Map Size (Open-World) What is possible, what is needed?

I know 4km by 4km is max in UE4, but I heard you can stack them next to each other, so lets say 4 in a square making it 8 by 8 or 9 in a square making it 12 by 12 etc…

Is there a simpler way than this?

For an Open World MMO, the average size seems to be around 50 - 100km square. Which would mean around 77 or 1010 at least.

Now to populate such a world, is there a simpler way rather than just manually sticking tree by tree, foliage by foliage, rock by rock, and then optimizing every single part of it? This would take thousands of man hours no?

What is the best way to create an HD terrain of a large size.

Reason is my idea needs to fit all player into 1 map, I only want 1 gameworld not multiple ones like other games, and zones won’t work since the world needs to be player controlled…

Meaning if I get 1,000 players online at a time it needs to fit them all. Therefore I need quite a large area.


  • How large would such a file be optimized? I’d rather stick to below 20gb files with all assets
  • What optimization other than culling, LOD, Mountains etc do you need?

I’d like the map to include fog, how would I go about objects behind such a fog not loading? Making the FPS higher.

You just answered your own question! This is the very reason why indie-MMOs either never take off, or have to radically reduce their artistic expectations.

You could populate the map procedurally, which would make making the content a lot quicker (post initial setup, of course).

LODs and culling are your best proven methods for reducing overheads when working with massive environments.

What do you mean by procedurally? And yea LODs is a must. Not sure what culling is to be honest.

What kind of a team do you need for an MMO?
1 concept artist -> 10 artists
1 team manager also working on putting everything together
1 Sound guy :slight_smile:
2 coders?

Culling -> hide meshes so that they are not rendered (you can do that with a culling volume + in the foliage tool)

Procedurally basically means you’re letting the computer do something through code. So for example, instead of placing trees individually, or even with a folliage painting tool, you tell the game to place a tree everywhere that conforms to a set of rules. Such as flat areas within x units of a body of water, or anywhere that isn’t steeper than a certain value.

Setting something like this up is a lot more involved than hand painting and finessing things that way, but it means you can iterate through terrains much faster since the computer is doing the grunt work of placing the trees.

For a true MMO, you’ll need a lot more than just two coders. There’s a huge amount of network backend stuff that needs to go into place to support 1,000 players - especially if they’re all on a single map. Tracking all of their movements and interactions in realtime is no small undertaking, and would need some seriously optimised and well thought-out code to support it.

I meant on-hand Coders.
In terms of the networking I am hoping that by then there will be a plugin or a 3rd party software with the core code, and then hire a few professionals to fix it up to match my game? Not sure how much that will cost.

Since for 20 workers, it will cost around 1,000,000$ per year.
If game production takes 3 years thats 3 million just for the workers.

This is not going to happen. Tools get more powerful all the time but what you are hoping for is unrealistic. It’s like saying " I want to create my own brand of car- what do I need? Tires, engine- steering wheel? Anything else?".

Look around - there is a reason you don’t see thousands of indie mmos. If it was easy, there would be.

I strongly suggest you start learning a segment of game development before asking these broad general questions. Because then it will become much more apparent instead of these macro questions which answers aren’t going to help you or anyone really at all.


I would actually disagree with this. There are a lot of users who have not really worked in game development before. Whether these are seasoned vets in other fields or those completely new users who have never used modeling software or game engines at all.

I’m going to be a little cliche and cheesy here but it’s just like level design. You don’t start out, or at least you should, start out with designing every detail about this one part of the level. You start by getting the broad strokes of the design done and testing for playability then you will refine and build on that until you get it to a polished shippable state.

@ anyone who’s still reading. :slight_smile:

There are a lot of specialties within development that others may not be aware of. Asking questions and then refining those questions to be more specific is a good way to learn.

The biggest thing is to research any and all avenues you wish to pursue. Find examples of companies that are releasing MMOs and see if you can find post-mortems for them that explain what hurdles they faced while developing their games. If you cannot find one for MMOs specifically Gamasutra has some for other games that will cover their successes and troubles. Every game you develop will have these issues. It’s bound to happen that there will be bugs, issues, or complications in developing some core feature you want to have.

It’s hard to plan for these, but having buffer time built into your schedule from the start is a good thing.

If you’ve never seen the web series Extra Credits, I would suggest taking a look. They bring up some valid points for their topics and they are fairly entertaining to go through.

Here is a good one to get your started since you’re essentially an Indie Dev. :slight_smile:

If you have any broad or specific questions feel free to ask! There is a great community here willing to help guide you for sure!


I understand, the idea of mechanics etc. But my game mechanics require a lot of content, the idea behind my game is all about the mechanics.

The problem is the mechanics don’t really work unless there are a lot of people working together in a large environment.

I have a few marketing plans etc, I’ll pm you

thank you Tim Hobson for what you posted, I’ve always seen someone ask stuff and others put them down for their ideal and dreams, for new ideals and dreams is what makes the best new games out there. As of UE4 every month, no every week I see new stuff for it, and it’s getting stronger and stronger. Plus the support team really helps out the user of unreal, standing behind there product. Plus should I say the weekly twitch, is very helpful. What can I say but Thank You for UE4 and keep up the good work.

I think you’re off by at least an order of magnitude in cost compared to the big-name MMOs. (As in: $30M and up.)
There’s a reason MMOs are knows as the money suck of the games industry. They need absurdly large amounts of content!
The only genre that’s even worse is block-buster AAA shooters and open-worlders – Call of Duty, Battlefield, Saint’s Row, Grand Theft Auto; you’d probably need > $100M if you really wanted to take those on. And there’s a significant risk you don’t make your money back!

Anyway, if you haven’t really done games as a business before, starting small, and learning from each step, is likely a much better way to get where you want to go. Even if you want to make a AAA shooter MMO with open world sandbox mechanics, your best (and fastest) way of getting there is to first make pong, then to make Pac-Man, then to make a single-person platformer, then to make a multi-player simple shooter (Quake style), then to make a single-person RPG (which is big enough on its own) and then, perhaps, you can start thinking about the MMO part. And with luck, profits and credibility from some of the previous games might help you get where you want to be going!

While I see why people warn new guys about taking on too big projects, saying that an MMO is impossible for a single person, I have to disagree with that.

First of all, as soon as people hear the word “mmo” they think of World of Warcraft, Guild Wars, SWTOR and such.
Taking on a game like that as a very small team would indeed be more or less impossible within any reasonable time frame, but you have to keep in mind that not all games are the same and made by huge companies for the AAA market.

Just because you play the most mainstream MMOs it doesn’t mean there aren’t smaller ones out there, that are still good and played by a lot of people (although of course not millions of people).

Xsyon which have a huge world and still sees success today, 3 years after launch and just recently got greenlit on steam have a team of (as far as I know) 10 people today, I can only imagine it being smaller back before the initial release 2011.
WURM online was made by 2 persons and now, 2 years after release they are 4.
LOVE might be very strange and different but it’s a still huge, open world mmo and it was made by a single person.
Runescape had 2 people before it’s success.
A tale in the Desert was made by one person.

You see my point, the list goes on, so don’t go crushing someone’s dreams just because you have a narrow idea of what an open world MMO is.

I agree that it’s a good idea to start small, but to tell someone to start by making pong, then going on to make Pac-man and so on sounds like a terrible idea and bound to put people off.
I don’t think a lot of people would find it interesting to work on stuff they are completely sure they won’t use at all, this is not how you keep motivated.

What you should do is split your project up in tiny, tiny goals.
If you want to create your game idea, what ever it might be, start by making one tiny part of it and learn what you need to know to make that happen.
I’m not saying make a finished starting area of the game, think even smaller:
Make your first prop, learn from that and create another one that you want in your game.
Keep going and before you know it you have learned some basic 3d, shaders(materials) and how to use the UE editor.

After that you might get into making some basic functionality, make that door you made open-able with blueprints, keep going, make a fireplace and learn some basic cascade for the particle effects, use blueprints to start the fire, keep going and so on and so forth.
After a while you’ll have a tiny part of your map made with assets you created and you’ll have something to look at, knowing you created it and it’s something that will make you proud and keep you motivated. I hardly believe pong will do that.

You’ll most likely come back to a lot of these first things you created later and make them better or completely redo them because they aren’t up to par anymore, when you are a lot better. But at least it was something that got you excited and you could see your game grow and a lot of it can probably still be used.

Don’t let yourself be put off by the enormous mountain in front of you, see it as tiny goals you overcome and sooner or later you’ll have climbed the whole thing. Don’t let the nay sayers get to you, work for your dreams.

When I started working on our project, we didn’t know smack about making games, now, years later we are still working, we’re still a very small team and we have learned tons and tons of techniques and skills and the game is taking shape, and it’s something I can be proud of which is the most effective motivation.

I am myself and a small team working on a big-scale project, although it’s a bit different because we all have experience in game development and are employed in game devs studios.

I would agree with most of what Sitrec wrote: working on something you are not interested in, in your freetime, is the best thing to get bored and give up after only a few weeks.

Don’t learn for the sole purpose of learning. What I would advise is the following:

  • Divide your project in extremely small tasks. If you have to create the level design of a city, with interiors and stuff, design an apartment first. Then a building. Then a block, then a street.
  • Do only one task at once, but be sure it will be relevant to your project.
  • Do a lot of research before going to a new task. Don’t ask to people, try to find by yourself. How did big studios dealt with the same problem as mine? What are the best ways to do this very specific thing?
  • Try to have a playable prototype first. Then an enjoyable one. It will be easier to advertise your project to future team-mates.
  • Find team mates that are at the same level as you, and MOTIVATED. That’s the big thing here. Your project can take months, years before reaching an alpha prototype.
  • Don’t push on yourself too much. It’s like listening to a song over and over again all day long, it doesn’t matter how much you love it, you’ll be bored of it eventually.

I’m not a senior lead designer in a multi-billion dollars company with 25 years of experience, so take those advice as “advices”, not rules. Although the only game design rule I would tell you is: Rules are meant to be broken.

Hey guys, I just woke up so I will read all these replies. But just wanted to post something budget wise.

Big studio needs: 100 Million to create the game.

This includes LOADS of advertising costs
Millions in Engine costs
Millions in fees / taxes / lawyers etc

EVERYONE working with them is Inhouse, they are paying millions for the building, social security and more
They are hiring people with 20+ years of experience and paying them 2 - 3 times as much as you’d pay someone with 5 years of experience
They probably have 10 different managers, loads of different concept artists, and have to redo work 5 times each.

This means they are wasting a magnitude of money. I estimated this cost to be 20 TIMES the cost of an indie producer who would use UE4.

This means to make a 100m game you’d need 5m. Now this may sound insane but hear my budget idea out.

First off, 1 concept artist. Stay close to this guy, try to get a good friend and stay close to them and talk to them. Make sure they understand exactly what is in your head, so that there are minimal redoes. On the first 10 concept arts work with them, don’t ask them for fully detailed art at first, tell them to make 5 - 10 rough muck ups so that you can see what direction this is going in. A rough concept design can be done in 15 minutes. A detailed piece can take 2 - 4 hours. So get 5 - 10 rough ones, throw away the ones you don’t like, stay with 2-3 and start working on them.

Once you have EXACTLY what you want, everything written down, organized and know exactly what your plan of action is and what it should look like, hire artists overseas. By that I mean, from anywhere around the globe, what you will be paying them for is COMPLETED WORK. That means you give them a realistic deadline, you tell them EXACTLY what you need.

You provide them with a file with the concept art, if you already have some of the rest of the scene it should fit into give them that aswell, give them a good .txt file with an insanely accurate description and maybe even talk to them. Then once they complete it you look at it, you don’t accept anything that doesn’t fit in, if you did what I stated it should look close to what you wanted it too. If it doesn’t it’s their fault or you were expecting way too much.

Now instead of having 1 artist go from 1 model to the next you can hire 20 - 50 artists or even more to work on specific scenes using your concept art.

  • Most of the time you will try to purchase models at 10$ - 20$ for smaller ones etc, don’t cheap out on main characters obviously. My game has NO need for any characters other than 1. A char creation screen to make your own 2. Animals. There is only 1 race etc, not that much content really.

I believe I can make such a game with a budget of around 5 million simply because the amount of content I need is less than half of other AAA MMOs, and I will be working hard on budgeting.


I of course have more “realistic” goals rather than go "oh I’ll open a royalty post on the forum and make a triple A MMO.

My goals are the following:

  • Learn for the next 2 years, focus on my studies and put together the game and how it will be budgeted realistically, keep reading from everywhere etc…
  • In 2016 aim to finish a demo, by a demo I mean a 15 minute gameplay which I am already starting to plan, it will be very advanced but not need much content, just really refined content.
  • In 2017-18 I finish my studies, so aim to release it around there, get a few people on board with you, NO ROYALTIES EVER.
  • In 2018 I will be in the USA hopefully, put together a kickstarter and if it fails, rethink business plan, if it works look for extra investments, investors are into putting money after a successful kickstarter

Meaning 2018 enters the production stage of the actual MMO. By then I am hoping to have 2 demos and a prototype.

Also in terms of marketing I already have ideas, on how to market the game for very low costs, I am talking a few thousand $.

Good to see you have a plan! I wish you all the luck in the world. It’s the Indie developers that are breaking new ground when it comes to radical gameplay ideas and concepts so I’m really happy to see so many new, enthusiastic people that are willing to learn despite the steep learning curve and the goal being far away. Would love to hear some more about your ideas.

Jeez, there’s so many varied opinions in this thread! Personally, I think with the right amount of time and dedication, anything is ultimately possible. However, I do think some of your expectations are over-the-top and some of your numbers are way off; but that’s me.

I do wonder how you’re going to achieve such a ‘budget’, though. Because relying on Kickstarter these-days isn’t the wisest of ideas. There’s so many fantastic titles that have attempted and failed through that method as of late, and they’re asking for a mere fraction of what you’re talking about. I hate to think what that scene will be like in a couple more years!

Of course I don’t want 5 million through a kickstarter. I think of kickstarters as a proof of concept. As in, to show investors that other people are interested too, the way I plan on doing this is:

  1. Hosting a blog, facebook, twitter of the whole development
  2. Having a YouTube Vlog series dedicated to the development of the game and the company in general
  3. Obviously having a website that is constantly updated, to give users ALL the information that I have available without ruining the game.

I hope to build a fan base just through the fact that I will be very open about the store, the monetization etc, I won’t be hiding the financials etc, in fact once I do have a plan I will tell them exactly what and where I will be using the money and guiding them through the development.

I want to be very open about all of it.

I think this will attract many people, especially when they see how unique the idea is, and hopefully after a successful kickstarter investors will take me “seriously” and are more willing to invest, helping me reach those 5million.

One of the things I like to mention when people are always asking me if something can be done or is generally possible (typically when I’m wearing my programmer hat) is that sure, anything’s possible…it’s just a matter of how much time and money you want to throw at making it happen. To expand on that, you need to be smart about the things you want to make happen, as well. Just because you could throw two years and millions of dollars into a feature, doesn’t mean you should. (The new Far Cry has a great example of this…they spent TONS of time and money on getting their elephants so you can ride around on them. Once it releases we’ll find out if it was worth it. Heheh. But it sounds cool!) Sometimes just because something will take 100-man hours you might think, “I’ll just add 10 people and they each do 10 hours and it’ll be done in a day!” And that may make sense, depending on what you’re working on. (Usually the more people you add the more administration becomes involved and at some point it swings the other direction than you’d expect.) But the reverse is true, as well. Something that might take one bunch of 10 people to accomplish, can be accomplished by one person spending a much larger amount of time focused on doing everything themselves. (And may also receive time-benefits because of the lack of administration and communication requirements for dealing with more people involved, but lose out on the different perspectives and strengths each person may bring to the table.)

Getting back to the other half of the original poster’s questions, though… FitnessRegiment, you were speaking about what’s needed in order to have 1k players all in a single play-space instead of having these areas separated by zones. There are always technical limitations to hardware and networking that are trade-offs in these situations. You can only devote so much CPU, GPU, and networking resources to so many different tasks. At some point you have to test and make decisions about what’s more important. Do you want trillions of fancy particle effects, meshes, and super-detailed textures that only one person can see, or do you want a billion people in a tiny space that see nothing but a few low-detailed textures and basic representations of the other players? The answer usually lays in the middle somewhere. UE4 doesn’t have any set number of multiple player limits on it to this point…they are leaving it to us developers to push the limits and try new things and see what we can make happen with it at the moment still.

A lot of MMOs tend to split areas into separate zones in order to achieve the balance they want. Do you really want 1k people trying to interact with a single object in a game space? Probably not, you probably want 1k objects so they can all play with an object at the same time. But now you’re talking 1k objects and 1k people interacting in that space, all at the same time. Is it going to be physics-based? Are they going to try to make a circle of 1k people and all kick a barrel into the center of that circle at the same time just to see how the game will react to 1,000 barrels all flying at teach other at the same time? That’s going to take a LOT of physics calculations. The new Crackdown leverages cloud-powered calculations in order to take the load off of the Xbox One consoles to accomplish more while also increasing performance. World of Warcraft balanced their graphics towards a low-end in order to make their game available the largest number of computers already in consumer hands at the time of their launch. Their game wasn’t the prettiest on the market, but it could play on a lot more computers than some higher-end MMOs that had higher minimum hardware requirements for play in order to support their higher-end mechanics, graphics, and audio features.

Take a look at your base mechanics and how your players are going to be interacting with the world. Without you having shared any of that information, it’s hard to come up with specific suggestions on what may or may not be a good way of making your specific situation balanced out for the loads you need. You stated that “…zones won’t work since the world needs to be player controlled…” and I wonder why that is, exactly. If the world is literally being controlled by players, such as perhaps having the landscape raised and lowered real-time as in a god-game style you could look into ways the new EverQuest game (now called Landmark, I believe) handles that. But even with their massive zones that are very Minecraft-y in the way players can dig away at landscapes and build up things…they have zones and still fit a lot of people into these larger areas. And you still won’t typically find 1,000 players standing all in the same spot for any particular reason due to resource balancing constraints.

If your game design requires that this happens, that 1k people are gathered together trying to play or accomplish changing something about your game world…perhaps you should think about that mechanic and whether you just THINK it would be fun. Run some tests with 1k AI running around and trying to do something. See if it’s actually fun when all those player representations are smashed in so close to each other that most players would probably be overwhelmed and not understand a single thing that’s going on in the smaller area. Or worse, 500 people trying to do one thing and 500 trying to accomplish the opposite, with 900+ being frustrated and irritable and perhaps 100 laughing at how crazy it is things aren’t working right. See if performance slows to a crawl or if you have some fancy algorithms in place that compensate for it and allow the gameplay you want to work at an appropriate playable speed.

You may find that streaming in nearby zones would work, though. You can make a bunch of maps and then stream them in to the player based on their distance from them. You have to keep in mind how you design areas so that maps don’t magically pop into place right in front of the player with that method. (Or disappear while still in the player’s view.) Often games will use winding tunnels or fog banks to cloud far-off areas to avoid that issue to pretty decent affect. Technically players would be in the same large play space “level”, but a lot of it wouldn’t be loaded up for individual players and can help with performance and resource management issues.

I hope something I shared helps you get headed in the right direction, solve some of your problems, or thinking about questions you never thought of. Best of luck to ya!

Oh sorry I misspoke, by ZONES I meant the term a game like Guild Wars 2 uses, where players can be in the same place, but not see each other. Kind of like on different subservers. I can’t have that.

Because how players interact with the game, like taking over a strategic point, is crucial that ALL other players can interact with that point. I can’t have the same strategic point 2-3 times… Not sure how to explain this.

I need it ALL to run on the same world for the mechanics to work. So that the world can evolve with the players depending on the choices they make. Not sure if thats any clearer.

The 1000 people example was not meant to be in the same confined space, but spread out over 50km square or even 100km square landscape. But on the same world server.

Finally someone that answered the guys question! Like you fitness i have dreams and don’t know my limits i’m embarking on a massive project (Alone) hoping that i’ll have a prototype in 3 years time! Good luck with your project!