GOTA MP Should I do it?

games are cool but 100 times cooler when you make them multiplayer, the core of my game is designed to be multiplayer already and the knowladge that I do have on multiplayer is very high, I wonder would it be a good move to use steam sessions with UDK, or should I go for open world feature, I believe that multiplayer would be cool enough just a duel mode… Just in case someone wants to play it, BTW, is steam sessions very cool with UDK, I saw that there is a tutorial is it truly valid ?
What do you guys think?

it all depends on your goals for it

are you doing it for fun and learning, for you to play yourself with friends, or for your actual players?
if it’s for your players, keep in mind the player count factor. how many players do you expect to reach? IIRC you released your game on different platforms (indiedb, steam, itch-io?) so how big is your player base and how reactive would they be to a MP version of it? do you think it would only be interesting for your existing players or do you expect to sell more copies of it? there’s few things more sad in the gamedev world than a full released multiplayer game with no players.
and if it’s for selling copies, as a player paying money for a game I would expect a lobby with matchmaking or at least a match browser, live servers with people to play with, etc. can you do all that?

also are you aware of the difficulties of an anim/timing based game on multiplayer? a game with timing-based blocking combat just fails at more than 250ms ping unless the attack animations are slow or unless you are an expert at game networking.
imagine I play with someone and we both have 250 ping. so I attack and 250ms later the other player sees it and after a little time reacts and presses block, which arrives to me another 250ms later, but at that time it’s most likely already too late for the block because I already landed the hit because the block arrived a full 0.5 seconds later without counting the player’s reaction time.
this scenario is already bad and becomes much worse if the player is not the server, because then the attack takes 250ms from me to the server, another 250ms from the server to the player, then the player blocks which takes 250ms from that player to the server, and another 250ms from the server to me. a full second of delay, for actions that are timing-based, just doesn’t work.
in case you wonder, this double delay doesn’t happen on shooters because there is no parry/block button needing hit confirmation, it’s just hit or miss bullets.

TL;DR, depends on what you want to achieve :slight_smile:

1 Like

One of the points of chosker is true, that FPS games are just better than Melee games for multiplayer, BUT titles like chivalry got it working… but those are very titles I guess. With a lot of knowladge behind and people and simplicity in mind.

Well I do know how would it be to replicate a Unreal what I don’t know is how to make it very good for networking performance.

Like The tool that monitors traffic I don’t even know how to use it, Like unreal should make an effort to make it easy to use… like everything else in the engine. To know irrelevant UDK and TCP packets and such… Because it is hell you get to try your game on your end and when you realese with steam sessions it goes very laggy full of errors and such when you try it with the teammates. Networking Insights Overview | Unreal Engine Documentation I think it is this… Anyways do you guys know of any steam sessions working tutorial that I can rely on to make it work on steam? just in case I would want to becasue GOTA 1 ( a game that I got to put up on steam ) Has the advantage that is just hit and protect… No special attacks or other fancy stuff… and that makes it an easy task to make it multiplayer…

What problem are you having. Is it the sword fighting and looking like it is not in time or right and not laggy?

That I don’t know if steam sessions and UDK works wll together.
And I don’t know where is the official tutorial.