So it's been a while since I've written a blog, and I apologize to those who have subscribed, but between personal issues, changing (and being promoted in) jobs, I haven't had as much time as I'd like to develop games. With my hours more normal these days, and especially having weekends off, I have been able to make a lot of progress on Star Requiem: Humanity's Last Stand over the last few months, and now it is feature complete and locked - an exciting time when you know you have a playable, working game system all your own!!
I wanted to take this time to look back at what changed during the design process, the tools used, and what I have learned so far from the experience - hopefully others who have been thinking about putting their design to paper (or computer, or iPad, or...) can take something away!!
Anyway, what have I learned so far???
#1: Prototyping, prototyping, prototyping
I estimate that in the year plus that I've been working on SR I've gone through about 6 or 7 different designs for the starmap, the boards, the system cards, etc. Usually when I'm working on a game I will write and scribble on a design for as long as I can stand it, then I'll print out on my home jet printer a workable version of the game boards and counters, and when I really feel like I've hit a design milestone (3 times so far) I'll trundle down to my local friendly Office Depot and have the whole thing professionally printed (about $10 or so) and mount and use that as a working prototype. Needless to say, this got old, and being a long, involved game, it was hard to find the time to play through even one phase, much less the whole game, before I created a slightly new version.
The solution: program the game and play on the computer - let it handle all the fiddly bits!!
This was really great for SR. Now, I could change an idea or a value much quicker and see the result play out - and more importantly, I could very quickly see what needed to 'drive' the game (more on this later). It was during the computer prototype that morale, the revamped lab system and the nanovirus concept were created, among others.
Now I fully realize that most people can't just up and port their boardgame design to a PC program from scratch. However, I do recommend using VASSAL if possible - yes you still have to do the rolling of the dice and such but there's a lot of automation you can build into a VASSAL module and it's much easier to create a VASSAL module than you might think. I encourage budding game designers to take a look at computer prototyping in some way. Once you get the initial work out of the way, it's fantastic!!
#2: Where's the exit?
So in the original version of SR, there was no 'ending turn'. You played until you captured the homeworld or destroyed all 4 'Grand Fleets', or the Xyl conquered your homeworld. While there was nothing inherently wrong with any of these victory conditions, as I playtested the design I realized how long it would truly take to get to that point, and starting thinking about how the player could 'break' the system. Having unlimited time allowed the human player in the end game to, among other things:
- amass 50-60 BPs per turn, making the production decisions virtually irrelevant in the end game; why fear losing a fleet when you can replace it on the slips each turn?
- research to the end of the tech tree
- overwhelm the Xyl with so many missions that you could drop the XCL (Xyl Confidence Level) 7 or 8 spaces in a turn if things went well, while they were limited to just a few attacks at the most, even during the late stages
- build so many defense units and mines on the capital world that it was mathematically impossible for even a Grand Fleet to get to the inner system
In other words, what happens to many 4X games in the end of days, where a player starts to ring up a runaway exponential advantage over everyone else (I call this the 'compound interest' syndrome) was happening to my test games. Who wants to play a game where you know whether you're going to win or lose in the first 20% of the game? Not me.
So the answer was to create a bookend - a rule that the game ended after 100 turns. If the human player survived 100 turns, they won. Yay. And that's how it stood for a while. Until...
#3: The Force(s) are with you
Most of us remember a game called Moon Rover. It was an old game where you were in a little rover with a jump button and a laser, and your goal was to get to the end of the level. On the surface, it didn't seem so tough or interesting, but the designers made the decision that the player, like it or not, would always be moving towards the goal. So you couldn't just stop and 'camp out' a situation waiting for just the right moment - you had to be proactive throughout the game. I felt that this concept is what made Moon Rover a classic game, and SR didn't have that same 'forces' moving people through the game. What could I use to push and pull players through the SR experience?
I had one force on the player initially: the XCL. All it really did was set the type of missions the Xyl would try and the size of the fleets the player could face, but there was no real way to control the flow of the game beyond just sitting and turtling - amassing tons of BPs, waiting out the occasional Scout mission that increased the XCL by one, and finally, 30 turns later, moving to the Attrit phase. By that point, you had 2 or 3 shipyards, were pulling down 35 BP per turn, and could begin the dreaded compound interest cycle. What could I use to 'push' the player?
Eventually, three concepts were added, but the most important by far was the reinforcement system. Each turn, the Xyl could potentially add a new fleet, which meant an additional opportunity to attempt a mission. Now, instead of the second phase arriving in 30 turns, it could arrive in 8-10 sometimes. Much better. But then there was an additional issue: Now there was no 'fear' from the player going all out at the beginning and basically going on the attack all the time, hitting the Xyl at their weakest (during the Scout phase), succeeding in driving off missions and hitting weak frontier systems, and a lucky player could keep the boot firmly on the Xyl for most of the game! So another 'force' was needed to keep the player balanced between offense and defense, and the HAL (Humanity Aggression Level) was born. Of all the things I added to the design, other than morale, the HAL had the greatest effect.
Now there was an opportunity cost for almost every mission decision. Attack the Xyl too often and the HAL would quickly move to max. What happens at max? The Xyl reinforce 3 times faster, their rolls for missions increase 40%, their fleet sizes and compositions change, and most importantly the amount of chances they have to assign a valid mission increase vastly. This last point compared to the others finally created the backswing to the hammer the humans had been enjoying. I had one particular game (and you can see it happen in my current AAR) where the Xyl were able to mount 4 successful scouting attacks on my systems, causing the XCL to go from middle Scout to Attrit (later Terror) phase in one turn. Now, there was incentive to leave the Xyl alone from time to time, but one more force was needed to create the 'decision triangle' that most games have.. and it was morale.
The idea of morale was the single biggest new idea for SR. Now, I had a mechanic that overlaid every decision I made, not only for this turn but for others down the road. My morale is dropping - I might lose the game well before turn 100; my people need to see some positive results! So I might take a risk and go all-in on a scout mission to show that we can make progress in the war. Consequently, my morale is probably going to get hammered in the next few turns (perhaps from moving into the Terror phase) so perhaps I need to build some more production centers to offset the potential penalties and NO PRODUCTION NEXT TURN events? Now, I have an incoming Xyl fleet; do I attack it and increase the HAL and (if I fail) XCL, or do I let it complete its mission and leave the HAL alone, but potentially drop morale? At last, I had a decision triangle: the HAL (offense-->defense), the XCL (steadily increasing overall difficulty on the player) and morale (actively moving to increase morale at the expense or detriment of the other 2 factors). Now we had something! But there was still something major missing.
At this point, I still had a hard turn limit of 100 turns. It still felt long, but it was a lot more fun and balanced, and there were significant decisions throughout the game. But it still felt more like Fortress America towards the end - load up the defenses and fleets on your capital, and grimly hang on until turn 100, and you were gravy. It was turtling on an endgame scale, and it was not quite what I was looking for. A final force was needed to pull the player to the end.
That force was Doomsday.
Cutting the turn limit to 75 initially (and finally 50) and creating a situation where there was going to be an Armageddon event at the end of the game, and if you weren't proactive you were most likely going to lose, completed the game's cycle of tension, pushing, and pulling. Dropping the turn limit also had the effect of compressing the production decision cycle, so that every turn meant more. A game like Eclipse or Risk 2210 A.D. where you only have 8 and 5 turns, respectively, mean that you simply can not waste a turn. And that was an issue - I felt like there were a lot of 'empty turns' in SR. At 50 turns, you literally can not waste a turn between the research/production/shipbuilding triangle, and you can not take too many reckless chances because you don't have the time to simply build a massive industrial complex or to replace fleets - now simply building a scout could take 10% of your entire game length! Having the Doomsday roll tied to how far in the XCL the Xyl advanced made it to the player's advantage to be proactive and for the first time since the very first design of SR, actively attempt to manage the XCL. Doing this, on top of the other considerations of morale and the HAL, under a strict time limit where defeat is entirely likely if the game reaches that point, completed the forces at work.
#5: Research? Feels like rocket science to me!
I was very proud of the initial research system for SR. It had over 60 different things you could research, dynamic research times, breakthroughs and wrong turns, random research paths and results, and tangible effects to everything you got. It was brilliant.
And for a board game, it was horrendous.
Sure, you could research 60 different things. How are you going to keep track of them all? And more importantly, how are you going to keep track of all the different bonuses you got? I can't remember if I have lasers now or kinetics, and do I get that +2 bonus to movement for just scouts or for all size 1 ships, and did I remember to raise my max Build Point cap because I just researched the right technology, and did I achieve all the prerequisites for that Size 3 ship? No? Whoops.
I toyed with the idea of placing counters for everything, but I realized that I would need a separate board just for those, and it would look like a submarine's 'Christmas Tree' by the end of the game. And you still needed to fill out the Encyclopedia Galactia by hand to make sure you remembered what you had researched already! Clearly, it wasn't going to work as a practical thing that people would put up with (my previous effort, Falcon Ace, was crushed by systems like these) so back to the drawing board I went. How could I make research fun, strategic, and most important, quick to do?
I cut virtually every technology and focused on only 3 areas: Starship Technology, Weapons Theory, and the Nanotech Virus research that now gave the player a more peaceful goal to victory. Furthermore, I increased the cost of labs and put a maximum of 6 in the game (though there are 10 slots they can be placed) so there is some strategy there. And I made the benefits easy to understand: all size 2, 3, 4 and Terra One can be built when a category is researched, and you get a bonus across the board to speed, stealth, and sensors or weapons and targeting rolls. Period. No more economic benefits, no more cross research to slowly gain weapon ability. It was a good idea, but it didn't work for a paper game. The new research system I feel nails the balance between choice and playability.
So there was a lot more that I changed in the year plus of development of SR, but those are the big decisions I made. This was a long blog, but I wanted to share my design thoughts, missteps, and start-overs. Now it's your turn - any designers out there want to talk about their most wrenching design decisions or ripping out entire game systems that you loved... but simply didn't work when they were put to paper?
Note: This blog entry will concentrate on several computer games for examples, so forewarned is forearmed.
Anybody ever play MOO3? Sequel to one of the greatest computer games ever, the expectations were through the roof. The designers were under intense pressure to create an encore that would rise the name to stratospheric heights in both sales and lore. What to do, what to do?
The answer, apparently, was to throw everything in but the kitchen sink. And then they added that later.
MOO3 is an easy comparison, but I want to examine the phenomenon of taking concepts that work and smashing them into an ill-fitting design with a little less-known game: Elemental: War of Magic. This was a game with a ton of expectation behind it - it was all but advertised as the spiritual successor to what most consider the 'pinnacle' fantasy 4X game - Master of Magic. It would feature many of the same concepts, including building cities, having heroes with their own stats and abilities, being able to wander the countryside and slay creatures of fantastical lineage, and generally building up your kingdom until you became the most powerful badass in the land. Sounds good, eh? I certainly thought so. Until I played the game.
What happened to Elemental is what can easily happen to many game designs whose reach exceeds its grasp. Elemental had a ton of good ideas, but when put together, the balancing came apart when the different systems had to work together. An example: Spells were fairly varied and the early spells weren't all that much to write home about. Spend just a little time in certain 'books', however, and within mid-game your heroes were death-dealing fire-breathing demigods that would have sent Hellboy himself cowering in fear. Not exactly the balance you're looking for, but pulling off so many systems and balancing them all against each other, especially when the overall game dynamic is changing and evolving, is extremely difficult to do well.
Taking a fairly obscure example of a game that did it right (in my and many other people's opinions) is a game called Midwinter. This is an interesting game from the DOS valley circa 1989. Essentially, the Earth is in a new Ice Age, a dictator has taken over and it's up to you, the player, to gather a motley band of freedom fighters and eventually reclaim the island for freedom! (and heat)
What made the game amazing was its combination of genres that (at the time) was far, far beyond its time. A first-person shooter, a driving/skiing/hang-gliding(!) simulator, a social simulator, and a strategic war planning game were all mashed together in a new, original concept - but it worked. Why? Because there was no one overpowering way to victory. It didn't matter if you weren't good at skiing - find a sled or recruit people who were. Not enough weapons? Get some more that you like. Not good at sniping planes? Use a grenade and let area-effect take care of things. The point was there were many paths to get to the same end, and the player had the right to expect that as as long as they played their style competently, they had a fair shot to win the game eventually. Unfortunately, that is becoming increasingly hard to design for in games today - more and more there's an obvious path to victory, because an otherwise great design idea or concept was implemented poorly, or a game fell prey to what I'll call 'sales-generating-feature-bullet-point-itis' and thus exploited.
In Star Requiem: Humanity's Last Stand, balancing the game systems is a huge priority. It's hard enough to do it with several systems, but when you have a variable 'time limit' to when the player should be at, they shouldn't feel like they're hopelessly outclassed at that point - it becomes much harder. What are your thoughts, if you design either boardgame or video games, about balancing multiple game systems together so that one doesn't throw the rest off?
Getting a game more or less complete is an exciting time: on the one hand, you've put a lot of work into getting it to that point, and there's a lot of pre-alpha work done to get it even that far and that balanced, and you have every right to be proud! But like anything that is made with quality, it needs to be refined, balanced, and judged impartially, and that is the point in development that can be the most nerve-wracking.
It's your game, and OF COURSE it's the greatest game since Catan. Why, who would DARE not love such a original concept? Who among us wouldn't love all the optional rules and variants that you oh-so-thoughtfully added? And who doesn't love a good story of murder, violence, betrayal, and revolution... all in the same game?
As it turns out, lots of people.
See, being creative is a double-edged sword. If you never intend for your creations to see the light of a jaded public, than you are free to create as your muse travels, without fear of reprisal or rejection. If you are among the majority of creative people who yearn to show their brainchild's off, then you must be prepared for brutal rejection. And to me, that used to be the hardest part of the creative process. I also write, and have written for several publications in the past, and to me the proofing process is sometimes excruciating - but after the fact, usually necessary.
The same goes with game testing. It hurts a lot less to find out your game concept really doesn't work early on than when you're trying to shop your game to publishers and/or self-publish at considerable expense to you.
So Star Requiem: Humanity's Last Stand is at alpha stage - meaning it works, and is mostly documented, and is more or less art complete, and most of the systems are in, but the balance is probably not so hot and the rules could use a major run and polish. I have sent my alpha files to 5 people who I have never met in real life, and it will be very interesting to hear what they think.
I love getting feedback, and even though a lot of the feedback will be how to make the game even better (and they volunteered to do this for free and at their own pace because they like the concept of the idea) it is still a process that never truly gets comfortable. But I want this game to work, and I have a very open mind, so like a newborn being shown off by their proud parents, SR has emerged from the womb... and like a real child, it will only get stronger and more able to stand alone over time.
First of all, welcome back, BGG!! We've missed you. Today, I want to talk about rules. In my opinion, bad rules can absolutely ruin an otherwise good concept. This is not a particularly earth-shattering observation, but what are exactly 'bad rules'? Are they complex? Are they poorly written? Are they redundant? Or are they simply... inefficient? To illustrate, let me present one section from the Falcon Ace rules that I feel are particularly illustrative:
126.96.36.199.2. FALCON ATTACK PROCEDURE. If your Falcon is eligible for attack, proceed to the ACTION TABLE – FALCON ATTACK MODE and roll against the enemy’s class and piloting skill. Make an Action Roll to resolve. The red die determines their fire range. If you have not already, roll for the enemy’s ATA rating and loadout on the AIRCRAFT LOADOUT TABLE to determine their maximum engagement distance and place the enemy’s missiles in the short and long range weapons boxes on their respective bogey slot. If the value rolled is equal to or greater than the current range between the Falcon and the enemy, move to the ATTACK SEQUENCE described below. The white die is the speed roll. If the result is ‘FULL’ use the highest WHITE value from the AIRCRAFT SPEED TABLE for that model plane and replace the speed value on the BOGEY STATUS DISPLAY. If the result is ‘AFT’ use the highest BLACK value from the AIRCRAFT SPEED TABLE for that model plane and replace the speed value on the BOGEY STATUS DISPLAY. If the enemy is not in attack range, move the enemy towards the Falcon using the special movement described in 188.8.131.52.3. When complete, proceed to the next enemy is applicable; to the Ground Attack Phase if not.
184.108.40.206.3 ATTACK SEQUENCE. Roll 1D10 on each non-blank column to determine the Actions the enemy will take in the attack sequence. The total Actions for the enemy depend on their Piloting Skill and type. The first Action will always be considered ‘TGT’. No roll is required for the first Action. If the result is ‘TGT’ make a targeting roll, using the most applicable weapon in regards to range, with those values and the appropriate table. If the weapon is infrared, use the INFRARED MISSILE LOCK TABLE and if the weapon is radar, use the RADAR MISSILE LOCK TABLE with all applicable modifiers. If the lock is successful, place an ‘ENEMY IR LOCK’ or ‘ENEMY RADAR LOCK’ counter on top of your Falcon counter for the respective type of lock. If the result is ‘FIRE’ without a prior successful missile lock then the Action is treated as a snap fire missile launch. Place the missile on top of the enemy counter and place a ‘NO LOCK – SNAP INFRARED FIRE’ or ‘NO LOCK-SNAP RADAR FIRE’ counter on top of the Falcon, else treat as a normal missile launch. If the result is ‘TURN XXX’, the attack is considered ‘broken off’; turn the enemy in that bearing and complete the enemy move. After a break off of attack, reroll for new pilot mode on the BOGEY PILOT MODE TABLE using the +3 modifier for prior ATT mode. When complete, proceed to the next enemy if applicable; to the Ground Attack Phase if not.
Wow! Are you slavering to shoot down a MiG now? I know I'd be running away if I had a rulebook with 47 pages of rules like that above - but as the designer, I couldn't see that when I was in the middle of writing the rules. It was hard enough just getting them onto paper and grammatically correct and complete that little things like readability and usability got thrown out with the rest of the bathwater. Taking some time away from FA and coming back to the game after a year or so really made me realize how much of a disservice I had done to my idea and to my game by making the rules so impenetrable. Which is not a mistake I intend to repeat with Star Requiem: Humanity's Last Stand. Here is a snippet from the game setup section:
SETTING UP A GAME
So enough reading background, let’s play the thing already! OK. Assuming that you have prepared all your materials and counters for play, first lay out the star map. This is the big 24” X 24” map with all the ‘DEEP SPACE’ boxes on it. Put it in the middle of your play area and place the ‘XYL CONFIDENCE LEVEL’ counter on the 0 space in the ‘SCOUT’ track on the right-hand side of the board. Next, take out the ‘HUMAN BUILDS TRACKER’ and place the two BP tracker counters on the ‘0’ boxes in the ‘TOTAL BPs’ section. Next, place the ‘HUMAN SCIENTIFIC PROCESS’ board and place the 5 category ‘TECH LEVEL’ counters on LV1 on their respective categories. Set aside the 5 category RESEARCH PROGRESS counters for now.
Next, take out the ‘HUMAN FLEETS TRACKING BOARD’ and place the two CP tracker counters on the ‘0’ boxes in the ‘COMMAND POINTS’ section. Take the BATTLE BOARD and place the ‘HUMAN INITIATIVE’ counter on the ‘0’ box of the CURRENT ROUND INITIATIVE track. Take the XYL FLEETS TRACKING BOARD and place it nearby.
Now, the star map needs to be populated. Shuffle the Xyl system decks and the Human system decks, separating out each side's green dot systems, the red dot systems, and the yellow dot systems. Place the capital system card for the humans "New Terra" in the lower-right system box (“10”) and depending on your game scenario, draw 2 green systems, 2-3 yellow systems, and 2-3 red systems. Place your green systems face-up in boxes 8 and 9 (your choice). Next place your yellow systems face-up in boxes 5, 6, or 7. Finally, place your red systems face-up in boxes 1,2,3, or 4. You may place your systems wherever you wish on their row; the only rule is that there must be at least one direct path to each system – in other words, you may not go through ‘DEEP SPACE’ boxes; each system must have a hyperspace path.
Now you will set up the Xyl worlds. Draw as you did for the human systems; equal number of cards per color. Xyl systems are placed face-down so that the ‘?’ side is showing. Place the capital system in the upper-left system box. The placement rules are the same as for human systems and in addition the red Xyl systems are placed directly across from the red human systems. In other words, they will be placed in the boxes directly above the existing red systems.
Almost done! Now you’ll set up the fleet draw cups for the Xyl. Place the ‘Grand Fleet’ counters, the ‘Major Fleet’ counters, and the ‘Minor Fleet’ counters in 3 small opaque cups. Place the ‘Scout Detatchment’ counters along with 5 ‘NO RESERVE FLEET’ counters in a separate cup, away from the other fleet cups. This is the reinforcement cup. Leave a 5th cup empty for now – it will be the Scout Detatchment draw cup. Separate the Xyl Seeker ships, Size 1, Size 2, Size 3, and Size 4 Grand Flagships into 5 draw piles.
Finally, set up your starting situation. As the Humans, ensure that your ship types are separated into unique draw piles. You will only need the Scout and Size 1 Warship piles to start. Draw your starting shipyard by taking the shipyard counters and randomly drawing one. Place the associated shipyard card and place it on the Human Builds Tracker in its designated area, and place the shipyard counter in the ‘SHIPYARD/STARBASE’ section in the New Terra system. Draw 2 Size 1 warships randomly from your draw pile and place them in the ‘PHOENIX FLEET’ fleet box on the Human Fleets Tracking Board. Place one ship in the FLAG section, but if this is your first game don’t worry too much about which one. Place the ‘PHOENIX FLEET’ counter in the SYSTEM FLEETS area on the New Terra system card. Draw a Scout from your Scout draw pile and place it in any red (frontier) system in the SYSTEM FLEETS area on the system card. Finally, place either a Level 3 Resource Installation or a Level 3 Science Lab counter on New Terra, and place 5 more points total of installations on any planet, provided that the level of Resource Installation does not exceed the System Resource Level OR that the total levels of placed installations in a system do not exceed the System Infrastructure Level.
Now that you’re set up… it’s time to save humanity!!
So, in a little less than twice the length of just 2 subrules in Falcon Ace, I have completely walked a player through how to set up a game for play, ready to start. You'll see no sub-sub-sub sections here - just sections in bold, easy English (like, um, Setting Up A Game) that I hope will make it much easier to get into Star Requiem - and be able to appreciate the engine behind the game, not struggle to gain entry and fight through learning how to take a single turn. I will be spending more time on the rules than anything else, and I can promise you that the game will not release - in any form for purchase - until the rules are clear as crystal and every overlong, unnecessary rule and procedure will have been pared away.
So I throw it to you, the reader. What do you look for in rules that make for ease of use? Lots of examples? Outline form? Pictures? Do you prefer 'casual' writing style, like in the SR example, or more what I'd call 'Avalon Hill 101' style, like in Falcon Ace? At the end of the day, what is most important to you in a good ruleset? I'd really like your thoughts on this!
Every game ever created, whether solitaire or a 16-player epic, has some sense of scope. In fact, I would argue it's the most important facet to consider as a game designer - in effect, it's a simple question:
"Who are you?"
Are you a five-star general, plotting grand strategy across Europe? Are you a solitary soldier running across the battlefield, trying not to get killed for God and glory? Are you a futuristic Marine squad commander, trying to lead your troops through an alien-infested derelict? Are you the supreme commander of a civilization, or a single-cell organism? It doesn't matter - IF the scope of the game matches correctly with the answer to that all-important question.
I thought about scope very heavily when designing Star Requiem: Humanity's Last Stand. Since my last game was so (overly) detailed, I was initially a little hesitant about making the game 'epic'. But as I really started to work on the backstory, I realized what the game was truly about. Originally the game was going to be a war between two roughly equal sides- the Order and the Lost. The Lost were a group of humans who split off from the main new colony of New Terra to practice their own form of religious expression (read: evil) and eventually, seeked to reclaim their soulless brethren in all-out civil/factional war.
Now, this was interesting, but it had two problems from my point of view: first, religion is always a touchy subject in any game and balancing the factions might prove tricky since I planned to have a heavy demonic influence on the units/systems that the Chaos used. But the second, and far more important reason, is that the scope just didn't feel right. The stakes didn't feel high enough. You're fighting a war against.... yourself, essentially, and if you lose... well... you'll be going to a different church in the future. While important, it didn't feel epic... until I came up with the Xyl.
Here, we have a force that is (until now) unstoppable, unnegotiable, and uncaring about eradicating humanity from the face of the universe. They're not looking for reconciliation, or even slaves. They want everything dead. Except for them. Period.
Now that's a threat! And in that context - where humanity must essentially 'learn to fight' again and essentially become what had nearly destroyed it (irony!) the scope of an epic game and feel made sense. Technologies would have to be uncovered; ships and fleets would have to be created; systems would have to be scouted and reconned, and offensives planned; and events would happen that would effect entire systems. Most importantly, the enemy threat could build slowly enough for the humans to have a chance, but without quite knowing when the Xyl will start actively destroying and conquering systems, there is a tension that stretches out well, even across game-years (each turn represents 3 months).
So, in the end, after much internal pontification, I decided that SR will most likely be a pretty long game to play. Somewhere around 12+ hours for a full game. And I worried about that. I know that will turn off a lot of people who are now used to Euro solitaire games that take 30-45 minutes. And there is nothing wrong with those games - you need some bite-sized diversions now and again - it can't all be Silent War if you hope to have time to enjoy the breadth and scope of gaming! But the stakes are what they are; and I just didn't feel in the end once I had nailed down the backstory and the enemy that the story - and the conflict - could be told in just an hour or two. Since SR is meant to tell the story of an epic struggle between two galactic empires who are waging a war of desperate annihilation, the plot deserves its time to grow and mature. And so it shall.
In the right hands (and the right game) tactical combat can be a great thing. In games like the Total War series, many people feel that there is a near-perfect balance between the tactical portion and the strategic portion. Sure, you can auto-resolve the tactical battles, but many people feel like you'd be missing out on the experience!
When designing Star Requiem: Humanity's Last Stand, my original plans called for an elaborate tactical space battle 'module' where you would essentially 'zoom in' on the action and fight an epic duel with your fleets, your commanders, and your researched ships.
This quickly sucked up the majority of my development time, because I was falling into the trap of hideously complex tactical situations (see Ace, Falcon for the bad example) and one thing that I told myself that I learned from FA is NOT to make the player's head explode!
So then, I decided to make the combats very top-level, ala Steel Wolves. And I was OK with this for a while. But the thing with Steel Wolves is that the combats have to be abstract by nature since each turn you might encounter 20-30 different combats - and there are almost 200 turns in a SW game!! With SR, you may go many turns without a single conflict (scouts do not attack) and even in the later game you may only have 2-3 battles a turn. So I hated to give up the unique units and 'feel' of customizing you fleets just to decide the whole encounter with a few dice rolls. It seemed almost like cheating the player's building and fleet crafting strategy.
So the Ranged Battle Grid was born. Depending on who the attacker was, the enemy places their ships randomly in 'Approach Range', and the other player places theirs on the other size of the board in their 'Approach Range'. Then there is a roll for side initiative, and the higher roll moves last and fires first. It's a very simple, but elegant system I believe. The Size 1 ships move and fire first, then the Size 2, then 3, than 4, and they move forward or diagonally in the grid towards the other side. There are 4 ranges that both sides come into: Approach, Missile(Shard), Kinetic(Flayer), and Laser(Reaver). Ships may only fire at enemy ships that are in the ranges of these weapons. Ships may not stop on top of each other, and ships may not target ships that are directly behind another ship. Damage system is very simple: Ships may take a # of hits equal to their size rating before they are 'damaged' (1/2 all ratings) and the next hit of any kind will destroy them. There are no die roll modifiers - you simply roll against the enemy's ECM rating to check for a hit; if it's equal or higher then you roll 1D10 and add your weapon rating to the die roll to get your attack value, and divide the enemy's defense rating into that attack value - each whole # divisor is a hit! That's it!
For example, if you are attacking a Xyl ship in missile range and your missile value is a '6', and the Xyl's ECM rating is a '3', you would roll 1D10 to check for a hit. If the roll is equal or higher than a 3 then you have hit; you then roll a 1D10 and add your missile rating (6) to get your attack rating. So if you rolled a '9' your total attack value would be 15. If the Xyl's defense rating was a '6', then you divide 6 into 15 to get 2.5. Only whole numbers matter, so this attack would score 2 hits. If this was a Size One ship, it would be destroyed, and a Size 2 ship would be damaged.
So it's a simple system, but there are lots of strategic possibilities. If you have a command ship in range, you can do coordinated attacks (simultaneous attacks of the same weapon type from multiple ships) to help punch through heavy ship armor. ECM ships can protect 1 ship around them from a missile attack. Aegis ships give all ships around them a negative modifier to hit from an enemy. Multitargeting ships can designate several targets at once (ships get a # of attacks equal to their size). You can use larger ships to protect smaller ships and create a 'wall' of battle (the board is 10 units across and each ship size unit takes 1 unit - look at the ship counters of the larger ships to see how this works) BUT since you can't move 'through' ships by doing this you slow down your faster ships since they can't move before the larger ships, and you can't 'wait' for the larger ships to move before you move - you have to plan your moves at the beginning of the turn if you want to shield your smaller ships well!
What makes combat decisive is that ships MUST move towards each other each round. You are not 'stopped' in space EXCEPT for command ships and dedicated missile ships. But even they must start to move lest they get overrun by fast-moving scout and reaver/screamer ships. You can retreat ships by moving them off the grid diagonally but this takes time since it takes 2 movement points to move 1 diagonal space.
I'll be posting more about the tactical system later but I wanted to give people an idea of how hard it is to 'scale' a good combat system on top of a strategic layer - too high level and you don't feel an attachment to your builds and fleets; too low level and you will start to dread each combat, and eventually quit playing the game.
Welcome back! If you've stayed with me this far, you already know that I dislike linear systems, and I touched on the idea of 'known' research results briefly in my last blog. When I was designing the concept of Star Requiem: Humanity's Last Stand, I wanted there to be an overarching story about the situation, and a deep and believable backstory to help explain not only why the player should be invested in the story, but to explain some of the design choices.
The backstory of SR is long, but suffice it to say for this post that humanity has found itself at peace with itself after being forced to find another home because of catastrophic ecological damage to Earth after the Final World War. Realizing that humanity would become extinct if they couldn't work together to solve certain pressing problems, like, say, finding a place to live that would actually sustain life over the long haul, humans managed to band together, get off Earth, and start a new society in a new corner of the galaxy. And for the last five hundred years or so, they've been very happy. Since war is what almost destroyed the species, war technologies have decayed to the point where it is little more than the functional minimum requirements to keep a modern civilization to recognize its mores to society, and little more.
So what happens when a remorseless, xenophobic, cybernetic species discovers you in their neck of the woods and decides they'd very much like you not only dead, but every last strand of DNA of your species sterilized at temperatures roughly found within the center of a fusion reaction?
In SR, you crack out the old Earth encyclopedias and get to work relearning how to wage war on a massive scale, that's what.
What makes SR compelling to me is the narrative (and irony) of mankind trying to get its war groove back. In most sci-fi games involving humans, they've already got some pretty impressive hardware and it's more a matter of getting a lot of it built, and maybe tweaked. In SR, you're essentially starting with the interstellar equivalent of a sling and some smooth rocks. Since each game turn is one month, and most games should take 75-100 turns (at least; that's my target), that's roughly 6-8 years for humanity to get off the mat and stand toe to toe with the Xyl.
To do that, I needed a research system that was:
a) not terribly long and complex - the scope of the game isn't long enough to discover 50 new technologies
b) easy to use and make strategic decisions for the player
c) difficult or impossible to 'tech rush' to the same conclusion every game
d) not so powerful that moderately poor decisions by the player would doom their chances of winning, no matter how well they played elsewhere
e) be fun to use for the player!
I hate the idea of basic technologies taking 8 turns while far-off technologies take thousands. Why should they? And for that matter, why should 1000 researchers in 1000 labs across an empire automatically shave 1000% off an arbitrary 'base time' to discover a certain tech? I hate logarithmic tech advancement just because you now have 2000 scientists instead of 200! There's no provision for a 'Newton moment' in most 4X games. I understand 'why' in game terms we have gotten used to this standard in 4X games... because it's easy to understand and because it's easy to program the AI to use.
Fortunately, when your enemy already has all the tools, designing a different system becomes much easier.
My solution was to create 5 broad 'tracks' of knowledge that only pertained to getting humanity back on its war footing. The basic idea is that you can invest research points (RPs) into one or several tracks, with each multiple of the minimum required giving you more chances (rolls) to progress down the track towards a discovery, somewhat like the shipyard system discussed yesterday. You do receive more RPs if you have more labs (more 'chances' to make a discovery) and you can pool them into one track if you want to focus your efforts. Here's a picture so that you can follow along with my thoughts:
So how it works: You roll as many times on each track each turn as you have multiples of RPs invested on that track, and if you roll at least the target # or above on the current node, you move your track marker down one node and if you have any rolls left, you roll against the new node's target number, and so on. You keep going until you reach the yellow node, which is the 'breakthrough' node. You get ONLY one roll immediately upon reaching it - even if you have invested huge amounts of RP - and if you make the roll, you draw for a technology at your current level of knowledge, and may then either move up one level in that track and start over at the top node or stay at the current level of knowledge and return to the track's REPEAT LEVEL node that is somewhere around the middle. If you don't make the roll you go back to the REPEAT LEVEL node to eventually try again or abandon the track (for now). That's the system in a nutshell. Easy, and elegant, I think.
Certain tracks are harder to start going down than others - for example, it takes a roll of 9+ on 1D10 (0-9) just to start progressing down the Xyl research track. Economics, meanwhile, can be started with a 6+. Unfair, Steve! I can hear you say! But think about that a minute.
Unlike just about any other game, I haven't taken away the chance for you, the player, to attempt to learn more about the Xyl. I haven't set some arbitrary time in the future for learning about your enemy. What I have done is set it up so that you can try, but it should be hard (after all, you've barely seen them and don't know anything about their systems at the start of the game!) but as you progress down the track it should get easier to do so (you are learning more, testing theorems, adding additional data to hypotheses, etc) until there comes a moment where either you were right - and you learn something useful - or you were wrong - and you back up to the point in your research where things went awry.
And - surprise - this is how it happens in real life!! There is uncertainty in all research, but there's no almighty force saying 'You aren't worthy to learn about X subject; therefore it may not be pursued.' So the SR system allows anyone, at any time, to research any category, as long as they understand the costs (in RP) and risks (wasted time).
So let's compare war economics. It's a short track (4 nodes), the starting number is 6+, and it drops to 4+ and 2+ with the 'breakthrough' roll being a 1+. That's a 90% chance of getting a discovery on your breakthrough roll! Why is Xyl a 6+? Why risk a 60% chance of having to go back 2 nodes and have to roll 8+'s AGAIN?
Well, first of all, economics have been done before. There's a lot of information out there about scale of economy. It's mainly about restarting a dormant pool of knowledge. But the Xyl are a new species. As humans, you literally don't even know where to start (hence a 9+ roll to get on the track). The same goes for weapons studies and space propulsion. You had the knowledge at one time, now you just have to adapt it! (Hull/Ship Design is so high at +9 because, again, humans never had war space ships and this is a field that they are literally starting from scratch, beyond basic 'habitable and moveable hull' designs. I may drop this value based on testing, though)
So let's say you put in the RPs and got to the magic yellow breakthrough node and make your roll. What happens? You roll for a Level 1 technology! Yes, finite random technology rears its head. Note that all discoveries will be useful in some way - just perhaps not the way you need. For example, to build screamer ships (sort of a destroyer-sized close range super-high-speed laser ship) you need Screamer technology. It's a Level 1 hull/ship design technology. So if you want to use Screamer ships and you roll for Screamers, great. But if you weren't planning on it, now you have a decision (yay!) to make: do you 'use what you gets' and adapt your plan, or do you repeat the research to try to make that discovery down the road?
And this is another unique feature of SR's research system. When you repeat a track, you don't have to start from the beginning! Each track has a REPEAT LEVEL that the node below the line becomes the start point instead of the top in 2 cases: when you fail a breakthrough roll AND when you discover a technology and decide you want to keep investing on that track - AS LONG AS YOU KEEP THE SAME LEVEL TARGET. So in other words, if you invested in Hull/Ship Design and you made your breakthrough roll and received, say, Space ECM Tactics and would rather get Interstellar Ordinance Navigation (for advanced missile systems) you have a choice to make. The nice thing is that if you decide you just have to accessorize your new warships with the latest in shiny missile tech you can start at the 6+ node instead of the more daunting 9+ node two nodes above. So you end up investing far fewer RPs for further study down a related line - AS IT WOULD BE IN THE REAL WORLD!!!
Here's the caveat, though. To prevent wanton jumping around levels, you only get the repeat level WHILE YOU ARE RESEARCHING THE SAME LEVEL OF TECH. If you decided to, say, move on to Level 2 discoveries and then decided that missiles are a must have after all, you can still go back to Level 1 to get them, but you have to start over at the top +9 node. Planning becomes essential - another DECISION is in order: Do I stay at a certain tech level that I've already invested precious RPs in to 'rack up' easy tech or do I move on to bigger and better things - knowing that I'll essentially 'lose' my previous investment beyond what I've already learned? You can cherry pick lots of slightly better tech or you can laser focus on hopefully some hard-core, potentially war-skewing Level 5 tech that you can use before the Xyl sterilize your new home world.
The choice is yours... just as it always should have been, in my opinion. With SR's research system, the power is yours! *cheese alert too late*. Thoughts? Likes? Boos? I'd love your thoughts on these systems. Any view is welcome. I just want people to know my thought process when I design a game system, and why certain choices were made.
When you play a solitaire game, in a sense you're playing a system that never changes. It's like playing a computer game against an AI that only has one strategy. When you play a game with people, good games have rules that let the players develop effective strategies - some that the game designers themselves would never have thought of!
In a solitaire game, this effect would represent a pinnacle achievement.
So as designers, it becomes important to change the 'structure' of the game in order to make the experience different and fun. You can do this by changing the starting conditions every game (which Star Requiem: Humanity's Last Stand does), creating conditions so that while the games starts the same every time, choices or events early in the game have a striking effect in the way the game plays after several turns (Raid on St. Nazaire and RAF are classic examples of this) or by creating a unique narrative throughout the game that tells a story (Ambush!) or even rewriting a set starting point in history where you are trying to get to the same result but perhaps in a different way (Where There Is Discord: War in the South Atlantic is a good example of this).
So fine. Somehow, obviously you have to create a new experience for the gamer, or they're going to get bored. In SR, I use the concept of what I call 'asset variability'. You know the type of 'power gamer' in strategy games - they play Civ and they're COUNTING on getting Pottery by turn 7 and Chariots by turn 25 and Iron by turn 41 and so on... to players like this, the game has been reduced to an algorithm that can be predicted, planned, and ultimately beaten unless the AI rules are changed or bent to unnatural lengths. Is that fun? For some people, perhaps... but not for me.
So as an example, when pounding out the basic design of SR, I looked at shipyards. Now, shipyards are a classic concept in almost every good 4X game worth its salt. Regardless of whether they are on the ground or in space, shipyards build the ships of fiction lore. And that's fine. No need to reinvent the canon wheel.
BUT... how do we KNOW every scout will take 2 turns to build? Why does every battlecruiser have to take EXACTLY 7 turns? More specifically, how do we even know as gamers how long something takes to build? In the real world, we'd get an estimate, but I assure you there's a big gap between a delivery date before the first keel is laid and when the ship is ready to go to a fleet. But this has been an accepted convention of games like this because it's easier to plan for. If you know you're getting that supermegadreadnaught in 4 turns, well, why NOT stir up the enemy or taunt them into attacking your home world, knowing to the turn what will be waiting for them when they arrive?
Great, except... what happens if the post-yard trials fail?
In SR, I want players to be able to have a general idea of things. Research, builds, enemy locations, enemy strength, etc. But like in the real world, things change for better or worse, and those contingencies are sometimes the most fun to plan for! In SA, the shipyard system is one way to introduce variability. Each shipyard has a 'rating' of sorts for how effective they are in building size 1, 2, 3, and (if they can) size 4 ships. This is represented by a target # that the player has to roll each turn to see if the build progresses to the next phase. Think it as passing a systems trial; roll over the check, systems are good and move on to the next building phase, otherwise rework.
The thing that makes this strategic besides not knowing exactly how long it will take a ship to move through the build funnel (in one game, I had a scout which should take 3 turns take 9!) - you have to pay a CP (construction point) every turn the ship is on the track! This is above and beyond your initial start cost (representing congregation of materials, manpower, energy, etc), as it would be in real life!
Now throw in the fact that you draw your shipyards at random when you build a new one (and when you get your first one) and you begin to see the strategic ramifications of just one small system. What if I get a terrible shipyard for size 2 ships (CAs, BCs)? Do I just keep cranking out smaller ships that I can currently build more efficiently? Do I save up over numerous turns and try to build a more efficient shipyard? Do I try to research technology that will improve my build rate of size 2 ships (you can)? Or do I grit my teeth and hope my sad-sack mid-tier construction slips can move ships through with some good and lucky rolls? And if I do this, can I afford in time and CPs to keep ships on the slip turn after turn after turn while the Xyl roam ever closer to my core worlds?
That's strategy, and that's what I want. Tangible, consequential decisions as a result of variability and imperfect data. Hopefully SR will succeed in this goal.
So one thing that is critically important in any game, especially solitaire games, is the idea I call decision cost. In other words, how critical is any one decision to the players winning or losing? As an extreme example, a game like Tic Tac Toe has a very high decision cost - playing between two equally competent people, one single mistake will likely cost you the entire game. At the other end, a game like War in the Pacific (second edition) has literally hundreds and hundreds of decisions that are made, virtually none of which by themselves will win or lose a game.
Every good game must make a player decision mean something, especially in a solitaire game. In other words, what's the 'cost' of doing A vs. B? I have found the very best games extract some sort of tangible cost for each and every decision, no matter how small, that challenged the player and that when reflected upon was a clear reason for victory or defeat. So based on that, anytime I am considering a new feature for a game (which manifestly involves a decision/s for the player) will the scope of the game and the feature allow those decisions to 'cost' in a way that is consistently relevant strategically? If the answer is no, then it's probably not worth adding.
Example. In my currently being designed game Star Requiem: Humanity's Last Stand, a major part of the game (especially early on) is for the human player to set a broad strategy: research more and better weapons and ships, or focus on industrial might to churn out more ships and defenses early? You can have both, but neither will be overly effective - the game does not really reward a 'down the middle' strategy unless other decisions elsewhere are made.
The game plays out very differently depending on the player's decision (high decision effect) but what is important is that either way can lead to a victory (low decision cost - potentially) IF the player plays their strategy well (and with some luck, of course!) Solitaire games can fall into the terrible trap of being 'one way to win' since the AI is in most cases predictable, and designers must provide additional 'doors' for the player to walk through and explore. It's a tricky balance, though, especially if one door is overpowered or underpowered (or meaningless). I believe designing an exciting, replayable solitaire strategy game is just about the hardest thing a game designer can do... which is why I love the challenge of trying!
Wed Mar 14, 2012 10:30 am
Well, Star Requiem: Humanity's Last Stand is well on its way to becoming a viable entity on BGG. It's been a long journey from beginning to now. My journey in board game design actually goes all the way back to my 6th grade year, when I was designing dungeon games for 50 cents a pop. They were pretty cool, actually - you moved to a room, then opened up a 'door' behind the room to let in the light and reveal what was in the room, like a key, a monster, or treasure. You either escaped the dungeon or you died trying. Hey, it was 1988.
Moving forward, I have been fascinated by wargames, specifically solitaire games. Some of my favorites include Raid on St. Nazaire, Ambush!, Patton's Best, and more recently Hornet Leader II, which really started to pique my interest in the whole 'one person can design a game now' if it was print and play!
Falcon Ace was my first attempt at a solitaire game. I actually wanted to take a more detailed version of a flight game and actually get 'in the cockpit' but not quite at the level of exact speed and maneuvers. So FA was designed as a hybrid RPG/tactical game of air combat. It was detailed as hell, it took me almost 2 years to balance and get all the bugs out of the rules, and it was redesigned visually at least 5 times. It was, and probably remains, the most accurate and detailed single-player flight game ever created.
And I consider it a colossal failure.
Not because it's not a fun game, per se. If you have to time to get into it and really learn the game, it can be an amazing game. But therein lies the problem. I was designing a game for a subset of gamers so tiny as to be almost invisible - namely me. I wanted a game where I could designate my pilot's skills and wingmen. I wanted to be able to arm my plane and buy better weapons. Add fuel, an ECM pod, weapon load ratings, plan my mission and range, do advance scouting for SAMS, and create a living battlefield where targets would move and act without my input, giving me the freedom to engage or not (or to order my wingmen to do it!) And for the most part, it was a success... once you got through the 40 page, Avalon-Hill-style 6 point manual. And nobody was going to fight through all that just to maybe get some fun out of the game. Lessons learned:
1) Make your rules simpler!
2) Make your counters bigger!
3) It's OK to not offer the player every option under the sun.
4) Feature creep exists, and it is malevolent.
5) Be true to the shining core of your game.
This last one led me astray. I started with a modest ambition - to create a top-down flight combat game - and ended up with War In The Pacific of solitaire games. I kept adding THIS idea and THAT idea and DOESN'T THAT SOUND COOL idea... until the game had lost its way.
So with SR:HLS I'm trying to streamline. There is ONE enemy. You must kill your enemy or it will kill you. There are a limited amount of options - but they are all important and will make the game play differently. The rules will be written in conversational format with lots of pictures. The counters will be 3/4 inch at least. Etc, etc.
Have I learned my lesson(s)? We'll soon find out!