If your slowest player in a real-time game is still connected at 56k, then everyone playing with them will be constrained to 56k.
The Star Citizen Thread v5
- Thread starter Brett C
- Start date
- Status
If your slowest player in a real-time game is still connected at 56k, then everyone playing with them will be constrained to 56k.
Okay - so how do you think they will go from 16/24 which they can't currently make work - to even 50 -100 players (LOL) - to 1000 players?
What exactly is it about their performance so far that gives you confidence they can do this thing that Planetside did years ago?
So you didn't disprove anything I have said. The average network bandwidth in USA in 2016 is 12.6 Mbps, and ADSL introduced at the end of the last century had speeds of 1 Mbps downlink. Do you really think that's the same improvement as in computing like the one between original Doom and Doom (2016)? Remember that bandwidth isn't everything, networking hardware (routers and so on) didn't improve that much, and the speed of light remains annoyingly constant. Use real arguments, stop being facetious.
Edit: Different explanation - networking improved in last 15 years, but it doesn't mean it has improved as fast as you think, and it doesn't mean it has improved enough for the game to lve up to CR's claims.
Last edited:
Gamer: I wonder if SC will have <insert literally any feature>.
Chris Roberts: Yes, it will.
CIG Staff: Wait, what?
Elon Musk, Peter Jackson... Chris Roberts is more like that guy at work that owes you twenty dollars than any kind of notable business legend. That he always makes the same mistakes usually has the same corollary: he NEVER learns from his mistakes.
He'll gladly pay you Tuesday for a hamburger today, in layman's terms.
He'll gladly pay you Tuesday for a hamburger today, in layman's terms.
Last edited:
Right, I keep hearing this "people say *thing* is impossible, we sure showed them" statement.
1. No one in the last 40 years at least have said creating an electric car was impossible, it's widely known to be possible but incredibly costly and also enviromentally challenging due to the current tech of batteries generally producing more carbon to create and having a short lifespan than what a normal petroleum powered car would produce over it's intended lifespan.
2. No one has ever said creating a multiplayer space game with ships and fps was impossible and there are many examples of multiplayer (both mmo and arena) spacey games with both fps and vehicles, but many have said that CIG would likely fail at trying.
CIG have spent over a year trying to make an airlock door function correctly, the complete and permenant fix still eludes them to this day. Time does not always mean better, some things when they were designed were flawless in ways that even the creators didn't expect and have stood the test of time with no one else being able to better the initial design or build. A lot of people think taking longer or spending more money on things always results in improvement, it doesn't and it is an incredibly naive and limiting view.
The creation of SC is likely not possible for CIG with it's current leadership and working practices in regards to sensible time and budget constraints, they spend to much money on faffing and waste too much time on superflous oddities in order to increase the cash and drag out the time. In other words, they are too busy trying to drag in money and placate backers to give them more time constantly rather than just getting on with making something that is actually quite simple. Maybe had they not spent far too much time making flash ship commercials and fluff vids they could have poured that energy into actually designing a solid tech foundation to build on.
People keep chasing this "they said it was impossible" thing everytime someone comes up with a lightbulb that is slightly dimmer and has a shorter operational lifespan than the one it was designed to replace and someone else had said beforehand that "probably not a good idea", as if somehow just saying the words makes it an incredible achievement over all odds. CIG are not competent, disiplined or knowledgable enough to make a door work, yet they are going to reinvent the very complex and tightly managed way in which someone else's server farms communicate with each other and client systems over a worldwide stage and give it 100x it's current efficiency for their computer game? HAHAHAHAHAHAHAHAHA, yeah, sure boss.
It's interesting that most PC gamers have some idea about the limitations of their graphics card, monitor and CPU but have never had much of an inkling when it comes to networking technology. Most of us are perfectly happy with the stock router our ISP provides and the network card built into our computer's motherboard. Yet we have no idea what the limits are regarding online games and we are perfectly happy to buy into the concept that an online simulation game currently throwing out huge amounts of real time data can apparently synchronise us with "hundreds" or "thousands" of other players using some some magic software.
It's the most difficult thing to get right and pretty disingenuous of a games developer to claim they can do something without having any deployed proof of concept let alone take, what is essentially, pre-order cash for it. It's almost as bad as saying that you are going to support VR and wanting pre-order cash without having first fully explored the pitfalls and realities of it and adjusting the game design accordingly.
I've had discussions with people on this very forum, maybe four years ago, about this very subject. It's long been claimed that SC will categorically somehow support massive instances of players experiencing real time simulation of some very busy cryengine levels, all because of some claims by developers that technology X would allow it and some back of a napkin theories about how it must work.
If you don't have the goods then you have nothing new to sell, it's that simple really.
It's the most difficult thing to get right and pretty disingenuous of a games developer to claim they can do something without having any deployed proof of concept let alone take, what is essentially, pre-order cash for it. It's almost as bad as saying that you are going to support VR and wanting pre-order cash without having first fully explored the pitfalls and realities of it and adjusting the game design accordingly.
I've had discussions with people on this very forum, maybe four years ago, about this very subject. It's long been claimed that SC will categorically somehow support massive instances of players experiencing real time simulation of some very busy cryengine levels, all because of some claims by developers that technology X would allow it and some back of a napkin theories about how it must work.
If you don't have the goods then you have nothing new to sell, it's that simple really.
Last edited:
Well, they kind of do. I'm currently shifting some stuff around at a rather pedestrian 32Gb/s. Networking has come on leaps and bounds in the last 15 years - unfortunately
the infrastructure to support such high speeds on external networks is usually hosed due to inept / greedy governments, companies without a clue, and misguided treehuggers putting the lesser spotted runny-bottom virus before decent internet
32GB/s is a goodly figure. My home internet is 40MB/s down, 10MB/s up, which is LAN speed over 10 years ago (athough it works out a bunch slower due to routing obviously). Ahhh, after-hours Quake deathmatch in the office in the late 90s. That was fun, and all at 10Mbps!
However what has also changed is usage. 15 years ago doing anything realtime outside a local network was very rudimentary.
We still have the problem of bandwidth vs fidelity vs tick rate for FPS. "Fidelity" requires, relatively speaking, a lot of data to be passed between server & client, especially if we start getting into limb positioning/bullet tracking. Then add voice comms and all of the other stuff. Then keep multiplying when you add users. There's only so much that you can get away with by slinging data across UDP however good you are with compression. EA have run the Battlefield franchise for years and you don't get more than 64 people on a server on relatively small maps. CSGO is five a side, with performance and hit accuracy as priority. Forget the Planetside folks, look at the folks doing competitive FPS and see what they have brought to the table.
Plus, as His Derekness pointed out, if you're doing a strict client/server architecture your bandwidth costs will be brutal.
Right, I keep hearing this "people say *thing* is impossible, we sure showed them" statement.
1. No one in the last 40 years at least have said creating an electric car was impossible, it's widely known to be possible but incredibly costly and also enviromentally challenging due to the current tech of batteries generally producing more carbon to create and having a short lifespan than what a normal petroleum powered car would produce over it's intended lifespan.
2. No one has ever said creating a multiplayer space game with ships and fps was impossible and there are many examples of multiplayer (both mmo and arena) spacey games with both fps and vehicles, but many have said that CIG would likely fail at trying.
CIG have spent over a year trying to make an airlock door function correctly, the complete and permenant fix still eludes them to this day. Time does not always mean better, some things when they were designed were flawless in ways that even the creators didn't expect and have stood the test of time with no one else being able to better the initial design or build. A lot of people think taking longer or spending more money on things always results in improvement, it doesn't and it is an incredibly naive and limiting view.
The creation of SC is likely not possible for CIG with it's current leadership and working practices in regards to sensible time and budget constraints, they spend to much money on faffing and waste too much time on superflous oddities in order to increase the cash and drag out the time. In other words, they are too busy trying to drag in money and placate backers to give them more time constantly rather than just getting on with making something that is actually quite simple. Maybe had they not spent far too much time making flash ship commercials and fluff vids they could have poured that energy into actually designing a solid tech foundation to build on.
People keep chasing this "they said it was impossible" thing everytime someone comes up with a lightbulb that is slightly dimmer and has a shorter operational lifespan than the one it was designed to replace and someone else had said beforehand that "probably not a good idea", as if somehow just saying the words makes it an incredible achievement over all odds. CIG are not competent, disiplined or knowledgable enough to make a door work, yet they are going to reinvent the very complex and tightly managed way in which someone else's server farms communicate with each other and client systems over a worldwide stage and give it 100x it's current efficiency for their computer game? HAHAHAHAHAHAHAHAHA, yeah, sure boss.
CR is too busy wringing the wallets dry of gullible and misinformed citizens to be bothered with creating the BDSSE these days.
Not that it's a SCAM or anything like that at all.
The jpegs must flow.
I meant infiniband 40Gb/s - there is something therapeautic about watching full .iso's pop up every second or two
Happy Hour: February 17th, 2017
TLDR
Source: https://relay.sc/transcript/happy-hour-february-17th-2017-summary
TLDR
- Today's guests were Senior Writer, Will Wiessbaum, QA Tester, Michael Falkland, and Streamer Clifford aka Miku
- The Xi'An apparently have different tastes when it comes to food, some of it would be considered not edible to humans, but delicious to Xi'An. Also the Xi'An favour texture more so than flavour.
- Also Will couldn't go into detail, but they seem to be planning some gameplay around how the Xi'An hold things in comparison to humans
- Artemis was referenced in the show, asking if it'll ever be findable. Artemis was a ship that was piloted by an AI which held a crew in deep sleep. They lost contact with the ship and it's been missing ever since, if you can find it, that's a question you might have to answer yourself.
- The next concept sale is the Hurricane which goes on sale next Friday. Price is unknown.
- If Imperator Kastigan and Messer were pitted against each other in a game of BopIT, Messer would win says Will.
- The Squadron 42 Script is hovering around 1200 pages and is still being tweaked.
Thanks Rolan!
Nearly three whole reams of paper for Genuine Robert's masterpiece.
I wonder how many lines there are of "Copy! Cover me! Run run run!" and <waves hands>
Nearly three whole reams of paper for Genuine Robert's masterpiece.
I wonder how many lines there are of "Copy! Cover me! Run run run!" and <waves hands>
100,000 players in an instance. Eezy peezy.
https://www.reddit.com/r/starcitize...ulti_region_server_going_to_work_for/ddvkdal/
https://www.reddit.com/r/starcitize...ulti_region_server_going_to_work_for/ddvkdal/
- Status