Well I finally decided what my new game will be all about - many thanks to all the people who helped with ideas!

Synopses:
You play a rookie cop who is trying his best to move up the ranks in law enforcement world. You start out as a Patrol Officer, cursing the streets and arresting speeders. As you complete missions, your rank goes up. Once you have enough rank you can become a S.W.A.T Officer, missions get harder and more interesting. As a SWAT officer you may be in charge of escorting and protecting important people, working as backup for other police officers etc. After a certain amount of missions as Swat, you may be eligible to become an Air Unit - this will allow you to pilot a police helicopter in order to bring down the bad guys (how you ask? I'll keep that a secret :)) Finally, once you acquire enough points, you will become a Commanding Officer - at this stage you are in charge. You get special missions, at your disposal are patrol officers, SWAT teams, and Air units, any of which you may call in to assist you.

This brings me to my poll question. Do you think it would be better to make the setting in Country roads such as in NFS series or City roads such as in MidtownMadness and Driver?

Pros of Country Roads - Less objects required ie: Traffic Lights, Pedestrians, Heavy Traffic.

Cons- harder to model, as they are twisty and bumpy.

Pros of City Roads - Easier to model, not many curves and bumps.

Cons - Much more work developing traffic lights, pedestrians and making the city come to life.

(Do you really need a poll's statistical breakdown for this?)

Country roads. To stay away (in my mind at least) from anything making one think of GTA. Off the gritty streets and the standard drug pushers, and into fields with weird country things along the roadside... (Crudely hand-painted signs advertising things for sale being the most common, of course. "Fresh Vegitables $1 Each" (spelling intentional)).

Since your "cracking down on crime" I think a city is the logical answer...

"(Do you really need a poll's statistical breakdown for this?)"

It's for the lazy people who prefer to just click on a radio button rather than typing a paragraph ;)

papillion has a good point about staying away from a gta look, but on the other hand, a city could be cool. If you do make a city, please please take into a count that real life cars cannot knock over all the lamp and signs posts they want and still keep moving. This really bugged me about GTA3 since it made the whole world feel like it was made from plastic blocks. If a car rams a traffic light with enough strength to knock it over, it should be out of commission or close to it. Country roads would make for interesting chases and can still be interesting if you have some signs of civilization. They are probably also much easier to do in the long run, despite their winding nature. I expect to be able to drive through barns and find a secret tractor driving mission ;)

Just wanted to say that "What do you prefer?" is a _very_ _bad_ topic which doesn't say anything about the contents of the thread.

"I expect to be able to drive through barns and find a secret tractor driving mission"

lol, alright Lizardsoft, just for you I'll do that.

"Just wanted to say that "What do you prefer?" is a _very_ _bad_ topic which doesn't say anything about the contents of the thread."

Ah yeh, sorry about that. I had a different point in mind when I started typing, but it changed and I forgot to change the topic.

Well personally I'd say city streets as they hold far more interest than the occasional 'Fresh vegitables $1 each' signs.

Also, given your game outline, I cant really see how you would justify most of it in a purely rural environment. To me, it would only seem semi-plausible and would make me think you had deliberately opted away from a cityscape because other than the winding roads etc it seems far easier to create.

-Z-

I dunno, as someone who grew up watching the Dukes of Hazzard I never thought driving Hazzard county looked all that boring. In the end it depends entirely on your design and execution.

Glad you were able to draw some inspiration from the replies in your earlier thread. I think both terrain types have their advantages, but I also think your game engines capabilities may also be a big factor, as you'll want the least number of polys for a nice frame rate. Maybe a rural setting would be cool with some small towns or gas stops for buildings. It would be fun having long high-speed highway or back road chases where you have to pull over a car before it gets out of your jurisdiction.

Then again, the chopper chases would be extra cool with buildings in the way. It reminds me of the old arcade game Thunder Blade where you could fly inbetween city buildings. In general, you were allowed to fly higher, but you didn't get to shoot as much stuff. It was more fun flying around the buildings anyway and was really challenging too. Sorry. I guess I prefer both. Hmmm... now that I think of it, I always loved racing games where you could pass through a city or town. It could be looming in the background if you drive far enough in the highway mode, but you only get to see it if you increase your rank to a city cop and actually patrol there. Hmmm... or your rural jurisdiction could increase to samll towns that you hadn't seen before either. -anyway, good luck. I would certainly download a demo of this no matter what the graphics look like. I just don't know of any games where you play the role of a speed patrol officer. How can you beat that?

Hmm, looks like the poll is leaning towards City roads, although a couple people bring up good ideas. However for development reasons I am going to do rural roads. Obviously I will have some urban settings, it won't be just plain roads - hopefully this way I will be able to satisfy both types of customers.

"I would certainly download a demo of this no matter what the graphics look like. I just don't know of any games where you play the role of a speed patrol officer. How can you beat that?"

Hey thanks Hydroaxe, you inspire me :)

There is one game I believe that is based around patrolling and it called Highway Patrol 2. However its very old, I couldn't even get it to run on my new machine - MS should have made a better DOS emulator on Win2k and XP :(

Originally posted by Mman
Hmm, looks like the poll is leaning towards City roads, although a couple people bring up good ideas. However for development reasons I am going to do rural roads. Obviously I will have some urban settings, it won't be just plain roads - hopefully this way I will be able to satisfy both types of customers.

"I would certainly download a demo of this no matter what the graphics look like. I just don't know of any games where you play the role of a speed patrol officer. How can you beat that?"

Hey thanks Hydroaxe, you inspire me :)

There is one game I believe that is based around patrolling and it called Highway Patrol 2. However its very old, I couldn't even get it to run on my new machine - MS should have made a better DOS emulator on Win2k and XP :(

You're welcome. :) I'm sure you're you'll be able to do many things that just couldn't be done the last time there was a game concept like this. I searched for Highway Patrol 2 and that was done in '89. There was also Chase HQ and its sequel. I used to love playing that in the arcade. It's very old too. There's also ABP, but all these games are so old that the concept would seem quite new again. Sure, games with cars in them and plain racing games have advanced a ton, you can advance in a direction that they are not going. Just hope that EA doesn't release a patrol game of their own. ;) (-don't know why they haven't done that already.)

Originally posted by Mman
Pros of City Roads - Easier to model, not many curves and bumps.

ROTFL, in Europe we have lots of historical town centers with curves and small roads.

For instance take a look at Heidelberg: http://www.stadtplandienst.de/map.asp?sid=cd09d5d233027076e5bc608b6f2dab8c&ix=21&iy=15&grid=heidelberggrid1


Cons - Much more work developing traffic lights, pedestrians and making the city come to life.

This is a non-issue imho you'll have to develop "much stuff" anyway to have some elements in your game.

I'd say country roads. Because city would be too complicated since there are a lot of features such as the air unit. Country roads is good enough for fulfilling the game without any risk of game play failure. City would be a challenge for a developer I think.

ergas

Originally posted by ergas
I'd say country roads. Because city would be too complicated since there are a lot of features such as the air unit. Country roads is good enough for fulfilling the game without any risk of game play failure. City would be a challenge for a developer I think.

ergas

I agree with Ergas. I think it could be easier to create good looking graphics for the country. (Especially for wide open spaces like in America). (But I would stay away from doing forests :) ... but its still a lot of work for an indie developer to create all the graphics for a modern driving game (at least one you hope to sell?).

Still an indie company can do a good game. Check out Small Rockets driving game. (They have a number of great games) :)
http://www.smallrockets.com/pc/chix76/index.htm

Glad to hear the tractor will make it in. I've been playing GTA3 today and am convinced that the city is just too hard. The large amount of textures, complicated terrain (any good city will have some bridges for example), pedestrians, street lights, etc. As if the framerate weren't stressed enough with all the minor detail and buildings, a city is expected to have many cars on the screen at once. Maybe I'm underestimating you, but I don't think an indie developer can pull off a city without an unprofitable amount of work and high system requirements. GTA 3 allows for movie-like chase scenes with burning cars rolling down hills and ambulances riding off ramps. This is obviously more than is expected, but if you're game is based around chases, not being in a city will help prevent the player from wandering why he can't pull off a particular maneuover that your engine doesn't allow for. Of course if you can clone the GTA world flexiblity, all the more power to you.

Trust me, you are not underestimating me lol. Like many people agree, I'm going to go with country, I'm not trying to make the next GTA and I certainly don't want to compete with GTA and the likes. I'm just trying to make my first indie game and hopefully sell it.

Mman, are you sure that starting with a racing or an automobile game is a good choice for the first go? I find this a little risky for an indie.

ergas

Originally posted by ergas
Mman, are you sure that starting with a racing or an automobile game is a good choice for the first go? I find this a little risky for an indie.

ergas

Well, I couldn't think of any better ideas.

I'm having problems with detecting collision between the car and the terrain. I got the car to go up and down sort of, but than I still need to figure out a way to calculate the angle of the car when its going down or up a hill. I'm thinking maybe making a City will be longer, but at least the roads are mostly flat so the coding will be much easier in the long run - I should have thought about that first, it just sliped my mind.

Hi-
IMHO you are going the wrong way about this project, and In the little expirience i had on starting projects (specially larger projects then a one screen puzzle game) you need to do a detail design document before starting to do the game- Making a game is a much larger issue then formulating the basic idea and then jumping into code writting and from what Ive read in this thread you are walking into a sack of problems-

my advice would be to take a lot more time and effort in constructing and formulating the idea, design and game with a pen and paper- In the long run it will save you a lot of headaches and ultimatly your project will be finished in less time- i guarantee

:)
there.

thatd own dude looking for dev (C++)

Sorry to be harsh, but if you can't even handle proper car-terrain collisions (the easiest kind) I would recommend you to find another project than a full 3D racing game. You don't seem to have any idea of the programming hell you are getting yourself into :) 3D racing games are not in any way easy to make.

Guess what... I've started a 3D racing game (à la Star Wars : Pod Racer), and then decided to finish a small board game instead :-) (a connect-4-like, not the most complicated thing around). And I've been working on the board game for many months... (even though I'm reusing many things coded for the racing game)

For the angle of the car on a hill, I suppose that you don't really need it.. 2 cross products will give you the "down" vector that follows the hill, for example. Oh well, I don't know what you need exactly...

Originally posted by Mman
Well, I couldn't think of any better ideas.

I'm having problems with detecting collision between the car and the terrain. I got the car to go up and down sort of, but than I still need to figure out a way to calculate the angle of the car when its going down or up a hill. I'm thinking maybe making a City will be longer, but at least the roads are mostly flat so the coding will be much easier in the long run - I should have thought about that first, it just sliped my mind.

Hi Mman, you are under risk! If you have any problem with detecting collisions, I (as a physical simulation developer) would advise you to use simple roads without any hills. There is a lot of work to do on automobile simulation even if it is an arcade one. But preferably use arcade, not physics at the beginning.

Good luck

ergas

Originally posted by ergas
Hi Mman, you are under risk! If you have any problem with detecting collisions, I (as a physical simulation developer) would advise you to use simple roads without any hills. There is a lot of work to do on automobile simulation even if it is an arcade one. But preferably use arcade, not physics at the beginning.

Good luck

ergas

Hi, yes I am using arcade physics; it’s just the collision that is tricky. I don't know why people say this is the easiest part, for me detecting collision between a hilly terrain is the hardest part. I am certain everything else is not that bad. I already have arcade physics, proper engine sounds (acceleration, deceleration, etc.) All the graphics effects such as billboards, particle system, etc. Although I might have to take away the 2d trees and replace them with 3D models because flying in a helicopter will make the 2d trees look fake. I'll see if I can come up with anything, If I can't I guess I'll use flat roads and just add a few ramps for jumps. In my opinion the game will have enough features to make up for the flat roads.

All you need to orient the car is a couple of cross products between the ground normal and the car direction to build a new orientation matrix. If you can't handle this, read up on linear algebra.

Originally posted by Mman
Hi, yes I am using arcade physics; it’s just the collision that is tricky. I don't know why people say this is the easiest part, for me detecting collision between a hilly terrain is the hardest part. I
I'd hate to patronize you, but the thing is that some of us is able to take a guess at your skill level from the kind of problems that are giving you trouble (we've all been there), and our experience is that you're headed for a hard-coded hell of special cases that are never going to work just right.

You need a general collision response system (it doesn't need to be realistic, just believable) to handle car physics, because when you figure out how to make the car follow the hills you'll probably want it to be able to lift off from the ground when going over smaller hills/bumps at high speed - so you add a special case just for that. Then you'll want the controls (brakes/throttle) not to work when you're airborne, so you add another special case. Then you want the handling to behave a bit differently on impact, so you add another, etc, etc. The number of different permutations of special cases is going to raise exponentially, and they probably won't look that good either - I don't think that is a good way to handle the problem.

I apologize if I wastly misinterpreted the way you're going about the problem - I'm just trying to help out and possibly save you months of work.

Originally posted by Mman
Hi, yes I am using arcade physics; it’s just the collision that is tricky. I don't know why people say this is the easiest part, for me detecting collision between a hilly terrain is the hardest part. I am certain everything else is not that bad. I already have arcade physics, proper engine sounds (acceleration, deceleration, etc.) All the graphics effects such as billboards, particle system, etc. Although I might have to take away the 2d trees and replace them with 3D models because flying in a helicopter will make the 2d trees look fake. I'll see if I can come up with anything, If I can't I guess I'll use flat roads and just add a few ramps for jumps. In my opinion the game will have enough features to make up for the flat roads.

Not only detecting is hard. The hardest part starts when implementing the forces or velocity settings on the car. In physical simulation forces are used and in arcade both forces and velocity settings are used. Its best to start by your own methods, then if they don't work, study math and physics. You'll do it.;)

ergas

Originally posted by mkovacic
You need a general collision response system (it doesn't need to be realistic, just believable) to handle car physics, because when you figure out how to make the car follow the hills you'll probably want it to be able to lift off from the ground when going over smaller hills/bumps at high speed - so you add a special case just for that. Then you'll want the controls (brakes/throttle) not to work when you're airborne, so you add another special case. Then you want the handling to behave a bit differently on impact, so you add another, etc, etc. The number of different permutations of special cases is going to raise exponentially, and they probably won't look that good either - I don't think that is a good way to handle the problem.


That is the principle approach that simulation design should do. Diverse and multiple cases are hell. mkovacic is right! It takes some time to understand the concept though. Writing more and more programs will eventually bring a developer to the understanding of mkovacic states. Math, second order differential equations and physics is important here.

ergas