The Fall of FiveM
fivem.team
fivem.team
I was involved in the development of a multiplayer mod for GTA5 nearly 10 years ago. I couldn't understand the technical direction of FiveM at the time because it was based on the existing peer-to-peer, client-authoritative architecture of GTA Online that caused countless problems with cheats/hacks. All it did was provide a way for Lua scripts to call the game's native script functions.
This is in contrast to other mods, such as Alt:V which is the target of much of this tirade. These other mods were inspired by MTA/SA:MP and provided a complete reimplantation of multiplayer in the game where the server is the ultimate authority. Much like any MMO or competitive FPS.
The server only tries to network a limited amount of players to the client (that are nearby) in order to bypass the GTA online protocol player limit. But ultimately this meant that a lot worked on FiveM compared to Alt:V (e.g. traffic, networked scenes, full character appearance etc).
I was one of the developers of Just Cause 2: Multiplayer, which was party to servers with over a thousand simultaneous players. I've also worked on the multiplayer for an unreleased open-world game. So pardon me when I ask this: what are you talking about?
> I couldn't understand the technical direction of FiveM at the time because it was based on the existing peer-to-peer, client-authoritative architecture of GTA Online that caused countless problems with cheats/hacks.
And yet FiveM succeeded where other projects didn't. It's much easier to build on something that already works and evolve it, which is what eventually happened: they switched to server-driven synchronisation half a decade ago.
> These other mods were inspired by MTA/SA:MP and provided a complete reimplantation of multiplayer in the game where the server is the ultimate authority. Much like any MMO or competitive FPS.
That's not what they did, because they couldn't have done it. It is effectively impossible to do that without access to the game's source code - it is intractable to convert a closed-source single-player open-world game into a server because the game is built around the assumption of the world existing around one player, not many.
What every open-world multiplayer mod - including FiveM, MTA, SA:MP, JC2-MP, and others - does is distributed authority: each player is responsible for their slice of the world, including their own state, and they send it to either a central server or other peers. In either case, the client is still implicitly trusted on its view of the world: there is no authoritative simulation to refer to, because that's not logistically or computationally feasible.
What can be done on top of that is validation: other peers, or the server, can verify state changes to make sure they're not too out-of-spec. That's not to be confused with what competitive FPSes do, which is actual server simulation, complete with rewind, lag compensation, and more [0].
> A huge amount of words, a lot which validate the perspective of the people it is directed against.
To conclude: does it really? I read through the entire thing, and it's consistent with what I know of FiveM's development, the people involved, and what happened post-acquisition. There's plenty of evidence that supports the timeline of events as presented here.
[0]: https://developer.valvesoftware.com/wiki/Source_Multiplayer_...
[0]: https://kotaku.com/just-cause-2s-ambitious-multiplayer-mod-m...
All gave the impression that there were a lot of talented working behind the scenes for the love of the game.
Being too young to legally buy MW2 and finding out about AlterIWNet for my not-so-legally owned copy of the game was life changing for me (for better or worse).
Thanks for those teenhood memories that just came rushing back.
Worth noting that he's now become a game publisher [0].
This could have really use some heavy editing. It repeats itself way too much, the writing style is sloppy, and after 4+ hours in reading “we will explain that in detail later” (for the 100th time) made me wanna chuck my phone across the room.
I think around the half-way point I read a line that was “Let’s get started” or “to begin” or similar, I about lost it.
It’s clear this was mainly written by 1 person (Disquse) but I’m sure other did work on this document since that nearly explains some of the repetition and bad copy/paste blocks (some paragraphs are repeated or almost repeated, as if it was a first and second draft right next to each other). Obviously we have to look at this through that lens but I walked away from this monster of a document believing it.
It just rings true, the evidence presented is staggering (both public info like the repo and private like personal chats) and you can really feel the pain in this document.
I was not really to jump on the “Ethan is the devil” train from the start but there is nothing that boils my blood more than incompetent people screwing things up for the hard working people. Even worse when they are paid more and/or put in a position over the people actually doing real work. It seems clear that all of this was reported to Ethan multiple times (ad nauseam, one might say) and he ignore it all, that guy sucks. Period.
I worry that the writing style and length with hinder the message this document is trying to spread. Hopefully the TL;DR (at the bottom) helps some.