Skip to Content

A weird CCG idea

25 replies [Last post]
hotsoup's picture
Joined: 08/28/2009

Here's something that occurred to me...

Would it be possible to create a CCG in which every card was unique? As in each card is only printed once, ever. No one else who plays the game will ever have a card identical to any of yours.

I'm not planning to make such a game myself, but it's been interesting to think about. Having different art on all the cards would be impossible, it seems, unless it was procedurally generated, somehow. In the same way, it would be impossible to write different text for all the cards, so maybe you could have a number of possible numerical values on each card, plus an assortment of pre-designed powers. There would have to be millions of possible combinations, though.

Could this be fun? It would make opening new packs awesome.

questccg's picture
Joined: 04/16/2011
Other fun ideas

I had the idea of having each card with a *scratchy*... As in some statistic remains unknown until you "use" the card in a game... I know my neighborhood printer has already told me it would be possible to do. I just inquired because I wanted to know if I could design my own scratch tickets.

Anyhow this is *in the spirit* of your one card for each player, you could use the card once without knowing the value of the card (it could go either way), next time you play you know what the value is. The first time, it adds an element of surprise.

starflier's picture
Joined: 01/09/2009
Neat idea

That sounds like a brilliant use for something like this:

Use a bit of scripting similar to this to generate the card text:

This is doable. Colour laser printer and a digital die cutter and you've got an infinite card game.

Joined: 08/13/2012
This is intresting

I just thought of how fun (and hard) it would be to program something to generate art for something like this.

A CCG like this would be very fun. Opening packs would be epic.
It would be practically impossible to sell cards online, share decks with other people etc.

I think it should cycle around so that there are a few more than one of a certain card in existence. For example if there are 1million cards and you print 5million cards there would be approximately 5 of each card in existence.

JustActCasual's picture
Joined: 11/20/2012

You could make a fairly basic property based algorithm for the art: so if toughness > power than it would look for hard things (wall or turtle) and then combine that with another trait (maybe it has evasion, so "sneaky" or "Ninja"). Then you have a bot search a selection of free art sites for the keywords and/or combine them into one image. So you end up with "sneaky wall" or "ninja turtle" as a card with appropriate art, and it fits the mechanic-ish. It worked for yu-gi-oh.

Another way would be to have different filters applied to stock art. So the bot would look at the 6 different dials for the card and set filters to appropriate level.

You could have 10 different line art bases, 10 stops for each of RGB for background, 10 stops for a spectrum from curvy/bold to spiky/thin to apply to the quality of the lines, and 10 different brush styles. You have enough art to supply a million cards, and it works to communicate the underlying mechanics.

hotsoup's picture
Joined: 08/28/2009
Space Invaders!

Man that space invader thing is cool. What's so awesome is that you tend to read a shape or character into the random pixelation, and they all could really be used as minimalism game art. Perhaps the cards would only have those shapes on them, and the gameplay would revolve around the way that different colors interacted...

starflier's picture
Joined: 01/09/2009
I hacked together a few cards

I hacked together a few cards

The names come from: and the numbers on the cards are random from O to 64, from:

No idea about Mechanics yet, but it might be best to do this like Icehouse, where there's no "official rules")

Joined: 12/25/2012
RE: A weird CCG idea

Hey Hotsoup

You have a great idea and I love the concept of each card being unique. I am a little stumped, however, about how this would be worked out on a large scale. Runedrake may have a good point about possibly having a few multiples so that there would be more cards in existence. I think this is key in that there are only so many combinations of affects, abilities, and overall differences that can exist in a "playable and fun" game, although there would be quite a bit. Let me explain:

You can only have so many mechanics and card features that go into a game. To clarify this, look at MTG. There are multiple types of cards: instants, artifacts, enchantments, creatures, etc. Let's say each type of card has their own unique mechanics. Now each mechanic can be made different from some numerical and/or categorical basis (such as damage dealt, multiple targets vs single targets vs global effect, use multiple times vs single use, etc etc etc).

To keep the game fair and playable, there would need to be some limit to each and every mechanic (some will be self-proclaimed, such as you cannot have 18 targets in a 4 player game). Using MTG as another example, look at creatures. While there are creatures that having enormously high Power/Toughness, the number of them that exist shrink exponentially as P/T's increase. If it didn't work this way, the game would not be fun as there'd be creatures with P/T absurdly high. The same works in the other direction, P/T's don't really go into fractions otherwise they are basically useless, within the context of the game. Limits to the cards and mechanics come natural to the game; it's part of what makes the game work, and fun.

Now back to your concept of uniquely individual cards. It would be pretty difficult to design a game in which you can create one million unique cards that are also playable within the game mechanics, but I'm not saying it can't be done. You can create more viable cards by having more card types and card type mechanics. The difficulty with this is, the more card types and card type mechanics you have, the more rules there are, the more difficult the game becomes to learn, and the less likely people will be apt to pick it up to play. This can be solved, however, by starting simple and expanding the game later. I hate to go back to the same reference, but look at Magic. It started out incredibly simple, got a huge following, and then expanded into new rules, card types, and mechanics. This was necessary to keep the game exciting and new for players, while also allowing the production of new card ideas. But MTG only has 11,441 different cards (not including basic lands), and it's been out for a very long time.

I'm not trying to deter you from your idea, as I find idiosyncrasy and character admirable, but am trying to do quite the opposite. Games like MTG have very simple and big limitations, whereas designing a new game is literally limitless. I am merely trying to put into perspective the task at hand and want you to question how much you want to take on such a big undertaking.

Anyways, good luck!

NASG's picture
Joined: 10/15/2012
Possible mechanic

From the example cards you've made, you could use the interaction of numbers on matching colours, or corners for example - the number has to be greater on the card you place than the card you place against, or less than, or one greater, one less, both even/odd. Perhaps you could score differently depending on how many of these criteria are met.

I think the problems arrive with publishing the cards. You'd be asking a printer to print ~100,000 different PDFs each with 12 of the random cards on. You could probably get some software made to create the PDFs along with printers marks etc.. but proof-reading would be impossible (arguably it's unnecessary anyway), and they might try charging you for 100,000 single orders, i.e. no bulk discounts. The printing would have to be digital, which might limit the quality.

The whole idea of completely unique cards/decks is cool though, and why not use the numbers on the card to create some kind of random pattern or as seeds into a fractal generator for the artwork on the reverse side?

Joined: 07/08/2011
You could meet half way

What if you had Characters who could be "upgraded" in some way, and each character interacts differently with each card you play? Thus, the same deck of cards used with two different characters would be a totally different deck.

Joined: 08/13/2012
3d scene generation

I got an idea for how you could create tons of art for so many cards.

- Find/make a ton of 3d models.
- Generate a random theme, setting, mood etc. for the art.
- Make a program to smartly place the 3d models in a scene based on the random theme, setting, mood etc.
- Now maybe render it with some really interesting filters and maybe program the computer to sketch a few more elements onto it.

- Create the text, name, and stats for the card based on the theme, setting, mood etc.

There should be a set released every year (100,000 to 1,000,000 different cards in each). Each set should be totally different than each other set (100 to 1000 random special mechanics for the set), Each set should have a teensy bit of story behind it to theme some of the cards on...

Something like this would easily make it into world record books.
The thing is... Once something like this is made, there is practically no work needed to make new sets, cards etc. because practically everything is randomly generated.

NASG's picture
Joined: 10/15/2012
2D artwork & game idea

Sticking with the four numbers, one in each corner:

Each number can be either Red, Green or Blue.

Each number is between 1 and 10 inclusive. This gives 30^4 possible combinations of numbers (apologies if my maths is wrong!), which is 810,000 cards.

The artwork in the centre of the card is in three colour, and is based on circles of a radius from 1 to 10, scaled to fit a standard playing card size. Where Red and Green converge you get an arc of Yellow, Green and Blue gives Cyan etc..

The mechanics are then based solely around the mathematical nature of the numbers in the corners and their colour, and perhaps a set of dice.
Colour D6: 1-2, colours must match, 3-4 colours must differ, etfc.
Number D6: 1-2 Greater than, 3-4 less than, 5 - is a factor, 6 - equal.

You then shuffle the cards and deal out X to as many players as are playing, turn one card over to start and lay it down, then take it in turns to roll the dice and try and match a card from your hand against what is on the table.

Maybe you have to match 1 corner, maybe match 2 corners, this could reflect in how you score.

Would be a great Maths tool in a school setting, where you could alter the rules depending on what you were teaching?

Fhizban's picture
Joined: 01/11/2009
this somehow reminds me of

this somehow reminds me of the good old "barcode battler". a handheld console that was able to scan any kind of barcode and generate a warrior or upgrade from it. then you let your barcodes..errr, warriors battle against each other. we where crazy finding the most powerful barcode!

the random card generation sounds really good. that could be expanded by creating not only attribute ratings, but also random mechanics for the cards as well. at least to some degree.

is see a problem with overpowered cards, generating them randomly will result in many useless cards as well.

and: the formula to generate them would have to be perfect, otherwise you start and *woosh* produce a million cards that all suck or are so imbalanced nobody plays them.

on the other hand, if one of you invents a formula which guarantees that only 100% tournament legal, powerful and "rare" cards are generated...well...this would be heaven *endless booster treasure hunting*

larienna's picture
Joined: 07/28/2008
Another way to achieve this

Another way to achieve this is to make a card game where players can design their own cards. In that case, all cards will somewhat be unique.

One of the problem is the learning curve. It slow down the gameplay because players are constantly reading cards to know what they do.

NASG's picture
Joined: 10/15/2012
The more I think about this

The more I think about this idea, the more I like it. Well enough to mock up what a few of my corner numbered, random artwork cards might look like.

The corner in each number is used as the dimension of a square of the appropriate colour, e.g. a blue 4 gives a 4x4 blue square.

The shapes are 50% opaque and arranged top left number at the bottom, then top right, then bottom left, then bottom right on top.

Perhaps the mechanics for the game could be similar to FLUX, in that there are say 3 mechanic cards in play at once. The mechanics relate to laying a card (in a similar way to dominoes) using number, colour or both, e.g. both colours match, colours are different, one number must be equal, both numbers must be greater, numbers and colours must not match.. etc..

starflier's picture
Joined: 01/09/2009
My thoughts on balancing the

My thoughts on balancing the cards was to make each number have pros and cons. Like, in a combat-centred game, one number represented "order" used for constructing more units, but when fighting, the card with the lower "order" would win.

With that in mind, I wrote up a set of rules yesterday for a set of cards with three statistics: Autonomy (1-16), Order (1-32) and Peace (1-64)

* Both You and your opponent shuffle your decks and draw four cards from the top, putting them face-up in front of you

* On your turn, first choose one card to sacrifice. It is put into the discard pile.
* You may then activate any number of cards on the table whose total Autonomy is less than the sacrificed card, and perform one action.

* Breed: Draw a number of cards from your deck equal to the total Autonomy of all activated cards
* Construct: Play a number of cards from your hand, onto the table, whose total Peace is less than the total Order of your activated cards
* Train: Put your activated cards under any of your other cards whose Peace is lower. Subtract the lower card's Order from the higher card's
* Attack: Put your activated cards onto any of your opponent's cards whose Order is higher. Your opponent must discard these cards

* Peaceful Victory: You have 500+ Peace on the table
* Violent Victory: Your opponent has no cards left on the table

Fhizban's picture
Joined: 01/11/2009
@NASG: Really like the FLUX

@NASG: Really like the FLUX mechanic idea! I had a similar idea a while back. It was concipied as a hacker card game and the players actually shaped up the mechanics of the computer system by themselves.

What i dont understand: do you want the squares to replace the art? I really liked the "random invaders" idea - the produced "creatures" look nice.

NASG's picture
Joined: 10/15/2012
Invaders would be cool

It was just to illustrate another way of unique artwork, based on the cards value. Nothing to stop Invaders being used :-)

You could also give each a name based on the four, coloured numbers.

Fhizban's picture
Joined: 01/11/2009
can't get the idea out of my

can't get the idea out of my head anymore!

maybe this would be best as a online game, as printing random cards would be very costly/risky.

and: I would not recommend to tie the names/artworks to the mechanics.

because if we do so, weird stuff could happen: like all invaders that have a red pixel on their upper left will feature special ability X. lets say this special ability X is very good: so players will start searching for all invader cards that have a red pixel in the upper left of the artwork...hmmm.....

EDIT a few more thoughts:

* the main point for me, for a game like this to be fun, would be that opening a booster pack will get you cards NOBODY ELSE has ever seen/played before and that the chance to draw a card a second time is nearly 0 (but not 0).

* I would model this ontop of a battle focused CCG system. I could really imagine this using the same system as Duelmasters (sounds cheesy, but maybe Larienna understands why - Duelmaster is just simple and plain fun)

* i think we wont be able to make all mechanics/special abilities random too. this would break the game and generate tons of nonsense. maybe we have to agree to a "core base" of abilities (like flying / first strike). and develop, say 100 of them. thats a lot! then the cards would feature combinations of abilities and numbers. (like attack 2, defense 3, speed 1, life 4 and first strike and flying).

just brainstorming here!

Fhizban's picture
Joined: 01/11/2009
sorry, double post. After

sorry, double post.

After work i put together a quick design document that explains how i could imagine a random card game.

please note that this represents my personal interpretation of what we have discussed so far

you find it here:

hotsoup's picture
Joined: 08/28/2009
Hey, great work, Fhizban! I'd

Hey, great work, Fhizban! I'd love to see how you develop this idea!

It also occured to me that there could be a way to randomly generate art that's more interesting than space invaders. It would take a lot of work, but it's at least in theory possible. First you would have to have a random terrain generator. I think these do exist somewhere, but I'm not sure where. Some games use things like this already, such as Minecraft, or the kickstarter game in development, Sir, You Are Being Hunted, which generates a random massive version of the English countryside to play in every time you start the game.

Create an algorithm to generate random CGI characters. Come up with lots of different articles of clothing and weapons for them to be holding, along with lots of random values for their height, expressions, proportions, etc.

Next, make a program that drops them somewhere random in the procedurally generated world, and then takes a screen shot of them from one of 100 different angles that you've determined look good. Add a cell-shaded effect if you want.

Doing something like this might be outside the abilities of a single person, but a company dedicated to making a card game like this could probably do it. Heck, there might be a way to use Minecraft to do this now.

Fhizban's picture
Joined: 01/11/2009
glad you like my

glad you like my approach.

yeah i was thinking about better art as well. but do to a proof-of-concept fun game, the invaders should be enough.

on the other hand, fractals are used quite often in computer games nowadays. they can be used for terrain generation. this could be developed into a more complex system. maybe we have to move away from civilised beings on 2 legs a bit. to a approach like seen in the game "spore". imagine fractalized mutant-zombie-monsters roaming a fractalized random world !

PS: just updated my journal with thoughts about infinite sets in a random CCG enviroment.

PPS: first output from my random invader algorithm (there is still a lot of manual work necessary)

Im just playing here okay? :-)

hotsoup's picture
Joined: 08/28/2009
Man, the Spore Creature

Man, the Spore Creature Creator would be perfect for making art, if only it had a randomizer...

pelle's picture
Joined: 08/11/2008
similar procedurally generated game idea

Had this wonderful idea this morning that was too good to be unique. I have it all sorted out in my head how it could work, but now that I googled to see what similar games already exist I found this thread. Of course someone here was on it already. :)

Will hopefully find a few free hours to make a proof-of-concept implementation this weekend. The idea is that players download a free application that can be used to randomly generate army packs (PDF countersheets for single-sided print'n'play). To find a good army you have to generate hundreds or thousands looking for one you like. Sort of like harvesting for bitcoins I guess. The random seed used to generate an army is printed on its countersheet, so any other player can confirm that an army is valid by generating it and compare, so no one could just make up their own army with any values they want. That way you can have a fair game of unlimited expansion packs without a central big company pushing out expensive booster packs and still if you arrange a competition you can make sure that all armies are valid.

I think collectible army packs would work better than cards, because in a card-game players expect more varied illustrations and card titles. Army units are ok with some slightly varied graphics (I'm probably trying simple 2D robots first) and different values.

Have some more specific ideas how to make it work, but I'm saving some for later (and its own thread).

X3M's picture
Joined: 10/28/2013
Hey, the game Warzone2100 has

Hey, the game Warzone2100 has over 3000 different units by designing them.
They also have over 3000 different [names (based on the parts)], [looks, again the unit parts], [unit stats, again the parts, but some are combinations],

The difference with your game would be, that every design will be in the world. While Warzone2100 lacks at the most extreme designs. Players go for the "best". And build a lot of them.
Every statistic should be useful in a sense. So the weak armoured should be strong against some weapons anyway.

So making a card game only? ok then.
Have you thought about the statistics?

- All the units are the same, in worth that is. So all the stats are the same, but shuffled.
- Armor types, A, which are several for 1 unit.
- Damages against each Armor, D, which would be a random table. But the score would be 1, 2, 3 etc.
- 2 Armor type options gives 2 Damage options, 3 Armor type options gives 6 Damage options, 4 gives 10 etc.

Number of cards would be A * D

The Armor types:
2 Organic or Mechanic
3 Small, Medium or Large
2 Slow, Fast
3 Melee, Short Ranged, Long Ranged

The Damage would be for example
1-2 anti mechanic
3-2-1 anti small, less anti medium
2-1 anti slow
2-1-3 anti long ranged, less anti melee

That is 2*3*2*3 = 36 Armor types to choose from
And 2*6*2*6 = 144 Damage types to choose from
36 * 144 = 5184 different units.

Now then, 2 cards fight.
You simply multiply the Damages scores and then compare them. The highest number wins.
The lowest score possible is 1, the highest score possible is 36. So this is still doable.

If you want more different cards. Simply add more Armor types or expand the list for 1 of the Armor types. The table simply follows by adding score.

You could also do something different, not a pc based game copy. But it can have random pictures too. Example:
6A and 21D Circle/Triangle/Square/Quintet/Hexagon/Octa
9A and 45D Red/Purple/Blue/Green/Yellow/Orange/White/Black/Grey
2 Normal/Stretched
2 Small/Big
2 Normal/Inverted

432 different pictures, 7560 different weapons, 3.265.920 different cards.

Please let me know what you think.

Toa Lewa
Toa Lewa's picture
Joined: 10/31/2013
Fhizban, I really like the

Fhizban, I really like the idea you are coming up with, and I also liked your card graphics. Just a quick thought. You could offset any disadvantages associated with lower leveled invaders by giving them "group" abilities. As an example, a lone low level invader is almost worthless. However, when you spawn many low level invaders, they gain abilities and become stronger together.

I don't know if you ever played LOTR TCG but there was a strategy deck known as a swarm deck. Basically you filled your deck with tons of low level cheap minions. By themselves they would be killed and not harm the other player. However, if you played your cards right, you could annihilate the other player by the sheer number of them.

I look forward to seeing how the development progresses.

Syndicate content

forum | by Dr. Radut