On the Flipside

Posted in Latest Developments on May 12, 2006

By Mark Rosewater

Working in R&D since '95, Mark became Magic head designer in '03. His hobbies: spending time with family, writing about Magic in all mediums, and creating short bios.

Hello, Mark Rosewater here. If you were expecting Aaron, then I'm guessing you didn't read “Making Magic” this week. (And really, is there any excuse for that?) This week, Aaron and I are swapping columns. No, it's not another Swap Week. Aaron and I are switching places because Dissension provides us with an interesting opportunity. You see, Dissension was Aaron's first time as a lead designer of a set and, as fate would have it, it's the first development team I've worked on in a number of years. (Although for the record, I was also on the design team.)

Realizing that we finally have a set where Aaron has insight into the design and I have insight into the development, we felt it might be fun to shake things up for a week. So, yes, this week you get to hear me talk about development. It's not as crazy as it might first seem. I'll let you in on a little secret: I've actually done a lot of development. I've been on more development teams than Aaron. I think more than both Randy and Brain Schneider (the former head developer and the current head developer). In fact, I was hired as a developer. I was on every development team from Alliances through Urza's Legacy and on and off development teams for years after that.

That said, my skills definitely lean more towards the design side. I'm horrible at judging power level and my playtesting skills are nothing to write home about. Luckily, a number of design skills overlap nicely with the needs of a development team, so every once in a while I like to stretch my development muscles. That is how I ended up on the Dissension development team.

Before I jump into the thick of it, let me begin by introducing you to the Dissension development team. (I know Aaron did this briefly but he didn't include photos.)

Matt Place

While Matt has been on a number of development teams, this was Matt's first time leading a team. (Just as it was Aaron's first time leading a design team. – not too shabby for two first timers.) I've known Matt for years. He and I first became friends back in the early days of the Pro Tour. In fact, Matt attended the very first Magic Invitational (at the time known as the Duelist Invitational) in Hong Kong. (By the way, the 10th Annual Magic Invitational is going on as we speak, check it out here.) I was there when Matt won Pro Tour--Mainz. Matt disappeared for a number of years, but then resurfaced a few Pro Tour seasons back.

Matt had heard that there was an internship opening up in R&D and he told me he was thinking about applying for it. I heartily encouraged him and gave a ringing endorsement for him back in R&D. Cut to present day. Matt has become a key R&D developer. I was quite excited when he asked me to be on his team (Brian and I both like to let our team leads have some input into who will be on their team). The development process was relatively painless and I'm quite happy with the set we turned out.

Brian Schneider

Brian is my equivalent on development. I don't envy him. While design is a lot of work, it has a lot less drudgery than development. As I always say, design's job is to make it fun and development's job is to make it fair. Making things fun is a lot more fun than making things fair. (Well, at least that's my perspective – good thing I run design, huh?) Creating a balanced environment takes a lot of work. I think so much focus has gone into how much fun Ravnica block is that many people overlook what a good job Brian and the developers have done to keep everything balanced. I don't know if I can remember another time in Magic where every format was running smoothly with significant diversity. This is Brian's doing.

In addition, I enjoy working with Brian. While our roles occasionally force us to butt heads, in general we get along great. Dissension development was no exception.

Steve Warner

So much focus goes on some of the higher-up people in R&D that I think the average player doesn't realize how much work gets done by the rank and file. Steve is one of our developers that specializes in playtesting. One of the major reasons that the environments are so good right now is that R&D members like Steve find the problems before we print them. Prior to joining R&D, Steve spent years working in our Customer Service department. I'm not sure, but I think Dissension was Steve's first development team. Not that you would know it. Steve rounded out one of the best development teams I've worked on.

Mark Rosewater

And then there's me. But I already talked about me, so you'll just have to enjoy the lovely picture.

So here's what I'm going to do for my one and only development column. At the beginning of development, the design team hands off a file known as the Handoff File (a.k.a. the 0.0 File). Today, I'm going to show you a few cards as they came into development. Then I'll show you the final card and I'll explain how the card got from State A to State B. (By the way, there's no rhyme and reason for the order other than my whim.) Sound fun? Then away we go.


Gain control of target multicolored creature.
Brainwave 1U (At the beginning of your upkeep, you may reveal this from your hand and pay 1U)--Target creature becomes the colors of your choice until end of turn.


Govern the Guildless

Here's the fun of development. We spent hours to change one word. As far as I'm concerned it was worth it. So why is “monocolored” better than “multicolored”? To understand, this we have to take a step back. One of the mini-themes that the design team interwove into the set was a “multicolor matters” theme. That is, the design team wanted the multicolorness of a card to mean something.

During development we examined all of these “multicolor matters” cards and we came up with an interesting observation. The cards that rewarded multicolorness proved much better than the ones that punished it. Why? Because the game of Magic already punishes multicolor pretty harshly. The mana system pushes players towards monocoloredness. What makes the theme interesting and different is that it makes the game reward you for playing multicolor cards. Plus, this was a multicolor block. We wanted you to play them. Making cards that punished that behavior seemed wrong.

As such, the development decided that we were going to make the vast majority of the “multicolor matters cards” reward multicolor. The design team, by the way, had stumbled unto the same resolution, but the development team took an even stronger stance. Which brings us back to the card. Both cards allowed the fun interaction we wanted, where the forecast effect could allow you to steal a creature you otherwise couldn't. The difference was how good the card was when you didn't use the forecast effect. Before the card punished multicolor play. After, it rewarded it. This is why all the bother over the change of a single word.


Flying Cockroach
Creature — Insect
Hellbent (If you have no cards in your hand)--CARDNAME gets +2/+1.


Demon's Jester

How can you have less change than one word? How about no words. What's interesting about this card is how much development time it took up to change nothing in the end. Here's what happened: Development was worried that this card was too swingy. The concern stemmed with a larger concern about the hellbent mechanic. Hellbent's closest cousin, mechanically speaking, is the threshold mechanic from the Odyssey block.

One of the lessons of threshold was that you have to be careful on the differential between the two states of the card. Both threshold and hellbent have the same issue in that when the proper state is met, all the cards with the mechanic change over to their other version at once. This means that in conglomerate reaching the chosen state can have a giant power swing. That was the concern that was plaguing the Flying Cockroach (a.k.a. Demon's Jester).

For the majority of Dissension's development, Flying Cockroach merely got +1/+1 at hellbent. What got us to change it back was my plea that the card just wasn't eliciting the same kind of responses it was earlier. During design, hellbent proved to be an adrenaline ride. But when we de-powered the common hellbent cards, it took the wind out of hellbent's sails. Other team members chimed in that they had similar feelings. Then one day, Matt just said, “You know what. You guys are right. We have to change it back.”

And like that, we were back where we started. Ah, the fun of development.


Master Bureaucrat
Legendary Creature — Vedalken Advisor
White spells cost 1 less to play.
Blue spells cost 1 less to play.
Black spells cost 1 more to play.
Red spells cost 1 more to play.
Green spells cost 1 more to play.


Grand Arbiter Augustin IV

Now let's get to a card that went through a little more change. We knew that the leader of the Azorius was a bureaucrat and we knew that Blue/White liked cards that shifted the environment. From a design standpoint, we felt that the card was elegant and seemed cool. Development had some other ideas.

The first strike against the card was that it was a little confusing. For example, the effects overlapped. This meant that multicolored cards were all over the map as to what this card did to them. Here's the effect on the following ten two-color gold cards:

White/Blue – costs 2 less
Blue/Black – costs the same (the 1 more and 1 less cancel out)
Black/Red – costs 2 more
Red/Green – costs 2 more
Green/White - costs the same (the 1 more and 1 less cancel out)
White/Black - costs the same (the 1 more and 1 less cancel out)
Blue/Red - costs the same (the 1 more and 1 less cancel out)
Black/Green – costs 2 more
Red/White - costs the same (the 1 more and 1 less cancel out)
Green/Blue - costs the same (the 1 more and 1 less cancel out)

Note that it did nothing on a majority of the color combinations. Next, we realized that it heavily discouraged players from playing the other three colors themselves. Third, we realized that five lines were going to take up all the text box space leaving us no room for flavor text. While we often leave off flavor text on wordier cards, the guild leaders definitely liked having a little something said about them.

The development team started by figuring out what was cool about the card. It turned out to be two things: one, it allowed you to play your White/Blue stuff cheaper; and two, it made many of your opponent's spells more expensive. Why not cut the fat and make the cards simply do that? So we did.


Hellbent Oriflamme
Whenever a creature you control attacks, that creature gets +2/+0 until end of turn and CARDNAME deals 1 damage to you.
Hellbent (If you have no cards in hand)--Whenever a creature you control attacks, instead that creature gets +4/+0 until end of turn and CARDNAME deals 2 damage to you.


Anthem Of Rakdos

I designed this card with the express purpose of making a card with hellbent that you occasionally might not want to get hellbent. Note I only made one. Mechanics that discourage themselves are dangerous territory, but I felt that hellbent was something that was easy enough to undo that the card warranted inclusion. The development problem? Everyone other than me on the development team hated it. Well, maybe hated it is a strong word. They just wanted to remove it from the file.

But I held firm. I liked the card and I liked the kinds of things it potentially could do. Every time the team would try to kill it, I would rally to the cause and fight to keep the card. Then one day Matt said, “Okay we've fought about this card long enough. No one but Mark seems to like it . I say we just kill it.”

Realizing that I couldn't win another fight, I tried a different tactic. “I think I can fix it,” I said.

“How?” Matt asked.

“What if I could find a way to hide the negative? What if I made hellbent on this card seem positive just like all the other hellbent cards? I mean, it would still have negative complications, but they wouldn't be so in your face.”

“And how are you going to do that?” Matt replied.

I have a habit of getting into arguments where I haven't thought my plan all the way through. I had faith that I could design something that would lessen the elements the other team members didn't like while retaining what I liked about the card. I just assumed I would have more than a minute to do so. Luckily for me, I think fast on my feet.

“Let's get rid of the numbers.”


“What if we didn't actually show the higher numbers on the card? What if we hid the extra damage?”

“You've lost me,” Matt said.

“Okay, let's use some Tempest technology. What if we turn the hellbent ability into Furnace of Rath?”

“We have it double all damage?”

“Exactly. This way it still does what it did before but it has this extra value as now players might fight to get to hellbent to help double their other damage sources. Overall it will seem more positive, but the element I wanted, the idea that sometimes you have to be careful of hellbent and work from reaching it, remains.”

The development team liked my suggestion and thus we ended up with the Anthem of Rakdos you know.


Growth Chamber
You may choose not to untap CARDNAME during your untap step. 3, T: Tap target creature. As long as CARDNAME remains tapped, that creature doesn't untap during its controller's untap step and has “GU: Put a +1/+1 counter on this creature.”


Evolution Vat

Several weeks ago, I wrote a column about the design of various Dissension cards. (“In the Cards”) In it I talked about Evolution Vat. While everything I said was true, as I was looking through the design handoff file for this column I realized that the Evolution Vat actually evolved significantly (ironically enough) in development. The original version of the card was modeled after Amber Prison from Mirage. It locked down a creature and kept it locked down as long as you chose to keep it locked down. The Simic activation was to encourage you to occasionally lock down your own creatures.

In development, we came to some important realizations. First, the lockdown effect was stronger than we wanted. The entire point of the guild artifacts was to tempt you to play it in the proper guild. The lockdown effect was so strong by itself, you always played it regardless of if you had blue and green. The development team didn't like that. This is when I pulled out the dusty leftover from Fifth Dawn (part Icy Manipulator, part Dragon Blood).

But we didn't stop with lessening the power of the first half, we strengthened the second. We wanted the player who was playing Simic to be really tempted. In my ongoing quest to get the word “double” into more textboxes, I came up with the new and improved activation.


Creature — Elephant Wizard
T: Target creature becomes 3/3 until end of turn.



The first detail I want to point out is a very subtle one. Look at the first line: RZ86_DEL. What does that strange bundle of letters and numbers mean. It's what we call a card code. Let me break it down for you. The first letter is the rarity. C is Common; U is Uncommon and R is rare. The second letter is the kind of frame. Most often that means the color of the card but it also refers to things such as artifacts and land. (W = White, U = Blue, B = Black, R = Red, G = Green, A = Artifact, L = Land) So what's Z? Multicolor. Why Z? Why not say M? The system predates me so I don't really know. Multicolor has to be specified as multicolor cards have their own frame. (And as of Ravnica, each two-color combination now has its own frame, but that's a discussion for another time.)

The last three letters are the expansion code. DEL stands for Delete. The codename for Dissension was Delete (Ravnica was Control and Guildpact was Alt). That leaves the two digit number in the middle. This is the interesting one. It's the card number. Usually that number is a low number between 01 and 21. The number of a particular color varies set to set but there has yet to be a Magic set with eighty-six of the same frames. 86 is a code saying that this card was in the file and cut. (It was 86'ed.) The 86s are sometimes left in the file if the lead designer thinks that the card has potential to go back in. (For those that care, the sets also have 99s that represent extra cards.)

So what did the last two paragraphs teach us? That this card was turned over from design not in the file. It had been cut prior to handoff. But Aaron liked it, so he kept it around (or maybe he just forgot to cut it before handing it off). So why was the card cut? Because it was missing something. We liked the mechanic of the Simic having a guy that can transform creatures. And 3/3 is an interesting number, as it allows you to use the card to both sometimes make creatures better and sometimes to make creatures worse.

Quick aside for a funny story (which was referenced once in a Card of the Day). The above mechanic was used on an artifact rare in Mirrodin design. It had two activations, one colorless and a cheaper one that had blue mana. I called the card Wand of Polymorph. In development, the team, inspired by the name, changed the card to have the effect of the blue spell Polymorph (from Mirage). The name was then changed to Proteus Staff (and the colorless activation was dropped). My mechanic was killed because I chose the wrong playtest name for it.

Anyway, during Dissension design I brought up the idea of using this effect on a creature. Aaron put it into the file, but it later lost out on numbers (which means that we had too many cards and had to get rid of the one that least fit and/or was least liked). During Dissension development, we cut a Simic card and made a hole. As this card was in the file (86'd though it may have been), the team took another look at it. They too liked it, but felt it was missing something, and that is when we hit upon the problem. A 3/3 what? What were you turning things into? If we could figure that out, we might just have a card. After studying the Simic a little bit, we realized that we had a perfect creature type that overlapped blue and green (although as Brady Dommermuth will tell you frogs are blue and toads are green): frogs. Once the card turned things into frogs, everyone liked it. How can you not like a creature that turns things into giant frogs? I don't think it's humanly possible. And with that little change, Omnibian was back in the set.


Blessed Changeling
Creature — Shapeshifter
W: Whenever CARDNAME deals damage this turn, you gain that much life. Play this ability only once per turn.
W: CARDNAME gets +0/+1 until end of turn.
U: Swap CARDNAME's power and toughness until end of turn.
U: Return CARDNAME to its owner's hand.



Time to play "Can you spot what development did?" As I mentioned in “In the Cards”, this card was inspired by Morphling right down to its 3CC (C stands for colored mana – aren't you learning all of R&D's tricks today? Technically this card is 3CD, as D means a second color of mana) mana cost to its five abilities. So why did we change the card from spirit link to vigilance?

Is vigilance better? It does a little better job than spirit link of feeling like “: Untap Morphling”. (The other four abilities all line up with Morphling abilities. Check it out on Autocard. I'll wait.) That's not why we did it. I chose this card as it demonstrates an important part of development. Sometimes the job of development is just to make it work. The problem that sunk spirit link? The proper template made it three lines long. We didn't have the extra line. The text simply didn't fit on the card.

You see, we have hard and fast rules about font sizes and what we can and cannot fit (this is partially about readability and partially about allowing enough space on foreign translations). If a card crosses the line, development has to change it. No ifs, ands or buts. I liked what spirit link did for this card, but in the end it just didn't fit and we had to change it. That my friends, is the kind of thing development has to deal with every set.


Wall of Hats
Creature — Plant Mutant
Mutato 4 (CARDNAME comes into play with four +1/+1 counters.
Whenever another creature comes into play, you may move a +1/+1 counter from CARDNAME onto that creature.)


Vigean Hydropon

Not too much changed on this card. Design turned all the graft creatures in as 1/1s with their graft number being one lower. Development thought it would be better if every graft creature could use its ability until it literally dropped over dead. But that change was not the giant fight on this card. Can you see what else is different? Yes, eagle eye reader, Wall of Hats was a defender. I mean really, how much attacking could hats do? The idea behind this card in design was that we liked having a beefy graft guy whose only real purpose was handing out +1/+1 counters (a.k.a. hats). To allow us to do this, we gave the creature defender.

Then development got its hands on the card and we realized that having a four-drop 5/5 wall was too good. That's when I suggest we give it the line “CARDNAME cannot block”. (And dropped the mana cost by 1.) It tickled me to no end that we'd have a Defender that couldn't block. Unfortunately, it seemed to annoy most other R&D members. How could a defender not defend? That's what the word means. It was weird, I admitted, but I just thought it was quirkily fun.

At one point I believe Aaron suggested that the card could read “CARDNAME cannot attack or block”. He felt this wording would seem less awkward to people. I chimed in that I liked the awkwardness. I thought it had a better impact. Yes, it was a little disorienting at first, but hey, I'm a fan of occasionally disorienting the fans (as my past and future designs will demonstrate). I even argued that we had a card in the set that let defenders attack. By making this a defender we increased in-set interaction.

Obviously this was a battle I lost. As I'm a stubborn little man, I still feel the team made the wrong choice, but part of development is compromise and you can't win every fight. This is one I lost. I'm curious in the thread if people agree with me or with the choice the development team made.


GU Guildmage
Creature — Elf Wizard
3G: Target land becomes a 5/3 land creature until end of turn.
3U: Return target creature you control to its owner's hand.


Simic Guildmage

I left this card for last because it's the one I'm most proud of. (I'd take credit for Experiment Kraj, but the design team has no idea who actually made the card – my best guess is that it was either me, Gottlieb or some combination of Gottlieb and myself. Yes, I believe the Johnnies of the design team made it.) As you can see above, the design version was rather stinky. Development realized this and set out to create a Simic Guildmage from scratch.

We had been trying to find a place to put an activation that moved around +1/+1 counters as it played nicely with graft. The design team looked but never found a good place to put it. We had talked about interacting with +1/+1 counters on the Simic Guildmage but felt that we'd be trampling on the Golgari Guildmage whose green ability made +1/+1 counters. This was on my mind as I set out to build the better Simic Guildmage.

I tried every ability I could think of that made sense in blue or green. My problem was that I was having trouble finding two that worked together that had a Simic feel. One of the tricks when you get stuck is to go back and look at old cards. Oftentimes you'll stumble upon a card you've forgotten about that will jump start ideas, which is exactly what happened. While searching through Gatherer, I found the card Enchantment Alteration from Legends.

I have very good memories of Enchantment Alteration. Fans of my Magic: the Puzzling puzzles might remember that it showed up in a lot of puzzles. A lot. As I thought about the card, it dawned on me that Ravnica block did indeed have an enchantment sub-theme. And the Simic were all about manipulating and changing creatures. But if I chose this for the blue activation, what could I possibly mirror it with in green?

This is the part were the thunderbolt nearly snapped my head off. Of course, moving +1/+1 counters! The one final tweak I added was to prevent you from using either ability from moving one player's stuff onto another player's stuff. The reason for this is the effect is very powerful. If we allowed it, we would have to cost the activations for it and then the only reason you would use the cards would be to steal +1/+1 counters and enchantments. And that wasn't the point of the card. (Yes, a Johnny through and through.) This change also made me feel that the activation would play differently enough from the Golgari one.

One last note on this card. If you open this card, play it (or draft it). This guy is the bomb in Limited. If you make the slightest effort to build around him (and Ravnica block has +1/+1 counters and enchantments all over the place), he is insane. I had him at the Employee Prerelease and I can't begin to describe the crazy things he let me do. He's fun and powerful – a potent combination.

I could go on, but I don't want to get Aaron's readers accustomed to any higher word count than this. (I have to work with the man after all.) I hope you enjoyed my peek into the world of development. If you enjoyed this column and somehow don't make it a habit to read “Making Magic” on Mondays, come check me out. I'm just like this, except I talk about design.

Join me next week when I'll be back on Mondays and talking about making time.

Until then, let me end with a little poll. This is “Latest Developments”.

Bye, bye now.

Mark Rosewater

Latest Latest Developments Articles


June 9, 2017

Changes by, Sam Stoddard

Hello and welcome to another edition of Latest Developments! Today I'm going to talk about several kinds of changes within R&D and how we deal with those. Card Changes From the day ...

Learn More

Latest Developments

June 2, 2017

Things I've Learned by, Sam Stoddard

Hello, and welcome to another edition of Latest Developments! This week is the five-year anniversary of me joining Wizards of the Coast as a contractor on the development team. My officia...

Learn More



Latest Developments Archive

Consult the archives for more articles!

See All