FIRST Video Game: Hammering out details

[Ryan M.]

Originally Posted by Astronouth7303

Here's how I imagine the final program will be:

• The primary EXE will contain no data about speeific robots, fields, games, etc. it will just contain the framework to run it.
• Every robot will have it's own file(s). This may (or may not) include driver's station, program, textures, and/or bot.
• every game will have it's own file(s). this may (or may not) include the field, the mobile parts, the field control code, and/or textures.
• every field may have it's own file(s). this may (or may not) include the field, the mobile parts, and/or textures.

Hear, hear! Kinda what I said. It will read it in from some sort of file.

This helps to keep everything nice and clean.

Originally Posted by Astronouth7303

This also allows for custom fields; like a street practice, the wood field, a brand-new field, or a well used field.

That's something I hadn't even thought of, fields that aren't "official."

Originally Posted by Astronouth7303

I think there should be seperate utilities for creating/editing these files.

Definetly at the beginning. Maybe there would be an editor built in at a later point, but for now a simple external utilitie will work.

Originally Posted by Astronouth7303

[I personally do not like the 'earn stuff by winning' idea.]

I don't like it much either. One reason would be that it doesn't simulate the real FIRST experience well. And, I don't like having pressure...

[phrontist]

The way I see it you have two options:

1) Slave away, doing everything from scratch, and if you don't kill youself out of frustration produce a crappy game.
2) Use ut2k4!

Seriously, thats a very, very good suggestion. You're not going to top their physics!

Actually, for the ultimate in pimp-age, use Half-Life-2, which is/will be mind boggling. Ragdoll physics for human players

[Astronouth7303]

Or use a open-source engine.

[Heretic121]

i know i have said this before and others have brought it up also... with UnrealTournament 2k4, you get an editor to make maps and script with... the physics is already there, all you have to do is tweak it... everyone who used 3dsMax or Maya for the FIRST animation can make thier teams robot... the only problem with this is making a hook work or ball grabber or ball collectors... this would be coded in, which wouldnt take to much time when you learn it... Fields are easily made and can be configured in many ways... yes the robots would be diffrent files as would the fields... but i dunno how you would implement the human elemnt... its just something to throw around if things become to (something) in making your own proggy and physics... i could make the field/fields (i make maps atm for ut2k4) and show you what it would look like in game if you all want to see how it looks visual wise =D

[Mike Ciance]

Originally Posted by Heretic121

the only problem with this is making a hook work or ball grabber or ball collectors...

this is why we should go with stack attack lol

anyway, i was thinking that, not for the first version, but for later versions, we could have a money system. it would be a continuous story mode. you start off as a poor team with hardly any money, spondored by a local car rapair shop or something like that. you use money register for a campaign (year) and you get the kit of parts. any extra parts will cost extra money. when youy win rounds, you get sponsorship points. the more sponsorship points you have, the more likely you are to get additional sponsors. here's where this idea would really work great: all the sponsor companies would actually be spionsors of the game. they would pay us to put them into the game as advertising. this would do wonders for the project. another possibility for this story mode would be to make a system that randomly generates a game. i would be willing to lay out the framework for this aspect. it would be a huge list of rules and game aspects with a permeation web so that there are hundreds of possible combinations, but so all make sense and work properly.

[Astronouth7303]

Originally Posted by Heretic121

...with UnrealTournament 2k4, you get an editor to make maps and script with...

My mom would freak at the idea of me owning UT.

Originally Posted by Heretic121

but i dunno how you would implement the human elemnt...

Good question. I think that the Game files will be dlls, or zips containing dlls.

Originally Posted by afflictionblade

all the sponsor companies would actually be spionsors of the game. they would pay us to put them into the game as advertising.

great, now we need a group to handle cash, too.

[JVN]

Guys,
Do you think you're getting ahead of yourselves here?

The most important aspect of any videogame is GAMEPLAY.

Why not start with a simple FIRST-like, 2D game. You could either use a 2D approximation of a FIRST game, or an entirely new game. See what you can do. See if you can make it fun to play. I imagine the 1997 game would adapt itself well. Lot's of strategies that could be done simply.

Craft in some decent AI. (This is probably one of the hardest parts, but one of the most important). With decent AI, the game can be challenging to play.

Put in a bunch of different robots. Different styles, different mechanisms, different capabilites.

Throw in some "videogame elements":
Unlockable robots.
Robot upgrades.
Special "boss" robots for the playoffs (the super robot that grabs all 3 goals and walks to the endzone!).

Make it into a 2D RTS type game.
Think... Warcraft with robots...sorta. For every move, there is a counter move. Have people play through several "seasons". Each season you progress until your robot/drivers get beaten. Then you get some experience points depending on how well you do. These experience points let you develop your robot further the next season.

Program in several such 2D games. Allow a "franchise" mode where you play through 14 seasons (with 14 unique games), gaining experience along the way. Program in some real life teams.

There are ENDLESS possibilites for a simple game like this.

Make it multiplayer online.

Think NES era, not Xbox era.


*shrug*
I'd play it.

John

[Astronouth7303]

Originally Posted by JVN

Guys,
Do you think you're getting ahead of yourselves here?

The most important aspect of any videogame is GAMEPLAY.

Why not start with a simple FIRST-like, 2D game. You could either use a 2D approximation of a FIRST game, or an entirely new game. See what you can do. See if you can make it fun to play. I imagine the 1997 game would adapt itself well. Lot's of strategies that could be done simply.

We could, but we'd still need a physics engine. And when we switched to 3D, it'd be a MASSIVE overhaul.

Originally Posted by JVN

Craft in some decent AI. (This is probably one of the hardest parts, but one of the most important). With decent AI, the game can be challenging to play.

version 0.3. (we're working on 0.1 Alpha)

Originally Posted by JVN

... Make it multiplayer online.

Yes, that's on the list, too. v0.5, I believe. (NES did online gameplay?)

I'm also half dreaming that this will double as a simulator.

[JVN]

Originally Posted by Astronouth7303

I'm also half dreaming that this will double as a simulator.

I think that's going to be your major undoing.


Simple, Versatile, Fun.

vs.

Fancy, Complex, Overkill.



I guess it all depends what your goals are, and what you're really trying to create here.

Does the fun of this game come from manipulating the arm of a robot into position, and dropping the 3D poly-mapped ball into the 3D poly-mapped, bump rendered, goal? Or does it come from playing through different strategies against different teams, outthinking and outplaying the opponent?

Or... are you going to try for both....

[Astronouth7303]

Accurate/fast modes?
Dual engines!

[Ryan M.]

I agree that it should be simple at the beginning, but for things like 2D vs. 3D, to be able to switch between them, no matter how modular and separated you are, is going to take a lot of work.

Also, slight side note:
If you would like to get the Crystal Space code from CVS, Astronouth has written a very nice bat file for Win32. Download that here, from my server. Extract this into any directory and double click on the "crystalspace.bat." And new directory will be created under it. Crystal Space is large, be warned!

[mtrawls]

Originally Posted by Astronouth7303

We could, but we'd still need a physics engine. And when we switched to 3D, it'd be a MASSIVE overhaul.

I believe his point was that a 2D game is still challenging ... but quite feasible to the dedicated group of programmers/artists/etc. A 3D game like you envision, however, is a MASSIVE undertaking, overhaul or not. I don't think you realize just how much work and how many professional programmers and man hours are required for something like this. I don't even know, fully. But I did try a mentorship project at my school to develop a 3D simulation, and ultimately I just couldn't complete the thing working 7+ hours a week (with school thrown in on the side, remember), for a semester. Granted, there were various problems and I was working on other things too ... but my point is that this is huge. I've said it and others have said it and others have said others have said it ... but it's worth one last repeating by me. As someone who's made the mistake before, Start out simple.

And before you start thinking about dlls and classes ... come up with a gameline, a storyboard. Write down everything you expect the game to be, keeping in mind to start out simple and add complexity later, if needed. (And it doesn't hurt to remind you now that premature optimization is the root of all evil). Then, once you've nailed the game description, start working on overviews of the code structure. Have you ever worked on Pre/Post conditions, or writing programs from detailed instructions? Before you write the code, come up with those (it's infinitely easier and less agravating to spot crucial errors in structure/heirarchy this way, not to mention less costly to fix than after you've coded it all).

[Ryan M.]

Originally Posted by Astronouth7303

Or use a open-source engine.

Right now, Astronouth and I are thinking about either CrystalSpace or something called Genesis3D. Both of these are open-source engines which might fill our needs.