Game design musing: It’s about time (Part I)

Gear up, spelunkers! It’s time for a dizzied descent into the dankest depths of game-design geekdom.

In a very old blog post I briefly pondered the topic of action pacing – especially combat pacing – in RPGs.

Below are some thoughts on how three major game systems tackle the topic. A caution in advance: while I know my GURPS, please accept my apologies where I mangle HERO; it’s been a long time since I last played. And I really risk disservice to D&D, as my only familiarity with 3e rules is from perusing the books, not actual play. Corrections to my text are greatly welcomed.

Timing Basics

In the three game systems, a character’s chosen combat action is typically resolved on the spot, and then is “over”, although the character’s turn effectively continues to run until the start of his next turn. That is, whether or not the given system calls his turn “over”, the character is still active in an effective “post-action” portion of his turn and may be called upon to take defensive action, or otherwise respond to surrounding happenings. Typically, the options open to a character in this “post-action” portion are affected by his preceding choice of action.

That’s a simplification, but I believe describes the basis of the three systems. On to more specifics of the three:

Visitors to this site are probably very familiar with GURPS play. GURPS uses one-second turns (and to properly understand those turns, think of a character’s turn broadly as a second of time alloted to the character, not narrowly as just an attack).

HERO has a pretty unique system, dividing a 12-second round into one-second phases. A character gets a number of actions during that round equal to his Speed. As the phases go by, he acts on given phases specified by his Speed, ending up with his alloted number of actions spaced out (more or less) evenly over the round. If you’ve run Car Wars‘ system for vehicle movement timing, you know how HERO‘s system for character action timing works.

Per my simple understanding of D&D 3e, the system uses six-second rounds. (My main D&D play experience was back in the days of one-minute (!) rounds, always a topic of intense debate.)

Simultaneity and Rounds

GURPS character turns are not simultaneous, and are said to “overlap”. Character A starts his turn, then Character B starts his, then Character C. With each change of turn from one character to the next, a fraction of a second has elapsed. That fraction isn’t specified, but what we can say for sure is that when the order comes back around to Character A, a full second has elapsed since the start of his previous turn.

A cycle of GURPS turns (i.e., each combatant taking one turn) does not form a “round” (though I see no particular harm, nor benefit, in applying the term). Thus, there’s no start of “a new round”, and no shuffling of turn order. (It’s a bad idea to shuffle: see GURPS House Rules Best Left Homeless.)

D&D uses explicit rounds, each round encompassing one turn for each combatant. Turn order is typically shuffled at the start of each round (or at least it was back in my day!). But at the risk of playing down any difference with GURPS, I don’t think the turns can be labeled “simultaneous”. Clearly, plenty of time passes between the first character’s attack and the tenth’s attack (by which I mean time in the game, not the table time that claimed the third bag of Cheetos). And although the round officially “ends” after all turns have been taken, I believe the actions chosen by each character on his turn will continue to affect his available “post-action” options during the new round, until his next turn rolls around.

In other words, although some GURPS players make a big deal over its “overlapping turns”, I’m not sure there’s a real difference in practice from D&D turns, or much significance to D&D‘s “rounds” (beyond bookkeeping) – if the D&D turns aren’t shuffled at the start of a new round. (And shuffling turn order sounds like a bad idea in D&D for the same reasons it’s bad in GURPS, such as a slow character getting two turns in a row and outrunning a fast one.) But, I’m talking past my knowledge; I welcome enlightenment from D&D players.

HERO turns overlap in a very different way from the other two systems. This is the only of the three systems (perhaps one of very few in the world of RPGs) that varies turn length itself by character. Thus, Character A may have two or more full turns for every turn of Character B. Or he may have four turns for every three of Character B. Any ratio A/B is possible, as long as A and B are integers between 1 and 12.

There’s no shuffling of turn order with new HERO rounds – again, character Speed rigidly determines timing and length of turns. Thus, I don’t believe the end of a twelve-phase HERO round has any special significance, other than as a bookkeeping measure.

Wait… Hold still so I can hit you…

Let’s take a look at slower-than-usual attacks. GURPS allows some slow weapons to strike only every other turn, creating an interesting new attack frequency that’s much slower than typical. However, I wouldn’t quite label this a change in attack speed itself. There’s a requirement that a slow Ready maneuver precede the attack, but once readied, the “slow” weapon strikes as quickly as any other.

GURPS‘s All-Out-Attack option for extra damage or a TH bonus could be seen as a “slower” attack, if you see it as denying defense by “taking up the whole turn”. That’s not a very satisfactory explanation, though, as the attack itself occurs no later than a regular attack would have. However you may wish to describe the action in terms of game color, it mechanically simulates an attack that happens as quickly as any other, but whose lengthy post-attack follow-up or re-readying disallows defense later in the turn.

I’m not aware of any HERO and D&D options for unusually slow actions. In short – and correct me if I’m wrong – it appears that none of the three systems offers common options for true slower-than-normal attacks (again, other than unusual readying time in GURPS, which is really a separate thing from a slow attack itself).

Whoosh whoosh whoosh

One of the reasons the games give short shrift to slow attacks is, of course, that slow or delayed attacks aren’t fun. (Even GURPS seriously sped up its slow-readying weapons in 4e.) Every combat geek will do almost anything for just one more attack. With that in mind, let’s head to the Matrixy motion-blurred side of things.

All three of the games allow multiple attacks per turn. Starting with GURPS, we have All-Out Attack for multiple attacks (or a Feint + Attack combination), as well as the difficult Rapid Strike action for multiple attacks that still allow later defense. (GURPS 3e had a separate option for automatic, penalty-free extra attacks with high skill. I was not a fan, and do not mourn the rule’s disappearance.)

D&D awards multiple attacks automatically with higher character level, and special extra attacks with certain double-ended weapons. These multiple attacks require the “Full Attack” option, limiting movement, but are otherwise easy to employ. The game also awards free “attacks of opportunity” in many circumstances.

I’m not sure what unique options HERO awards; maybe nothing. But I believe that all three systems, HERO included, allow extra attacks for mundane circumstances such as using the off-hand. In short, if you want to fit an extra attack or few into your turn, all of the systems will accommodate you to some degree.


That brings us to the subject of how attacks interleave – that is, in what order they take place. When attack frequencies vary from the one-per-turn base, it’s an important consideration.

As expected, all of the games interleave attacks that take place at the standard one-per-turn pace. HERO does this explicitly through its varied turn lengths, whereas in GURPS and D&D, one-per-turn attacks will be neatly interleaved in “you-go-I-go” fashion (or less than neatly in D&D if turn order gets shuffled with each round). No surprises there.

Further, all three systems interleave attacks that take place at a slower-than-standard pace, through the simple means of allowing (or at times requiring) non-attack actions. If I sometimes break from my one-per-turn attack pace to make a readying action, to spend a full turn on defense, or to run around for a full turn, then I’ll make Y attacks per your X attacks over a given length of time. That naturally adds spice to combat. All is good.

But when it comes to the faster-than-standard pace of multiple attacks per turn, all three systems (unless I’m again wrong about HEROfollow the “all at once” model. No more interleaving: if you and I each have three attacks to throw during our turns, you’ll make your three all at once, then on my turn I’ll make my three all at once.

In other words, as soon as the action shifts to faster-than-standard “multiple attacks”, all three systems shift to an abstract attack pacing in which attacks no longer interleave. The word “multiple” itself points to the break in handling: two attacks in two turns is not considered “multiple”, while two attacks in one turn is. It’s an interesting quirk of RPG design. (It’s also wildly artificial from a realism standpoint, but I’m not criticizing; it’s difficult to make this factor both realistic and playable.) 

Weapons and varied attack speeds

What’s faster, a knife or a two-handed sword? The former in real life, of course. But how about in the game?

As mentioned above, HERO is the only of the three systems with explicitly varied turn lengths. But that varies by character, not inherently by weapon variable (such as mass or length). To the best of my recollection, little knives and heavy halberds attack at the same speed in HERO, unless the latest version and/or Fantasy Hero inject a difference in some way.

GURPS hands every character the same one-second turn, though the Ready requirement for some weapons nicely adds variety to attack frequencies. (It’s not fine-grained variety, by any means: your attack frequency is either one attack every other turn with the slow weapons, or a base one per turn with everything else. There’s sort of an in-between speed, represented by unbalanced weapons that can attack once per turn but not parry on those turns, though I’m told that this is an odd feature of the skills used, not the weapons per se.) Otherwise, there’s no differentiator for attack speeds. (There is an interesting differentiator in parry speeds for certain light weapons used with certain skills, as represented by the lessened penalty for multiple parries.)

In D&D, I believe there’s nothing at all to vary character turn length or weapon-dependent attack speeds.

In short, there’s little or nothing to differentiate attack speeds in the three games, other than GURPS‘ ready requirement for some weapons. However many attacks you get on your turn, you get that number whether you’re wielding a wee dirk or a hefty claymore.

Where this is going

It’s about time we considered that. This is heading to – you guessed it – some gearhead thoughts on tweaks to the above and alternate approaches. Come back for Part II in a few days. If you’ve got the time.

Part II
Part III


  • Kuroshima

    IIRC (and I’m no D&D Guru), rounds are no longer randomized in D&D. However, IIRC (and I’m even more likely to be wrong here, or simply expressing a local house rule) the rounds DO have a specific significance: in order to cast a spell, you must succeed in a concentration check determined by the amount of damage you suffered since the start of the round. This means that the round “border” creates a discontinuity.

    • tbone

      I don’t know; a reading of the D&D rules doesn’t make things crystal clear for me, but my impression is that the round border doesn’t affect things here. Taking damage during casting appears to be an issue for castings that take 1 full round (or longer), which, if I understand this correctly, is the equivalent of a GURPS spell requiring a Concentrate action (instead of instant effect).

      D&D full-round casting time runs from the start of the caster’s turn in which he announces and begins the casting, until the instant before the start of his turn in the next round. It appears, then, that all wounds taken after the start of casting, whether by foes acting after the caster “this turn” or foes acting before the caster “next turn”, will force the concentration check. I don’t see the round border affecting anything.

      Then again, that’s pure rulebook-reading, not play experience!

  • dar

    It occurred to me that it would be appropriate for players to all lay down a card or token stating which maneuver they are going to take this round. All at once. Then resolve things in normal order. It would signal that they should be thinking of what next and it would better suit the quick rounds of GURPS. I’m just not sure how unwieldy it would get if the GM had to do the same for the NPCs and antagonists. Anticipating part II.

    • Kuroshima

      Personally, I don’t like that idea. not only it creates a discontinuity in the round boundary (something that greatly damages my suspension my disbelief) but it creates a sort of declaration phase. Consider in GURPS, that you see someone at the exact distance necessary to reach them with an all out attack, and so you declare. However, since they move first, they take a step back, and ready their weapon, you however, are now defenceless even if you didn’t get the benefits of AOA…

      This should go to the home rules best left homeless IMHO

    • tbone

      I think you and Kuroshima are both right. No, I’m not being wishy-washy.

      Simultaneous action is something I’ve been playing with. Setting aside practical considerations (“does it play well?”), simultaneous actions (or declaration of actions) is a perfectly sensible thing to allow in a combat sim, as it of course can happen in real life. In fact, that’s a (minor!) failing of typical RPG combat sims, in that the rules simply don’t allow for two combatants (two crazed or dumb combatants, perhaps : ) to go at each other simultaneously, and run each other through at the same instant.

      BUT, there are some special ingredients needed to make such a thing work. One, likely, is the ability for a character to perform two or more actions simultaneously, such as twisting aside (i.e., dodging) while attacking. Two damn-the-torpedoes swordsmen could each try to skewer the other guy while avoiding his stab – doubtless a difficult thing to do, and very dangerous for each, but it could be attempted.

      Another is the addition of an “abort” action, so that when two hotheads jump at each other at once, one could lose the “game of chicken” and revert to panicked defense instead. Again, it’d no doubt make for a difficult defense, but might be smarter than the above crazed move.

      And despite cool capability for simultaneous shish-kebabing under the rules, a sword battle still needs to allow for plain old back-and-forth thrust-and-parry. That doesn’t necessarily mean a return to ABAB turns; simultaneous action could still work, but again, with the right ingredients. It may require tracking strikes and re-readying as two separate parts of any melee attack, like GURPS does for halberds (but not swords).

      Picture it under simultaneous actions. Fighter A makes a thrust, while B (more or less) simultaneously parries it. Done. Then we simultaneously declare actions again, but this time, A can’t thrust, he’s still “extended”. He needs to pull back, while B is free to take some sort of swipe (we assume his parry left him in reasonable position to do so). So A will have to parry that while withdrawing his extended arm (as above, essentially two actions at once: pull back arm + parry).

      On to the next actions. After his parry, A is presumably able to launch an attack from his position. B is “extended” and can’t attack; he has to withdraw while parrying. And so on, back and forth, clang-clash-clang in classic swordfight style.

      All very interesting stuff – as a thought experiment that I haven’t seen yet brought to live play. Whether it could work, or whether the minutiae and bookkeeping would kill a real game, are big questions, and that’s where Kuroshima’s concerns may win out. (Or similar concerns; I don’t see any problem with “round boundary”, just overall futziness.)

      As always, that was longer than I expected. : ) I’ll try to expand/expound a bit more in the upcoming second part of the article.

  • dar

    To Kuroshima: I see what you mean, though. Still, I think that it could be a feature. The slower player would be mad to take that manuever. But then again your right, it’ll beg for another tweak, like saying the faster player has to take bieng attacked anyway, bleh. Or the slower player has to take the dumb manuever anyway, bleh. Still, should the slower player get the benifit of observing the faster player and changing plans, as in the normal rules?

    Maybe they declair in order?

    Never mind.

  • Juballa

    In the section titled “Wait… Hold still so I can hit you…” you write:

    I’m not aware of any HERO and D&D options for unusually slow actions.

    Then in the section titled “Interleaving” you write:

    Further, all three systems interleave attacks which take place at a slower pace[….]

    The second sentence certainly implies a contradiction with the first. As I said, a minor thing, but it made me page back and forth a couple times.

    • tbone

      An article of this sort really calls for some clarification of vocabulary first, which, to the reader’s dismay, I haven’t done. Might be something for interested parties to hammer out.

      I think my point is on solid soil, but needs some clarification. (And in all of this, it’s actually a little hard to peg HERO, with its unique turn system. The best I can do is pigeonhole it where it seems to fit, and add notes as necessary.)

      I’m not aware of any HERO and D&D options for unusually slow actions.

      Here I refer to an attack (and I should have said “attack”, not “action”) which itself requires more time than does the standard “one per turn” attack. An example would be a rule that has a heavy polearm swing… very… slowly, and spend two turns to make an attack: the first slowly building the swing, and the second completing the swing and hitting. As far as I know, there’s no such thing in HERO or D&D. (HERO will have one character taking slower actions than another, but it has nothing to do with dagger vs polearm use, or otherwise adjusting the time required for a given action by a single character.)

      GURPS sort of has this type of “slow attack”: the polearm will require a Ready action, reducing its attack frequency. But that readying isn’t the attack itself. You can ready and hold, and from that position, swing the polearm as quickly as the dagger. So if you ask me whether a polearm is slower than a dagger in GURPS, I’d have to respond with a deeply meaningful “kind of”.

      Further, all three systems interleave attacks which take place at a slower pace[….]

      I think the key here is the reference to slower pace of attacks, not attacks that are themselves slower. Again, my use of the vocabulary may not be up to the needs of clarity. If I may babble more:

      For all three systems, the only significant means I know of for a weapon to attack at a slower frequency than one-per-turn is for the fighter to simply not attack on some turns. GURPS enforces that for users of certain heavy weapons, which adds a little realistic detail. I believe that HERO and D&D don’t enforce any such thing for heavy weapons.

      But all the systems freely let you spend turns running, or defending, or taking some long action, or staggering stunned, or just hanging loose, without attacking. By taking actions other than attacking, at the end of 10 turns, you may have have made 4 attacks and me 3 attacks, and they’ll be interleaved realistically in whatever order they occurred in.

      (Yes, that’s blatantly obvious stuff; the intent is only to contrast that with actions that take place faster than one-per-turn, where you’d take your 4 multiple attacks all at once, then I’d take my 3 all at once.)

      Anyway, I don’t think there’s a contradiction. Let me see whether I can state it more clearly: For most weapons and attacks, the games don’t model an attack itself requiring more time than the standard one-per-turn attack time (as in a weapon that swings… really.. slowly). Attacks can of course be launched at a pace slower than one every turn; such attacks by multiple fighters will interleave in logical “it happens when it happens” manner. But at a pace faster than one every turn (“multiple attacks”), attacks shift to the “all at once” model.

      Which is all a way of saying “games fit actions into ‘turns’ “. Hmm, I wonder whether I can think of an even more obvious point for next installment. : )

  • Juballa

    I know I’m kinda late to the party, but I just thought I’d throw this into the mix.

    AD&D 2nd edition did have a mechanic for varying the attack speed by weapon type. The weapons table in the Player’s Handbook (pp 68-69) includes a Speed Factor for each weapon, ranging from 2 for a dagger to 13 for a pike. IIRC, this number was added to your rolled initiative, with the lowest number going first.

    Additionally, missile weapons had a Rate of Fire statistic, listing the number of shots your could achieve in a combat round. These ranged from 1/3 for an arquebus to 3/1 for a dart.

    I seem to recall a Dragon magazine article back in the Stone (Paper?) Age that had an alternative combat system that took into account weapon speeds and lengths.

    • tbone

      Yep, I remember those AD&D 2e Speed Factors, and used them myself back in the day.

      But they are outside of my main topic. I’m yammering about attack speeds in terms of how much time it actually takes to make an attack, and how many attacks fit into a given time period/turn/round. Speed Factor, IIRC, only affects initiative, and initiative doesn’t interest me much. For typical RPG uses of initiative, however you determine it, it doesn’t change how long each subsequent attack takes or, or how many you get in a round/turn. In AD&D 2E, the dagger with its Speed Factor of 2 is likely to attack each round before the pike with SF 13, true, but at the end of 10 rounds each will have had opportunity to attack 10 times. That’s the attack speed I’m talking about, and Speed Factor doesn’t make the dagger any faster than the pike in that regard (that I recall).

      The Rate of Fire for ranged weapons that you mention does create a weapon-based difference in attack frequency. Thanks for the reminder. Though once again, I’m sure that multiple ranged attacks per turn fell under the abstract “all at once” model, rather than some more realistic model that allowed interleaving of your three dart throws with my three dart throws.

      Re the Dragon article you mention: Do you remember whether it only adjusted melee weapon initiative for weapon type, or actual attack frequency as well? The latter would be the kind of speed adjustments I’m talking about, which seem to be very rare in RPG melee rules.

    • Juballa

      tbone wrote:
      Re the Dragon article you mention: Do you remember whether it only adjusted melee weapon initiative for weapon type, or actual attack frequency as well? The latter would be the kind of speed adjustments I’m talking about, which seem to be very rare in RPG melee rules.

      Why rely on mere memory when I can just look through the stacks (and stacks, and…) of old magazines? I knew I kept them for a reason. Well, a reason other than making my wife crazy.

      Dragon Magazine, issue 71 (March 1983), “Who gets the first swing?” by Ronald Hall.

      It’s largely a modification of initiative for AD&D 1e, centered around determining a new set of modifiers to the traditional d6 initiative roll. You factor in weapon length, dexterity, size, experience, and strength to get modifiers for closing and in-range initiative rolls, as well as rules for charging, fending, retreating, and pressing.

      However, more to your point, there was also a system for replacing the standard ‘multiple attacks based on level’ that AD&D had.

      The article’s author wrote:
      If the figure’s AP number [modified initiative roll] is between 1 and 6 inclusive, one attack is allowed at the specified time (with results of 6 attacking first, 5 second, etc.).

      If an AP number is greater than 6, a second attack by that character or monster is possible. The first attack is executed ahead of any “1-6” results (see the preceding paragraph). Then 12 is subtracted from the initial roll, yielding a result of -5 or greater, and another d6 is rolled. If the sum of those two results is greater than zero, the number is then treated like any other AP roll and given its place in the sequence of attacks during that round. With this system, a third attack is theoretically possible for those with high AP modifiers (+7 or greater).

      • tbone

        Ah, old stacks of Dragon. I used to have some of those…

        Thanks for the Research roll, Juballa. Interesting, a mix of a heavily-modded Initiative roll, and a multiple-attack scheme derived from that. Essentially a system of the “action points” type, many examples of which I’ve seen.

        Initiative rules themselves don’t interest me so much. For any game, you can easily make an initiative system with as many factors as you like: start with some roll, and add mods for DX, unusual reflexes, IQ, readiness, encumbrance, everything and anything you think is necessary. You can hack a rule out in one minute. Roll the dice, and you know who “goes first”. And that’s all. At the end of n turns, every fighter will still have had a base n chances to attack.

        The multiple-attack tie-in is trickier stuff, and more interesting. I’ve written part of my subsequent article, and will add notes on action point systems.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.