Damage ticks, boons, fields/finishers, runes that operate on hit - Yeah I think you give ED too much credit in the calculations department.I would assume that the amount of dataflow in a 6DOF game like ED is several magnitudes higher compared to a relative simple game like GW2. Physics, mainly flight model and collision detection, are a lot more complex in ED. So you're basically comparing apples with peanuts (wouldn't mind oranges).
Why did Fdev choose to go the way they have with the netcode?
- Thread starter stevesideas
- Start date
- Status
Unlike the forum.........
But I'd assume that at the very early stages a decision was made to go P2P for financial reasons, which if you consider how the game was originally funded seems the sensible option.
I'd say P2P is here to stay but who knows what the future may bring.
Last edited:
Its entirely possible to change. Probably not economic to do so though.
Last edited by a moderator:
Sadly no collision detection since it would bring every ones computers to their knees in a 80vs80vs80 fight in world vs world and PvE folk would scream their heads off if you added it in their mode. As for just a few numbers I would encourage you to look at a player build with all of the RNG elements (yuck) that they have then multiply it by 240 (the amount of players in 1 WvW map.
I've posted about this several times and i do believe its one of FDs biggest design choice mistakes but essentially yeah...You only use P2P if you lack funds, expertise or a mixture of the 2.
Unfortunately by designing the game on a P2P multiplayer premise, it has categorically ruled out several very good potential features in the future that cannot be done on this kind of system.
Unfortunately by designing the game on a P2P multiplayer premise, it has categorically ruled out several very good potential features in the future that cannot be done on this kind of system.
Dunno why they did it originally, but among forum white knights it seems to have become the go-to excuse for why certain features or bug fixes are fundamentally impossible and therefore we should never ask for them. It seems to render "impossible" (and therefore unrealistic/unreasonable to ask for) persistent NPCs, non-RNG based game systems, persistent, changeable/destructible structures, and prevention of Combat Logging, among other things. So that's a few more things the devs don't ever have to work on. I guess that's as good a reason as any.
Destiny 2 is using peer 2 peer an activision doesn't lack funds.
Having changeable, destructible structures has nothing to do with Netcode, it all about how you do that in a MMO type game, where you are playing along with thousands of other gamers playing across different instances, not all visible at once to all players.
Figuring out the game play mechanics of that is fairly tricky, not recording the state of a structure in a instance after player activities and then storing that state in a database so it can be sent to other players when they enter into a system, you could even have a reset time for each system as you don't end up with a stupidly large database.
If you allowed players to destroy stations across the bubble their wouldn't be many left. There would difficulty with keep everything in synced across different instances in real time, an what do you do if there dozens of different instances running at the same time for a single system like Lave?. You face that challenge no matter what type of network design choose. Instead it all down to how you design the game mechanics, most MMOs like STO do it by having events which you queue for that are more like levels in a offline game that resets every time you enter them.
They already drastically improve persistent NPCs from one instance to the next compare to the early days but this has little to do with the netcode and more to do with what data you are collecting an storing server side and how you are using that build players instances.
This is the problem people like you haven't got a clue what you are talking about, so it impossible to have a sensible discussion.
Last edited:
Which is exactly why they choose the cheaper option, in this instance peer to peer.
In France the government announced that the whole country will have the very high speed with the optical fiber (or Vdsl2 or cable) at the latest for 2022. Hope that the whole world follows the same way
Last edited:
Because there aren't enough players geographically near you? That's hardly Elite Dangerous' fault... (And resorting to VPN to trick match-making will only result in both you AND all other players in your instance having a lousy experience...)
Which is why ED's match-making code *attempts* to avoid putting people with poor internet connections with others who have good internet connections. It's not that great at it (although I think has gradually improved), and it can also be tricked by a VPN from the sound of it.
No doubt in 20 years time, when everyone has faster internet connections, IPv6 is here (optomistic I know),, all those dodgy routers and modems have been replaced, we will be able to have larger instances with better performance.
Until then, be happy that we can get it working this well.
A central server, as others have pointed out, solves nothing. Just means more hops for the data.
Until then, be happy that we can get it working this well.
A central server, as others have pointed out, solves nothing. Just means more hops for the data.
Cosmo and Robby are correct in my opinion.
It's nothing to do with 'netcode' magic and every thing to do with crapola connections, wacky routers that have been diddled with and worst of all, potato boxes.
Monthly payments would cause alot of folks to give up on our game.
Cheers, I got 8 edges and 4 blades to put in-service tonight. Thank you for choosing Bell.
o7
It's nothing to do with 'netcode' magic and every thing to do with crapola connections, wacky routers that have been diddled with and worst of all, potato boxes.
Monthly payments would cause alot of folks to give up on our game.
Cheers, I got 8 edges and 4 blades to put in-service tonight. Thank you for choosing Bell.
o7
A lot of people here don't seem to know that Cloud-Computing exists, and can be scaled automatically to a games needs. If no players are logged in or aren't close enough to each other, no cloud-based server would be in use and thus it would cost exactly ZERO until it's actually started.
P2P isn't the future, our internet connections are too different for it to be reliable. It's just super cheap, that's it. The future is Cloud as a service, e.g. AWS or the dozen other proven services out there.
The sole reason why FD won't do that is because they would have to hire one or more experts to manage the setup of the backend (plus the obvious rewrite of the game engine). And that is costly, but only one time. But obviously, keeping the P2P "solution" is not going to really cost them, so they keep it.
And don't you guys think FD isn't making enough money through cosmetics and the sales of their other games? They don't need a subscription based model, the costs are not astronomical, especially with CaaS.
P2P isn't the future, our internet connections are too different for it to be reliable. It's just super cheap, that's it. The future is Cloud as a service, e.g. AWS or the dozen other proven services out there.
The sole reason why FD won't do that is because they would have to hire one or more experts to manage the setup of the backend (plus the obvious rewrite of the game engine). And that is costly, but only one time. But obviously, keeping the P2P "solution" is not going to really cost them, so they keep it.
And don't you guys think FD isn't making enough money through cosmetics and the sales of their other games? They don't need a subscription based model, the costs are not astronomical, especially with CaaS.
- Status