Search Unity

  1. Welcome to the Unity Forums! Please take the time to read our Code of Conduct to familiarize yourself with the forum rules and how to post constructively.

wantz mak MMORPG

Discussion in 'General Discussion' started by MTracerStudios, Dec 4, 2010.

  1. MTracerStudios

    MTracerStudios

    Joined:
    Aug 24, 2010
    Posts:
    118
    Let's start with the basics.

    I AM a noob, I am NOT a retard.
    I DO want to make an MMO, I do NOT think I can do this (yet).

    Now, with that out of the way, let's all be civil, Ok?

    I have a few questions.

    First what would you guys recommend for networking? Unity internal server system? Electrotank? Photon? SmartFox? And, much more importantly, how would you recommend I host the game? At home? Amazon?

    And, the question that has had me really intrigued for the past few days: What part of development do I cut to make this game craftable? Blizzard takes years, millions of dollars, and many employees to make WoW. I recognise that I do not have these things. The two things I want to keep strong are: shear playtime length of a single story arc, and diversity of quests/immersion. What do I sacrifice to achieve these two goals?

    Thanks,

    MTracer.
     
  2. Nikolay116

    Nikolay116

    Joined:
    Mar 21, 2010
    Posts:
    421
    why don't you just use search? this have been discussed at thousands locations
     
  3. tatelax

    tatelax

    Joined:
    Feb 4, 2010
    Posts:
    1,167
    Well the first thing I would do is get a budget of some sort. While you probably don't have a budget...you should use Unity's networking to set up a test.

    Also this should have been posted in the networking section.
     
  4. MTracerStudios

    MTracerStudios

    Joined:
    Aug 24, 2010
    Posts:
    118
    Mods, if you agree with tatelax final comment, please move this thread.
     
  5. stimarco

    stimarco

    Joined:
    Oct 17, 2007
    Posts:
    721
    There are a number of options for making a project like this feasible:

    a. Cut the first "M" in "MMORPG". It's the "Massively" bit that makes these projects very, very hard to do well. And you'll need large quantities of cash to support the necessary infrastructure anyway.

    1. Further to point (a), you won't get masses and masses of players for an indie game right away anyway, and there are inherent problems with large-scale "virtual worlds" that limit your quest design. Very few players will want to play anything other than a hero of some sort. And your world will therefore have hordes and hordes of these heroes cluttering up each and every town and village, wandering the landscape, with entire gangs of them all trying to complete the same quest at the same time. Single-character quests where you can actually provide a high quality of story ingredients is hard in this situation as you'll also have to design 'gatekeeper' elements to prevent dozens of questing Princes all trying to save the same Rapunzel.

    i. Every game that puts the player in the "role" of a character—and that means anything from "Super Mario World" to "Assassin's Creed", by way of "Grand Theft Auto"—is technically a "Role-Playing Game". Virtual world games like "World of Warcraft" and "Runecraft" actually see very, very few customers playing the game as role-playing games in the traditional sense. (If you don't believe me, take a look at the many servers for "Runecraft", and notice how many are set up for things like "grinding" and the like, compared to the pitiful number of servers supporting genuine "role-playing" game modes!)

    So, make damned sure this really is what you want to do.

    *

    The "Massively Multiplayer" part is, in my view, the best bit to cut. Just make it a "Multiplayer RPG". This dramatically changes the design and gameplay dynamics too. World of Warcraft and Runescape effectively give you a virtual world swarming with heroes and wannabe heroes. What's so special about being a hero when practically everyone you meet in this world will also be a hero?

    Instead of these vast armies of supermen roaming your world, you can make heroism something special again and design accordingly. There's also a lot less need to maintain a massive, persistent, virtual world simply to have somewhere for all those players to go. You can create a smaller, but more detailed world—perhaps just a single country or island. Segment it. To scale the game up to meet demand, you simply spawn parallel worlds all on the same server—sharing the same assets—and give each group of players the place to themselves for a change. Let being a hero feel unique again!

    That's how I'd do it, anyway.

    *

    Technically, if you take my advice on-board, there are ways of reducing the head-count:

    1. Tools, tools, tools!

    I can't stress that enough. If your game will have new quests added on a regular basis, you need tools to build those quests, test them, and make sure they're rock solid and won't break anything. I've seen how the folks at JAGEX create the quests of Runescape, and it explains why they have such a huge headcount. (Hint: if you're using a scripting language, you're doing it wrong. Scripting is a terrible UI for quest design and development: better to create visual tools.)

    Basically, the more you can automate, the better.

    2. Procedural assets.

    Procedural assets are those created primarily by computers rather than people. Tools like Allegorithmic's Substance suite are an example of how to reduce reliance on artists and the long process of hand-painting textures. (This technology is coming to Unity sometime in the near future. Probably early next year, at a guess.)

    Terrain and landscape generation tools are also a handy option, but can cost money, although you can find tools that will take DEM files and turn them into bitmaps suitable for use in Unity's terrain engine lying about. Similarly, scripts that let you lay down roads and create entire towns and villages will save a lot of time. If you can afford it, take this route. Generating towns and villages will be tricky as you'll probably need to write the scripts yourself, but there's a lot of research on the subject already, so Google is your friend here.

    The more work the computer can do for you, the better.

    3. Off-the-shelf assets.

    You can literally buy entire enough models to create complete medieval villages if you're willing to spend a little money. Don't be fooled by people telling you they'll make your game look like everyone else's: most gamers really won't care, as long as the models themselves look good. (You can always customise them too.)

    Rolling your own is a terrible idea for games like these anyway: medieval buildings weren't all that varied anyway, so similarities in style is normal. You were much more likely to see variations in the more expensive, formally-planned stone-built structures such as churches, cathedrals, defensive walls and castles. But, minor cosmetic changes, like paint, aside, your typical 13th Century English pub looked much the same as its neighbours.

    Ditto for grasses, trees and other scenery items. If you can find a couple of oak tree models online, use 'em! Save your custom artwork cash for where it's really needed.

    The same, obviously, goes for music and sound effects. You can get loads of the latter online, and even copyright-free music is readily available from a number of sources. The creators usually request only that they be credited in the game.

    *

    In summary: spend and work smarter, not harder. Look at the resources available to you already. Do some research into game design and production—by which I mean spending some money on some books on the subject, not just asking questions in forums!

    Forums are great places for asking questions, but if you don't even know which questions you should be asking in the first place, a forum is a poor choice. They're reactive sources of information, not pro-active. They're also designed to favour short replies, rather than long essays. But the real problem is that people who make a living at this simply don't have the time to distil the contents of multiple textbooks and a few years of experience into just a few posts.

    (Before you ask, no I don't make a living from games any more. I'm still a grizzled industry veteran, but my primary income now comes from technical authoring and translation.)
     
    Last edited: Dec 4, 2010
  6. MTracerStudios

    MTracerStudios

    Joined:
    Aug 24, 2010
    Posts:
    118
    Dude! That was possibly the most epicly awesome post I've ever seen! THANK! YOU!
     
  7. lochlainn1066

    lochlainn1066

    Joined:
    Nov 22, 2010
    Posts:
    12
    My advice for anybody looking to make a MMORPG is to get a copy of Neverwinter Nights or Neverwinter Nights 2 and learn the NW C# based scripting languages. There are scripts available for mysql (and probably other) database connections. I learned C# doing that and was able to get a persistent world working fairly easy.

    Once you have a working NW persistent world, think about removing the graphics, rule coding, and base engine and coding your own engine and making your own art. Ask yourself if that is what you want to do. Run your world for a year with a playerbase. Then ask yourself again.
     
  8. TestPilot

    TestPilot

    Joined:
    Nov 13, 2010
    Posts:
    55
    My advice is to take a look at Esenthel engine. It is dedicated to MMORPG worlds and have built in networking. I don't know how good is it, but if you ask same question there - you will get response from people actually trying to create MMO games.

    And it is an emerging threat to Unity, which hopefully would make UT to extend built in networking to the point that it would be capable to support MMO. :)

    Edit: not sure, but if it is not ok to mention here competitors that might have some advantages over Unity, please feel free to remove this post or notify me, I will remove it myself. Not associated with Esenthel in any way.
     
    Last edited: Dec 4, 2010
  9. raymix

    raymix

    Joined:
    Aug 25, 2010
    Posts:
    192
    wrong engine

    try Ryzom engine, its fully featured pay to play engine, but they are giving source code for free, and most (if not all) assets with it
    http://dev.ryzom.com/
     
  10. TestPilot

    TestPilot

    Joined:
    Nov 13, 2010
    Posts:
    55
    AGPL for source code and CC-BY-SA for assets pretty much kill ANY chance for developing commercial project based on that code or art assets. It is possible to recommend it, but you, raymix, probably assuming that OP will be trying to create non-commercial MMORPG... I seriously doubt that is a case.
     
  11. gooongalooo

    gooongalooo

    Joined:
    May 6, 2010
    Posts:
    140
    @stimarco - really epic^^
     
  12. JRavey

    JRavey

    Joined:
    May 12, 2009
    Posts:
    2,377
    Esenthel looks pretty neat. I've played around with Big World, but was tired of their Indie Version not supporting indie tools.
     
  13. MTracerStudios

    MTracerStudios

    Joined:
    Aug 24, 2010
    Posts:
    118
    In case you guys were wondering, I asked this on a unity forum.
     
  14. appels

    appels

    Joined:
    Jun 25, 2010
    Posts:
    2,687
    get the community edition of SmartFox 2 if you know Java programming and start on your gamelogic server side.
     
  15. JRavey

    JRavey

    Joined:
    May 12, 2009
    Posts:
    2,377
    I still need to get off my butt and get a matchmaking server going in SFS2. Would you be willing to help with that, Appels?
     
  16. MTracerStudios

    MTracerStudios

    Joined:
    Aug 24, 2010
    Posts:
    118
    Do I really need to learn a new tool? What does smartfox do that isn't trivial to program into unity's server system?

    I'm thinking, on each server, have a file with player information, (each player on that server will always log into that server) and a Unity instance for each map in the game. What's wrong with that?

    BTW: We have Dynamic DNS at our house, and technically, our house has a constant URL, not a constant IP. Will this be a problem?
     
  17. simone007

    simone007

    Joined:
    Oct 30, 2008
    Posts:
    221
    @Stimarco: awesome answer
     
  18. Demi

    Demi

    Joined:
    Dec 6, 2010
    Posts:
    20
    Let me put it this way..

    I been working on a MMO for years and years. The MMORPG is the most difficult piece of engineering you will every attempt in computer science. It is not to steer you away but to point out to you what is the internal workings of this type of software engineering.

    Now you're somewhat on the right track here. You need a client that can do graphics and has the capability to connect to the server. But, before you get into just a client you need a server. The following are a MUST have.

    Your need a simulation engine that is capable of maintaining its state machine no matter what happens. If you take it down it has to come back exactly like it was before it was taken offline. Unity can not do this so you will need to write your own which means you need to understand state machines, artificial intelligence and physics.

    You need a database infracture that can handle 1000s of transactions per second and is bomb proof. Loose your players data and you loose your players. Unity can not do this but mySQL can so you need to learn database programming and how to interface database to client efficiently. Smartfox is a start and is saleable but you still need to learn how to program your database. Nice thing with a java based server it is platform independent so when you grow to the point you really need the horsepower buy a mainframe.

    You need hardware that can handle giga-bit connections and is capable of clustering. Most small home based servers can take on a few players in real time simulations but to get maximum connectivity you need as a minimum a T1 line but a G6 be better. Good place to get this is Rackspace. The prices are reasonable. You can go with clod servers and pay on demand. Best pricing you can get probably 600 bucks a month for a dictated server that can handle the state machine and 3000 clients.

    Now that you have the server you need a storyline that can grasp the player and keep them coming back. You need as a minimum at least 2000 quests. You need a world size that is reasonable and populated with enough content that keeps other interested.

    Making a MMO is work, work and more work and once you get it going then the real work starts keeping it running.


    I will comment on several other here. Esenthel is a great piece of software. It has a basic networking design and is not even close to being capable of being a state machine and server. Esenthel does not have a database for world design so you need to remember where you put things so you can navigate back with go to x,y position. Esenthel does not have any foliage or tree capability unless you get the source for 100,000 bucks.

    Ryzom is that complex piece of software. It is beyond a novice capability to even grasp the internal workings. I have worked with, compiled and set up a Ryzom server. The downside of Ryzom is exactly what the OP is trying to get away from. Ryzom requires 3DSMax to even get a mesh into. There are no other pipelines into the engine except through 3DSMax. Every piece of interface to get things into the engine require command line input. There is no easy button called (make/install). The toolset is more difficult than the C++ programming of the engine.
     
  19. MTracerStudios

    MTracerStudios

    Joined:
    Aug 24, 2010
    Posts:
    118
    I'm not sure if I understood your verdict on Unity. Can I use Unity and SmartFoxServer to create a good MORPG? (dropped massively, for the time being anyway) I'm guessing I'll also need the MySQL extension for SFS?

    I will not make this game until I am more confident in my skills. But, please help me plan ahead.
     
  20. toby90

    toby90

    Joined:
    Aug 2, 2010
    Posts:
    42
    SmartFoxServer's site says that Unity 3D can be used, so I guess you can use them together, yeah. You'll need all the MySQL details to store user's data.
     
  21. Demi

    Demi

    Joined:
    Dec 6, 2010
    Posts:
    20
    I rate Unity as one of the top picks for indies. (My rating does not really mean much here LOL but then I have had my hands on every useable game/graphic/physics/networking and AI engine I could afford for the last 15 years.)

    I rate them on how easy they are to use, pricing, overall toolset, streaming worlds, holes in terrain and so on. Even though Unity does not have built in holes it still is capable of them -BTW. Unity is the only engine I rate beginner to advanced. Engines like UDK are complex and require a much greater understanding of the mechanics and code structure since you have to rip them apart and re-write a lot of functionality. UDK is not a engine out of the box that can do a MMO but is well capable in the hands of a skilled programmer. Bigworlds Indie is a great engine but it is also complex. You have a BWT license but you are probably overwhelmed at the complexity of using it since you don't have the experience level needed.

    Unity is well capable of making a MMO with Smartfox server. Just look a Multiverse and how a scripted server running Java and Python work. Multiverse is based on the same technology as Smartfox with the Axiom engine (C# port of Ogre engine) driving the graphics. Unity engine has a much better toolset and is further advanced than MV. I had my hands in MV for 5 years.

    With a game engine like Unity your time to completion is less since you have a better tool set but you still need the skills to use the tools.

    I just want you to be aware of how complex a MMO is compared to a single player game. Be prepared for a long development cycle. Find a good team and work through one development cycle. Learn the things well I talked about and you will go very far in the game making business.

    As for a good MMO that would be the developers responsibility. If you do it right and make the design count where it should, you can make a smashing AAA quality MMO with ANY engine.
     
  22. MTracerStudios

    MTracerStudios

    Joined:
    Aug 24, 2010
    Posts:
    118
    May I ask a technical question?

    What should I look into in terms of doing player location control server side with SmartFox? I want to give players smooth movement clientside, but I also don't want them to hack and skip through terrain or teleport.

    BTW: Thanks for the help so far, demi and stimarco. It is much appreciated.
     
  23. Demi

    Demi

    Joined:
    Dec 6, 2010
    Posts:
    20
    There are a number of algorithms to use. I would go dead reckoning with a anti teleport hack. This is basically a section of code that says player moved further than physics allowed therefore the data is void and unacceptable. I am not going to give you the code as your task is to learn by research.

    Speaking of research look for these books. Lot of good design information

    Ultimate Game Design Building Game Worlds - by Tom Meigs

    Programming Game AI by example - By Mat Buckland


    You will have to search the internet for these as I can't remember where I got then as it was several years ago.

    Scalability Study of Massively Multiplayer Online Games - This paper is a very interesting read on scalability of a java based server. It covers a different server but the rules apply to all java based server technology.

    High-Level Development of Multiserver Online Games - Deals with clustering design.

    Architecture for a Massively Multiplayer Online Role Playing Game Engine - Interesting read and well worth your time.


    After you finish these you will have a much better understanding of the complex engines that are used in MMOGs
     
  24. MTracerStudios

    MTracerStudios

    Joined:
    Aug 24, 2010
    Posts:
    118
    I will look into these papers, have no fear. But will a very small scale multiplayer rpg need these things? Couldn't I just use SmartFox on amazons cloud or something? Guys like BurgZergArcade make it look so easy.
     
    Last edited: Dec 10, 2010
  25. MTracerStudios

    MTracerStudios

    Joined:
    Aug 24, 2010
    Posts:
    118
    Okay, I've looked into "Scalability Study of Massively Multiplayer Online Games" now, but it seems to only a apply to people who want to do a lot more than me. I don't need something that complicated! Nor can I achieve that.
     
  26. Hirosaki

    Hirosaki

    Joined:
    Dec 10, 2010
    Posts:
    1
    This has been the most helpful MORPG development thread I have ever read, we need more people like Demi and Stimarco out there.

    I have asked questions like this and pretty much had my face spit in telling me that I can not hope to achieve making an MORPG as an indie.

    The books that Demi recommended help me get a better grasp on what I need to accomplish but I do agree with MTracerStudios in the fact that this does seem to gear more for companies developing very large scale games.
     
  27. stimarco

    stimarco

    Joined:
    Oct 17, 2007
    Posts:
    721
    Well, the problem is that the two "M"s in "MMORPG" stands for "Massively Multiplayer". The exact meaning of "Massively Multiplayer" has never been precisely defined, but it's generally assumed that the number of players in the game could conceivably be described as "Oooh! Lots and lots!"
     
  28. Demi

    Demi

    Joined:
    Dec 6, 2010
    Posts:
    20
    Yeah a lot of MMO threads get slammed on because it is probably the most difficult design of game programming. Most of the new people in this field are discouraged because no one will help them get started. I been there done that so I know. I believe that by giving the real information is the key to making it possible for a small indie to get past the hump and at least get a grasp on the task of making their dream.

    Game engines like Unity, UDK, Bigworlds, modeling programs like Blender and paint programs like Gimp, open source development like RedDwarf are a major Godsend since they allow those with talent and very little money a chance to express themselves. Some of the artist and programmers are beyond the level of novice but to afraid to say I have this dream to make this game that is a massive multiplayer online game because others slam them so bad and make them feel like a total idiot. Back in 1998 I went to a board devmasternet and said the same thing. I had been working on a engine for awhile and wanted to expand it to networking. I wanted pointers to MMOG technology and wanted to make this Massive Online World. I got slammed so bad I was discouraged and never went back. My 18 years of programming was evidently not enough. :(

    I started researching other game engines, network infractures, AI engines, Physics engines, database engines. I got so wrapped into finding the best engines 4 years passed before I got back to my game. Then like so many others I went to mmorpg.com and started discussing the design of a MOVW called Whren I wanted to make. I hit some major houses with the design and got file 13, ignored or laughed at. I was invited to this board called mmorpgmaker.com the guys and gals there are like me they have a dream to build a MMO. I started a team with Multievrse as the game engine. But then I decided the Multiverse platform development was dead and wanted to switch engines. My team fell apart. My game is on hold again probably indefinitely. I am getting old and have not the time left in life to complete such a dream. It is the cold hard facts here. I went back to research of engines since much has changed over the last 6 years.

    I came to the conclusion that the big developers will never invest in such a design because it is a high risk and it would be an indie that makes the next generation game.

    I now watch for threads like this to help young potential developers get a real grasp on technology What they need to aspire and put out the this next generation game.

    I am glad I am not alone and I wish others like stimacro that have knowledge would share.

    Understanding the thing about scalability - some here are not experienced enough to grasp the scalability thing. I will point it out as this is the key to making what I call a Massive Online Virtual World. MOVW - Yeah, I coined it :)

    this is only meant as an example. You will not be able to do this alone as the time to complete is is beyond your years.

    You and 9 other get together and all agree on a design theme, rule system class system, world design and story. You all devote time to making assets, writing stories (have a writer make a story of "The World" that is grasping and has real meaning) and programming of the say RedDwarf intfracture to interface the game with a meta-server we will call "The World". "The World" is a collaborative server that contains all information. You connect and modify "The World" remotely put in all the assets until it is populated with enough that is looks lived in. Once you have "The World" built to its design. Each one of you put up a server (low cost VPS/CLOD or even homebased IF you have a giga-bit connection, dictated if you have the money but you can always expand) with a replication pointer of "The World". Now there is 11 servers with enough distributed programming to maintain 5 to 10 thousand connections. If one server goes down "The World" is still there. If someone leaves, "The World" still lives.

    You now put out a client with any game engine, Unity is excellent since API programming will be easy, so other can connect but can not access the Meta server (The World). They can login and create an account to play for free or for subscription in "The World". Some people just want to play for free, some want to support the efforts and get additional Perks. However, some people want to get involved and display their skills in programming, art or story so you sell off sections of "The World" to other players that can modify "The World". They must use the same style assets, stick to the same theme, follow the rules and any assets they provide are part of "The World". As more and more players buy into "The World" it will take on a life of its own. Your story line will grow the quest system will grow and "The World" will become a living breathing thing of its own design. The more servers that come online means the greater the distributed programming which mean the AI can get better and more complex and every low cost server will add additional connections. The quest system can become multipathed, the events will become interesting and role playing can achieve new heights. The monies revenue is distributed across the investors (developers).

    Now say another developers did not like "The World" but wanted to make another World that was different theme. Another Meta-server world is created that gains those interested in that theme. They use the same design but not the same assets however they are still tied to your development. It is a self sustaining revenue here everyone benefits mostly the game playing public.

    Of course this is a very rough draft of the principle of scalability and what it can do but I think you get the idea of what I mean that a indie can put the real Massive back into MMOs.

    If only they would try. :(
     
  29. SirGive

    SirGive

    Joined:
    Sep 27, 2010
    Posts:
    384
    :eek: Demi, that is a MASSIVE idea. And a great one, I must admit. Lol.

    MTracerStudios, try building a log in system to get a hang on how to send data back and forth between client and server. I am currently developing an online game myself. I just finished a log in system complete with a chat. :3 It saves all the players information to an XML and retrieves from said XML. The server is the only one who can create these files and such as well as validate the information being sent. I'm just doing this all in Unity, no backend servers or anything yet. I don't feel like dishing out monehs for players when I currently have no one interested as well as no viable play or show. :p But thats just how I'm starting with my team.
     
  30. MTracerStudios

    MTracerStudios

    Joined:
    Aug 24, 2010
    Posts:
    118
    Sirgive, do you realize that most server backends (ElectroServer, SmartFoxServer) have free versions for testing with a limited number of players? You could use that, and then, say, use Kickstarter to get enough money to buy a version with more players.
     
  31. SirGive

    SirGive

    Joined:
    Sep 27, 2010
    Posts:
    384
    I do, but this is a senior project. Should, for any reason, we get some financial support from anyone(players or investors). Then we'll code in a backend server. I'm more curious as to how well unity can act on its own as client/server, though. I haven't seen anything on the forums about anyone doing it.

    Edit: And from what I've seen, its like 20 players for free or so(could be better ones). I'd venture to say Unity could support that as well.
     
  32. SrBilyon

    SrBilyon

    Joined:
    Jul 13, 2010
    Posts:
    290
    This thread is awesome and Demi wrote the best posts I have ever seen related to this matter.
     
  33. emergence

    emergence

    Joined:
    Aug 25, 2010
    Posts:
    386
    Thanks for the information.
    I have a simple question for Demi... on scalability (probably not the scalability he was talking about)

    I start with a simple game, which is singleplayer or very small multiplayer (4 to 10 players, max).
    This game has only 2 characters, and 2 classes each, and only one "zone" to play in.
    There are no quests, items are simplified (being only a handful, and static statistics on the items), and skills/abilities are few and limited to the two classes. There are only a few monsters, so AI isn't too complex, and most of the gameplay is player-created via PvP, competing player vs player combat, with NPC's in the world fighting alongside them.

    I then add to it. More character, more classes/skills, more zones, more items, more abilities, better AI, more monsters, and begin to add in quests (a bit complicated. Quality over Quantity.)

    Is this realistic? Smart? Or will it be added trouble to not make the game MORPG from the beginning? I want to start ORPG or just RPG. In fact, it'll be more like a Tower Defense Game or a Browser Game than an RPG at first. It slowly scales with player interest and demand. If players do not like it-- it never goes beyond a simple demo of a singleplayer experience.


    The idea for when adding PvE is to have quests/adventures to be based on a "type" of quest or adventure. These are self-contained adventures, meaning I create ONE adventure, which creates MANY adventures. The beauty of computers is how it can do the work for you. You can create machines, to create machines. I have eventually hundreds of variables in ONE adventure type, which creates unique and fresh content. The goal is based on variable as well-- just the shell is ONE type. And I do this several times (3-5 times) creating 5 different types of gameplay, all with randomized, variable content (It will have to be playtested until it FEELS like it creates DIFFERENT adventures, even when it doesn't)

    And IF and ONCE the game gets to a bigger scale, I add more variables and gameplay types, more characters, items, monsters, zones, and a bigger world.

    But it all starts off with only 2 character types, 2 class types, very limited abilities and items, NPC's derived from the character types, and only one battleground/zone to play in- supporting 4-10 multiplayer.

    Then 4 characters, 2 class types, a few more abilities/items, and a second battleground/zone-- supporting 10-25 multiplayer.

    Continuing in this scale until there are 40 characters, 10-25 class types, 20 armor types, 20 weapon types, about 200 abilities (some of which are hybrids of other abilities), and 4 to 20 zones. Supporting 100-1000 multiplayer.


    I don't know what an MMORPG is on this forum or in people's opinion. I've seen games with less than 20 ppl ever interacting being called MMORPG's, while games which support millions being the same genre. Quite different in design, but both considered "MMORPG's". The term is vague.
    I just want to make a ORPG, or MORPG. Also, my take on what an "MMORPG" would be is nothing like what I see today. I refuse to make a copy or slightly-altered version of the same non-innovation I see in gaming today. I'd want to make something that would arguably not even be considered an MMORPG, but an entirely new genre branched of of MMORPG's. Something closer to MOVW but focused on the fun of MMORPG's PvP systems and PnP RPG's PvE systems.
     
    Last edited: Jan 12, 2011
  34. Demi

    Demi

    Joined:
    Dec 6, 2010
    Posts:
    20
    I was talking scalability of the server. The clients are not an issue at all since the client runs on one machine for one person. Server design on the other hand is very complex. Some server engines are only capable of handling maybe 1000 to 1500 CCU (Concurrent Connected Users) and are not capable of 10K to 20K CCU. These server are one server shards so once they have reached the maximum population you have to send any new players off to another server. Servers like RedDwarf, Bigworlds and believe it or not Multiverse are built to cluster so when you reach a point where server load is to high they can pass off the overflow to another server. What this means is all players can play in the same world.

    There is a possible problem though in if there are to many players in one tight area (all visible to each other) the clients would not be able to keep up and will lagg (100's of visible players can have a huge impact on the client end). Under this condition the servers may well split the area and instance it to a limited number of players visible at any given time.

    You're on the right track to just get a basic building blocks. A little forward thinking here can save you a lot of grief later so keep this in mind as you build the game. If you make a single player game and just program it all in one lump then you will have a major stumbling block to make a MMO if you decide to do it. If you build a single player game and split the server side and a client side to the design then making a MMO would be easy since you followed the basic requirements up front. Suppose after making the game you decide it is not worth making a MMO then you can release it as a single player game anyway.
     
  35. emergence

    emergence

    Joined:
    Aug 25, 2010
    Posts:
    386
    Thank you for taking the time to respond.

    Well, either way I want to make it multiplayer (minimum 2 players, but hopefully 4 to 10).

    Should I start with one of the three MMO middlewares right from the start (Photon, SmartFox, or Electrotank?)

    Or should I start multiplayer with Unity's multiplayer?


    Okay. So all I need to focus on is to make sure that the entire game is split server/client, and that will make it easier to add large or simple multiplayer later?

    Alright. I will make it multiplayer from the very start then. Now just to decide if I need to implement one of the middlewares immediately or if I can do that later, when Unity's multiplayer begins to fail me.
     
unityunity