5 Graphic Design and Typography Tips for your Card Game
Chris Farrell of Illuminating Games just wrote a thorough critique of card games setting their body text to be so small that they cannot be read at arm's length, let alone across the table. Here's an excerpt, but the whole thing is worth a read for any would-be card game designers out there. (Myself included.)
Break out your copy of the base set of Dominion, and look at the Chapel. This has a text box roughly 3.5cm by 4.5cm. It's got a single line of text. That line of text is 1 (one) millimeter high. 1mm! For me, it's only clearly readable at half an arm's length even in the bright light of day. (...) Compare this to a more sane game like Glory to Rome, where the font size is 2-3 times as large (text is 2mm high, keywords 3mm and usually highlighted). I can generally read Glory to Rome cards across the table, and can certainly see the important keywords. To heap insult upon injury, not only does Glory to Rome have far more legible text than Dominion it also has larger art. San Juan also thankfully starts at 2mm, although it could still easily be larger with no loss of aesthetics.
Here are some tips for designers who want to stay on Chris' good side.
1: If it can be said in fewer words, say it in fewer words.
Alright, before we get into any matters of layout, the first thing you have to do is look at how well you're using your text. Survey all the cards in your game and set a benchmark for the amount of text acceptable on any card. I recommend no more than two or three lines. Whenever I come across a card with more text than that, I tend to just ignore it. There are probably other, easier-to-understand cards available elsewhere and I'd rather get to play than puzzle over a miniature wikipedia article.
2: If it's said repeatedly, say it in symbols.
Depending on the kind of game you're making, you'll probably have certain phrases that come up frequently. Instead of repeating yourself and wasting valuable text real estate, just make a symbol that stands in for that piece of text.
"Deal two damage to any monster. Deal an extra damage to any Fire-based monsters." can be replaced with [SWORD] [SWORD] (+[SWORD] vs. [FIRE]).
"Draw two cards. Keep one and discard the other." can be replaced with [CARD with an X] [CARD with a CHECKMARK]. There is always a danger in overcoding your visual language, but if executed well, it'll make your game more accessible and playable.
3: Set the right style for the context.
The first thing young designers learn is that 12pt Times New Roman is the Devil. Young designers then overcompensate for years, insisting on 8pt regardless of context. We all have to get over that eventually and learn that there is no one golden rule that fits all contexts. A little card is different from a game board is different from a rulebook.
Here's a simple rule of thumb, though. Up close, set your body text at 10-12pt. Arm's length, 12-16pt. Across the table, 24pt. Always make your leading at least 1.5x the text size. And always, always choose a classic, legible font without any flourishes. I personally recommend Trade Gothic or Frutiger, but both have a distinctly modern feel that might not jive with your game's theme. You might be happier with Garamond or Caslon.
And lastly, if you simply must have flavor text, it's okay for that to be 8pt. It's an easter egg, not critical to actual gameplay.
4: Choose the right background.
Most card games have consistent block backgrounds in which you'll find the rules text, flavor text and any other important words. For the sake of efficient production, that background block is kept to the same size, regardless of its contents. That means cards with little text may have lots of unused space, as Chris notes in his critique.
That being the case, I highly recommend first taking steps 1 and 2 to make sure you're using text as efficiently as possible. With that base established, set your background block to fit your text. You can do this manually card-by-card, but that's a bit of a pain. Try setting a thick stroke around your body text style so your text effectively creates its own background.
Alternately, you can use a 0-distance, high-spread drop shadow. Either way, the background should be high contrast with very little texture or pattern that would interfere with the text.
5: Choose the right art.
And last but not least, the remaining 80% of card real estate is probably taken up with some gorgeous art, right? Just make sure the art actually communicates what is said in the text. In the chaos of the gaming table, the text "Firehose: Deal 7 damage to fire-based monsters" might get covered up by some other cards or a nacho or something. But if your art actually shows a bunch of fire elementals getting fizzled out by a firehouse, the message is still effectively communicated.
And that's that! Some simple steps you can take to keep Chris from complaining about *your* card game. Personally, I like the design of the cards in King of Tokyo. The titles are all custom designed and there's plenty of room for the awesome, illustrative art. The actual body copy is still a wordy on some cards and could afford to be larger, but for the most part KoT gets card design right.
I would argue that Race for the Galaxy is a case where overcoding makes the game opaque and difficult for new players to get in.
ReplyDeleteYup, indeed! When I spent more time looking at the glossary of symbols than the cards, I knew something was off. :P
ReplyDeleteI also think it's important to place information key to the game in the right location on a card. In my opinion, too many modern games takes cues from Magic the Gathering instead of traditional playing cards. Look at a playing card. It's form was developed over centuries of iteration. Rank and suit are coded by symbol and color, and placed prominently in the upper left corner. Even when you have a mitt full of cards, it's easy to scan your hand and know exactly what you have.
ReplyDeleteA Magic card, by comparison, hides the relevant information in the upper right and dead center of the card. You are forced to shuffle through your cards one by one in order to know what you have. Yes, there is a minimum amount of information on a playing card when compared to a Magic card, but there is no reason that modern games like Dominion should be copying Magic's formula for layout. A better example is Race for the Galaxy. While Race is guilty of over-coding, the designer at least made an effort to put all the symbols on the left hand edge of the card. This choice means that a player can fan his cards (sort of), and quickly get an idea of what everything does.
With so many modern card games coming out, we need some innovations that allow for better information parsing. I agree that better text placement and sizing is important. However, the location of key information on a card is just as important.
Oh definitely. And actually, if tips 1 and 2 are followed well enough, then all game information might be symbolized and placed in the corner like a playing card. I ran into that design challenge when doing the cards for Belle of the Ball, actuallly.
ReplyDeleteI think it's actually not quite as bad as that in the case of Magic, because the *name* is in the upper left, which is going to be the key for most people. But a lot of the Magic-derivative games running around today (like Dominion) the upper-left corner is completely unused, which is strange. I think it would have helped the genre if Dominion's design hadn't been so totally awful, as it would have given people something to copy.
ReplyDeleteI'm conflicted on Race. As a non-graphic-designer, I actually like the design of Race for the Galaxy. It's a sopsticated game with a lot of information to present, and it seems to do it in a pretty clear, comprehensible way, with clear iconography and a minimum of illegible text (in the base set, almost none). I sympathize with the folks who find the layout and all the symbology daunting, but I also feel like it does a good job of making a sophisticated game more approachable than if it had gone the Dominion route of describing everything with tiny text.
ReplyDeleteAs a non-graphic-designer, I think this is good advice, but if I had to guess, I think the root of the problem might be elsewhere. My suggestions would include:
ReplyDelete- Make sure you play-test for significant periods of time with your final choices for font choices & other graphic design elements
- Make part of your play-testing process testing the graphic design. At minimum, make sure all your text is readable under real game conditions (artificial light, real game distances, etc.)
- Always ask yourself what more you can do with the design that will make your game require fewer or less complex rules, and of you have rules that are putting and undue burden on the physical design, strongly reconsider them.
In the gaming space, it seems clear to me that graphic design issues have now gotten so bad, in general, that designers would do well to have a non-trivial "evaluate the physical design" portion of their play test process. There are so few good examples to simply copy from now.
Yuppers! Agreed on all counts. This is what I call Gamma Testing. It's when you have the game completely baked, but now you want to test its presentation. I should probably write a post or three on that subject.
ReplyDeleteI have the strong sensation that Race for the Galaxy would've been better served in a medium other than a pure card game. With a board, chits and perhaps some pawns, they might have been able to sequester some visual language to the appropriate contexts instead of putting the burden entirely on the cards and glossary.
ReplyDeleteNice article. You said, "Always make your leading at least 1.5x the text size." Which means 12 pt type on 18 pt leading. Which is not what I think you meant. Maybe 1.5 pt leading?
ReplyDeleteIn your example, 18pt leading is 18pts between the baselines, which for 12pt type looks pretty balanced.
ReplyDeleteIf you mean 18pts between the actual rows of text, that would be something like 30pt leading, which would indeed be overkill.
Magic has tried to rearrange the card frame with the CC in the UL corner but it didn't work well in testing. That might only have been because it wasn't what players were used to, but it's not like they haven't tried. (Instead they went with the 'new' card frame that is just a re-skin.)
ReplyDeleteFWIW, I think that the name of a card in Magic may be more important than the casting cost. Personally, I usually sort my hand by type, not by casting cost. (Mana Leak and Storm Crow both cost 1U, but do very different things.)
Interesting. I hadn't realized that Wizards had experimented with different designs. I'd love to read their results.
ReplyDeleteWhile I agree that the name of a card is very important to an experienced Magic player, that's only because he already knows what the card does. For a new player, or one who hasn't familiarized himself with all the cards in his deck, the name is almost meaningless. In a game with as much variety as Magic, I'm convinced there's a better way.
And of course none of this addresses the points brought up by Daniel (e.g., how east or difficult it is for other players to spot a card across the table and at least have an idea of what it does). I know that from my Magic days, I was constantly asking to see an opponent's card because everything was new. Now I have the same frustrating problem with Dominion.
Magic's Time Spiral block has cards with every face they've ever publicly tried: http://gatherer.wizards.com/Pages/Search/Default.aspx?page=4&action=advanced&block=+%5B%22Time%20Spiral%20Block%22%5D
ReplyDeleteHere's a direct link to the card Colin mentions.
ReplyDeletehttp://gatherer.wizards.com/Pages/Card/Details.aspx?multiverseid=136054
Magic is a tough nugget to crack because of how much unique rules text each card requires. In this card-text in particular, there are so many clauses and conditionals that it'd be tough to translate it symbolically without just being a confusing rebus.
M:tG is sort of a different beast than Dominion, though. Dom runs on small card sets with high repetition; M:tG is about constant expansion of cards, and as a result it needs to teach players to memorize card text.
ReplyDeleteInteresting discussion on a complex issue to balance. I'm designing a card game that has just slightly less complexity as Race for the Galaxy, but with more interaction between cards on the table and cards in hand. Cards kept in hand need to be briefly numerically analysed in terms of challenge attribute totals to similarly laid out challenge locations on the table. It's a layout challenge. Here's a functional prototype in Word. All icons yet to be properly designed.
ReplyDeleteThe top left icon is gold cost to play the card and the following 4 icons represent fundamental attribute stats used throughout the game as part of particular challenge totals and also help determine power of certain abilities. Faction symbol is top right and VPs (end game only) bottom right. Other post-challenge modifiers are along the bottom.
Some argue gold cost should be separated from attribute icons stylistically or in position. I appreciate the idea but find that left edge vertical layout helps do fast maths of the 6 - 8 cards typically in hand and both cost and attribute totals across your hand of characters is a key part of that.
How would you try this?
If folks are telling you to change the gold look, you might consider putting the other icons on a special background or container. Otherwise, you can just create more space between the gold and the other icons. Or you can have an icon of gold coins next to a numeral rather than a numeral on a symbol as with the other symbols. Lots of solutions to that problem!
ReplyDeleteyeah thanks, that's in line with my thoughts. Might also try a version with Gold / Name / Faction (pretty standard fare in such games but probably for good usability reasons) on a header block at card top. that would also distinguish them.
ReplyDeleteSo is there a magical maximum threshold of a number of symbols memory and usability wise, per card / per game? ie you want to minimise repeated text by symbolising, but not have players needing to constantly cross reference to an aide...
ReplyDeleteEvery game will be a little different i guess and the context of the gameplay and system may impact on this too. But any thoughts?
The common wisdom is that humans can only hold 7 bits of discrete data at a time, such as digits. I dunno if that has been debunked yet, but it's still a decent rule of thumb. So those bits could be sentences, symbols, key terms, whatever, but it's probably best to keep them under 7 total. Kind of fluffy, but at least gives you a place to start? :)
ReplyDeleteI agree King of Tokyo got layout right and card look and feel (80% image) totally on target with the core theme of big monsters with even bigger power ups. That's 90% of the game right there as the game play is very light. Its game and graphic design in destructive harmony.
ReplyDeleteSome games feature far more interlaced complexity. How about use of "Keywords" in such games?. They are functionally similar to symbols.
Games like MtG and the A Game of Thrones CCG/LCG and many others have opted for a mix of Keywords and Symbols.
MtG uses basic keywords for original (?) card abilities and for newer abilities adds the keyword with a functional description in italics in brackets, which i always thought odd as they tended to be limited in number and memorable from set to set. I presume its because MtG is mostly about competitive play where rule books are not on hand. Of course it uses the tap symbol and 5 man symbols too.
AGoT, a more casual but complex game generally, mixes text based Traits and Keywords (14 all up) which are quite descriptive of the ability - eg 'Deadly', 'Stealth'. But they also use an array of symbols; House / Gold / Military / Intrigue / Power / Influence / Initiative / Unique and crests; Noble / Holy / Learned / War / Shadows on cards outside of text.
MtG can effectively be played by those 12 and up (and is it!) but enjoyed by adults. More text on cards is arguably a factor there...
AGoT is effectively an adult game, but adults are and i think always will be its core (often fanatical) audience. The keywords and names of symbols, icons and crests reasonably reflect themes in the books. But the wide array of these adds another layer of complexity on top of comparatively more complex rules. That's not inherently bad as it gives the game greater depth its often retired MtG players find appealing.
The degree of interactivity between cards is another factor worth exploring.
Games with set building want symbols. Limited interaction games like RftG also thrive on symbols so you can more easily discern opponents probable strategies at a glance (really the core of player interaction in that game).
However, in games like MtG or AGoT where cards interact directly in more complex ways it is often handy to have the text at hand while you are smashing one card into another because the rules of their interaction are right there in the interaction, so it actually speeds game play. The conundrum there is that from across the table you cant possibly read the text. In such games art > text memory is often key to success. But so is the mix of symbols with immediately meaningful Keywords and text abilities.
Inversely for our group, the fantastic game 7 wonders' Leaders expansion essentially broke the game for us. The core game has no text, symbols that make sense and is an exercise in fast paced minimal downtime gaming elegance albeit with limited interaction. Then Leaders arrived. There are i think 30ish leader cards that can interact with the game state in a large variety of ways. All expressed by symbols which you really need to constantly reference the single rulebook to understand. 5 - 7 players needing 1 rulebook? The game stalled. We tried it 3 times and haven't played it in ages. Overcoded Leaders killed the 7 Wonders.
I guess I'm arguing there are a wide array of factors to balance in more complex games. And there is a sliding threshold of symbol vs text utility based on degree of card interaction.
Daniel can hopefully express in 1 infographic what ive taken 1000 words to!
Re: Rulebooks. Part of why Small World is so successful is that it comes with four or five reference sheets for their numerous powers. Those powers are presented as illustrated iconography and keywords so you can easily find them on the reference sheets. The powers themselves are fairly brief and intuitive. Imagine if Small World didn't come with those sheets! I suspect it would run into the same stumbling block as you found in the Leaders expansion.
ReplyDeleteIt certainly would have. No doubt numerous reference sheets give small world a usability advantage over the similarly sybolised 7 wonders leaders. I guess the lesson is; if you are going to have a symbol fest then give a reference sheet (or booklet in the case of Leaders) per player (hello production costs!). However, in both games most of those symbols represent unique powers not 'repeated' powers, so neither strictly follows your laudible tip 2 and I think both games can be argued as potential cases for over codification. They could both have worked more on simple ability wordings which DO fit / work on the boards playing tiles / cards.
ReplyDeleteI think designing for internationalisation plays a big part here as a completely symbolised game only requires multi language rule sets, thus keeping down production costs for a truly global audience.
Cyclades is a case in point. Apart from card names, its 100% symbols across the board, tiles and cards. Its a great game but to be honest our group just found there just were too many symbols to effectively memorise. After 6 plays we collectively remember abilities symbolised on pieces and detailed in the rules and abilities booklet mainly from card names / pics (like any MtG player) NOT their symbols... and we still find ourselves pausing the game to refer to the ability listings in the rules to a frustrating degree. So in this case, the internationalisation / symbolification strategy comes at the cost of in game usability, which through online or word of mouth game reviews has gotta hurt sales.
So where is the sweet spot?
The award winning Kingdom Builder finds a better balance between symbols and words on cards and tiles. The 8 different types of special power hexagonal location tiles contain extremely simple and representative symbols for their rules. 8 is a magic number in web usability and I suspect other areas requiring easy access memory retention, and of these 8 only 4 are used in any one game. My wife and non gamer friend found these KB location tile abilities easy to understand on the first reading and then remember from the symbols. Then there are 10 Kingdom builder cards, only 3 of which are used per game, which alter the game Victory Point objectives. These rules are inherently more complicated (hard to symbolise) and are reasonably well explained on the cards with further info provided as required in the rulebook. So in Kingdom Builder I think they only need to vary the rules and those 10 kingdom builder cards. That sounds like a sweet spot from both a usability and production point of view.
Apologies for another essay, I seem to have a disorder.
Great article / site by the way Daniel.
The background tip about shadowing reminds me of unsharp masking, which Imperial Steam, Roxley Brasses, and No Thanks did well
ReplyDeleteArk Nova has a great example of featuring both icons, keywords, and text descriptions