We play according to a homemade-ish freeform-ish system. I, as a GM, have a lot of fun with the system as I morph it. And the benefit for the players is that the whole thing is a one-pager.

But we have a problem: the turn dynamics. The problem is that I have two contradictory requirements for that sub-system.

  • Firstly, some actions should be playable very precisely. For example, an exchange with edged weapons should include at least several turns / "decision points" for both players. This also provides the chance to describe in great depth the life-or-death situations.

  • Secondly, there are other actions that take incomparably longer than the aforementioned. For example, a large crossbow could take a minute to crank. Leaving the reloading character out of the action is somewhat bearable. However, people get offended not knowing exactly when they are going to be able to act again.

Of course, I will talk to my players for some kind of compromise and what they really want (as those up there are my thoughts). However, a tested, mechanical solution, would be much better than making compromises with the current situation.

Basically, I want to steal the turn dynamics from some successful game. However this turn dynamic has to be highly flexible, allowing both very short and quite long actions to happen together. Basically, it should be able to switch from fast-pace to slow-pace and back. If possible, it should provide mechanisms for mixing those different paces.

  • \$\begingroup\$ What's up with the tags?? I remember the tag "system-design" existing here. I definitely do! \$\endgroup\$ – Vorac Dec 5 '12 at 10:25
  • \$\begingroup\$ Have a look at current fights: fencing, MMA, and boxing will do nicely for hand to hand combat. Reports on police fire fights and military lesson learned reports will give you a good idea of modern combat. You can then do some statistics to see how long a fight is, how long between hits, and the length of time between interesting/game changers exchanges. From personal experience and previous, in hand to hand, this would be order of seconds and for fire fights, it will be in the order of minutes. These times intervals would give you generic rounds. \$\endgroup\$ – Sardathrion - against SE abuse Dec 5 '12 at 11:23
  • \$\begingroup\$ @Sardathrion, do I understand you correctly in your proposition to define different types of "turns" e.g. close combat turn == 1s, ranged combat turn (crosbows, rifles; primarely sneaking) 1 minute etc. ? If so, maybe a useful simplification is that longer type of turn in infinitely longer than the shorter one, that is all action of the shorter turn type should complete before someone gets a longer turn type. Or am I completely off? \$\endgroup\$ – Vorac Dec 5 '12 at 13:23
  • \$\begingroup\$ You are right. This is why I would not model combat as a blow by blow account: it is unrealistic, unwieldy, and takes too long to run. I won't even go into critical/fumble statistics aberrations. I tend to have longer "round" of several minutes those result is a aggregate of attacks, defences, and counters. For example, a boxing match round (5 mins) could be three game rounds with specific modifiers for good/bad role play and descriptions. A gun fight could have a game round as long as 15/20 minutes or 1/3 minutes depending on where it is taking place. \$\endgroup\$ – Sardathrion - against SE abuse Dec 5 '12 at 14:29
  • \$\begingroup\$ My two comments are long enough to be an answer (with a little elaboration). However, as an answer it would not really answer the question. So, worth adding or not? \$\endgroup\$ – Sardathrion - against SE abuse Dec 5 '12 at 15:38

The Swedish "Mutant 2" game had a 5-second "combat turn" and 1-second combat phases. Typically, you'd end up using the combat phases exclusively and simply read "combat turn" as 'five seconds'.

The way it worked was that various actions took a different number of phases. That could be "fixed number of phases", "multiple of the character's speed" (a number that got lower the faster the character was, with 1 as the fastest possible and, um, something like 6 or 7 for the slowest) or "character speed plus a fixed offset".

The way this worked mechanically was that in the phase after a combatant completed an action, the next action was stated and the phase it would happen in computed, then that was noted down in the combat log. Multiple combatants stating actions at the same time do this in (numerically) decreasing speed rankings. You can then skip to the next phase that has any action in it.

It was possible to abandon an action part-way through, but there were penalties to that. I don't recall what the exact penalty was, but I think it was "start the next action in the following phase" and "no partial completion of the previous action".

  • \$\begingroup\$ I was going to answer something like this (though my model would have been The Riddle of Steel and The Burning Wheel). If the players are frustrated just not knowing when their long action will happen, put numbers on everything and make it unambiguous. \$\endgroup\$ – SevenSidedDie Dec 5 '12 at 18:05

A free-form system my friends and I made up for non-system-specific role-playing involved a "timeline" that tracked how long before players could take another action.

Basically, we had a meter long ruler with 100 centimeter marks and made each one worth one tenth of a second, then placed tokens for each player at the zero mark and would roll initiative. The first player would declare/resolve their action and the GM would determine how long it took, putting that token on that spot on the ruler. So if the player does something very complicated/time consuming, they would go many seconds up, and if they did something very quick, they might only go up a few tenths of a second.

Note: we used this for a near future setting with automatic firearms and were trying to accurately track things like firing 10-20 rounds a second with automatic weapons. Fantasy settings might be better served with full second tick marks.

Once everyone at zero had resolved their action, everyone would move down the timeline towards zero until someone reached zero. So if the nearest player to zero was at the 5 tenths mark, everyone would move their tokens down by 5 tenths (actually, we used a long floor area and just moved the ruler up until a token was at zero. Each time we ran out of space we flipped it around and went back the other way!). Any time there was a conflict (2 players reached zero at the same time) they would re-roll initiative (this had the added benefit of making higher initiative more useful than once per fight).

It had the downside of being "front loaded", that is, all the result of a long action happened the second the action was executed. So if a player declared they were unloading their SMG's entire clip on full auto, they'd basically roll all the attacks and damage right then, even though the action was ruled as taking, for example, 3 seconds.

We contemplated having a delcare-and-resolve system where you'd declare your action and only resolve it when the token reached zero, but it was hard to remember exactly what you declared, especially if you declared a long action and multiple short actions happened in between. You can try this system either way and decide what works best for you and your group.


The only option I've seen in published work was an optional expansion for 2e AD&D with Weapon Speed. Every weapon has a speed rating for how long it takes to point and shoot or swing and recover the weapon. The smaller the number the faster it is. A knife is speed 2, whereas a polearm is up in the 10-15 range (books aren't in front of me, this is my best recollection).

From my recollection, speed works in that you roll initiative (let's say 2). You can swing your knife on actions 2, 4, 6, 8, 10, etc. Your opponent rolls a 3 and can swing his polearm on 3, 13, 23, etc. I don't remember at what point you stop adding to your speed rating for your next attack and when you roll initiative for the next round, but as a starting point I've always thought this would work well for balancing weapon size and damage.


Burning Wheel has finely grained time in combat. If you decide to use the detailed combat subsystems for (ranged) Range & Cover and (melee) Fight, each divides combat down into volleys which are then broken down into actions.

The Fight system has the smallest time resolution. Each exchange of combat contains three volleys; each participant has a number of actions equal to their Reflexes which are evenly distributed across the volleys. The more complex the act, the more actions it takes up. A single action is on the order of a "heartbeat".

Effectively, unless you have a bow or crossbow ready to go at the start of a Fight, it's a bad idea. From rest, a hunting bow takes 5 actions to Nock and Draw and a final action to Release. That's likely an entire exchange's worth of actions for someone with a sword to be stabbing you. Don't bring a bow to a sword fight!


Check out Scion

Their Tick system breaks down combat in one-second intervals, with different actions having varying Speeds. It's White Wolf, so take everything with a grain of salt, but it might just be what you're looking for.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.