive to do this thing right.
I wasn't going to give the game a second thought but damn...
I'm gonna be honest. I couldn't care less about your referral money.
But noted that you do not care.
All time classic MY NEW FAVORITE POST! (Keep laying those bricks)
"I should point out that no other company has shipped out a beta on a disc before this." - Official Mortal Online Lead Community Moderator
Proudly wearing the Harbinger badge since Dec 23, 2017.
Coined the phrase "Role-Playing a Development Team" January 2018
"Oddly Slap is the main reason I stay in these forums." - Mystichaze April 9th 2018
No-one is perfectly honest, but Ashes of Creation devs seem to be pulling a lot of dirty tricks when it comes to money and advertising.
I'm sure the "glorious" developers at Intrepid will make it all look rather easy much to Chris Robert's embarrassment.
"True friends stab you in the front." | Oscar Wilde
"I need to finish" - Christian Wolff: The Accountant
Just trying to live long enough to play a new, released MMORPG, playing New Worlds atm
Fools find no pleasure in understanding but delight in airing their own opinions. Pvbs 18:2, NIV
Don't just play games, inhabit virtual worlds™
"This is the most intelligent, well qualified and articulate response to a post I have ever seen on these forums. It's a shame most people here won't have the attention span to read past the second line." - Anon
That sounds really technical.
So I watched about half of it and skipped around and skimmed some of the rest to understand what they are doing at a high level and I never heard them talk about making Unreal 5 multi-threaded. Matter of fact I would be shocked if Unreal 5 wasn't multi-threaded out of the box. It's difficult to imagine a high performance client engine being single threaded. Now that may just be a misunderstanding on your part. I don't know how technical you are. Not big deal if you're not. However multi-threading is simply the ability of a program to execute multiple threads (or commands) against the CPU simultaneously. It's how a program can benefit from a multi-core CPU which every CPU these days is multi-core. I don't know that you could even buy a single core CPU anymore. Even your phone has multiple cores.
Anyway...
What I did hear that they had to go tear the guts out of Unreal 5 was that they had to make it multi-server capable. That is, the client has to interact with 1-n servers, not just one server which is how Unreal 5 is coded out of the box. That was the complicated bit from the client side coding.
What they're doing server side is VERY interesting. I'm pretty sure the Star Citizen team is taking notes because basically they are solving the same problem Start Citizen has. The difference being that where Star Citizen has some slides and lots of talks about what needs to happen, AoC has videos of it working in real time with debugging enabled to show you where the server boundaries are. This corroborates my thought on the Star Citizen team that they are basically just not up to the technical task before them. I don't want to use the word incompetent because it's not some easy task to solve, and I'm sure individually they have a lot of smart people on their team, but it seems like maybe they didn't have the right smart people on the team (or poor leadership) because here we have the AoC coming along and solving the exact same problem with less budget and fewer developers (that I am aware of).
My quick short version of how this works so you don't have to watch an hour-long video follows. Feel free to correct me if I mess anything up.
First of all, everything is server authoritative meaning they will have much tighter controls on hacking. "Actors" is the terminology they user for anything in the game world that can have an interaction. The only thing that wouldn't be an actor would be like the terrain itself. Even the trees are actors because the wind will blow in a direction and the trees will react to the wind. Trees can be cut down, etc. If it can act, it's an actor.
Now understanding that, each server is responsible for a geographic area and all the actors in it. As an actor starts to approach a server boundary, the server starts negotiating with it's neighbor to spawn what they call a "proxy" of the actor which looks and acts just like the actor itself. The server that owns the actor starts replicating its data for the actor to the neighboring server which is how the proxy is indistinguishable to the actual actor because it will be in the same location, play the same animations, take damage, get healed, etc just like the actual actor does. This way you can have a real-time pvp battle with someone across the server boundary in real time. The video shows no perceptible lag in this scenario. When the proxy crosses the server boundary, the actor is basically handed off to the neighboring server and it now becomes the owner of the actor and the process is reversed until the actor moves out of visual range an no longer needs to be replicated across servers. This can happen in a corner situation as well where there are 4 servers in a corner boundary situation. The proxy/replication will happen between all 4. Finally the server boundaries are not fixed. They can be dynamically reallocated to apply more server processing where it's needed in real time. For example if there's a large pvp battle going on, the servers will reallocate such that the battle itself is split up among several servers to divide the load while relatively empty areas will be consolidated into a single server since it would be plenty to manage a large, empty area.
Pretty cool and sophisticated stuff. I'm not a big pvp guy, so I'm not sure the game is going to be my cup of tea, but the technology behind it is really cool and interesting for sure.
I think they meant the multithreading part at 1:08:20 where they talk about multithreading the Unreal replication graph to improve server performance and how Unreal replication is one of the most single threaded systems. I've also heard about how Unreal Engine is very single threaded.
As for comparing to Star Citizen, my understanding is they want individual servers down to room scale or individual ships. All while the planets and ships themselves are moving because they are so stubborn in wanting every tiny object to be persistent. Meanwhile Ashes of Creation seems to work on grids since they call it dynamic gridding and from what their slides have demonstrated. I don't have the technical knowledge for how much of a difference that is but the amount of physics simulation in Star Citizen does seem to make things more difficult. Both games are also at the stage where they can be demonstrated and tested with Star Citizen having already done an 800 player test, so it will be interesting to see how AoC does.
Yeah sure Star Citizen is big but 99.9% of that is empty space or procedurally generated planets with no wildlife or anything. Even if it wasn't, I think that would just be a matter of scaling up. I don't see why the same tech wouldn't apply here.
Capital ships being its own moving server definitely sounds unique though but honestly... I am skeptical on that one ever working.
Every object being persistent is also probably unique to SC but whats going to stop trolls from piling tens of thousands of objects in random spots on a planet? I have doubts on that one too.
I am not fully defending Star Citizen and their development fiasco, but even so in the recent few months they have been ever so slightly justifying the prolonged development cycle.
At their "CitizenCon" event in October they have shown off an interesting amount of content, which as far as I understood, nearly half of it has already or is on it's way to be implemented and during one of their presentation they have shown off a functional demo of the "Replication Layer" and "Server Meshing" technologies, which does seem quite intriguing and showcases that even while on different servers, entities and players are able to interact with one another seamlessly (or that is at the very least the intent). Furthermore they have conducted multiple test of the early meshing technology this year and judging by the many posts on reddit, the players have been pleasantly surprised by it's functionality.
While the Star Citizen developers are notorious for letting their fanbase down and we are used to judge Star Citizen as a money laundering scheme, the recent few months have been interesting in observing as they are slowly progressing in development. On the other hand AoC seems to be tackling a similar task and while i couldn't agree more that their progress is much faster, thus proving the lacking organization of Star Citizen dev team.
Either way i am looking forward to this competition between the two. It will be interesting to see who will be able to bring this technology into the hands of the players first and how it will affect them. So far it seems that Scam Citizen has a bit of an advantage with the already performed tests.