Event-based timing (as opposed to turn-based timing)

[tex]

For my own roleplaying system (Orp, http://zuggers.org/orp.pdf, CC licensed, see the pdf for details), I was hemming and hawing about how to write the turn mechanics.  I wanted to capture a lot of realism, but avoid the usual cruft and bother that piles up around turn rules (d20 has its bestiary of action types, for example, while Shadowrun has that wacky countdown mechanic).

I thought about it a lot and suddenly a crazy idea hit me:  Why not just let everybody go at once, and do whatever they want, subject only to the limitations of what human-like individuals can do?

So I wrote these "event-based" rules, basically recounted here using the canonical medeival fantasy combat situation:

0) Time is kept using seconds.  Events are divided up discretely as beginning on a given second, and finishing at some second in the future, and generally don't take effect until the end of their duration.  One good example is a mage casting a spell:  The mage decides to cast a spell on some second, and starts chanting, and then finishes at some second in the future, when the spell will then take effect.

1) At the beginning of combat, all the PCs and NPCs independently decide what they want to do (if they want, they can do nothing for a little while and wait and see what develops).  This decision happens in the instant the characters realize combat is imminent, and represents the character committing to the action in question.

2) The GM decides on how long it'll take for each character to accomplish what they want (usually, secretly, and randomly:  d12 seconds for most actions in the typical chaotic combat environ).

3) The GM describes what occurs as the first few seconds expire and events begin to unfold:  Characters draw swords and approach enemies, begin casting spells, start to take aim at eachother with missile weapons, and so on.

4) The GM ticks off enough seconds to cause the first event to conclude (say, a mage finishes an incantation and casts a spell).  The character can now pick a new action with, probably, a new random duration.

5) The GM repeats this process until the combat is concluded:  The next action resolves, and the acting character may choose a new action.

So, that's it, basically:  Stuff happens "when it happens".  There are more details (such as a mechanism for allowing characters to interrupt eachother and themselves), but this is the essence of it.

Obviously this is a really weird approach.  I'm concerned that it may be hard to learn, and also too cumbersome at the gaming table.  Well, I feel like I could GM this, but I thought of it, so of course it seems very natural and intuitive to me.  : P

[J. Scott Timmerman]

I think it's a great idea.  Both Exalted 2nd Edition, and my system, ERA d6 do something similar.  I sympathize with the fear that it may be hard to make it un-cumbersome, and have had some criticism about that here on the Forge.  A lot of effort has been spent on my part trying to get that fixed in my games.

Combat in Exalted 2nd Edition is measured in seconds (or minutes, in special cases), and every action takes a number of seconds.  Initiative for purposes of Exalted is actually an action itself (i.e., the action to "get oriented" on the situation), and the number of seconds this takes is rolled (0 to 6 seconds).  After that, pretty much all actions are a set amount of seconds, based either on the type of action or sometimes on weapon statistics, and battle advances to the second where the next character gets to do the next thing.

I've found that Exalted 2nd Edition is hard for some players to wrap their heads around.  An option called the "Battle Wheel" that a fan came up with seems to explain it a little better; but there are other cumbersome aspects to Exalted combat as well that are hard to work around.

In ERA d6, the roll for the success of the action and the time the action takes are subsumed into one roll.  Again, like your system, all participants determine their actions "simultaneously," then they roll the time it would take them to complete the action successfully.  The mechanic you use to "interrupt" actions is actually built into the resolution mechanic in ERA d6.  Seconds aren't "ticked off" by the GM; they just happen as an effect of the mechanic.

Another problem I've had is simply breaking people out of turns.  I've found that players that are used to turn-based combat don't like Exalted ticks.  Originally, I had called the spaces of time between events (for instance, the completion of an action would be an event) in ERA d6 "Plays", but I ended up going with "Rounds" due to popular demand (even though they seem to work more like plays in American Football).

Of course, each of these two games have their own goals in using their respective systems.  When you say realism, are you referring to the feeling of randomness of a battle?  If that's the case, I think the d12 might work there.

In avoiding the "cruft and bother", do you mean doing away with the way D&D expresses time as a resource as broken down into action types which have different uses, and instead using a number of seconds for each action?  If so, will you grant a character the ability to perform more than one action simultaneously?  I think I already have an idea of how you might handle movement in the system you described, but how were you thinking of doing it?

I know I'm repeating myself here, but my main point is, more than being cumbersome, I've had trouble with players' fear of, or at least distaste for, the unknown.  Then again, I'm working with players that don't often move far from d20 without a push.  I hope you have a broader group to work with, though.

-Jason T.

[tex]

Wow, my very first foray into The Forge and I'm already getting interesting, insightful feedback.  Thanks very much.

Thanks for the heads up about Exalted.  Maybe I'll buy it now to see what the specifics are.  Is your game, ERA d6, published anywhere?  I'd like to look at it.

Another problem I've had is simply breaking people out of turns.  I've found that players that are used to turn-based combat don't like Exalted ticks.  Originally, I had called the spaces of time between events (for instance, the completion of an action would be an event) in ERA d6 "Plays", but I ended up going with "Rounds" due to popular demand (even though they seem to work more like plays in American Football).

The word I use in my game is actually "pulse", and its meaning is only connected to a second of real time as a convention or suggestion, so that the GM can have some flexibility for reasons of poetic license or mechanical convenience.  (Maybe during a chase scene with sailing ships, for instance, a minute or even an hour is a better "smallest" unit of time.)

Of course, each of these two games have their own goals in using their respective systems.  When you say realism, are you referring to the feeling of randomness of a battle?  If that's the case, I think the d12 might work there.

That was certainly part of it.  I think this kind of system also fixes a number of really weird realism problems that turn structures raise, that folks are so used to accepting that they don't think about it.

In avoiding the "cruft and bother", do you mean doing away with the way D&D expresses time as a resource as broken down into action types which have different uses, and instead using a number of seconds for each action?

Yes, exactly.  d20 has something like six or seven action types, and that's not counting iterative attacks, which are effectively a special kind of way to take extra actions.  I'd rather have one kind of action that can be used for anything.

I want my system to be lightweight by default, so I don't really define a bunch of different actions and how long they ought to take, but I do leave the possibility open for people who want to adapt Orp to more precise playstyles (I also want the system to be adaptable to a wide variety of needs).  In the base game though, I mostly leave it in the hands of the GM, with d12 pulses as the default suggestion.

If so, will you grant a character the ability to perform more than one action simultaneously?  I think I already have an idea of how you might handle movement in the system you described, but how were you thinking of doing it?

Yes, characters can either rush through actions in rapid succession, or truly multitask if they wish.  My tentative plan is to penalize each additional activity by -1 per additional action.  But I could see moving this up to -2, given how bad human beings seem to be at "true" multitasking.  The example in the rulebook is a ninja running along a rooftop and throwing shuriken concurrently.  : ]

As to movement, I'll be honest and say I haven't really pinned down rules for that yet.  I'll tell you the vague ideas I have at the moment, though.  First of all, assume there's a standard rate of travel for human beings; 3 m / sec is pretty close to the d20 "double move" so let's use that for now.

I want to avoid letting players drive their characters around on a second-by-second basis; instead, I'd rather pin them down on their characters' intent.  People usually don't wander around for the fun of it, they always have a purpose (which may be "approach that guy," for example).  So, there's now a definite distance, which means a fixed duration:  One second for each trio of meters.  Now movement can be just another kind of event, with a duration and an effect, though in this case no die roll by default. 

I think the rest of the game system can now be used to emulate most of the stuff commonly imagined in action scenes when it comes to movement and position.  For instance, I can imagine ways to account for moving faster at need, catching up to retreating characters, preventing other characters from moving through one's own area, and so on.

Thanks for asking about this, I've been putting off thinking about it for awhile.  : P  How does ERA d6 work?

I know I'm repeating myself here, but my main point is, more than being cumbersome, I've had trouble with players' fear of, or at least distaste for, the unknown.  Then again, I'm working with players that don't often move far from d20 without a push.  I hope you have a broader group to work with, though.

Unfortunately I've mostly gamed with people for the last few years that don't really want to play anything but d20, specifically D&D.  I'm not hostile to d20 per se, but I will say such a large degree of homogenization in the community tends to leave people less interested in other directions.

[Vulpinoid]

I hate reading these threads, because they always make me go back to my own game with a critical eye.

I keep thinking, That is a really cool idea, how can I incorporate in into what I'm doing...

Now I'm reassessing concepts of initiative and turn sequence.

V

[Darcy Burgess]

Hi tex,

FIrst off, welcome to the Forge.

Second, I prefer to use my real name here.  It'd be awesome to be able to do the same with you.

Third, I have a suggestion for handling the "counting" of seconds.  Here goes:

1) we need to assume that every actor (pc, npc, or mob) is only capable of one action.  (This isn't entirely true, see below).
2) each actor gets their own proprietary marker (miniature, coin, whatever)
3) time is now measured in poker chips.  in fact, any stackable item of uniform thickness will do.  Poker chips are nice because they're thick.
4) an actor declares an action.  make a stack of chips that many seconds high, and put her token on top of it.
5) repeat step 4 for each actor
6) the shortest stack goes.
7) once an actor has gone, she immediately declares a new action, and adds the appropriate number of chips to her stack.
goto step 6.

At some point, it would be good to shorten stacks -- but you don't have to do it each and every go.  Maybe do it when your'e running low on chips.

Finally, actors could have multiple actions, they'd just need individual tokens to represent each one, with each getting its own stack.

Cheers,
Darcy

[J. Scott Timmerman]

Aloha, Tex.

Is your game, ERA d6, published anywhere?  I'd like to look at it.

It's gone through several revisions.  Since the last big revision, I still don't have it in coherent form.  Secondly, it's not really a game; it's all just system.  My documents don't even try so much as to explain what I'm talking about; they assume a lot, and give no examples.  In my first playtest this was pretty rough since I was basically trying to explain everything verbally, rather than asking them to read the rules in advance.  In some cases, I just have the theory of how something should work without saying how people can actually play it.

Then again, I excuse myself for being slow on getting it done, because I'm working and going to school.

But hey, maybe this weekend is a good time to straighten things out to a reasonable degree, and get a copy to you by email?  I'm not afraid of posting it online, I just don't want to go to the trouble.

I took a look at your document, and I notice you've done a good job of keeping everything straight there.  I've got to bring my writing style up to par.

The word I use in my game is actually "pulse", and its meaning is only connected to a second of real time as a convention or suggestion, so that the GM can have some flexibility for reasons of poetic license or mechanical convenience.  (Maybe during a chase scene with sailing ships, for instance, a minute or even an hour is a better "smallest" unit of time.)

Rounds/Plays in ERA d6 actually vary in size within a given scene, and from play to play, and are not arbitrary.  I can see why Pulse works great for Orp, though.  Keeping it arbitrary but constant is rather convenient.

I think this kind of system also fixes a number of really weird realism problems that turn structures raise, that folks are so used to accepting that they don't think about it.

I think I see what you're saying.  For instance, in d20, there's no possible way for a character to do certain things, whereas if actions are more generally defined, it allows for more possibility?

I want my system to be lightweight by default, so I don't really define a bunch of different actions and how long they ought to take, but I do leave the possibility open for people who want to adapt Orp to more precise playstyles (I also want the system to be adaptable to a wide variety of needs).  In the base game though, I mostly leave it in the hands of the GM, with d12 pulses as the default suggestion.

Still, it would seem that after playing the system in one playstyle, it shouldn't be too hard to shift to another playstyle, as long as you know the basic system.  If the GM changes too much, then it practically becomes another system.

Yes, characters can either rush through actions in rapid succession, or truly multitask if they wish.  My tentative plan is to penalize each additional activity by -1 per additional action.  But I could see moving this up to -2, given how bad human beings seem to be at "true" multitasking.  The example in the rulebook is a ninja running along a rooftop and throwing shuriken concurrently.  : ]

As to movement, I'll be honest and say I haven't really pinned down rules for that yet.  I'll tell you the vague ideas I have at the moment, though.  First of all, assume there's a standard rate of travel for human beings; 3 m / sec is pretty close to the d20 "double move" so let's use that for now.

Ok, keeps things pretty general.  You're erring on the side of player options rather than player restrictions.

I want to avoid letting players drive their characters around on a second-by-second basis; instead, I'd rather pin them down on their characters' intent.  People usually don't wander around for the fun of it, they always have a purpose (which may be "approach that guy," for example).  So, there's now a definite distance, which means a fixed duration:  One second for each trio of meters.  Now movement can be just another kind of event, with a duration and an effect, though in this case no die roll by default. 

That's how ERA d6 handles it.  A problem I've encountered is, what about when you have to decide how far a character is along their movement when another event occurs?  It doesn't look like it would be as big a problem in your system, though, since you don't account for acceleration and pulses are all the same length.  You could just say that, if an event occurs 3 pulses into a 5-pulse movement, that a character is 3/5 of the way, for instance.

I think the rest of the game system can now be used to emulate most of the stuff commonly imagined in action scenes when it comes to movement and position.  For instance, I can imagine ways to account for moving faster at need, catching up to retreating characters, preventing other characters from moving through one's own area, and so on.

Thanks for asking about this, I've been putting off thinking about it for awhile.  : P  How does ERA d6 work?

Everything is tracked in exponents of 2.  So 0 = 1s or 1m, for instance.  The Duration of any movement action is equal to Distance/2 or Distance-3, whichever is higher.  Explaining it further might take a while, so if you have any specific questions on what I'm talking about here, I could go on and on into another thread.  I think there is another thread about this.  I'll post a link to that thread after I find it.

Unfortunately I've mostly gamed with people for the last few years that don't really want to play anything but d20, specifically D&D.  I'm not hostile to d20 per se, but I will say such a large degree of homogenization in the community tends to leave people less interested in other directions.

I'm sorry to hear that.  I don't mind playing d20, either.  It's a decent system.  I know the rules.  I don't like running in it, though.

I want to comment some more but don't have so much time right now.  You'll probably see another post from me before the day is out.

-Jason Timmerman

[tex]

Hi tex,

FIrst off, welcome to the Forge.

Second, I prefer to use my real name here.  It'd be awesome to be able to do the same with you.

Thank you!  I don't mind using my real name.  I changed my profile, I think.  My email address gives it away in any case.

Third, I have a suggestion for handling the "counting" of seconds.  Here goes:

1) we need to assume that every actor (pc, npc, or mob) is only capable of one action.  (This isn't entirely true, see below).
2) each actor gets their own proprietary marker (miniature, coin, whatever)
3) time is now measured in poker chips.  in fact, any stackable item of uniform thickness will do.  Poker chips are nice because they're thick.
4) an actor declares an action.  make a stack of chips that many seconds high, and put her token on top of it.
5) repeat step 4 for each actor
6) the shortest stack goes.
7) once an actor has gone, she immediately declares a new action, and adds the appropriate number of chips to her stack.
goto step 6.

At some point, it would be good to shorten stacks -- but you don't have to do it each and every go.  Maybe do it when your'e running low on chips.

Finally, actors could have multiple actions, they'd just need individual tokens to represent each one, with each getting its own stack.

I really like this idea.  It's efficient, easy to understand, and cheap to implement.  Would you mind if this crept into Orp at some point?

Also, Vulpinoid, thanks for the compliment.  : ]  What's your system?

. . . and while I was typing, VoidDragon posted.  Thanks for the additional thoughts.  Like yourself I'm pretty busy too; I'll reply eventually.

-T

[Darcy Burgess]

Hi Tom,

As far as I'm concerned, anything I post here at the Forge is for public consumption, so, yeah, do what you want with my idea.  All it was was a specific tactile technique for implementing your concept anyway.

But...  (there's always a catch, isn't there?)

I'd like you to think about this question.  It's just a question (it's totally delivered in a completely neutral tone of voice).  You now have a pretty swank initiative system.  How are you going to leverage that system to make your game stand out?  What about your game cries for such a system?  Why is initiative critical to what you're doing?

Ok...that was three questions, but they're really just one question posed three different ways.

Cheers,
Darcy

[J. Scott Timmerman]

Aloha Tom,

Here's a link to an online tutorial of Exalted combat, as well as one version of the Exalted Battle Wheel, which you might want to take a look into as well, and weigh the wheel option with the tokens thing.  You'll probably have to adapt the wheel to have thirteen spaces instead of 7, but that wouldn't be too much work.

And here is the link I said I'd post.  It's a thread about Movement Abstraction.  That thread was started before I shortened the name of my system.  The project which was previously called AVERA is now ERA d6.  The thread is pretty hard to follow.  As you can see, my issue with movement is more complicated than yours seems to be.

-Jason Timmerman

[tex]

Aloha, Tex.

Heh, my brother is stationed in Hawaii right now, as it happens.  Aloha!

Then again, I excuse myself for being slow on getting it done, because I'm working and going to school.

Yeah, same here.  Orp has been kicking around my brain for a couple of years in the same way, a project I work on when I have time and the inclination.  In fact . . . the oldest dated record of Orp I have is from 2004, and back then it was scattered useless notes, and had already existed for awhile besides.

But hey, maybe this weekend is a good time to straighten things out to a reasonable degree, and get a copy to you by email?  I'm not afraid of posting it online, I just don't want to go to the trouble.

Please do!  I'd be much obliged.  I'm happy to provide motivation to you for working on it too, heh.

I took a look at your document, and I notice you've done a good job of keeping everything straight there.  I've got to bring my writing style up to par.

Thanks!  The truth is, I have LaTeX to thank for that, mostly.  LaTeX provides a really useful little toolbox of ways to structure text.  I'm not kidding when I say the table of contents was produced by typing "\tableofcontents" in the source file.  : P

I get into this on Page 9, but part of the reason I do it that way is I miss the way old military wargame manuals were so tightly organized and concise.  I have an ancient copy of James F Dunnigan's The Complete Wargames Handbook, which includes The Drive On Metz, which is an excellent example.  The reasons I like my roleplaying manuals to read like wargame manuals is a thread in and of itself, I imagine.

I think I see what you're saying.  For instance, in d20, there's no possible way for a character to do certain things, whereas if actions are more generally defined, it allows for more possibility?

That's definitely a strong motivation.  I'm a computer nerd by trade (if you haven't figured that out yet, heh).  In computer programming, "expressive power" is a desirable quality of programming languages.  I think it's also desirable in roleplaying systems.  Flexibility is one key way of achieving more of it.

That's how ERA d6 handles it.  A problem I've encountered is, what about when you have to decide how far a character is along their movement when another event occurs?  It doesn't look like it would be as big a problem in your system, though, since you don't account for acceleration and pulses are all the same length.  You could just say that, if an event occurs 3 pulses into a 5-pulse movement, that a character is 3/5 of the way, for instance.

That seems like a good approach.  Also, I think you've more or less discovered a way for characters to collide with eachother.  Heh, now I've got the giggles.

Everything is tracked in exponents of 2.  So 0 = 1s or 1m, for instance.  The Duration of any movement action is equal to Distance/2 or Distance-3, whichever is higher.  Explaining it further might take a while, so if you have any specific questions on what I'm talking about here, I could go on and on into another thread.  I think there is another thread about this.  I'll post a link to that thread after I find it.

K, I'll read and get back to you on that and the other concern you had about interrupted movement, once I read the thread.

[John Kirk]

tex,

We've been playtesting this kind of system for well over a year now for 8th edition Legendary Quest (7th edition, which is the edition currently available for download, uses a turn-based system).  We even use an Initiative Wheel very similar to Exalted's Battle Wheel.  I was actually unaware that Exalted had a Battle Wheel like this prior to reading this thread.  (Thanks, Jason, for pointing that out.)  This kind of system works quite well, even for large groups of people (10-12).  Many players have told me that they have an easier time keeping track of what's going on with this system than a turn-based system, because they have the wheel in front of them to see when everyone is going.  One important point here, that is kind of glossed over in the Battle Wheel description, is that the tokens you use to track players must be instantly recognizable by everyone sitting at the table.  The best ones I have found are alphabet beads that I bought at Hobby Lobby.  These are cubic beads with letters on the side.  You can assign each player a different letter based on either the character or player name.  I would imagine that Scrabble letters would work just as well.

Darcy has a good point, though.  This kind of system excels in timing actions for crunchy, tactics-oriented games.  My impression is that you are creating a tactical game, so you probably have a good match here.  If you don't envision your system being primarily tactical, though, you might want to try something else.

[Darcy Burgess]

Hi Tom, Hi John,

John's comment about the "good point" that he seems to think I was making is exactly what I was afraid of.  I didn't ask those questions with an eye to saying "that sort of a system suits such-and-such a game".  Rather, I'm much more interested in Tom's answers to those questions, not in him guessing what I may think the answers are.

Seriously, when I said that those questions were delivered in a neutral tone of voice, I meant it -- there's no hidden agenda, no test, no right answer.

They're strictly questions to provoke thought.

Cheers,
Darcy

[tex]

Hello again Darcy,

These questions of yours I'd planned on answering tonight:

I'd like you to think about this question.  It's just a question (it's totally delivered in a completely neutral tone of voice).  You now have a pretty swank initiative system.  How are you going to leverage that system to make your game stand out?  What about your game cries for such a system?  Why is initiative critical to what you're doing?

I guess it's sensible to first tell you what Orp is meant to be; the answers to your questions will be simple and easy after that.  Orp is intended to be a general-purpose roleplaying system, useful for extensive adaptation to a variety of needs.  I want to give GMs a large amount of freedom when it comes to customizing the Orp to suit their campaigns.  So, Orp "out of the box" is a spare and airy thing, which is playable almost immediately but doesn't answer many specific questions.  There are other design goals, of course:  I want the game to be streamlined but still realistic for its level of detail, for instance.  But the main thing is its adaptability.

So, this is why I'm actually pretty okay with Orp not standing out.  I like to think of it as a humble system, willing to bend this way and that, but most importantly, getting out of the way if the GM and players want an unobtrusive system.  In fact, once I'm comfortable with Orp's event-based timing, I intend to write turn-based timing rules too, so that way people aren't forced into one or the other.  The way the timing rules are written (and this is common to Orp generally), GMs can use them independently for different roleplaying needs, freely mixing them at need.  My original post is actually all of the rules used together.

Curiously, I'd never really thought of Orp as a "crunchy" game nor a "tactical" game until John mentioned it.  In fact, I generally go out of my way to keep the crunch to a minimum.  Really the timing rules aren't for tactical situations necessarily, it just happens tactical situations typical to roleplaying are a good proving ground for that sort of thing.  I could see using them in a social situation just as easily, modelling, say, the characters travelling around a city gathering information about some mystery prior to some important impending event.

I will say though that I do want Orp to be capable of being crunchy and tactical if desired.

[John Kirk]

Seriously, when I said that those questions were delivered in a neutral tone of voice, I meant it -- there's no hidden agenda, no test, no right answer.

My apologies.  My statement was poorly worded.  I was not trying to say that Darcy was implying that the system Tex describes is best suited to tactical games.  I was saying that Darcy had a valid point in asking the pertinent questions about how this fits into Tex's game concept.

I am the one saying that this kind of system suits tactical games.  While any game pattern can be used in the creation of a variety of game types, some patterns are better suited to certain specific purposes.  I believe that to be the case here.  So, while this pattern may be used in non-tactical games, I think the pattern promotes tactics-oriented thinking. Tex is here asking for advice, and that is the advice I am giving.  It has been my experience that creating a game that works for both tactical and non-tactical play is difficult.  I'm not saying it is impossible, only that I have spent a good deal of effort creating and playtesting a system very similar to what Tex describes.  In my opinion, this system clearly leans in the direction of tactics.  And, what's more, it's a damned good system for doing it.

[Bossy]

Interesting idea. Designing such a system while keeping it playable is even more interesting... I have some random remarks, I'll anchor them to your plan:

0) Time is kept using seconds.  Events are divided up discretely as beginning on a given second, and finishing at some second in the future, and generally don't take effect until the end of their duration.  One good example is a mage casting a spell:  The mage decides to cast a spell on some second, and starts chanting, and then finishes at some second in the future, when the spell will then take effect.

Depending on the setting, other time units could be more practical:
   - dozens of seconds for underwater combat
   - CPU cycles for matrix/metaverse combat
   - minutes for space combat
   - ...

I concede this is not really important for the design, anyway your rules mention "pulse" as a unit which could be variable from one situation to another.

1) At the beginning of combat, all the PCs and NPCs independently decide what they want to do (if they want, they can do nothing for a little while and wait and see what develops).  This decision happens in the instant the characters realize combat is imminent, and represents the character committing to the action in question.

Realistically, all characters do not realize combat is imminent at the same time. The faster characters should be allowed to start as second one before the others because they are slower to grasp the situation or because they were surprised.
In fact, this is related to how you plan to cope with reactions, typically dodging and parying. How many seconds does it take for a character to spot an attack and decide which parying movement is adequate, indempendently from the time it takes to raise a shield between the sword and himself? What happens to the action the reacting character was doing, for instance what happens to the spellcasting if the mage decides to dodge a missile? Does he have to interrupt the incantations? The answer depends on how much simulation and how little clumsiness you want in the system. As far as I skimmed through ORP rules, reactions are incorporated in success rolls so they don't have to be actually declared, right?

Conversely a character may want to delay the effect of a finished action in order to synchronize it with external events. For instance Hardbrat wants to delay her kick to let Frostboy to freeze the villain before (because everybody knows the villain will then sprinkle like a smashed ice cube). Another example: John Soldier points at the door and aims 20 centimeters above the knob, but he doesn't shoot yet (until the door opens).

2) The GM decides on how long it'll take for each character to accomplish what they want (usually, secretly, and randomly:  d12 seconds for most actions in the typical chaotic combat environ).

I would feel very uncomfortable to make the duration of an action totally random, especially using a single die. I'd prefer either a bell-curve roll (2d6, unthinkable in ORP) to make durations less variable or a stat-based roll (d12 - stat, I know substraction is evil) to take character's forte into account. In general I would strongly advise the GM chooses wisely the duration instead of rolling.

5) The GM repeats this process until the combat is concluded:  The next action resolves, and the acting character may choose a new action.

So, that's it, basically:  Stuff happens "when it happens".  There are more details (such as a mechanism for allowing characters to interrupt eachother and themselves), but this is the essence of it.

Obviously this is a really weird approach.  I'm concerned that it may be hard to learn, and also too cumbersome at the gaming table.  Well, I feel like I could GM this, but I thought of it, so of course it seems very natural and intuitive to me.  : P

That were basically my suggestions and remarks from the top of my head, though I didn't give much thought about it and never practiced something similar (never played Exalted). It seems that if you incorporate all subtelties beyond simple actions (reactions, multitasking, delaying, etc.), the system could quickly become clumsy. As it is I don't think it is hard to learn but it looks like quite demanding for the game master since he has to keep track of the time of each character's action. Have you already actually practiced this system? I'm curious to how long (how many pulses) a single combat sequence can be.