Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

How fighting games use delay-based and rollback netcode (2019)

ki.infil.net

Post reply on HN
Print view View original

31–40 of 107 posts

Re: How fighting games use delay-based and rollback netcode (2019)

#31

Post by chowells »

stubish wrote: post #9
frutiger wrote: post #3

Earlier quoted context omitted.

The author of any piece should be permitted a certain set of assumptions about their audience. I think as long as a term has a relatively unambiguous Google result (which is true in this case), it’s fair game to be used in public texts.

The arguments the author made make no sense without some sort of a baseline though. 'The majority always play online' for instance points to a fairly tight definition, and may even be tautological (No True fighting game is designed for offline play).

Every fighting game is designed for offline play, and has been for 25+ years. But that doesn't mean that's where the majority of players are, and that's the point the author is making.

Re: How fighting games use delay-based and rollback netcode (2019)

#32

Post by reitzensteinm »

whateveracct wrote: post #15

Relevant: GGPO, the age-old gold standard of rollback netcode, was recently open-sourced! https://github.com/pond3r/ggpo I haven't used it myself yet - mostly because I'd want Haskell bindings first hah

If your game state is entirely immutable, you're actually most of the way there.

I built a rollback multiplayer system in C#, and most of work goes towards having classes that are mutable, but remember their previous state so they can be rewound to any frame in the past second or so.

This involves enforcing invariants, persistent collection types, generating code to efficiently walk your entire game state, etc.

In Haskell, you just have a list of game states at the end of every frame, pick what you want and go.

Of course, you're not going to get C# performance this way, but you should still be able to build something pretty advanced if you give it a dedicated CPU core.

There are over a thousand people at this moment playing my indie web games built with rollback networking - if anybody has any questions, happy to answer them!

Re: How fighting games use delay-based and rollback netcode (2019)

#33

Post by ThatPlayer »

witherk wrote: post #24
zucker42 wrote: post #10

Super smash bros melee is having a bit of an online renaissance due to a combination of the COVID 19 pandemic and the recent implementation of rollback netcode by a community member that quit his job.

The fact that Fizzi was able to get implement rollback netcode for a game by just directly editing the assembly code without any access to a the game's source code is insane. The delay based netcode for the newest smash game, Smash Ultimate(2018), is so bad that it went from being the biggest fighting game of all time in terms competitive playerbase to basically having no scene at all due to covid killing lan tournam…

Was it done in-game? I figured it was done with the Dolphin (Gamecube) emulator, the same way GGPO was first implemented for arcade emulators.

Re: How fighting games use delay-based and rollback netcode (2019)

#34

Post by anandoza »

thaumasiotes wrote: post #14
shalmanese wrote: post #5

Prediction code for rollback is somewhat akin to branch prediction code in that the dumbest solution works surprisingly well but there's incremental efficiency gains to be had. I wonder if any fighting games have thought to train a neural network per player to try and predict the player's actions N frames ahead. The neural nets could be used for smoother netcode but if the accuracy got high enough, they could, eg: al…

> I wonder if any fighting games have thought to train a neural network per player to try and predict the player's actions N frames ahead. The entire point of playing a fighting game is to attempt to solve this problem. A good player, by necessity, can't be accurately predicted; if they could, they'd be a bad player.

There are many situations where a good player can be predicted because there is a clear best option (or a good option that advance knowledge won't invalidate).

Re: How fighting games use delay-based and rollback netcode (2019)

#35

Post by Impossible »

vxNsr wrote: post #2

I didn’t understand what the author meant by “fighting game” until the end of the first page. Maybe a definition of that term would have helped those of us who aren’t so familiar with different types of gameplay.

The issue with this is if you don't know what a fighting game is, explaining the genre would take up a lot of space in an article targeted at gamers that know the genre. Sure they could have opened up with "A fighting game, like Street Fighter or Mortal Kombat", but you might not know what those are either. This is like an article about React getting posted and complaining that you don't know what a UI or Javascript is. At some point Google is probably better at filling gaps in assumed information than the article itself.

Re: How fighting games use delay-based and rollback netcode (2019)

#36

Post by Jyaif »

shakna wrote: post #13
Jyaif wrote: post #6

This "Infil" dude just stole the entirety of the article from Ars Technica ( https://arstechnica.com/gaming/2019/10/explaining-how-fighti... )

> This article has been cross-posted on Ars Technica. - The infil site > Ricky "Infil" Pusch is a long-time fighting game fan and content creator. He wrote The Complete Killer Instinct Guide, an interactive and comprehensive website for learning about Killer Instinct. This article was originally published there. - ArsTechnica Both sites clearly reference where the original source came from. Nothing has been stolen.

Thanks. Don't know how I missed the reference to Ars Technica in the article.

Re: How fighting games use delay-based and rollback netcode (2019)

#37

Post by enneff »

zucker42 wrote: post #27
CarVac wrote: post #23

Earlier quoted context omitted.

The fact that they were able to optimize two frames of input lag away to make the online gameplay have the same latency as offline is itself incredible. As far as I know, even other rollback-equipped fighting games still add a little bit of latency in order to play online without requiring constant rollbacks. That said, Melee is also a game that is very hard on the rollback, because movement and attack startup are fa…

AFAIK they didn't do anything special to save two frames, just a CRT (which is what's used on LAN) has 2 frames of lag.

CRTs don’t traditionally introduce any latency. It’s digital displays like LCDs that use buffers to pre process images and that add latency.

Re: How fighting games use delay-based and rollback netcode (2019)

#38

Post by andai »

thewarrior wrote: post #29

This piece was really interesting. That you can hide network delays by building a “fake” model of the other player that replicates what a real player would do and rolling back when this prediction fails. So you can train some neural nets on what players do in such situations and you get a certain accuracy. And yet you can keep making this “fake” player better and better until it’s indistinguishable from fighting a hu…

Where did you get the thing about neural networks? I read your comment before reading the article and was really disappointed the prediction "algorithm" (pioneered by GGPO in 2006 [0] and still used today) is literally "assume nothing changed", ie. the opponent is still holding down the same keys as the last frame.

[0]: magazine article by the author of GGPO [pdf] https://drive.google.com/file/d/1cV0fY8e_SC1hIFF5E1rT8XRVRzP...

Re: How fighting games use delay-based and rollback netcode (2019)

#39

Post by Blackthorn »

enneff wrote: post #37
zucker42 wrote: post #27

Earlier quoted context omitted.

AFAIK they didn't do anything special to save two frames, just a CRT (which is what's used on LAN) has 2 frames of lag.

CRTs don’t traditionally introduce any latency. It’s digital displays like LCDs that use buffers to pre process images and that add latency.

Yes, we know. And melee is almost always played on CRTs locally.

Re: How fighting games use delay-based and rollback netcode (2019)

#40

Post by Galanwe »

thewarrior wrote: post #29

This piece was really interesting. That you can hide network delays by building a “fake” model of the other player that replicates what a real player would do and rolling back when this prediction fails. So you can train some neural nets on what players do in such situations and you get a certain accuracy. And yet you can keep making this “fake” player better and better until it’s indistinguishable from fighting a hu…

Not sure you would want the prediction to take non-dumb actions. You need to maintain the hypothesis of least surprise for the local player, otherwise you local player could start to act based on the wrongly predicted actions of the remote player, and that's even worst than nothing.

For instance, say the local player tries to hit the remote one. If the prediction for the remote player is to evade, the local player can choose to chase him. However if you now rollback and the remote player did not evade but charge in, the local player has been fooled.

Also, don't forget than in these games, input could be polled every 1ms. So a player pressing down "left" for 1s in fact is considered to have 1000 down inputs on left. Since players don't change inputs very fast, just replicating the last input is in fact 99.9% accurate.

Post reply on HN