Everybody spawns in at the same 4 locations, and even if each runs its own server, thats still hundreds of players on a single server for these big tests. Theres no dynamic sharding, so meshing will always have bottlenecks. Those bottlenecks will either manifest on the simulation server or the replication layer. Theoretically players could stay spread out enough or separated in such a way to avoid bottlenecks, but the game will have to encourage that, and since theres often no rails to keep players in their place, bottlenecks can and will still emerge unless the game can matchmake players into other shards as a fallback. They are going to have to at some point balance the player expectation of persistence with performance. I don’t expect star systems to grow in player size much until they do this, or at least have more granular dynamic meshing, but hopefully we’ll start to see better server tick rates.
they need to give up on a realistic persistance. the hangars should be the only true persisting environment and the persistence that we have outside just put a 2 hour time limit . They never thought about server hopping guys jhust leaving massive ships on the landing pad. griefing peoples vert hangars by placing objects on the doors etc..
There are so many edge cases to account for in this game. It must be an absolute nightmare to make a change without affecting something critical to someone’s gameplay. As a developer, this both excites and terrifies me.
29:02 for QA, tests are still written by people. If an edge case wasn't thought of, it generally isn't tested for. This is why there's a general saying that the testing phase takes an equal time as it takes for development. So for example, if it takes 10 hours of development, it'll take another 10 hours to test it. So total time estimation isn't 10 hours to complete, it's 20 hours.
I'm hopeful that this game will be THE example for space simulation moving forward. It's incredible to me, the vastness and different paths we can take, both physically and career-wise. If this game ends up even 75% of all its been dreamed to be, I'll be just fine with that. Side note: ATC issues are at the forefront of my issues with the game. Along with server volatility.
Starcitizen will be the TRUE testbed of a TRUE seamless MMO that ALL the biggest studios out there want to license or buy. CIG is driving the innovation where their technologies put together are game changing stuff. Chris Roberts is the right guy for the job regardless what people say, because he's a hopeless pioneer that wants the best of the best of the BEST! That's what we in the end will get.
@@TruthIsKey369 I sincerely hope so! I can say, without a doubt, this game has captured my attention moreso than ANY other game I've played in recent memory. I haven't played anything else since downloading this game. It scratches the itch for me as far as space simulation goes. And after seeing the rework of StarEngine, I'm so hopeful that this game will encompass everything I love about video games. Let's hope they can get it done!
World of Warcraft has had that server tech for a while now. Merging of servers and sharding is not a completely new tech. The main reason it's taking so long is because they have to work around all the limits they already created In the engine themselves.
@@arne1881 Admittedly, I don't know a whole lot about development. That's probably obvious. And my comment comes from the desire to have this game develop into something great. If it does, great. If it doesn't, hopefully something else comes in it's place.
@@Randaddy25 yea, if this game comes out as anything close to what they promised, it might be my main game for many years. I'm very optimistic, just very annoyed at how slow the development is. Being a games programmer myself, it blows my mind how long time they spend on certain features. I know a few people working at CIG, and the way they operate atm is not optimal.
This is why I back this game financially. The whole idea, the sheer size of these people's balls to even try this whole crazy project is mind boggling! They have guts, they have vision and they are insane!! Its the sheer audacity that I love so much about this project. Keep going CIG and never let the doubters deter you! 👍
Exactly. I want the game that Chris Roberts is building. I want all of it. I don't care about "feature creep" or pushed back promises. I want it all and I am patient to wait until it is all done right. It is rough going, but I am happy with what I am seeing. Some things will improve. Some things will get worse before they improve. Some things will change entirely. We are not really playing Star Citizen. We are playing in a test bed. I am just along for the ride. I am enjoying the ride despite all the bugs that are expected in a testing environment.
I understand why people say CIG doesn't test patches before release, because when you get into the game and try to do the most basic things, everything is bugged and doesn't work. You can't complete half the missions without something breaking.
They really dont have the manpower to have 1000 devs on a server together.. Thats why they do extensive testing with the community coz u know its an alpha
In regards to bugs in the code; in programming, the more complex your systems, the more likely you are to get bugs. That seems obvious, but then you have situations where you have already complex systems that are prone to bugs or edge cases (by which I mean unusual user behaviors), and you couple them to other complex systems that have their own problems (microservices, if you will). That interaction between systems is it's own nightmare of bug spawns, and can be very hard to diagnose. This game has TONS of complex systems that interact with and rely on each other, meaning an exponential increase in complexity and chance of bugs. Changing one system can (and almost certainly will) have unnoticed and unwanted knock on effects in another system that you might not notice during initial development, because it's easier to test a system in isolation. The real kicker is when you have a bug that you think you've fixed, but you later realize your fix didn't solve the problem; not because your fix didn't matter, but because the same issue can be caused by different conditions in the code. Maybe you successfully diagnose one cause in one system, but another goes unnoticed because you were testing the system in isolation. Maybe you solved the issue, but a fix to a different issue someone else does causes a regression that breaks your fix. Maybe you're just getting woefully unhelpful bug reports that don't accurately reflect what circumstances occurred to cause the issue, and having read some of the bug reports on Spectrum I can say that this is common (this is uniquely frustrating, and can result in days of wasted effort trying to diagnose and recreate problems). All of this to say, when you wonder if they tested for bugs before pushing the code, they almost certainly did; there's just a limit to how thorough any one dev and their reviewer are, when they have multiple competing tasks and priorities. Not excusing pushing bugs, but that's a quick rundown of just a few reasons it happens. No one does it intentionally. No one full sends bad code into the world maliciously. Every one wants to think that they're doing their best under whatever conditions in which they're working. I get the fatigue players have with it the recurring problems, but it's easier (and will make you happier in the long run) to extend some grace (and maybe take a break) in the face of bugs.
@21:22 I think in a PTU test with Server mesh, less players, that dont spawn i the same location (pyro/stanton) will happen really soon and will problably work pretty good. The big question are if RSI needs more rmq testing first. Fingers crossed that they do both of them soon.
Hardware not there yet at feasible price. Software work around to help process the immense data especially with so many players and systems running. Hopefully the hardware will be there and by that time. The game with be epically optimised and the new hardware will blitz through it all and make it cheaper to run the game.
I've heard this example of mission cargo being "staged" in stations and it makes zero sense to do so. There should never be a situation where cargo boxes are stored in an "inventory" when they aren't being directly managed by players. There is no reason to spend the resources beyond holding a token amount of data until it is about to be rendered to a player. Until actually setting up cargo into a rendering situation (being put into an elevator, populated in the environment, etc) there is NO reason to put resources into it. Clone a box and slap an id on it for a mission shouldn't be a major process. For example, if a hauling mission states 96 scu of iron to haul from point a to point b, there is zero reason to create 96scu of actual boxes until they are about to be seen by a player. It sounds like they are wasting tons of resources on unimportant guff to theoretically save time down the line (which is ass backwards development). The same goes for just about everything else being managed by kiosks. If they're storing more than the absolute minimum in player inventory, they're bloating their system from the ground up. The cargo kiosks should only be lagging long enough to receive a simple list of all the stuff in inventory (any information that's actually displayed in the kiosk interface). If a player moves items to the elevator, the server should then start worrying about getting all the necessary data to handle rendering it into the game world. In the case of cargo missions, there should be no actual cargo crates until a user actually stages them to go on the elevator.
@@criseichenlaub6527that depends on how you see things and let it affect you negatively. I'm very positive because I understand this kind of development takes a lot of time, and I don't have time to whine about stuff that is out of my control. Also, I have to little time and to much to do, and no time to waste on being a little shitkid that lack patients.
Why? You will be able to get them back. Sucks we have to wait for a reset now, but there will be special claim terminals for pledge items in the future.
@@Brigadier_Beau your pledge and purchased items should be a baseline the moment you respawn it should already be on your person but all your non purchased or pledge items should be on your old body to get unless some how they add a form of insurance for those items
@@Brigadier_Beau how is that if it's not equipped on your person? How is that when you die only the stuff you had equipped that was purchased or pledged is still in your inventory when you would normally have nothing? When you respawn In my scenario
All other game Studios work in complete darkness..... Hence garbage games like Starfield, Concord, oh my...... That is DARKNESS, while CIG is just below the sunset in comparasion.
You guys are ruining a $330 ship in the redeemer and why? Because you don’t want us to have an opinion to take out a Capitol ship? I’ll never spend another dime and I’ve spent plenty. What it needed was more flares because it’s slow but now it’s finally dead. One of the coolest ships in the game.
You guys are ruining a $330 ship in the redeemer and why? Because you don’t want us to have an opinion to take out a Capitol ship? I’ll never spend another dime and I’ve spent plenty. What it needed was more flares because it’s slow but now it’s finally dead. One of the coolest ships in the game.
Everybody spawns in at the same 4 locations, and even if each runs its own server, thats still hundreds of players on a single server for these big tests. Theres no dynamic sharding, so meshing will always have bottlenecks. Those bottlenecks will either manifest on the simulation server or the replication layer. Theoretically players could stay spread out enough or separated in such a way to avoid bottlenecks, but the game will have to encourage that, and since theres often no rails to keep players in their place, bottlenecks can and will still emerge unless the game can matchmake players into other shards as a fallback. They are going to have to at some point balance the player expectation of persistence with performance. I don’t expect star systems to grow in player size much until they do this, or at least have more granular dynamic meshing, but hopefully we’ll start to see better server tick rates.
they need to give up on a realistic persistance. the hangars should be the only true persisting environment and the persistence that we have outside just put a 2 hour time limit . They never thought about server hopping guys jhust leaving massive ships on the landing pad. griefing peoples vert hangars by placing objects on the doors etc..
There are so many edge cases to account for in this game. It must be an absolute nightmare to make a change without affecting something critical to someone’s gameplay. As a developer, this both excites and terrifies me.
They are getting paid to be pioneers, and love what they do. Perfect match, and for me very exciting!
29:02 for QA, tests are still written by people. If an edge case wasn't thought of, it generally isn't tested for. This is why there's a general saying that the testing phase takes an equal time as it takes for development. So for example, if it takes 10 hours of development, it'll take another 10 hours to test it. So total time estimation isn't 10 hours to complete, it's 20 hours.
I'm hopeful that this game will be THE example for space simulation moving forward. It's incredible to me, the vastness and different paths we can take, both physically and career-wise. If this game ends up even 75% of all its been dreamed to be, I'll be just fine with that. Side note: ATC issues are at the forefront of my issues with the game. Along with server volatility.
Starcitizen will be the TRUE testbed of a TRUE seamless MMO that ALL the biggest studios out there want to license or buy. CIG is driving the innovation where their technologies put together are game changing stuff.
Chris Roberts is the right guy for the job regardless what people say, because he's a hopeless pioneer that wants the best of the best of the BEST!
That's what we in the end will get.
@@TruthIsKey369 I sincerely hope so! I can say, without a doubt, this game has captured my attention moreso than ANY other game I've played in recent memory. I haven't played anything else since downloading this game. It scratches the itch for me as far as space simulation goes. And after seeing the rework of StarEngine, I'm so hopeful that this game will encompass everything I love about video games. Let's hope they can get it done!
World of Warcraft has had that server tech for a while now. Merging of servers and sharding is not a completely new tech. The main reason it's taking so long is because they have to work around all the limits they already created In the engine themselves.
@@arne1881 Admittedly, I don't know a whole lot about development. That's probably obvious. And my comment comes from the desire to have this game develop into something great. If it does, great. If it doesn't, hopefully something else comes in it's place.
@@Randaddy25 yea, if this game comes out as anything close to what they promised, it might be my main game for many years. I'm very optimistic, just very annoyed at how slow the development is. Being a games programmer myself, it blows my mind how long time they spend on certain features. I know a few people working at CIG, and the way they operate atm is not optimal.
This is why I back this game financially. The whole idea, the sheer size of these people's balls to even try this whole crazy project is mind boggling! They have guts, they have vision and they are insane!! Its the sheer audacity that I love so much about this project. Keep going CIG and never let the doubters deter you! 👍
I agree! (Don't say this on the subreddit)
Exactly. I want the game that Chris Roberts is building. I want all of it. I don't care about "feature creep" or pushed back promises. I want it all and I am patient to wait until it is all done right. It is rough going, but I am happy with what I am seeing. Some things will improve. Some things will get worse before they improve. Some things will change entirely. We are not really playing Star Citizen. We are playing in a test bed. I am just along for the ride. I am enjoying the ride despite all the bugs that are expected in a testing environment.
I understand why people say CIG doesn't test patches before release, because when you get into the game and try to do the most basic things, everything is bugged and doesn't work. You can't complete half the missions without something breaking.
They really dont have the manpower to have 1000 devs on a server together.. Thats why they do extensive testing with the community coz u know its an alpha
We are the testers. That is the point of the alpha.
In regards to bugs in the code; in programming, the more complex your systems, the more likely you are to get bugs. That seems obvious, but then you have situations where you have already complex systems that are prone to bugs or edge cases (by which I mean unusual user behaviors), and you couple them to other complex systems that have their own problems (microservices, if you will). That interaction between systems is it's own nightmare of bug spawns, and can be very hard to diagnose. This game has TONS of complex systems that interact with and rely on each other, meaning an exponential increase in complexity and chance of bugs. Changing one system can (and almost certainly will) have unnoticed and unwanted knock on effects in another system that you might not notice during initial development, because it's easier to test a system in isolation. The real kicker is when you have a bug that you think you've fixed, but you later realize your fix didn't solve the problem; not because your fix didn't matter, but because the same issue can be caused by different conditions in the code. Maybe you successfully diagnose one cause in one system, but another goes unnoticed because you were testing the system in isolation. Maybe you solved the issue, but a fix to a different issue someone else does causes a regression that breaks your fix. Maybe you're just getting woefully unhelpful bug reports that don't accurately reflect what circumstances occurred to cause the issue, and having read some of the bug reports on Spectrum I can say that this is common (this is uniquely frustrating, and can result in days of wasted effort trying to diagnose and recreate problems).
All of this to say, when you wonder if they tested for bugs before pushing the code, they almost certainly did; there's just a limit to how thorough any one dev and their reviewer are, when they have multiple competing tasks and priorities. Not excusing pushing bugs, but that's a quick rundown of just a few reasons it happens. No one does it intentionally. No one full sends bad code into the world maliciously. Every one wants to think that they're doing their best under whatever conditions in which they're working. I get the fatigue players have with it the recurring problems, but it's easier (and will make you happier in the long run) to extend some grace (and maybe take a break) in the face of bugs.
Time to get my coffee and listen.
@21:22 I think in a PTU test with Server mesh, less players, that dont spawn i the same location (pyro/stanton) will happen really soon and will problably work pretty good. The big question are if RSI needs more rmq testing first. Fingers crossed that they do both of them soon.
12:11 my first gaming PC 486/33 with turbo button and 130 meg hard drive! 🎉
Now, with the "radartech" in 3.24 i really hope we´ll get a new 1000 test soon.
The most interesting things about SC are the TH-cam creators
Hardware not there yet at feasible price. Software work around to help process the immense data especially with so many players and systems running. Hopefully the hardware will be there and by that time. The game with be epically optimised and the new hardware will blitz through it all and make it cheaper to run the game.
I've heard this example of mission cargo being "staged" in stations and it makes zero sense to do so. There should never be a situation where cargo boxes are stored in an "inventory" when they aren't being directly managed by players. There is no reason to spend the resources beyond holding a token amount of data until it is about to be rendered to a player. Until actually setting up cargo into a rendering situation (being put into an elevator, populated in the environment, etc) there is NO reason to put resources into it. Clone a box and slap an id on it for a mission shouldn't be a major process.
For example, if a hauling mission states 96 scu of iron to haul from point a to point b, there is zero reason to create 96scu of actual boxes until they are about to be seen by a player. It sounds like they are wasting tons of resources on unimportant guff to theoretically save time down the line (which is ass backwards development). The same goes for just about everything else being managed by kiosks. If they're storing more than the absolute minimum in player inventory, they're bloating their system from the ground up. The cargo kiosks should only be lagging long enough to receive a simple list of all the stuff in inventory (any information that's actually displayed in the kiosk interface). If a player moves items to the elevator, the server should then start worrying about getting all the necessary data to handle rendering it into the game world. In the case of cargo missions, there should be no actual cargo crates until a user actually stages them to go on the elevator.
Good video, ‘Mater
"Ben-Noir" .. :D
Resume @7:29
Stability??? 🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣
4th is elevator :D
After last years Citcon people were saying we will get pyro by the end of 2023 😂😂😂
To be fair, they sort of did.
Toxic optimism. This project is plagued with it.
@@criseichenlaub6527that depends on how you see things and let it affect you negatively. I'm very positive because I understand this kind of development takes a lot of time, and I don't have time to whine about stuff that is out of my control. Also, I have to little time and to much to do, and no time to waste on being a little shitkid that lack patients.
Pledge items and paid items should be NON LOOTABLE
Why? You will be able to get them back. Sucks we have to wait for a reset now, but there will be special claim terminals for pledge items in the future.
@@Brigadier_Beau your pledge and purchased items should be a baseline the moment you respawn it should already be on your person but all your non purchased or pledge items should be on your old body to get unless some how they add a form of insurance for those items
@@YEAROFTHELOTUS If all of my pledge items appeared on my person, I would not be able to move. If you include the ships, it would crash the game.
@@Brigadier_Beau how is that if it's not equipped on your person?
How is that when you die only the stuff you had equipped that was purchased or pledged is still in your inventory when you would normally have nothing? When you respawn
In my scenario
Aka: p2w
Did you have a speed increase on? man that was painful to watch if not my computer is jacked lol or I am losing my mind
All other game Studios work in complete darkness..... Hence garbage games like Starfield, Concord, oh my...... That is DARKNESS, while CIG is just below the sunset in comparasion.
Master Modes is shit, server meshing is still years away and CIG is doing cash grabs every chance they get.
It failed.
it was a crash test, dummy
44:00 How can someone working for CIG make that mistake! I'm an amateur and I know better. And how did CIG not notice when they were handed the asset!
You guys are ruining a $330 ship in the redeemer and why? Because you don’t want us to have an opinion to take out a Capitol ship? I’ll never spend another dime and I’ve spent plenty. What it needed was more flares because it’s slow but now it’s finally dead. One of the coolest ships in the game.
Its a gun ship, its not designed to take on that weight class alone.
Bro really commented the same exact thing twice💀
"Finally dead" as if it won't be changed again before release anyway, you're dramatic.
You guys are ruining a $330 ship in the redeemer and why? Because you don’t want us to have an opinion to take out a Capitol ship? I’ll never spend another dime and I’ve spent plenty. What it needed was more flares because it’s slow but now it’s finally dead. One of the coolest ships in the game.
Why are you commenting under here like you are talking to Dev? What's the point here?
The redeemer will be rebalanced again don’t worry it will be better again
@@UnknownMoses doubt it. CIG will more than likely sell a new gun boat
I mean I agree that it's a dead ship but how is this at all relevant to this video or it's creator?
You have no idea if it’s ruined or not. Everything about combat is about to change.