I’m planning an MMORPG. I have a good story framework, waiting to be filled with Quests and the likes. Seeing however the trends in games these days, everything seems to be 3D and shiny, but honestly, lacking in gameplay options and isn’t as fun as could be (IMHO). So, my question is, if You were making an MMO, what would you choose to make it like?
Seems to me that you could make it Top-Down and get it all set up, then switch to 3D
Maybe in the far-off future, but 3D graphics isn’t that important to me, never was, really.
It depends entirely on my coding experience.
If I had little to no graphcis experience Id go text.
if i had some 2D graphcis expereince, and acess to sufficient 2D art Id go 2D.
If I had 3D experience and access to sufficient 3D art, Id go 3D.
If I had NO coding expereinbce, and just a “story”, Id give up on the diea of MMO and buy Neverwinter Nights and do a module.
-
Well, I’m an experienced programmer, at least commercially but I have to admit, this is my first attempt at a Java game, although not my first attempt at writing graphics code as I’ve done that especially with MC68000 ASM.
-
I can’t draw to save my life, so basically I’ll need lots of artwork… and I’m currently always on the look-out for good artists who are of course willing to work for the sake of the work itself.
-
I prefer 2D/Iso since its simpler to work with than 3D, but also since it’s much cheaper and produces results that are closer to my taste… (I know, I’m a retro-gamer…) ;D
-
I also would like to do the programming myself rather than script something with someone else’s engine simply because I’m a DIY kinda guy when it comes to Software.
Thanks for the link, I was just going through that site a coupe of nights ago 
I also know what I’ll need in graphics, tiles, sprites and effects, so having a graphic artist (or team of artists) is the way I have to go.
For purposes of testing my code, I can manage with tiles and sprites found on the Net, so any such links as you’ve posted would be
helpful if they’re not already mentioned in these Forums. 8)
I would really consider if a MMO Game is a good choice. You will need a server to run it on, usually permanent administration of players and quests and such, it is basically a full time job, at least for the big ones. And dependent on the Gamedesign, you need a certain number of players online.
I have never seen a hobby based MMO-anything, but whatever hobby project I have seen that based on multiiplayer was almost anytime short on players. You would need a few hundred fans in order to have a few dozens playing at the same time.
Maybe a good game suited for single and multpilayer should be your choice. Multiplayer should be hostet on a client, you would only need a server to bring people together, one would make a server, others would join.
When it comes to a MMO, a big problem would be testing. Performance testing on the one hand, how do you want to test 100 clients? You cant test this on the fly, the people will leave your game if it runs too buggy while you fix the last things. Gameplay testing is the second problem. Will your game be stable and fun after a few months of permanent runtime? Is there no memory problem? Where do new quests come from, when a Player has done everything you brought in at start? Is the game well balanced for a Level 99 char?
Many people start to make MMORPGs or such but never finish. A working network code for a small RTS can already be a bitch, to get synchronized gameplay with an internet latency of up to 200 ms, while there are only 4 to 6 players. I would say, if you dont have a few teammates and if you dont already have made a few big games like rts, ego shooter or such with a complex gameplay and a working network mode, you shouldnt start with an MMO. Get something “smaller” working like a Warcraft, Diablo or Doom and add a multiplayer mode. If this works, maybe you can take the next step.
-JAW
Er, I’d start with Pac-Man myself, or something similar. Prove to yourself you can write a complete, finished, perfect product before even considering an RPG of any kind, let alone multiplayer. And just forget about the first M in MMORGP. And if I were an experienced artist I’d run a mile when I saw really how much work was involved for almost certainly no gain. And if I were a n00b enthusiastic l33t teen artist I’d volunteer, produce 3 sprites, then give up when I realised there was no hope in hell of ever finishing it or I discovered beer/pot/girls/life.
Cas
JAW and princec, you’re both very right in your advice. However, I’m well aware of the amount of work involved, the commitment to the game and its users that is required from everyone who participates and the problems of hosting and getting people to play the game.
About my commitment to this, I can only say that I already proved it to myself that I can write complete, finished, prefect products in the Hightech Industry. Of course we’re talking about commercial applications/components and not a game, but I also identify the differences between commercial applications and game-applications. These differences aren’t going to be difficult to bridge.
I’ve already had the experience with a n00b, l33t, teen member of my team who’s not there anymore and it was a lot uglier than just producing 3 sprites and disappearing, but I can also say princec, that I don’t have much choice in this anyway. I can’t really draw, and I have to trust that others with more graphic talent than me will be willing to help.
About the hosting problem, I’ve already addressed that and if all goes well, the game(s) will be hosted and produced by a third party who is a silent partner. I cannot, and you don’t need me to, say anymore about this issue.
JAW, I’m not going for an RTS. One of the reasons I’m going for a TBS is that I know how difficult it is to synchronize entities between more than 2 clients. Another reason for going TBS is simply cause I don’t like Real-Time games so much and prefer TBS RPG’s.
About the issues of balancing the game and working out stuff like what to do with lvl99 characters, I’ve got that too under control, although this is much too early to talk about.
The thing is, in this thread, I was hoping to get people to tell me what would be their choice if They were making an MMORPG, and especially about their taste in the graphic-engine part. I’m well aware of how difficult it is to make, and how time-consuming and how annoying it can get. I’m also aware that many projects never even get published, or even hosted. And I’ve read these forum’s threads enough to know that a lot of programmers here would advise me not to aim so high. But, I already wrote several graphic demos in ASM. I wrote PacMan on my AtariST and Dragon32 when I was in HS and I wrote professional Client/Server software for the last decade almost. I’ve been analyzing RPG’s (partly to hone my abilities as a DM in pen&paper D&D games, and other systems) and learned a lot about the inner workings of most systems involved in a game including scripting-engines (even wrote some myself).
To sum up guys – don’t take it the wrong way cause I don’t mean to offend anyone – my mind’s made up and I’m going to push through with this for as far as I can go, I just really wanted to know what would most people prefer to work and play with, 2D top-down, the Isometric view, 3D scenes or even just Text based.
I dont know if there is a great majority for any kind of graphics. For me i think that even 2d appeareances can sometimes be easier done with 3d and fixed camera. I’d just think of a Diablo style game. An isometric game always comes with min 8 directions to face for each model. Diablo brought in different looks for armor and weapons. I think this is better done with a 3d model than handling dozens of “real images”. When it comes to heights, 3d can be better to take than 2d optical tricks that make things look like height and depth and such.
What I really would find interesting is a comic look game. There was an ego shooter called XIII which used a 3d engine but made a comic style look. I am personally interested in Animes and these are very often in a Fantasy setting with different “classes” of charactes and a roleplaying style story and development. So I would personally try to make an anime fantasy setting come to life. It would at least be different from those all out 3d stuff currently used. A little more AI, more gameplay, less super light and shadows and such.
In my eyes, the main requirement for the look is, it shouldn bother or distract the player. The first role of the graphics is to be the interface between game and player. Game and Player interact, the Player with the keyboard and mouse, Game with screen and sound. In both ways, this interaction must be seamless, fluid and straight, without the smallest annoyance. You dont have to win a 3d contest, and the screenshots dont make the game good. A game is played as long as it makes fun. Fun either drops with gameplay or with graphics or controls. It can be 2d or retro, looking like 486 times, as long as it works and you get everything.
I think the best graphics are those that arent noticed at all. Great visual effects make the player go “wow” once or twice, thats it. There are many great games with moderate graphics. My personal interest is said. I’d like a comic like look, like in old adventures as Monkey Island 2 or 3, Full Throttle, etc. Or a 2d look, even if it is made by a 3d engine. Just make the camera fixed and dont go for super light effects, transparency, trillions of polygons and so.
-JAW
Thanks JAW, that’s exactly the sort of answer I was looking for!
Well, given the, er, givens in your post, I could help you prioritise a few things, based on my past experiences:
- Don’t worry at all about server hosting. It’s cheap, trivial, and only relevant very late in the production.
- Turnbased: a very wise move, fantastically easier to code, you can even use RMI as your protocol and it’ll work beautifully.
- Gameplay tuning: also, largely irrelevant compared to technical problems you are likely to encounter. So don’t worry about it yet.
- Graphics: this is your biggest problem, however it is entirely surmountable if you use really bad programmer art and do the entire game with crap art. Once the game is actually up and running nicely - and that means more or less complete - you will probably find artists much more willing to cooperate.
- Writing games as actual products is much harder than business software. The tolerance for faults and rough edges is far slimmer; the technical requirements usually much greater; and as complex as anything else you’ve ever written - except you’ll have only 1/10th the time to spend on it.
Cas
Well the usual solution for this is “paper dolls”. Done right its actually a lot easier (at least on the coder) then peicing together 3D models.
Thanks for the advice. I’ve gone over your list and its more or less what I figured I could do. I plan to use RMI and my crappy graphics and whatever I get from the people helping me. You’re also right about the Business v.s. Entertainment (i.e. Commercial Application Products v.s. Games). I code in two “modes” if you will, first I attempt to understand a problem and figure out how to solve it. Later, once I have something working, I optimize, divide into classes, modulerize and whatever…
“Paper Dolls”, is what I was thinking about too. My graphic-artist also understand there’s a LOT of work to be done for that. :o
If anyone’s interested, I’ve got a very modest demo of an Isometric Display which I coded after reading and experimenting with lots of Isometric Articles found on this site. I’d appriciate any comments, of course.
I’ll second JAW’s idea (simply because I didn’t get to suggest it first.) Even if you want to make a 2D iso game, using 3D with a camera from an isometric perspective will mean less artwork and more flexibility. Also, using a 3D environment will make some spell casting special effects easier to employ, torch lighting, overhead map views, decals for shields, etc. And, when/if you want to swap to an orbitting camera the coding is trivial while such an effect is 2D is nigh impossible. Once you get your props created for walls and such, level building is like working with an unlimited set of miniatures.
TAZ
zircher, you’re right in everything you say, and so of course is JAW, but I’m not into the 3D scene yet, this is also a learning experience so even though I’m serious about producing something to be proud of, I want to start with something I understand, and can currently control.
To your applet:
When I do nothing, its about 100 fps or 120 fps. When I make a selection with leftclick dragging, fps go down. The greater my selection, the more it goes down, down to 10 fps when I select almost every tile.
I run on 450 MHz, so I do not have state of the art performance, but it helps seeing certain perfomance issues, becaus my slow machine really makes a difference where 2 GHz CPUs just need 1 ms more or so
What I want to say is, whatever your “selection” code or graphical display does, it is extremely slow compared to the “nothing selected” graphics. I would guess you do some on the fly computing every frame and change the appearence of a tile every frame, which brings it down. A one time precomputing would then be better, because selection or no selection should be have equal fps.
-JAW
The same slowdown happens on my machine, 220 -> 48 fps
P4 2.8 ghz
geforce 6800
I have a feeling he is doing a translucent operation with the selection area, which is causing the big slowdown.
First of all, guys, thanks for the input.
Secondly, what I’m doing when there’s a selection in place is drawImage(…) of a translucent tile (the red-selector tile) for every tile that is “selected”, so yes, I’d expect a big difference between the selection-on/off phases.
What’s more, when there’s no selection, and you don’t even move the mouse, I simply do nothing because I only re-draw the background and “trees” where the cursor-tile has previously visited.
Another thing is, if you right-click, the “trees” should disappear and appear again in different locations… this action (even when repeated) shouldn’t bring a major FPS change either.