Advice on a 3D RPG game development

#1

This is my final year before I get my Diploma of Multimedia Computing. Together with 3 other friends, we propose to make a 3D RPG game for our Final Year Project (or In-House Project). We will like to make a game of Final Fantasy 7-8 quality and likeness. This is our first 3D game. We have around 5 months time to do this project. We plan to use DirectX 9 and C++ to do this project. We will be using Photoshop and 3D Studio Max too.

I would like to have advice on general game development, C++ game coding, DirectX 9, modeling, texturing and some others you guys think would be useful to us. :slight_smile: I really appriciate your help. :slight_smile: Thanks in advance.

Edit: Oh yeah, is it advicable to create our own 3D engine for a Final Fantasy 7-8 like and quality game? About how long would it take to create a 3D engine enough for a game like that?

#2

5 month is damn short for such a project.
Just for comparison:

Our “porjektpraktikum” consisted of creating a virtual world of the “Mittelrheintal” (a few kilometers of the river Rhein with a couple of castles). 26 people, 6 month, using OpenSG as a base system, while the “logic” (event and statemanagement) was created by our team.
The result was … rather bad. The main problem was content creation because people started with 3dsmax / photoshop mainly from scratch and at the end most parts of the story and models were skipped because of not enough time. And that would not have changed very much if all people would have worked all the time on the project (the main work was done in the last 2 month).

What I learned from this project:

  • Start early
  • Make a timetable with milestones (and try to keep them)
  • Make plans of what you want to do (UML-Case / Statediagrams, storyboards, etc.)
  • Don’t spend too much time on discussions
  • Be realistic in means of time and manpower
  • Don’t reinvent the wheel if you don’t have enough time (meaning - use an existing engine)
  • Look up programming patterns (singelton, factory and stuff) BEFORE starting the project, because discovering at the end is rather … annoying :wink:

The people after us did another Projekt, the “Virtuelle Marionette”. I watched this projekt because some of my friends were involved here. This project was a success (IMHO), although they did “finish” it over half a year after the official end of the project.

What they did right:

  • Enough key-players that worked EVERY day
  • Extensive use of existing libraries (some of our project, the boost-library, SDL, OpenSG, OpenAL, etc., etc.)
  • Extensive use of patterns
  • They did reinvent the wheel as they wrote their own physics engine, but they had enough people with enough knownledge to do that (result is named “open Simon” and AFAIR is available at sourceforge)
  • They used a wiki for project management (very nice idea)
  • Weekly meetings of the different teams to keep people working (that was one thing they learned from us, because we did not do so :slight_smile:

Conclusion (aka. my advice):
You do not have that much time (5 month is short) and not many people for you project.
Ergo:

  • If you’re from a more art-related department: use an existing engine and focus on content creation (more content, less programming)
  • If you’re more the programming people, take an existing scenegraph (that’s the hardest part to program) like OpenSG ( www.opensg.org ) and focus on writing the gamelogic (meaning you have to reduce the content a bit more)
  • Regardless which way you choose - keep content small, meaning do not try to create an epic story - you will NOT have enough time for that
  • Maybe think about doing an extensible demo (reduce content and keep the project open for more).
#3

I know nothing about engines or game development so I can’t tell you anything there. But to me the hook is the story and coolness factor (very hard mix of a lot of thing). Since the coolness factor is too ambiguous to define, what is your story? The story and, in my opinion, especially the villian(s) has to kick arse. I am trying to work on a short story now after not having written in years and it is hard to get the pen flowing. Not sure if this is the place to put story design though.

#4

i havent worked on many 3D projects, but I know the worked involved. I agree with arnecis.

5 months may be too short. However if you use existing engines (game or just graphix) and you work hard, why not?

Create only, say the first episode or act or playable area, first in the given time period. Then you have at least a working demo of the game and all you need to include is the rest of the game story and depth after the 5-month period.

#5

Thanks! Thats a piece of good advice. :slight_smile: if we had went into developing our own engine, we may fail our last semester. Well, I think the final product would be a short playable part of the game without much content, depending on our progression. Anyway, any recommandation on game engine? We still need to learn game engine from bottom up. :stuck_out_tongue:

Oh just for your info, my team consist of:
3 Coders
2 Graphics Artist

I’m one of the coder but I voluteer to do some of the graphics. :slight_smile:

#6

yo ppl,
Um if you want a pretty good game engine that doesnt cost alot of money it would be torque game engine. Its only 100$. Google it. Good luck. ull need it.

#7

Hmm… I think, maybe you don’t want to use Torque (www.garagegames.com).
I’m about to develope an own FF with some Friends in a Forum (we aim for a full-fledged one, but rather work just-for-fun). Thus am interested in the stuff myself and found this article on Engines in the c’t (magazin for computer technic). It says:
“und so steht man als Neuling dann vor Torque, das einen prompt, völlig, nachhaltig, gründlich und komplett überfährt”
roughly translating to:
“and then, as a newbe, you sit in front of Torque and it instantly, absolutly, thoroughly, very and comletely squishes (flabergasts? confuses? not sure how to translate it) you”
Torque is very good for long and thorough projects. If you plan to make an entire FF, and your diploma-work here is only supposed to be the first step, you might consider Torque. But taking your tight shedule into account, you maybe would rather like the Q-Engine (http://qdn.qubesoft.com/). It doesn’t have quite as much kontent, but it’s easyer to get into (very good step-by-step-documentation) and it’s free.
If it’s only for this projekt, maybe you’d be happier with 3D GameStudio Commercial (www.conitec.net/a4infod.htlm). Its easyer to get into, but it’s not as good for longer projects and costs 200$… (pro edition is 900$)

But I’m realy interested in that story of your’s… would you share your ideas? :hmm:

Greetz, Ky.

#8

We haven’t discuss about the story yet, only gameplay. Its not that we are slacking, its just that we haven’t reach the planning phase. I’m still at my second last semester. So I had nothing in my head right now. :stuck_out_tongue: But story won’t be the key I think. cause we are doing the demo. We would emphasis more on the technical part. Like modeling, coding, etc. :slight_smile:

#9

5 months is not enough time. You could spend your entire 5 months just working on getting an engine together and tools (exporters, editors etc.) for development leaving you no time for development. I would highly recommend looking for some “off the shelf” solutions so you can spend your time developing the game, and not the tools/engine (unless thats your goal, which in that case keep it more of a tech demo and focus on the tools/engine).

Even if you go with off the shelf solutions, you are still going to find yourself at very very steep odds to get the quality and quantity of content together you will want to even get near the FF games in 5 months. I would say in 5 months you should be able to have a few characters done, and perhaps in the engine and about one level.

What I’m getting at with all this rambling is you have to try to be realistic about your goals in your time aloted. If your have 5 months, instead of trying to make a complete FF game, perhaps focus on trying to make the 1st level of the game, have the main characters in and the scripting/gameplay in for one level. Game development is very very time consuming. Just getting level content in place can take months.

Research your goals, and try to come up with a schedule/plan. Work with your team members to get time estimates on thier tasks, and then pad the heck out of them because if it’s your first game, then you’ll need all the padding you can get.

Good Luck!

#10

I would suggest you use a direct x editor language like dark basic. it only took me a couple weeks to learn the language and it is very useful for 3d games plus it can easily be used with programs such as 3d max, maya, lightwave, ect since it can load objects in .x format
(which almost any 3d program can export to via a plugin)

darkbasic is a very simple language to learn, but also very dynamic and fluid you could get a trial and the full version here. also if you need any plugins that aren’t in direct x 9 sdk, check here.

I am also making a 3d game, but about an action movie my friend directed and starred in {Captain blasto (only shown at lowes theater)} i am using darkbasic, though i did run into a minor problem with the controls which should not be too hard to correct, but to give you an idea of what you can do with it you can look at the start of my game from here. Sorry about the popups, on my site, i cant control them but there are no viruses or porn or anything like that just regular advertisements so don’t worry

hope this helps:shrug:

#11

These people are correct, 5 months is a joke. A team of 50 experienced people couldn’t make a decent game fro scratch in 5 months, even if they were working full time and had an engine to start with.
With 5 amateurs you are looking at 2 or 3 years work, if you put every second of your spare time into it and are so talented you can work quickly and keep the quality up at the same time.
Things always take far longer than you think they will.

#12

I’d say either concentrate on developing a new graphics engine or make a game, not both! There are plenty of good graphics/game engines out there, some of them are even free. Check outOgre3D. I found it very easy to get into (once you got it installed :P) and there are a lot of helpful tutorials to start out with. However, I wouldn’t recommend it unless you feel comfortable working OO, with classes. You can use Direct3D9 with Ogre if you want.

Don’t create both a graphics engine and a game. You’ll be busy enough creating the game engine (unless you decide to use an already made game engine) + game content.

#13

Sister Ky – thanks for the q-engine link! I didn’t know it even existed until I read your post. This page here got me very interested in the Q-engine:

http://qdn.qubesoft.com/docs/1.1/doc/qdemo2/synopsis.html

I can’t wait!

Another excellent free 3d engine is irrlicht:

http://irrlicht.sourceforge.net/

It has a huge engine-user supported forum and simple tutorials to get you started (I got a demo going in a couple of hours) It’s fairly easy to tweak the included examples to do what you want. If you have a question, chances are they’re already answered in the forum – just do a search in the forum.

I have one project going using the Irrlicht engine (3d remake of the PC lion game – part of the the now defunct santuary woods predator series). I’m just starting on another one (3d remake of GBA’s Fire Emblem the sacred stones).

3D GameStudio is great for fast prototyping but I wouldn’t necessarily make your dream-game out of it. The 3d engine feels kind of slow and bloated the last time I tried it.

arnecls – thanks for your “post-mortem” post. it’s really insightful. i love reading game developer’s post mortem articles (they’re invaluable source for lessons learned – or don’t make the same mistake that we made).

DarkenedFire – i’ve mucked with darkbasic before. though it is easy to use, i’m more of a c++ programmer than basic (although I have programmed in vb basic before). so, i didn’t go that far with darkbasic.

#14

This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.