Show HN: Trading cards made with e-ink displays (wyldcard.io)
I made a thing!
In 2014, I was holding a stack of iPhones and thought to myself:
"Hey, if I had each phone display a playing card, I could click a button and they'd shuffle themselves"
Right now, each card costs me about $20 each, but with only a bit more scale, I think I can get that down to $10.
In doing this project, I learned how to design electronics and circuit boards. I learned Rust and wrote my first driver, I upped my CAD skills, 3D printed, and did my first resin casting. I generated the images on the cards using stable-diffusion.
HN always seems to appreciate new uses for e-ink. Thought I'd share :)
295 comments
[ 3.5 ms ] story [ 313 ms ] threadThe cards themselves are completely passive and unpowered. Base does everything.
Mostly wanted to inspire something new.
There's a thin and flexible e-ink display I could use instead, but they're $20 each instead of $4. I agree they're a bit chunky in this form :)
I agree the docking station detracts a bit from the naturalness of just playing on a table.
I decided to build what middle-school-me would have wanted to play with. I took a lot of inspiration from Yu-Gi-Oh which has the same docking station concept (just executed much better, where you wear it on your wrist).
[1]: https://starwars.fandom.com/wiki/Sabacc
I took inspiration from Yu-Gi-Oh, Card Captors, Angelic Layer...
https://i.redd.it/wufrxonvm0t41.jpg
My son was really into it for a while, but they included betting “credits” with the game that weren’t nearly as fun as normal chips would have been.
For anyone curious, the symbol made by the contacts is the Tree of Life from the kabbalah
Magic the Gathering has a monopoly on fantasy, and the cyberpunk genre is played out. Plus I want the inner-workings to be somewhat mysterious.
I had a lot of fun thematically laying out the contacts. For example: Keter is VCC and Malchut is GND
Beautiful
In terms of making a game with them, I think a design where the pins are on the face of the card may be useful. I want to be able to 'draw' a card without knowing what is on it.
Certainly can be useful with pins on the back (and I totally get how this orientation is probably more size-efficient), but I think front pins would be more 'playable'. Maybe a design could be achieved with holes that pass through all the way so it can be written with either orientation?
Being able to 'power up' an existing card in the upwards orientation could be really cool for situations where you kind of want 'counters' applied to a portable card. Could have a base-station that allows you to 'add' the qualities of one card to another target card, or 'evolve' a pokemon, etc.
Maybe I should raise the priority on that.
Definitely I'm into the idea of evolving, or breeding cards like pokemon. Powerup or 'combine' would be cool too.
That would work. But you can also make that work with current hardware (I think): you just need a delay between pressing the button and the screen changing? So that the player can press the button, pick up the card, and five seconds later, it's revealed.
So in a way, you could "leave" traps or powerups on spots on the base, and your character activates them when it lands there.
(obligatory: 'you've activated my trap card')
That's why I wanted to get this concept out there, to inspire the next thing.
[1] https://news.ycombinator.com/item?id=31191850
If they took off and were produced at scale, we could definitely do color, and make them thinner and flexible as well!
Good Display GDEW029T5D
May I ask what e-ink displays you're using?
Good Display GDEW029T5D
Turning 90 degrees... maybe I could do some magic with magnets.....
Very impressive!
I want kids to be able to play at school or at camp, so didn't add any connectivity. Plus I'm trying to stay cheap. But I can imagine adding bluetooth to the base or to your phone or something. Then I can go all "BILLS PC" and let you store your spirits on the cloud, let you load them in and out of cards, etc etc.
Except I don't want to turn it into an online game. Computer games are awesome, I think it's lame to emulate physical cards with a computer, a'la' Hearthstone.
The best advice I got was "Read the datasheet"
The biggest step for me was deciding to spend money on experimenting. With a budget, suddenly it was fun to order stuff, buy tools, try things out.
Rust Book was the best for learning Rust.
Is that what that's supposed to be?
I knew no Rust at all before starting. Yikes was it frustrating to learn a completely new language. I had to go through the whole rust book, not skipping any basics, in order to get there. I also assigned myself a homework assignment to get used to the concepts. See the other blog post on wyldcard.io/blog
Honestly, by the time I got to the finished prototype, I could have just done the whole thing with Javascript. Although intending to use an SDM32F7 for the finished project, it was really convenient to use a raspberry pi for development. Using VS Code, I could program remotely on the pi from my Mac and iterate really quickly.
the awesome-embedded-rust[1] repo was very helpful.
TBH, embedded rust really feels like it's not ready for hobbysists who aren't embedded experts already.
[1] https://github.com/rust-embedded/awesome-embedded-rust
These prototypes are running on a raspberry pi, though I wouldn't want to put them into production like that.
Since I only made two, it was easier just to keep it on the raspi for now.
Sorry for adding more work. :)
Kind of like the Blinks game system, these little hexes with colored lights in them that each have a separate game in them and can 'teach' the other hexes they connect to.
One of the Blinks Kickstarters: https://www.kickstarter.com/projects/move38/blinks-smart-boa...
Yeah, it's a weird in-between. I don't think a game publisher would take a risk on it, but neither would silicon valley VCs.
Kickstarter would be my bet too. I was going to put this project down for a little while and start a "real" startup though.....
You might still try for the latter. Especially if you can think of ways to make the business bigger.
Can intro me to anyone? email me: jonah@wyldcard.io
I was thinking, very possibly, if I found some angel investors who also love boardgames.....
By the way, I've been following your posts here for years! I'll trade you cards for candy!
That said, this might be a good "real startup." The problem domain seems modest (e-ink playing cards) but... could be a bridgehead to interesting territory.
These might be designed/used as playing cards, but it's actually a computer with lots of little portable screens. The actual thing is general, a proverbial "computing paradigm."
These are playing cards, but could be concert tickets, conference badges, security doohickeys... They can open a door, clock you in and display your in/out status. If you want to go full "SV Pitch:" these cards are money. Transfer 69 FTX coins onto a card at a secure terminal, and pay by handing it to the hooker. A casino could give you one of these to be your wallet.
Solutions looking for problems sometimes find them. See apple/msft.
Nah. Part of the strip club experience will always be showering the strippers in dollar bills... and why would you do regular payments for hookers with a special token?! Almost all credit and debit cards (=EMV cards) already can do this by NFC and you can also use watches and phones for this, the problem rather is:
- sex workers are pretty much banned from conventional payment methods and networks because sex work is illicit in many countries and even where it's legal, many sex workers prefer hard cash because chargeback fraud aka "post nut clarity" or actually stolen credentials is so common
- whenever you start a new payment scheme - because you are doing precisely this! - you WILL have to follow banking laws and regulations: customer identification, anti-smurfing and other money laundering measures, compliances for data protection, reports to banking authorities... an insane mess to do right. Of course, you can also hope to do a Bitcoin... but given the penalties if you are ever caught by the authorities, it's not worth it.
Sure. Problems problems. The FTX coin reference was supposed to tongue-in-cheek over specifics. Yes payment systems have payment system problems.
My point is that these cards have all sorts of potential uses. They're a programmable physical tokens that display a fixed image until updated by physical contact with the plinth. Playing cards are just one set of use cases. The device itself is more general than that, potentially.
What does epaper get you that paper doesn't? A microprocessor and persistent state. RFID cards on a board game might be a good middle state for most cards. Scan them as played. Build gamestate or dynamics to be populated to epaper.
Epaper would be good for storyline branches, timeline progressions, evolving characters. Make them traveling characters being owned by different players. Differentiate them from becoming a static scoreboard that could be represented on a single tablet.
When watching TV shows that have cool card games, like Yu-Gi-Oh, it seems like the game doesn't have rules, but instead the kids reason about the characters by looking at the pictures. Weird unexpected stuff happens all the time, I was thinking about how to implement that in the real world.
For example, imagine a vampire card that has its eyes closed. You try using it, but it doesn't do anything. But, you notice that when you play at night the eyes are open and now it's a powerful vampire!
So you can do time-based mechanics, I could add location-based mechanics. You can also make the cards do different things based on what other cards are in play.
> When watching TV shows that have cool card games, like Yu-Gi-Oh, it seems like the game doesn't have rules, but instead the kids reason about the characters by looking at the pictures.
This is basically how it worked writing-wise in the original Duel Monsters series (though the characters acted like these were mostly all known effects/interactions), after that came GX where they toned down the creativity and mostly used real effects, then in Zexal and afterwards I think they stuck almost entirely to real effects with occasional exceptions.
If you're into this sort of thing, I highly recommend checking out Inscryption, it has some really fun twists on this idea.
I'm friends with a bunch of people in the board game industry, as well as being a game designer (with a game signed witha publisher that still hasn't been published yet after like, four years), and everyone thinks Marvel Snap is super fun and well designed (and addictive). It's the only mobile game I've really gotten into (that wasn't just a port of a physical board game) in several years.
Slay the Spire is another one to consider as well. It's a rogue-like game where you fight with cards and build up your deck based on the choices you make in the run. The cards are fairly static (they even made a board game adaptation on Kickstarter very recently), although they can all be upgraded, which makes the cards better. I play that on PC but I know it's out on mobile and Switch as well.
I still maintain that it's lame to implement the concepts tied to physical playing cards in a computer that can simulate anything. It's funny how Slay the Spire is a card game which got popular as a computer game and then later made the transition to a physical card game, bringing us full-circle.
For instance in MTG day/night cycles were featured in the original werewolves (I want to say innistrad), location mechanics can be continuous effects on lands or ETBs, likewise for reaction to other cards in play.
For what it's worth, the actual Yu-Gi-Oh trading card games has rules and the anime uses a very loose interpretation of those rules.
For interested parties needing to program, I could probably partner with them and do the development myself (or have employees do it). The real hard part is designing a game, the software to run it will be pretty basic to start with, once I write the general framework for taking turns, etc.
When they released the real-world Dueling Monsters card game, they tried to keep the rules consistent with the manga/series, but changed them in places to make a more playable game. Ex.: there is an upper limit to the damage you can do with the Berserker Soul ability.
[1] nsfw https://shop.kingdomdeath.com/collections/in-stock/products/...
If we're talking business models, the one which immediately comes to mind as appropriate is that of a gaming console. Sell the hardware (e-cards, maybe a mat?) at a loss, then sell games for it at no marginal cost. Let other people build games for it because hit games are what sell platforms.
Since this essentially a portable gaming device, consumers may compare it to e.g. a Nintendo Switch - if you undercut the Switch and you have a blockbuster title or two, you could have incredible product on your hands.
Presumably the lucrative economics of trading card games could be applied here as well...!
I think I'd need funding for that though, can't bootstrap a platform without that first blockbuster game.
---
Sample game idea: CAKE DECORATION game - players have to try to decorate as many cakes as possible. Requires 10 cards.
One card is the 'design recipe' card and never leaves the base station. It shows an amount of ingredients, ex. "1 cup of buttercream, 3 shakes of sprinkles, 4 squirts of whipped cream".
Players start with either 3 blank cards or 3 random low-amount ingredient cards.
On a players turn, they select from three cards: each card is an ingredient, an amount, and a modifier. So cards could be "1 shake of sprinkles (2x)" "2 squirts of whipped cream (+1)" "1/4 cup of buttercream (+1/4)" (could make variations or add other kinds of things that might go on a cake)
The player draws a card, base station detects which card is missing, and the other two ingredient cards increase in amount based on the modifier value displayed on the card. ex. "1 shake of sprinkles with a 2x" will fully complete the recipe if not drawn for two turns (because 1 x2 x2 = 4 shakes, which is enough for the recipe)
With 4 cards in their hand, if the player can pay for the whole recipe, they win the round and get a point. On win, new recipe appears and all cards in the winning players hand and on the base station get rerolled to new low-amount ingredient cards. Score could be displayed on the margins of the recipe card.
If they cannot pay for the recipe, the player places a card back on the base station. And their turn is over. (Should ingredients be re-randomized when replaced on base station? Should that be a player decision whether to reroll it?)
First player to decorate N cakes wins the game.
---
I think something like this could make for a viable game with a low number of cards, but could be more fun with a greater number (17 ideally?) which would allow for 3 active recipe targets (+2), using the 4th slot for another ingredient and displaying recipe as a disconnected piece(s) (+1), and larger hand sizes (5 -> +4) to allow for more complex recipes
An interesting game design question is how random you want the cards to be? Fair random would probably be viable, but since the 'deck' can know and make decisions based on the state of cards not connected to the base station, you could deal unfairly if less randomness would make the game more interesting or fun.
I'm glad you kept the total card count low, that's a limiting factor because of their price and size.
I'll try it out!
From a long time ago I'm trying to build a e-ink magnet for the fridge with relevant updated relevant information about some school programs.
But I haven't really found a cheap e-ink display.
Edit: I see that you answered it:
>Bought off Alibaba, I think they're so cheap because they're designed for grocery store shelf price tags. Good Display GDEW029T5D Where do you get yours for $20?
[0] https://www.adafruit.com/product/4777
[1] https://www.waveshare.com/product/displays/e-paper.htm
But you need a base station to update them, and a way to connect to your network to upload images/text.....
If you email me your address, I'd be happy to send you one for your fridge :D