Please Delete

#1 If you search the forums, you'll find post after post of reasons why you shouldn't even attempt such a project vs. those who say go for it.

Thing is, generally, if you have to ask all those questions about making a mmog, you shouldn't attempt it. Start with a small game, work your way up.

#2 You make some good points - I should have considered this before posting this thread.

#3 This has already been discussed a lot, as NfoCipher said.

1. It depends on your game. Potentially no changes are required except some additional scripting and a database. Or you might need to restructure the way Torque does client/server connections.

2. It depends on your game. The connection management code would probably need tweaking.

3. It depends.

4. This works the same as in all other programming projects. Programs are, generally speaking, holistic. That said, Torque is modular, and you can change some things without affecting others.

5. Run a multiplayer game. (for a <128 players definition of small.)

6. See above.

Torque's networking layer is handily superior to nevraxes. This comes with a concordant increase in learning curve. You will spend a year learning Torque before you can do any serious engine development.

You'd get a better sense of the community by lurking and reading forum posts, or coming onto IRC and hanging out there. We'd be more interested in answering your questions if we knew you :)

#4 Can you explain this in a bit more detail? Why specifically it's better, etc.?

#5 LOL

Torque is cross-platform. It has no dependencies on DirectPlay, etc. etc. etc.

The networking code has no intrinsic limits. It's a matter of CPU/bandwidth usage. I consider it better than Nevrax for a game because it's tightly integrated with the object database, and because it's a very efficient protocol.

The problem with an MMO is not so much the technical requirements, as it is the content.

A MUD is an MMO... its draw is not the technical foundations of the game (which are primitive) but the content. The same holds true nowadays. A bad rendering engine, a bad networking system - these are detriments. But the thing that must be positive, is the content. And that's the rub for any indie.

#6 .

#7

Quote:What kind of methods would be best used to create large, seamless outdoor environments, while keeping in mind torques ability to only handle a certain ammount of space.

As is, you have to "zone". You can only make something so big before it takes a year to load up and require the client to have a gig of ram.

Quote:One method might be to stich the edges of each space of land together, and then load the next maps data uppon approach. But this doesn't seem like a good idea. Theres lots of methods I'm sure.

Would require some heavy modifications to get that to work.

Quote:Another thing that of course important to consider is that of the editing involved with getting torque to support more than 128 players. Are there just a few adjustments that need to be made? Or is the entire framework going to need to be torn appart and reworked to support more players.

There is no limit on number of players. The limit will come from your server hardware and/or bandwidth.

Quote:I was also thinking about the legal issues involved with using the network components of NeL, and using with toruqe if one so chose to do so. Is it legal to charge an access fee, and at the same time make a minor profit without breaking the law?
I'm pretty sure that using NeL will make any software its used with free, and therefore not legal for anyone to charge money, or be associated with code you paid for in the first place. IE The torque engine.

And I'm pretty sure you haven't read the GPL at all. Nothing just magically makes software free if it wasn't GPLed in the first place.

#8 I am no programmer, but many MMOGs have 'seamless' worlds, where different areas are loaded up dynamically so there is no 'loading' (like there is in everquest) other than the initial load up/login. DAOC is seamless I believe, Im sure there are papers written up on this on flipcode or gamedev.net would be a good place to start researching.

-Justorama

#9 No, I haven't looked at the GPL myself, I'm working off information given by me from different sources. I'll take a look at that right now. :)


NfoCipher, understandably it's a possibility that your definition of a zoning system may differ from mine, so would you care to explain in a little more detail?

#10 Zoning meaning EQ type zoning. You have a large area connected in various places that will move your character to another game server.

#11 If you were going to use Torque for an MMORPG then you'd really want to use an EQ style zoning system. There's even a resource to do it.

Just loading the level as you approach isn't the only problem, you'll need to do server handoff etc. And what happens if someone shoots across a boundary? Even worse what happens if there's a fight across one?

Consider your priorities, you can spend an eternity changing torque to work like this, you can use sometime else, or you can live with the minor problem and spend all the time working on your game.

Honestly the moment people stop hanging up on stupid little things they might get some games done :)

#12 And thats only the beginning. What everyone else has said so far plus
MMP type games also require a fair amount of infrastucture support, its not
just a bunch of clients and some game servers.

patchers/loginservers, possibly even database servers, groupmanager/chat server ,
customer support tools, monitoring systems, billing(unless its free),
possible shard selection(if you have more then one cluster of servers that make up your world), authentication, security to do, ability to easily add ongoing content,
load balancing capabilities, etc, etc, etc and the list goes on and on.

pickup the book called massively multiplayer game development, just came out a few months ago, bunch of good articles in there.

Torque is by no means even a close solution for a serious MMP/PSW style game, its only one small piece of what is generally a much larger system.

It can be, however, a great techonolgy to start with :)

#13 .

#14 You have to remember that people have faster computers and more powerful graphics cards than they did back in the Torque 2 days, so I bet it can handle a lot more models by default now.

Anyway, if I put my game designer cap on for a moment, I think you could easily redefine this whole problem. People get too stuck on the "fantasy continent" mode for MMOG's. Design your game world to take into account the server hopping. Maybe the maps are islands floating in the sky, and the only way to travel from one to the next is through portals (also known as loading screens)? Just an example. Design the game world around your technology, instead of the other way around -- The Way of the Indie?

#15 Also is not just getting a lot of models on the screen(and getting it to work) all those collisions, AI, physics etc all have to be tuned to be able to handle more people
(if they werent already designed that way)

so it could take more work then just art or LODS, it may take a lot of performance tuning and optimizations of those other systems to get them to support the increased number of people. There are lots of ways to accomplish this that would get into the nitty gritty programming side of things.

There is a bit of give and take with a MMP type game, supposedly you know what you want the game to be (the game design) and you have some idea of what the different techonologies offer. Also you usually have a time frame/budget to work with.

So the give and take come into play once you know all that, so maybe the game engine doesnt do something you want it to do (based on the game design) you might alter the game design to compensate, or decide to spend time on the engine and change it to do what you want. But if expect to ever get the game done(especially a playable mmp) you going to have to make consessions on both sides.

or if you have infinite time and money, write all the techonolgy from scratch and make it do wha tyou need. But in this day and age its not really practically to do that anymore. although as an indie developer you probably have more flexibility to do this then the mainsteam game developer working for ea, sony or whomever.

As a side note its a lot easy to get a single player or just plain multiplayer game up and working to test out the game design to see if it will work then to write a full blown MMP/PSW type game. Just a thought, but your first game probably shouldnt be a MMP
style game, get some experience in something more scaled down first, test out your game ideas at that level first, you get results faster, you get something to play sooner and you'll learn a ton.

start small, build up, learn, try, play, test, resdesign and repeat. MMP games need all that your going to learn doing that, plus MMP's have a bunch more stuff to learn when migrating you design to such a large scale.

But no matter what you do, good luck, Im hoping the next awesome MMP games come from indie developers because the ones coming from the market driven, bean counting bottom line, underscheduled, underfunded MMPs from the big guys dont seem to be getting any better. They just dont take risks like indie developers do and there games are refelecting that fact.

#16 Anyone interested in attempting an MMO title using Torque might want to check out the newest version of our massively multiplayer networking middleware 'Eterna', available at our DevZone.

For independent and unfunded development, it's available free of charge, and offers technical support through our DevZone forums. All we ask in 'return' is that you provide commentary on the technology on our forums, and if you have any helpful code samples regarding the interface between Eterna and Torque, that you provide them to other DevZone users (and here, really) as well.

If anyone has any questions about the Personal Edition license, the technology or other Quazal products (such as Net-Z, our middlware for 'normal' multiplayer games), feel free to contact me at mike@quazal.com.

Thanks,

Mike Drummelsmith
Developer Relations Manager
Quazal

#17 Also see the GORPE Project, it has code available for stitching multiple terrain blocks together up to a pretty decent size. For a 300 person game you might be able to get away with having it all on one server this way, though you'd certainly need to make a lot of networking optimizations (something the GORPE project is also working on)

I would work on getting a decent persistent world worth of content going on a standard Torque server (or series of servers linked together as "zones") then go from there.

#18 I took a peek at Net-Z and Eterna. Whew those license prices uh kinda on the higgggh side of indy. :) But those features are great.

Having spent a good bit of time reading and studying a lot of things I notice that the same solutions that expands torque appeal have always been third party (rock-solid)solutions that interface with the stock torque code.

Torque already has a kinda ghosting ability built in. So I kinda believe in Ben statements. If a programmer were to spend time on one aspect of torque and improve it you probaly could get similiar ablities out of the stock code.

A specialize torque server comes to my mind. a torque server design solely for the purpose of serving up MMOG goodness. Datablocks, Terrain, Objects and other Clients get handle a special way by this server. It sole function is to serve heh. I not talking about just a dedicated server that host. But a specialize torque-coded MMOG server.

Clients becoming pretty much just windows into this super (MISSION),(SIMULATION), (ZONE) or whatever. I not going to say I think it can be done, I know it can, somebody going to do it. One day.

John H

#19 @Mike Drummelsmith: Great resource, something I've been looking for for my own PW project... =)

@Raymond: Here's a few things that I did, though time's really going to tell whether I make the right decisions or not(I can tell on one or two already):

1) Design wayyy before you look for a team. You don't need 20 people to come up with backstory and features and concepts. Just you and one other person to bounce ideas off of.

2) Search the forums, take a good look at what TGE can do compared to other engines and what you want your game to do. Then gather a list of code snippet links that you'll need to integrate into your engine so you can track your progress. List other things that aren't covered by either TGE or the snippets as well, so you know what you have to hand-code.

3) Spend just about all your free time on it. You have no idea how much stuff you'll have to think up. Every layer of the game adds an exponential increase in thought(I do give myself headaches regularly now, lol).

4) Develop strategies for asset production, poly counts, naming conventions, folder trees, tasking, mailing lists, forums, chat, phone conversations, etc. with your team. It's a beast, and you'll have to ride it.

5) Write everything down. I have a blue book that I'm infamous at work(restaurant) for having with me all the time, where I write down ideas and stuff. Since I get to stand in one spot, I spend a lot of "boredom" time thinking through different aspects of my game.

6) Design around story, not features. Oh God, please, if you take advice from me on anything, take it on this =)

7) Documentation is king. Revise it constantly, keep track of it, grow it, and then when you have a good amount, get a team and drop it on them like an anvil on a cartoon character. They'll appreciate the front-loading of information(as long as it's good information).

8)Learn to be a jack of all trades. Giving orders won't get you anywhere, and if your team starts with that ever-present team attrition(like what hit my team a few times), then you need to be able to take on those lost tasks. Trust me, it's a crushing load of work when that happens, but if you just muddle through, you'll make progress.

9)Don't settle. Hell, if you're going to do something this hard, aim for the stars. If you don't hit the stars, you might still hit the moon. But that's just me, lol.

10) Become a book worm. Get every book you can find on MMO's. If you don't like what you read, then pay damn good attention to it, because it's likely a very important piece of information that simply doesn't fit with your viewpoint, and thus needs to be burned into your head(people have a tendency to not listen to things they don't like to hear, but if you look the ugly in the face, you'll be that much better off from the start).

11) Finally, be sober about it. It'll take a long time to do, and you need not only motivation, but endurance. My design started in 2001 and in Feb. '03 I started this. I probably won't launch until the end of next year, but I'm making good progress and I didn't come all this way to give up. Fail? Sure, that's what risk is all about. But I'd rather try and fail than try and give up, so if you're going to spend all this time on it, then steel yourself for a long, drawn out fight.

Oh, and read the forums A LOT. Even if you don't agree with what the other guys here say, pay attention to it, because there's wisdom in their words, even the ones that may not be perceived as right.

In the end, you'll either try and fail, try and succeed, or give up, and it all depends on you...

#20 Definitely agree with almost everything Ted said... except one thing. I can't totally agree with the idea of designing around story rather than features. A great backstory can go a long way, but if the game is 'more of the same', users are more likely to stick with a universe they're familiar with rather than invest their time starting up with a new one. The major aspect I think that's going to need an overhaul in the next generation of MMO titles is combat. It's just BORING for the most part now, across the board. Even a small wrinkle in the established pattern like combat stances or group attack techniques, or learnable defenses (something I tried to stress to the EVE Online team back during my Matrox days) adds a strategic element and a mental element that completely overhauls the traditional MMO combat system.

Obviously it's going to be tough going to a fully action-based combat system (not sure if that's even something that's wanted), but it CAN be done. Traditional client-server MMO systems, though, aren't the answer (which is where Eterna hopefully comes in).

Oh yeah - and to Johnny Hill, you're right. The pricing for Eterna is pretty high, for published retail games (that have the budget behind them). If you stay indy or hobby, it's always free. If you get lucky and manage to gain the financial backing necessary to bring an idea into a full-fledged professional production (which in the MMO world takes MILLIONS of dollars), the money you save on having a robust networking package behind you covers the cost of Eterna many times over. On the other hand, if you were trying to make a game like A Tale In The Desert, which is a very small (but 'retail') MMO title, there's a ton of options that we just couldn't list of the page that could drop the licensing price.