Game Concept Idea Thread

Game development specific discussions.
Posts: 5492
Joined: Sep 12, 2005 20:06
Location: Ohio, USA

Postby Pritchard » Jan 08, 2008 17:34

Alright guys, because I'm not smart enough to make this game:

Have a game where you can plug in your guitar (real guitar) to the computer and play - Like Guitar Hero but for people with electric guitars and the proper adapter.
Posts: 1759
Joined: May 23, 2007 21:52
Location: Cut Bank, MT

Postby notthecheatr » Jan 08, 2008 18:12

You'd probably need a MIDI port on your computer, which most people don't have except professional musicians. But yeah, that's an interesting idea.
Posts: 5492
Joined: Sep 12, 2005 20:06
Location: Ohio, USA

Postby Pritchard » Jan 08, 2008 23:13

I'm talking about the microphone/input sound port, which is standard...
Posts: 1759
Joined: May 23, 2007 21:52
Location: Cut Bank, MT

Postby notthecheatr » Jan 08, 2008 23:29

Yeah, I guess that would work. I understand it's possible to detect the note from the frequency (I would imagine it uses a FFT, but I don't know anything about FFT), so that might be possible.
Posts: 7808
Joined: May 28, 2005 3:28

Postby D.J.Peters » Jan 09, 2008 0:57

in the time of (8 bit home computers with only 1MHz-4MHz.)
i wrote a simple frequence detection without slow FFT.
the result's are perfect for solo notes but not for polyphone accords.
record in a loop 512 samples 22.025KHz should be ok
now test every position where the amplitude changed from up<->down
or down->up red circles (it's simple only if the sign is changed from - to + or + to -)

save the distance from curent position of sine change to the last
(x axis) in a frq table (distance/sample rate=freq*0.5)
increment a second value in the table by the amplitude (y-axis)

after ~50-70 ms you will see the higest value in the table is the curent frq/2.
all are rounded integers for example
a value near of 220 is a played note of 440Hz "A"

you have not to clear the freq. values for every 512 samples
it's better to fade it out

pseudeo code:
record 512 samples
detect sign changes save it in frq() array
calc curent freq (from higest value)
fade out the old values
for i=0 to 511:frq(i).amplitude*=0.75:next

(i'm with a friend was the developer of DIGI seven the first and only
7 bit ADDA card for CPC64/128 home computers, the trick was to
use the keyborad cable as AD In port and the printer port as DA out
but this is long time ago)

(sorry about my bad english)
Posts: 5492
Joined: Sep 12, 2005 20:06
Location: Ohio, USA

Postby Pritchard » Jan 09, 2008 1:58

That's AWESOME Joshy. Jesus Christ, you've been everywhere man.
Posts: 236
Joined: Feb 23, 2006 6:04

Postby Conexion » Feb 29, 2008 19:55

Pritchard wrote:Alright guys, because I'm not smart enough to make this game:

Have a game where you can plug in your guitar (real guitar) to the computer and play - Like Guitar Hero but for people with electric guitars and the proper adapter.

Posts: 1759
Joined: May 23, 2007 21:52
Location: Cut Bank, MT

Postby notthecheatr » Mar 03, 2008 18:46

One really sweet game concept is that of On the Run. It's like a racing game, but instead of trying to beat your opponents to a finish line you're trying to escape from your opponents (whose primary goal in life is to destroy you by crashing into your vehicle and making it explode).

A clone (whether 2d or 3d) would be great, as it's a fun game and very original.
Posts: 34
Joined: Mar 13, 2008 20:18
Location: Auckland, New Zealand

Game Idea - Technique for huge worlds with tiny program

Postby RayTomes » Mar 15, 2008 0:29

About as many years ago as when people were using BASICA to write programs I started to make a game I called WOW or "World Of Wizards". It was a dungeon exploring game based on a similar thing available on PDP11 computers at the time called something like dungeon. Instead of just one dungeon with 16x16x16 rooms, this one had 1000 of these dungeons per city, 2000 cities per country, 3000 countries per world and 4000 worlds. All the dungeons are different and reproducable, there is nothing random in them. The program size was 11 KB in 1994. Yes, nearly 10^17 rooms in 11 KB and there is another feature that means that in each dungeon you can have a wizard who makes another 10^13 dungeons making 10^30 in total (unless you do the same in those).

What is unique about my game is a technique for making complex features that satisfy almost any desired criterion without having to store or make vast amounts of stuff. The game itself is not perhaps worth doing much with, although someone might want to take it up, converting it from BASICA and adding to it. But I thought that I would describe a bit about the method because that can be used in other games to have the same result of enormous detail with almost no memory usage. There are several elements.

The first is a shuffle algorithm that will take any power of 2 objects (n=2^i) and shuffle them in a controlled way using multiple parameters (in my case the world, country, city and dungeon numbers) to give a new order for each. So you can have things like 1 golden orb (or some special thing) in each dungeon which when a person finds it they know they have done that dungeon. The algorithm only works for powers of 2 objects, so the number of rooms each way has to be a power of 2. In effect, each room in the dungeon ends up with a number which is in the same range as the room numbers - in my case 16x16x16 = 4096 so each room has a result number in the range 0-4095 or 12 bits long.

However it would be boring if every dungeon had the same things just shuffled. So there needs to be a means to make variations on the theme also. That is achieved by using not just the resulting room number but ones from the adjacent rooms when more than 12 bits of information are needed. Because of the shuffling, the adjacent room numbers in any dungeon are different. That means that if the bit pattern for the best thing in the dungeon is all 1s in the room itself, you then go to x+1 room to find the bit pattern there to determine what that thing will be.

The openings between rooms (doors) are dealt with by having 1 bit from the 12 meaning :"door to E" and one for "door to S". So to find door to N or door to W you must look in the appropriate adjacent room. Stairs are more tricky as they need to have walls to be set against, so stairs only happen if both the N and W (from memory) walls do not have doors and the same is true in the room above (fopr stairs up) or below (for stairs down).

Then we use the other bits from the 12 bits to tell about features and creatures and treasures etc in that room. You can have a couple of bits to say 00=empty, 01=feature, 10=creature, 11=creature+treasure. The remaining bits then tell the type of creature (or feature or whatever) and the subtype, colour, size (within its type) etc. If toy need more than 12 bits, no problem, take the 12 bits from the next room in the x direction or whatever. This makes consistency within the one dungeon and difference from all others.

When a creature is killed and its treasure taken a bit is set corresponding to that room. This means that in future visiting that room results in a message about a dead monster and no additional treasure. That also means that only 512 bits are needed to describe the state of the dungeon plus the inventory of the adventurer.

Some of the features are maps and teleports. A map may be of the current level or another level of the same dungeon. Teleports take you to another room in the same dungeon consistently, so once you work out where you are again afterwards they can be a short cut. There is much more, but you get the idea.

My old BASICA program works and has save and restore and a lot of stuff, but I never finished doing all the magic items.

Back to the monsters, you would have genus dragon with subtypes gold, platinum, silver, bronze ... etc and each comes in sizes very large, large, (medium), small and baby. So there are many variations. The game was test only except for maps etc. However I also thought about how to generate creatures using the parameters to make 3D forms and the sizes and colours and variations would all affect the generation of the basic form (making them more stocky or more lanky etc). I even did some real animated flames for the dragon on a different computer back about 1980ish when I first wrote the program.

The shuffled numbers can also be used (or additional nearby ones) to determine the type of material, textures and colours of walls and other features, so that near infinite variety can be generated from almost nothing.

If anyone is interested let me know. I am happy to upload the old BASICA code and explain the shuffle algorithm and its use in generating things. I do believe that it is possible to make a fully animated fully textured and realistic 3D interactive game that is very rich in features with more dungeon rooms than particles in the real universe and still takes up somewhere between 100 KB and 1 MB.

Have fun

PS - I should mention that I am aware that games are now on a different forum. I tried to join that, but as I experience with so many sites that I join the email never came, so I cannot confirm. For some reason my email does not function properly when I join sites. I have no idea why this is, and once a site does not work even 10 reties do not work either. Maybe I need to get a temporary email to join.

PPS I now joined the games forum, so if you want to discuss this, please go there to: ... 36.msg1613
Posts: 131
Joined: Sep 09, 2005 19:17
Location: Temple University, Philly pa

Postby Paragon » Apr 25, 2008 19:22

I have been thinking about a game recently but I don't think I would be able to do it right now.
I was thinking of a game that really amplifies my favorite part of rainbow six (the original). In the original whenever you started a level it would show you a blue print of the building you were going to be raiding, then you could choose start points and team setups for all your allies. After that you could give them a route to follow, and actions to do at diffrent points.. for instance you could create a 2 ally team with a sniper and a spotter, then create two three ally teams with standard assalt equipment. You'd be able to tell the two assault teams to get into position and wait while the sniper team took out some of the sentreys and such then give a go code. Then the two assult teams would simultaniously blast through the front and back door while your sniper team repositioned them selves. In rainbow six you could take control of any of the allies you wanted (kind of like possessing them.)
I was envisioning a game where the planing of the attack was the main draw point. Each ally would have a helmet mounted video camera and you would be able to sit back at the command center giving orders to the various teams you sent into the fray.
Of course I'd throw a few monkey wrenches into the equation, such as the blue prints being wrong or the terrorists remodling the building since the last intelligence report. Terrorists attacking your base so instead of planing an attack you'd have to plan a defense. Maybe even do some missions were you had control of a few tanks and choppers instead of just infantry... maybe give the ally AI the ability to request that you drop mortars where they ask you to and such.

Basicly just sit there and be an armchair general but not in the way that RTS's currently do it.
Posts: 96
Joined: Nov 07, 2006 12:42

Like the Sims

Postby wolfstar » May 18, 2008 11:53

Here's an idea that's well beyond me but I think might be popular. An environment creator that allows me to create an environment to interact in or upload to show off.

Say I wanted a bedroom with a supermodel in my bed. Well then I'd dimension a room, type "Bed" in, and the program would trawl the internet for pictures and dimensions of beds. I'd choose a 'love heart' bed that has enough pics of it and dimensions to be able to render, and place it somewhere in room. Then likewise for the model, maybe it could scan videos for mannerisms. Decorate the room and upload it somewhere to show it off. Maybe you could interact in it yourself if you got some video of yourself. I know the ppl in it wouldn't move realisiticly but it's something.
Posts: 11
Joined: May 28, 2008 1:08


Postby j8h9 » May 28, 2008 19:36

Count me in on the this software. SuperModel... outstanding
Posts: 1148
Joined: May 08, 2006 21:58
Location: Crewe, England

Postby jevans4949 » May 29, 2008 0:59

I'll be right down on the holodeck, Mr Worff-star!
Posts: 109
Joined: Dec 06, 2007 16:43
Location: Cokato, MN

Postby SamLemke » Dec 21, 2008 3:05

im working on a game right now but ill give you my ideas i dont care if you know them because it can help me to decide from your replys

heres my game ideas so far
massive multiplayer online (think runescape)
diablo style fighting click to hit and many monsters at once

im looking to make the game so there will randemly apear holes in the ground and when the hole apears gobblins rush out and start attacking the nearby village ( easy gobblins for the lowlevel areas and hard for the higher level areas) there will be "safe" areas so your not afk and die

a level up system that is more bassed on how many kills with a certin weapon.

you might be in the noob area and have the aztec sword now there might be 10 levels to this sword(also aztec armor peices) as you make kills your sword tracks somany kills and the sword gets stronger so you cold give it to some one and if they can handle the powerfull sword of 4000 kills then they will be more powerful.

if they strat from bottem up they clear the noob cave there is a drop from chest or boss and they will get a set peice (arrmor or wepon of the set area) that is one level stronger then there current peice so they upgrade.
the more kills that weapon makes the more powerfull it gets say aztec sword with 100 kills gets a +1 bounus 200 kills +2 and so on..

the sets(weapons/armor) ave a max level say 1 to 10 after that its time to move on to a harder area.

also if you wonder out into the "wild" it will progrsivly get harder and one might come upon bad guy town that are only possable to attack in groups.

also cylner shaped world or two ways to one location. to sail the ocean or to risk the wild walk across the lands

also crafting abilites speicalty combonations of crafting.

if a weapon breaks say your bloody sword of 5000 kills you can use the metel from the broken sword and combine it with other broken weapons/armor and have even better weapons. the idea behind this is if you loose you can still gain later so you may want to save your broken metels and such.

also charter classes basic humanoids but say if you want to have a paldin first you need to level up a warrior class and a magic class to a certin level

now this is an idea so far and im still loking for better/funner ideas!! so feel free to shoot me ideas. im almost to the point i can start making game content but also only so far that i just learnd picking :P so im saying maybe 6 months down the road i maybe getting closer and once in beta ill post here first for beta testers. but 1st things first!
Posts: 1253
Joined: Nov 06, 2005 5:19

Postby stylin » Dec 21, 2008 4:08

I really like the idea of "experience points" for weapons and armor, and the combining of different metals to make new weapons. The PSX game Vagrant Story had something similar, where the effectiveness of certain types of weapons/armor would vary depending on the enemy class you're fighting at the time, and of course different metals could be forged together to create new metals. It added a sense of uniqueness to the gameplay, though in reality the metal combinations were entirely deterministic (by a simple table).

Do you have a storyline at all ? These "holes" are intriguing; who or what, if anything, is controlling these monsters, and why are they attacking nearby villages ? Is there some central entity or entities that you are up against, perhaps demons, or a Dark Lord ?

I'd like to hear more about the character classes. You mention "leveling up" a warrior and magic class to achieve a paladin-like result. So, a player can focus on one, many or all of the different types of fighting styles if s/he chooses, not being limited to one particular class of character which is chosen, say, at the beginning of the game. Is this accurate ?

What types of rewards will players receive when completing a "quest" -- will there be "quests" at all ? You mentioned working as a team.. how will experience and "leveling up" factor in to the varied amount of players that may undertake a "bad guy town" ?

An MMO will be a mighty challenge. Have you done much research on networking so many players together ? If you're almost to the point of creating the [multitude] of assets for the game, I presume you have at least a rough outline of the game engine ?

Return to “Game Dev”

Who is online

Users browsing this forum: mrToad and 3 guests