• Advertisement
Sign in to follow this  

How impossible is a 3d M-G? for a single programmer

This topic is 4381 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

If you intended to correct an error in the post then please contact us.

Recommended Posts

hi, I'm currently in my 2nd year of my degree and I am flattering myself to go with a implementation type final year project. I think a 3D game will the best way to go, as some suggested that doing anything 3D these days will never hurt you. There is alot of breath and almost too much depth in a game, where you have treading, openGL, collision detections, tree structures etc.. to improve my amateur skills and will be a good project to stick it on my CV. And I thought it'd be cool to design something that works across the network, not nessesary internet but a simple LAN-game-type-thing. It'd do good for my CV, by adding networking aspect of my project on it. or rather something that would give me a great chance of getting a spot in the research scholarships (for masters) in UK. Do you guys think it's possible to do all of this in 1 year's uni term?

Share this post


Link to post
Share on other sites
Advertisement
im on my 2nd year n thinking about making a game for my final project as well. Its a simple 3d turn-base strategy Advance wars type of game. I already have a outline design of the game done. But I am not sure if multiplayer would b too ambious, consider i will have to pass other modules as well :S

Share this post


Link to post
Share on other sites
Quote:
Original post by superpig
A networked 3D shooter or something? Sure, perfectly doable in a year.


aye, or probably something RPGish (concept's the same, something with a 3D, have interactive characters, etc..).

Share this post


Link to post
Share on other sites
For a final year uni project you might find it better to have a more focused goal than just a '3d game'.

- Lecturers are usually very focused, and won't really comprehend the sheer volume of work that it takes to do a full game and all it's subcomponents
- To get anything of sufficiant quality/completeness done you'll want to use a decent chunck of libraries for the lower level stuff. This confuses lecturers even more over which bits you did and which you didn't
- Better defined project goals and targets make it much easier to say "I aimed to do this and succeeded"

You might want to pick something like networking methods, or terrain rendering techniques and go for something less broad but where you can try multiple different styles and compare them.

Good luck!

Share this post


Link to post
Share on other sites
Quote:
Original post by superpig
A networked 3D shooter or something? Sure, perfectly doable in a year.

You think?
How much time and how many people worked on battlefield 1942 for example

Share this post


Link to post
Share on other sites
Quote:
Original post by The C modest god
Quote:
Original post by superpig
A networked 3D shooter or something? Sure, perfectly doable in a year.

You think?
How much time and how many people worked on battlefield 1942 for example


He said a *networked 3D shooter*, not the Battlefield 2 or whatever. You'd think would learn to read before programming...

Share this post


Link to post
Share on other sites
Quote:
Original post by Outworlder
Quote:
Original post by The C modest god
Quote:
Original post by superpig
A networked 3D shooter or something? Sure, perfectly doable in a year.

You think?
How much time and how many people worked on battlefield 1942 for example


He said a *networked 3D shooter*, not the Battlefield 2 or whatever. You'd think would learn to read before programming...

He asked another separate question. He did not ask if it would take him a year to make BF2 alone.

You should learn to read. You should also be less of a dick.

Share this post


Link to post
Share on other sites
Quote:
Original post by The C modest god
Quote:
Original post by superpig
A networked 3D shooter or something? Sure, perfectly doable in a year.

You think?
How much time and how many people worked on battlefield 1942 for example


I don't know how much time or how many people were involved in the creation of that game. But I'm sure that it would take you more than a year to achieve this by yourself. Such a game requires more ressources that one person can provide during that period of time.

For a quick estimation... I met a guy that had a small studio and they had a team of 20 programmers and the developpement took 2 years (say about 48-50 weeks of work a year) it means 80000 man hours. Now say that you want to accomplish this (programming only) by yourself it means that you will have to work a bit less than 1600 hours a week (80000/52) over a complete year... and I didn't mention anything about overtime and those who had to come to work the week ends and other "free" hours. Then you have a game that you can't play because there is no art, no characters, no maps... So this is a lot of work for one person ;). If it was the kind of quality that you wanted to achieve... you might want to think smaller than that... unless you have 200 hours to put on that project every day for a year.

Otherwise you could think smaller... so now it all depends on how much time someone wants to work on such projects and if you can team with other people. It doesn't need to be as optimized and polished as games that you would sell. Since this is something to put on your CV to show you completed a project and can work with 3D API you can cut corners... don't support the newest video cards features, the artwork doesn't need to be extremely good quality.

If you can, I would suggest you to use a free/open engine. I had a course in which we had to make two games as homeworks during the semester, a MUD and a 3D game. For the 3D game we used an opensource engine and that really helped a lot mostly because it was already developped and tested (and maintained by a community). Try to use as much as possible third party libs... in an efficient way (if it takes too much time too learn it than to develop it... you shouldn't use it). We had 1 month to show something playable to the teacher (2 month for the complete project). So we had a first person game where you had to find cubes in a map and to shoot them as fast as possible. This is what we had time to achieve in 2 month with all the work needed using (and learning) an open source engine (3d modelling of maps, guns+hand, sounds and music, programming and all the artwork needed for the menus and others...).

We were 3 working for 2 month (plus the homeworks needed for the other courses that I'm sure you'll get too) and we achieved something that is really really far from a commercial game (and one was really experienced in 3D modelling).

Usually those school projects are there so you have something to show to employers and prove them you can actually be good to something so using third party libs will show them you can work with the code you didn't wrote. Also, it might be a good idea to team with others to make such a project. Making yourself the project manager would be something interresting to add on your CV too.

JFF

EDIT:Also that project helped me get a job in the industry.

Share this post


Link to post
Share on other sites
Quote:
Original post by The C modest god
Quote:
Original post by superpig
A networked 3D shooter or something? Sure, perfectly doable in a year.

You think?
How much time and how many people worked on battlefield 1942 for example


BF2, I'd guesstimate a team of at least a hundred people over a period of a couple of years. (Mobygames credits at least 120 on the development side).

However, BF2:


  • is cutting-edge in terms of graphics quality, requiring massively more art content (normal maps, detail maps, etc)
  • does not use any middleware, at least as far as I can tell from info at Mobygames and on their site and press releases
  • supports 64 players and a large number of gameplay elements that are not "required" for an online shooter, such as vehicles.


If you use middleware like Ogre, Irrlict, RakNet, HawkNL, FMOD, OpenAL, SDL, etc... and you've either got an artist or the rights to some free models, then you can make a basic Quake clone in a week if not less.

If you write the engine yourself and don't use any middleware beyond DirectX/Win32, you can do it in a month.

Single map, couple of player models, two or three weapons... minimize the amount of content you need, and you'll be done very quickly.

Share this post


Link to post
Share on other sites
Quote:
Original post by superpig
Quote:
Original post by The C modest god
Quote:
Original post by superpig
A networked 3D shooter or something? Sure, perfectly doable in a year.

You think?
How much time and how many people worked on battlefield 1942 for example


BF2, I'd guesstimate a team of at least a hundred people over a period of a couple of years. (Mobygames credits at least 120 on the development side).

However, BF2:


  • is cutting-edge in terms of graphics quality, requiring massively more art content (normal maps, detail maps, etc)
  • does not use any middleware, at least as far as I can tell from info at Mobygames and on their site and press releases
  • supports 64 players and a large number of gameplay elements that are not "required" for an online shooter, such as vehicles.


If you use middleware like Ogre, Irrlict, RakNet, HawkNL, FMOD, OpenAL, SDL, etc... and you've either got an artist or the rights to some free models, then you can make a basic Quake clone in a week if not less.

If you write the engine yourself and don't use any middleware beyond DirectX/Win32, you can do it in a month.

Single map, couple of player models, two or three weapons... minimize the amount of content you need, and you'll be done very quickly.


He said BF 1942 not BF2, the amount of art needed in the second one was probally MUCH less then the new one.

Share this post


Link to post
Share on other sites
thanks for all the advice. I was thinking it's impossible too. I'm actually to focus more towards my planning & design than I were on the programming part, so there has to be atleast some chunk of work in order for me to showcase.

how bout, designing a game engine? with some simple design just to compliment the game engine, enough to show how the game engine works.

Share this post


Link to post
Share on other sites
Quote:
Original post by Ainokea
He said BF 1942 not BF2, the amount of art needed in the second one was probally MUCH less then the new one.


My mistake. In that case I'd guesstimate about a 60 people for two years.

I would advise against trying to design a game engine when you don't have any particular games to try and support with it. A game engine is a very nebulous, cloudy thing, that has no real clear boundaries; it's just a library of utility code and components that you reuse across games.

In your position, I'd be more interested in picking a particular technique or idea - let's say "smell" - and building a game around it, showcasing computation techniques that relate to smell. You've got visualisation of smells (smell clouds), physical behaviour of smells (diffusion/dispersion, etc), and stuff like the reaction of AI agents to smells.

So, for example, you could write a game where you're a stray dog on the streets of a small town. You can 'see' smells, with different colours representing different things, so you can find food by following a yellow smell, or find other dogs by following red smells, or find humans by following blue smells, etc. With that basic gameplay in place you could build a story on top, like a sort of RPG - interacting with other dogs, maybe doing quests to establish yourself as a leader within the dog society, build your dog street cred and earn respect, etc.

From a technical point of view, you could pick up an existing engine - e.g. Ogre or Irrlict - and add the smell stuff to it, allowing your project's documentation to focus on the interesting stuff that you've written instead of the mundane "this is how I initialised Direct3D" type stuff.

Share this post


Link to post
Share on other sites
Quote:
Original post by Takaloy
Quote:
Original post by Thevenin
PONG; THE MMORPG! NOW WITH 3D GRAPHICS


??


Why not try something unique; something that doesn't require an armada of graphics artists?

Share this post


Link to post
Share on other sites
I've never used another game engine before, is it difficult to implement another's game engine into mine?

I like your idea. I think I'll go with the smell thingy, sounds like a game I'd want to play!

Share this post


Link to post
Share on other sites
Quote:
Original post by Takaloy
I've never used another game engine before, is it difficult to implement another's game engine into mine?


It is as hard as to learn it and make a wrapper for it. So I'd stick with learning it. We worked with crystalspace at school and didn't really like it, mostly because we often found "todos" in the documentation. So from my experience I would say that it's not really hard to learn an engine if you have all the documentation for it. Also, if you use another game engine, look at how big the community looks like and how much feedback it is possible to get from it (look in the forums specific for these engines). That way you make sure that you won't be stopped because you don't know how to use it.

If you make your own engine... the only way to prove that it works would be to have a game for it so you'll end up making the engine and a game. I also would suggest you focus more on some part like particle systems, an audio library or a rendering engine. Using an already made and tested engine will save you time and have more time to work on the game and you won't need to learn a 3D API and all the concepts needed to build a game engine.

Define what you want to make in the beginning and try to stick to this as much as possible this will help you fit in the one year requirement. I suggest you make a concept for a game, a story, design the gameplay, draw sketches of what you what to make and finally make minimized version of that.

JFF

Share this post


Link to post
Share on other sites
thanks for all the advices, where can I find Torgoe / Orge game engine? What game engine would you think will be the best for me? Hopefully not too complicated but sufficient depth for my application (which at this point still deciding).

Share this post


Link to post
Share on other sites
Hi,
Why don't you try something easy that can be networked. Like the original Gauntlet games but each player networked. Add some 3d instead of the original 2d tiles and you are ready.

Luck!
Guimo

Share this post


Link to post
Share on other sites
Guest Anonymous Poster
I do not know how skilled you are with programming, But, I started programming when I was a freshman in college. With only 2 years of Computer Science under my belt there was a lot I was missing to accomplish what you had in mind to do. The internet was just in its early childhood, at least it was not totally in it's infancy, so googling for answers was not very helpful.

My first project I wanted to do was a RTS game similar to StarCraft. But one of the requirements was that I wanted it to be 100% my own code so I did not have to worry about licenses using 3rd party code. I would work on it a little until I found a spot that I did not know how to do, so I started another smaller project that I also wanted to do and go back to the RTS when I gained more knowledge about what I was stuck on. I scheduled the next CS class which was OOP, Wow. OOP is heck of a lot better than the programming style I was using. So I scrapped the project and started over again using OOP. As I scheduled more classes in AI, Networking, a simulation class, etc. in my Junior and Senior years, I was able to try to apply those to my RTS. Since AI was fairly much an intro course, I needed to obtain a little bit more knowledge for that RTS.

So I was stuck once again. So I started another project and another and another, until I found the needed resources to learn from. Now, many years has past and 3D graphics have caught on... So making that RTS project using open GL is the way I am heading now. Because my artistic abilty in graphics stinks, at least with 3D I can kind of get away with it. So I need to learn more about Open GL. I have models that I am rendering on screen and so forth but there are many aspects I need to work on. But now I have other responsibilities and the RTS is taking a back seat most of the time.

I even considered making it a MMORPG a few years back. But I said that is way too much work and I would need a supercomputer for the server to handle all of that and at the time broadband was not catching on and dialup was around 14.4 or 28.8kbps. But I thought wouldn't it be nice to have a MMORPG/RTS where each player was controling a different military vechicle and a party / guild / raid could communicate to each other to perform strikes on different targets or even other parties / guild's / faction's military bases leveling their units and gaining more armor plating, better weapons, etc as they level.

Share this post


Link to post
Share on other sites
I would do something specific instead of the whole 3D game thing. Something innovative that makes our lives easier or more exciting. It could be something simple as designing a gfx algo/data struct or a neat gfx technique.

Share this post


Link to post
Share on other sites
Takaloy if I were you I'd write a kernel or a compiler or something. Something that would show off your skills without neccessarily requiring you to make tons of art. If you want a decent bootloader for a kernel, you can just use Grub. It's not really difficult to make a bootloader though, which is about the only part that needs to be done in ASM. What you do is basically;

1. Load the bootloader into BIOS memory.
2. Make the bootloader call your initialization function in C.
3. Write in C...

Share this post


Link to post
Share on other sites
Sign in to follow this  

  • Advertisement