Why are indie competitive multiplayer games so rare?

I've always wanted to make a competitive but simple FPS with Matchmaking and Ranking systems. From a quick glance a lot of what I need (3D models, Characters, Guns, etc) is already available on the asset store. So all I need to to is some coding, which is what I know how to do.

But before I try to make a project like this, is there a reason it's so uncommon for solo-devs to make these type of games? I don't see it being that much more work than any other type of game. Especially with how convenient multiplayer is in Unity.

 

Because these games are harder to make than people think. They are rarely "simple" if you want the gameplay to flow, the movement to feel responsive, have the game look nice, and also have the game stand out.

 

I would argue that no good game is "simple" to make. But I agree it is hard to stand out when only using things from asset store.

 

Simply put: most people don't make them simply because they are locked into place from these cloud networking packages that offer these nice matchmaking services, etc, that make it well, impossible to do anything like COD or Battlefield. So most people never take the jump and learn about Dedicated servers. because they, are beasts. As in, if you don't understand at least what you're trying to do, the concepts with how server/clients work, then you will be lost.

Which isn't to say you can't learn it, but it is by no means, a first timer's project unless you're really willing to sit down and learn the architecture's you are wanting to build.

Aside from that, there's not really much from stopping you.

 

We are making a MP shooter though for VR so most of the animations are made by the player. I would say the hardest part is syncing everything correctly while maintaining high tick rates. VR also has more tracking points than desktop so more data needs to be synced for each player

 

Ya good way to put it. It's a good niche to know as an indie. It's one of those areas that most indies aren't good at, so it's a good area to compete on if multiplayer is of interest. Compared to trying to compete in areas that most everyone is somewhat good at.

 

We have written our network engine from scratch , right now it's self hosted but the plan is to use Unity docker containers and host headless servers in the cloud

 

I feel like there's probably a place in the market for a pure Quake style multiplayer game (no classes or heroes) but you'd ABSOLUTELY need a strong aesthetic for it to work and you're not gonna get that from the asset store.

 

Only genre that is doable from the store is realistic shooters

 

Because to make a game it takes X time. And to make multiplayer game it takes 3*X time.
Also, balancing, maintaining servers, comunity magement are not that easy as people might think.

Also, without a proper advertisement, there's no real point of making competitive game. Otherwise you'll end up with 10 peoples playing your game at best.

 

The fact that you think that speaks volumes.

 

Your just a very unsecure person that's all, which will never amount to nothing

 

Agreed, I think there is a place for a good 2v2, 3v3 shooter with smaller maps, something like what halo 3 team doubles used to be.

 

Is everything ok? I feel like I notice a recent change in your behaviour, it's almost like you're trying to piss people off. I remember a much friendlier version of you, which I hope to see coming back soon.

 

Me? Have you seen that guy go on? For example towards that guy making vlogs?

He has a bad attitude and it has nothing todo on a business related forum, he can move on to Reddit or similar.

If you are nice, I'm nice.

 

To be honest, I dislike the vlog guy as well. Because he's shamelessly self promoting his channel on the forums without making any decent content.

I'd rather watch GDC for those topics.

But yeah, chill out guys.

 

Post literally every post I have made in that guy's threads and see how well your argument holds up.

 

Its not his opinion I disslike, I think the guy should use one thread for them too, since he posts so often, its how he state his opinion I do not like. But its how miserable people are, the S***post and try to make other miserable too. I dont personally get miserable from it, but it do get tiring some in the end. Specially from a guy that basically never have shown any of his own content.

Go out and make your own games instead of bashing on other peoples work are my pro tip.

And in this particular case he does not even know what he talks about. If you go realistic its very possible to make a game. Its when you go stylish and art assets dont match.

If you go realistic you can buy from Unity asset store, artstation, cgtrader, etc, etc. Done just that myself. Even some custom stuff from people I find on artstation.

 

I feel like Shootmania was that. Was a good game, didn't do that great. Although it may have come out at a bad time and definitely didn't have enough marketing.

 

It was also kind of a visually indistinct mess and filed the concept down far too much so they could appeal to an esports crowd that really didn't exist.

 

So my thoughts on competitive multiplayer for indie just revolves around player counts.

I don't agree that MP games are necessarily more complex. In some ways I think they can be easier, even much easier because pure MP doesn't require as much raw content.

I think the reason most indies wouldn't design a competitive MP game is because it's too hard to lock in the player counts needed to really sustain a game like this.

 

I agree on the content, but the code is much more complex. Well depends on how much you sync etc. But still more complex than no syncing.

 

because in competitive multiplayer, you need to satisfy all player.
Let's say that your game lags...
Player A is shooting player B while player B is running to cover
Player A sees that he actually shoot player B while player B still out of cover
Player B sees that he actually is in cover when player A shoot
now which choice will you choose? whatever you choose, most likely one of the player will down vote your game
I think, it will be more forgiving on cooperative multiplayer.

 

Just in general though, I think it's much easier to design a competitive MP game because you need to only really focus on 20 minutes of game play. Refining that down over and over. SP needs to focus on having enough variety provided to keep gamers engaged for many many hours, and once they consume the content they get bored.

Competitive MP is really an entirely different kind of "game design" problem from SP. There isn't a lot of overlap in most cases.

 

We disable lag compensation over 50ms. This way high pingers can stay on servers but they will have a dissadvantage. I think its better than letting the high pingers have all the advantage

 

The core mechanics are they same, movement, shooting, etc, etc, whats different is the "game modes".

 

What's different is the content and expectation. I am a competitive mp gamer at heart, I spent thousands of hours playing competitive 1v1 games.

The amount of raw content required for 100 hours of competitive MP is worlds apart from whats required for 100 hours of SP. You can't really compare it.

In terms of game design, they are entirely different problem sets.

 

oh... one more thing that indie multiplayer game usually forgot, bots.
imagine you have a battle royale in 5km x 5km level, and only 3 players joined...

 

I would say that the core mechanics are more than 50 procent of the game design. Offocurse a bit different depending on how complex core mechanics you have.

Also when we talk content, balance is a thing that's really important in MP, in SP you could create random worlds but in MO you need manual world's with alot of thought into placement of stuff. I still think SP is harder on the content side with motion capture, story etc..

 

We haven't , soon 1 year dedicated to only AI programming

 

Old age of empires was a proc gen RTS competitive multiplayer. The original Kohan was as well. You can do proc gen multi player, it's just again, a different game design problem.

For FPS - I'm not sure why proc gen couldn't work, I think it's because players feel that map knowledge is a core game skill. But I'm not an FPS guy, so I can't say.

"I would say that the core mechanics are more than 50 procent of the game design. Offocurse a bit different depending on how complex core mechanics you have. "

The thing is, I think SP actually requires more core mechanics than MP does. For single player to stay engaging without human competition you need more systems for the player to interact with. This is also why a lot of single player games can't be ported to multiplayer: the mechanics simply wouldn't work, or they become very awkward.

 

I don't know much about Unity multiplayer. But during the Unity conference in Berlin (I think) they introduced their own matchmaking system etc. While I understand there are still complexities, it seems it's at a stage now where you can let Unity deal with the more complex backend stuff. No?

 

UNet HLAPI is being deprecated with relay servers being shutdown @ 2022 (I think?). Same goes for their matchmaking solution.

So, nope. It's either Photon now, or other library solutions, or dedicated servers via LLAPI + own architecture.

 

In a FPS you use the world much more dynamic than a RTS, the cover when advancing and falling back etc is hard to get right with procedural world's. You can create the a base procedural and then adjust it manual add buildings etc. You can do some stuff automatic with tools, but alot of placement needs careful thoughts.

I think MO can be more advanced like COOP mechanics requiring you to work together to solve a puzzle

 

I'm not an FPS guy, so I honestly can't say why proc gen wouldn't work in comptitive MP fps.

I am honestly quite surprised that battle royale shooters aren't proc gen (like fortnite, pubg, etc). It seems like this would be pretty ideal.

If XCOM 2 can generate tactical turn based maps that look/feel good and play well, I have trouble accepting that the reason FPS can't use proc gen is technical. I think they just don't think the players want it badly enough to invest the huge amount of time and effort required.

Proc gen provides increased variety, but again in pure competitive MP, the variety comes from your human opponents. Again, it's a different sort of game design problem.

 

I'm not the right person to answer that, we wanted to scale for open world so we wrote a custom network solution. But the out of box syncing most of the time only work for the very basic stuff.

In our game we sync every little detail, even the cables when another player are inserting bomb wires are synced

We have networked physics with recursive ownership

It all boils down to how complex your syncing requirements are

 

Same reason, you need custom balanced theatres of war, I'm sure the terrain is proc made, and alot of assets proc placed. But road systems, towns etc are pretty much manual.

 

Well, there are two main reasons:
1) Multiplayer games are more difficult to make than single player games. This makes it much more difficult for a small indie dev to even successfully build a decent working multiplayer game.
2) Multiplayer games are extremely hits oriented. What I mean by that is millions of people will buy one game and play it for hundreds or even thousands of hours, while the rest of the multiplayer games are almost entirely ignored. Multiplayer gamers want to play the best, most popular multiplayer game. It is extremely challenging from a technical perspective and a marketing perspective to achieve the number one spot in a multiplayer category. It is nearly impossible for an indie dev to do it.

If we look back at one of the most popular FPS multiplayer games of 2017 as an example: Playerunknown's Battlegrounds (PUBG). PUBG is an example of where an established studio brought in the developer of a popular battle royale game mod and turned that concept into a full game. The PUBG team chose a readily available game engine (Unreal Engine 4), and then used a bunch of pre-made assets from the UE4 asset store to save time. Even with all of that, it still took a dedicated team of 50 people a year to bring PUBG to early access on Steam. Simply using an existing game engine and a bunch of pre-made assets is not enough. It still took a lot of effort on top of that.

Most indie devs don't have a team of 50 experienced developers, a years worth of funding, and a somewhat unique idea for a hit multiplayer game.

Now let's consider what it would take for a indie dev to convince millions of people to stop playing PUBG and start playing a simple FPS game they built quick using some assets. And to be honest, that is the real question. How to convince existing PUBG players to play something else instead of PUBG. What would it take? It would need to offer a new experience that people were hungry for and did not realize it yet. What would that be? It couldn't just be a simple 2x2 or 3x3 dealthmatch. At a minimum, it would need to have massive maps and lots of players, since that is expected of multiplayer FPS games right now. And even just delivering that part would be a serious challenge for most indie devs. But it would also need to deliver an obvious new amazing experience beyond that, so existing PUBG players would consider the new game.

 

Well, some aspects are proc gen. Loot, flight path, blue zones, red zones, etc. Those are already proc gen in PUBG.

The maps are not proc gen. A static map is more memorable to the player and makes it easier for people watching their favorite streamers to relate to the scene. But by having such a large map combined with the other proc gen aspects I listed above, each experience actually does feel unique despite that map itself not being proc gen.

If the maps were proc gen, I actually think the games would be far less viewed. I personally believe the static maps are a key to maintaining viewers for streamers.

 

You can proc gen at design time too.

 

Absolutely. Proc gen can be used as a tool during design time as well as run time. It is up to each dev to decide if/when they want to use proc gen.

Anyway, I don't think proc gen should be used in competitive multiplayer FPS games for map design at runtime, such as terrain or building placement. Like I said above, I personally think stream viewers need a static map design to relate to when watching their favorite streamers. But proc gen could be used as a tool at design time to help construct the maps.

 

This is not always true.
Making interesting and functional single player can be hard, or even harder. Specially when complex AI logistic is involved.

 

I agree but I also think it cant be used too extensively at design time either. It can be used to create the overall terrain base that you build your world on. It can be used to place most trees, rocks, plants etc. But large structures need to be manually placed.

 

Or you do both MP AND complex AI I'm tired of AI to be honest. In october its one year in the making

 

@OP:

There are some big problems with making "competitive" games:

1. You need to make a game which will be attractive for a decade.
2. You need to build a community around it or it will fail.
3. You need to maintain your game possibly for a decade.
4. You need good paid servers in various regions to minimize latency.
5. Competitive games are attractive mostly to hardcore gamers who are hard to serve by indies.
6. You need to design a game which will be fun while playing for hundreds of hours.
7. You need to actively fight cheaters, hackers, botters, and trolls.
8. You need a lot of people for customer support.

Finally, competitive games compete VERY hard among themselves for players. It's almost zero-sum business game. When somebody plays DOTA2 a lot they are unlikely to play a lot of LoL. When somebody plays a lot of CS:GO they are unlikely to spend much time with Rainbow Six: Siege. This is because in order to become "competitive" you must be dedicated and you can't "dedicate" yourself to several games because it takes to much time.

 

Indies like building games that have very little upfront and maintenance costs. If you make a competitive game with matchmaking, highly responsive netcode, replays, maybe esports integration, spectator support, etc then it really starts turning into a services and maintenance nightmare.

The barrier to do all that becomes much more resistant to your entry. You need bigger teams to spread out across all these features, really solid, modern, fast, optimized and specialized netcode for your game and that all just doesn't fit in the indie scope.

 

Super interesting to read everyones take on this, thanks for some good responses

 

Multiply the time and effort it takes to create a single player version of your game by 3X to get a rough estimate of the work involved making it multiplayer.

 

Like the recent Unreal Tournament that epic made with UE4? What about the aesthetic do you think would need to be different from UT to stand out?

 

Ultimately this. Competitive MP games rely on a critical mass of players. You need to be able to log in to the server at a random time of day, and have games running that you can join. Without other players, competitive MP is dead in the water.

Do the math and work out how many players you need total to ensure that someone logging in at 10am US time has a decent chance of finding a full game. Then work out a marketing budget required to sell that many copies of your game. Its a number that no indie studio is ever going to come close too.

The technical challenges can be overcome by a smart, experienced indie. The marketing challenges are pretty much impossible to beat.