PDA

View Full Version : So I'd Like a Way to Play Transistor on Tabletop...



JBPuffin
2014-06-04, 09:39 PM
Transistor (http://supergiantgames.com/index.php/transistor/) is this fricken' awesome game on the PS4. Essentially, the game's character fights with people's...souls, i guess, except they're action abilities. Here's the kicker: they're all also Passive Abilities AND Upgrades to existing ones. Now, I don't know of any other game that might manage this...so, what game can this be done easily with? Essentially, it needs to have Attacks/Combat-related Actions that can also be Passives and Ability Upgrades, but only one at a time...

For example, Crash is an area Attack that hits anything adjacent to you; it also can be Armor as a Passive, or add stun effects to most abilities (MOST; there are exceptions) as an Upgrade. Jaunt, on the other hand, is like a Dash Attack, except without damage and you can do it when you can't do anything else; as a Passive it gives you more time to use in your special "I'm planning" mode, and as an Upgrade it makes the Attack usable when you've got no plan time left (I loved this + the machine gun ability...ah, Ping, I love you so).

If possible, it can be a generic, malleable system; it that isn't, I just want this combat system, b/c it looks AWESOME *fanboy screech*. So, can someone hook a brother up without resorting to Fate?

DoomyDoom
2014-06-05, 05:27 AM
I would say it kind of depends on how much of the original system you are trying to preserve. Let's be honest, the game with realtime components (Turn() recovery and multiple over time effects) won't translate into tabletop all too well. Existing solutions as far as I know are:
a) split time into turns, which will result in having nested turns and other silly stuff like having Turn() on cooldown during your, uh, turn. Well, it's already weird, not even 2 sentences in.
b) aggregate results from multiple independent decisions. Well, not really independent, since people are going to describe them one by one... which will inevitably result in metagaming at least to some degree.

So I'll say take the ability system from the game, take a straight turn-based system + initiative, translate over time effects to turns somehow, rewrite some abilities (everything related to realtime mode), enjoy the combat. By "straight" I mean remove the Turn() recovery mechanic altogether and give some actual execution time to actions during Turn() as opposed to it being instant in the game.

Madeiner
2014-06-05, 08:00 AM
I only spent 2 hours in that game and i already wanted to create a system for d&d that resembled that.

I stopped very early as i wasn't satisfied.
However, here's what i thought:

IMC we have Materias, similar to FF7. They provide abilities, spells, effects.
I envisioned that each materia had one or two powers only, and each power could be used in different ways, but only one could be active at the same time.

So for example, you had Ice materia. Ice materia slows enemies.

You could "slot" it as an active abilitiy, so for example it was a spell that you could cast to deal damage and slow enemies.
Or you could "slot" it as passive in your weapon; every time you attack, you have a chance to slow enemies.
Or you could slot it in your armor; when you get hit, you have a chance to slow enemies.
Or you could slot it to a specific spell or ability of your class (say you are a sorcerer and like magic missile). When you use magic missile, you have a chance.. you know.

NichG
2014-06-05, 08:34 AM
So, here's a sort of top-down view of how I'd build this:

Upgrade system

For this to work well, the notion of making or receiving an 'attack' needs to be standardized and made very modular, so that its easy to see how things combine without explicitly writing out all 400 combinations

This can be done in part by judicious choice of the numerical aspects of the game, but some base rules need to be established to keep things from being too dominated by a few effects. Having a more 'continuous' type of numerics seems advantageous for a system like this.

To do this, lets separate things into 'attack vectors' - these are individual projectiles/strikes/etc created by the attacks. Each attack vector has a power scale associated with it, allowing us to do things like 'create 3 extra projectiles at half normal strength'. Every numerical effect of an overall attack will be scaled by the power rating.


Tactical grid

As much of the computer game centers around positioning and movement, attack ranges and movement speeds should be fairly limited, so that its not always possible to get into range for an attack or move to negate cover when your turn comes up. On a square grid, having attack ranges of 3-6 grid spaces and per-phase movement of 3 grid spaces (upgradeable with passives) might be a good first attempt. Some enemies might move faster or slower, so 3 gives us a reasonable range to play with to have sluggish (2 spaces) or speedy (4 spaces) foes.

Move+Attack is probably reasonable, with attacks whose time cost is greater than 4 requiring no movement that round. Move+Move probably should not be permitted to make maneuvering during between-Turn times remain very important.

One consequence of this is a sort of 'he's on my six!' situation where an enemy can chase you and shoot you every phase once they've maneuvered you into range, but they basically cannot effectively shoot you before then. Essentially, as long as you maneuver well, you avoid all attacks, but once you mess up then you might be subject to a whole series of attacks in quick succession. This is consistent with the in-game feel I think (given the mechanics for things like the Creeps, Young Ladies, etc).

Forced-movement attacks also become significant because movement is so expensive.


Timing

Much of the actual game centers around timing - using stun locks, masking, etc to buy time for your next Turn(). So a strict alternating turn-based system seems insufficient.

Instead, lets break the game up into 'phases'. Every phase, you gain 4 (maybe 3?) points of Turn Gauge at the start of your phase. The Turn Gauge saturates at 10 points. You can only use Turn() when maxed out. Effects can slow or speed up the rate at which you gain Turn Gauge, can increase the maximum or decrease the point at which you can use Turn().

When your Turn Gauge is not filled, some actions are disabled. We should probably be more permissive here than the actual game, so I'd propose that most basic actions stay enabled but lose the benefit of their upgrades for example.

Because of the finer-grained time, stun attacks become more reasonable than in other tabletop games. However these should probably be restricted to Turn() actions, not between-Turn() actions. Stun attacks versus an enemy would simply prevent it from acting for one phase; beefier enemies/bosses might have a stun gauge where they lose a phase of action for every 2 or 3 stun attacks delivered to them, to adjust for the fact that there's going to be a party worth of PCs and not just one protagonist.

Stun attacks against players should probably stop or reverse their Turn Gauge rather than deny them actions.


Turn

Once the Turn Gauge is filled, a player can choose to discharge it and make a series of attacks and movements all at once. Each movement has a cost of 2 points (so a player can move twice normal speed during Turn() if they don't attack), and most actions have a cost of 4 points, with big actions costing more and things like Ping costing much less. These can be modified by upgrades as usual.

One gimmick of the game was that your last action in Turn() could exceed your available time by a bit and it'd still be allowed (as long as it was less than something like 50% of the action cost), which I think is kind of neat and makes planning the order a little more important, so I think its reasonable to retain that.


Multiple Players

This is really the hardest thing to design for and predict. In Transistor, you had a single character who could take on swarms of enemies because Turn(), lots of stun-lock abilities, and better movement abilities gave enough of an advantage to outweigh the fact that the enemies were more numerous. If you retain those advantages across six PCs, then something needs to be adjusted to keep the number of enemies small enough that people can think about individual-level tactics, while also making sure the enemies don't just get stomped in three rounds.

Also, with multiple players having things like Turn() planning can bog down. You're basically doing two Turn()'s every phase on average in the system I've described so far. One solution may be that Turn() has to be taken as a party-wide thing, rather than on an individual level. The way that would work is that the party as a whole shares a Turn Gauge, every party member's phase increments it by 4, but it caps out at something like 10*# players. When Turn() is called, each player can plan a sequence of actions in a round-robin fashion where the players each do one thing at a time in sequence but the enemies can't act, and those individual player actions are taken out of the shared Turn pool. I have a feeling that might get frustrating if e.g. your Turn() is delayed by your ally getting hit with a stun lock or if a lot of the gauge is used up by the guy with Cull() - still, it may make it tactically more interesting.

My intuition is that this is the hardest part of the conversion to do in a way that feels satisfying.

JBPuffin
2014-06-05, 08:56 AM
Madeiner, NichG, I like the way you think. That makes things a whole lot simpler once I've seen it written out like that, so I might be able to make something out of that. Thanks!

banthesun
2014-06-06, 07:47 AM
People above have already said most of what I'd say, so all I've got to add is it might be useful to have little card tiles of varying lengths to represent the different moves. That way, when you need to plan your Turn() you can fit them together to save time and paper.