Iron Tower Studio ForumsRPGSystem Crash (Moderator: Gareth)System Crash design discussion
Pages: [1]   Go Down
Print
Author Topic: System Crash design discussion  (Read 3877 times)
Gareth
Moderator
*
Posts: 3268


Indubitably


View Profile WWW
« on: February 21, 2013, 10:18:15 am »

I'm going to make a separate thread from the feedback one to go into details discussing System Crash's design philosophy. I'll discuss why certain choices were made, how certain designs are intended to work (which may not be how they actually work in practice), where I think there are weaknesses, etc. I'm going to be totally honest, hopefully some of my reasoning doesn't cause people to sour on the game.

I'll start with certain points brought up by testers.


On Probability and Starting Hands.

One of the designs I took from MtG is the use of specific resources cards. In this case, Credit Chit cards. Playing one of these cards adds to your "pool" of credits, which is what you spend to buy other cards. Credit Pool is refilled at the start of your turn, you increase the pool size by playing Credit Chits.

There is a problem with this, as you can be a bit fucked if, through bad luck, your starting hand doesn't contain any Credit Chit cards. Generally, you want to make your deck around 33% Credit Chits, so your chance of drawing on is 1 in 3. But you can still have bad luck there.

There are designs that can avoid this issue. For example, some CCGs do away with specific resource cards. Instead, at the beginning of your turn you can sacrifice one of the normal cards in your hand for an extra resource.

The reason I didn't go with this, and this might anger some folk, is because it would complicate the AI.

I knew that AI would be a huge challenge in SC, and without multiplayer I had to have an AI that could give a decent challenge or everything else would be in vain. So I made some decisions with the design that make it easier for the AI to evaluate the state, and the possible actions it could take.

If I used the "discard a card to gain a resouce" design, the AI would have to make decisions about which cards to keep and which to sacrifice, per turn. It would need to have a good idea of what strategy it is working toward, a sense of the history of play so it knew which to expect and what the other players deck seemed to be focused on, so as to make intelligent decisions about which cards to keep and which are disposable.

With the specific resource card design, it just needs to evaluate whether it's got any Credit Chits in its hand, and then play them if it does.

That's the long and short of it.

There is another option. Separate resources and cards entirely, make it so that players start with credits, and they simply grow over time automatically. But I felt this was less satisfying and perhaps too predictable.

Based on the feedback I'm getting, I do think that I need to add in the option to Mulligan, for both the player and the AI. Mulligan gives the player an option to evaluate their first hand, and if it doesn't contain any Credits, to discard it and draw another starting hand. This would alleviate the frustration when you just have really bad luck, and reduce the odds that an AI opponent which should be a challenge is reduced to a cakewalk because of a similar bad starting draw.

On Deck Building Progression

To make introduce the element of "difficult choices" into SC, I made gaining new cards a 2-stage process. First, you "unlock" the card by beating certain story missions. This doesn't give you the card to use, it simply adds it to the Black Market, making it available for purchase. You have to use Credits, earned during the campaign, to purchase the card, which finally makes it available for deck building.

Two things are clear from the feedback :

First, that this process is not very clear to new players.

Second, that I am unlocking cards too slowly. Right now, you are often in the position where you have only one or two "available to purchase" cards in the Black Market, and you're earning money at a rate that will allow you to buy one new card every couple of missions. This means you're basically buying cards at the same rate you unlock them, which means no real interesting choices.

I need to up the rate of unlocking cards, while keeping the credit gain rate around the same. This will make it so that players have more choices, even if they are gaining new cards at around the same rate, one new card every couple of missions.

There is a final issue here. The starter deck isn't very exciting. Because of my small budget, I had to try be as efficient as possible with my art. In a normal CCG, you are limited to how many copies of a specific card you can include in your deck. In SC this number is 4. One of the ways I've reduced my art budget is by introducing the "Limitless" keyword. Limitless allows any number of copies of that card into your deck.

I use that keyword for certain cheap, generic cards. In MtG, there is a lot of card repeats. You may have a lot of different cards, all with different art, that are all mechanically the same, say a 2/2 creature with no abilities that costs 2. You need these to ensure that players have enough "stock" creatures to form a solid base to build a deck from. Limitless allows me to ensure that you have enough stock, workhorse cards to build a deck around, without using up art budget I need to ensure that there is a good selection of more interesting cards in the set.

There are Limitless cards at each "tier" of usefulness, up to medium cost cards. They are :

Very Cheap - "Innocent Bystander"
Cheap - "Neonmonger" & "Series D Blaster"
Medium  - "CorpSec" & "MetroSec"

This does mean that there are a lot of repeats in the starting deck. You have very few of the cooler cards to start off with.

It's a balance I struggled with, because I want to make progression exciting by dangling new cards to unlock in front of the player, and enough of those to give you interesting choices, without making the starting deck so boring that anyone who starts playing finds the experience lacking in all excitement and quits. I'm not sure I've nailed the balance.

Logged

“The power of accurate observation is commonly called cynicism by those who have not got it.” - George Bernard Shaw

My blog
GhanBuriGhan
*
Posts: 1636



View Profile
« Reply #1 on: February 24, 2013, 04:09:09 pm »

One thing I was wondering is whether you couldn't offer an option to get a bit more control over the deck. I can see a lot of balance problems, but at the same time I was thinking the performance of a deck is very dependent on the order of the cards, and some deck stratgies seem to hinge on having a few cards at the right time. So I was thinking about something like an ïnfiltrate"option: For a price,ahead of the mission  you can select e.g. two cards from your deck that go into a special stack. You pay e.g. 10% of the credits originally needed for unlocking the cards. You can then access these cards anytime during the duel. If you fail the duel, the credits are lost.
Whether this makes sense depends on what the penalty for loosing a mission is, of course. But I thought it could both alllow playing a bit more strategically, and creates an additional decision regarding what to spend your credits for.
Logged

------------------------------------------=====================---------------------------------------
"Merely killing those being mean to me. It's not my fault it's everyone in the world of AoD". (Vahhabyte)
mkbonde
*
Posts: 187



View Profile
« Reply #2 on: February 24, 2013, 05:00:20 pm »

Why 60 cards in a deck? Would 40 or 45 not suffice? That essentially cuts down on the number of cards you need art for, rather than making starting decks full of limitless filler cards.
Logged
Gareth
Moderator
*
Posts: 3268


Indubitably


View Profile WWW
« Reply #3 on: February 26, 2013, 05:17:02 am »

Quote
One thing I was wondering is whether you couldn't offer an option to get a bit more control over the deck. I can see a lot of balance problems, but at the same time I was thinking the performance of a deck is very dependent on the order of the cards, and some deck stratgies seem to hinge on having a few cards at the right time.

Mmm. There are two options there. The first is a meta-game element, as you mentioned. Those are difficult to balance though. At this point, I'm not adding another element to the game, though I might in future expansions.

The second is tools to achieve that within the current rules. You're right, having the right cards at the right time is a powerful advantage. Which is where event cards like Hired Gun and Weapon Stash come in.

What these cards do is basically say "if you have this card in your hand, you can substitute it with any card of type X that you have in your deck, if you can pay the price."

This is a way of manipulating the deck-draw probability. Normally, for a card you want that you have 4 copies of in your deck, you have a 4 in 60 (6.6%) chance of drawing. By introducing these substitution cards, you can have an effective probability of drawing that card of (4+4=8) in 60, or 13.2%.

The cost, of course, is that substitution cards take up slots in your deck which could be filled with more Agents or Modifiers or whatever. But still, it's a deck building choice. Higher chance of getting a card that you want (or can give you a card that you want) in exchange for a smaller range of options, and slightly lower resource efficiency (you have to pay first for the substitution card and then for the card you wanted to play).

There are other types of cards that support this kind of trick. In Magic you have cards that let you look at the top X cards of your deck and then order them in a way you want, so you guarantee the order of the next X cards you draw is to your liking. Or do it to the enemy, to screw them over.

System Crash doesn't have enough of these cards in the core set to make them REALLY viable. As I said, I had to make hard choices about what the core set would include. I will expand these kinds of options in expansion packs.

I'm looking at introducing manipulation cards in an Expansion that focuses on a plotline involving Multimedia Conglomerates and Reporters, as I think that theme of "manipulating public perception" fits the mechanic.

Quote
Why 60 cards in a deck? Would 40 or 45 not suffice? That essentially cuts down on the number of cards you need art for, rather than making starting decks full of limitless filler cards.

I tried having fewer cards in a deck, but certain other designs had a knock-on effect that caused that to be problematic. It's actually interesting, that chain of cause and effect in design decision making...

AI coding in a strategy game is really tricky. In a game like MtG, where you choose a set of cards to attack with and which ones to hold back in case an opponent launches their own attack, and the AI chooses which creatures to block, which to let through, which creatures to hold back so it can attack with them later, and whether there is any danger of a creature attack being buffed by a spell card after choosing to block would result in that creatures death, the decision space for combat is massive and complex.

I made a number of decisions to limit that complexity. Limiting playing cards to the phase before combat was one of them. Another was to use fixed slots and automatic attacks like Spectromancer. This dramatically reduces the number of decisions the AI needs to make about where to place Agents. It looks at the best place to play a card at the moment it plays that card, then the rest is mostly automated.

Fixed slots resulted in a limited number of slots per side. The addition of the Modifier display slots meant less screen real estate available and further constraints on number of slots. So I ended up with 4 slots for Agents.

Originally, the ratio of Attack to Health for Agents was lower. Agents had 3-5x the Health as they did Attack, for the average of that 'tier'.

But this had the effect of making the gameplay slow down to a drag. You'd quickly fill up your 4 slots, then sit with a hand full of Agents while you waited for a new slot to open up. You could help alleviate this by putting more modifiers and tactics in your deck, but this meant you'd often draw hands filled with nothing but Modifiers and Tactics. Not what I wanted.

So instead I upped the lethality of Agents. Now, they tend to kill each other in 2-3 blows, against an equivalent tier creature. Add in modifiers, and it's even quicker.

This solves the first problem. Now you have a high Agent turnover, which means you generally have a slot you can play something to. But this higher turnover had the effect of burning through your hand quicker. Now we another type of boredom. Now you'd get into the situation where you'd quickly empty your hand and spend the later part of the game drawing a card and then immediately playing it. Few interesting decisions to make.

So I upped the draw speed to 2 cards a round instead of the original rate of 1 per round.

Which means you burn through your deck quicker. Smaller deck sizes, you tend to run out of cards too soon unless you win decisively.

60 cards seem to be the best balance, and I think the game feels good as it stands. I could unravel the stack of design decisions to do it differently and maybe reach a different conclusion, but that would require a lot of time.

It's not quite as bad as it seems. About 20 of those will be credit chits, and of the 40 cards left, you will tend to have (on average) 3 copies of any unique cards in your deck. So around 14 unique non-credit cards per deck.

There are 57 cards in the core set, with 5 distinct strategies (and possibly others that people will discover). I figure this distribution gives enough distinct art per deck (just).

Logged

“The power of accurate observation is commonly called cynicism by those who have not got it.” - George Bernard Shaw

My blog
GhanBuriGhan
*
Posts: 1636



View Profile
« Reply #4 on: February 27, 2013, 07:58:48 am »

Yes I have meanwhile discovered these special cards in the SenseSim duels. Of course getting them is still subject to the same deck-order related randomness, but it does address the idea to an extent.
 
I fully understand focusing on the game as is now, and not adding new features. For future expansions (hoping you have the success needed to justify them), I would however encourage you to play and bend the rules in the campaigns occasionally to enhance the storytelling aspect. Different genre, but I'll again bring up puzzle quest here - I liked how they added different variations of puzzles for various aspects of the game there, such as training mounts (time limit per move), forging items (reaching specific goals of gems) or capturing enemies (solving static puzzles).
I could imagine you could similarly have missions that are time based (number of rounds to win limited) e.g. "get in and out before the alarm goes off". Require special scores (minimum of 10 points through hacking). Exclude certain cards (no hacking on this mission). Survival / retreat (survive x number of rounds) etc. - to spice things up.
« Last Edit: February 27, 2013, 08:32:54 am by GhanBuriGhan » Logged

------------------------------------------=====================---------------------------------------
"Merely killing those being mean to me. It's not my fault it's everyone in the world of AoD". (Vahhabyte)
Pages: [1]   Go Up
Print
Jump to: