Jon of the Wired
-
Posts
89 -
Joined
-
Last visited
Content Type
Profiles
Forums
Blogs
Posts posted by Jon of the Wired
-
-
Yeah, it's a bit perplexing. When I first saw the announcement that a TB mode was being added I assumed they were going with an FF X style system because I couldn't think of any other way to preserve the action speed / recovery economy. I was surprised when the beta came out to find that they just... didn't preserve it (well, nominally they did, but Initiative is weak enough that you don't have to care about it). I never considered that possibility because action speed seems so critical to PoE's mechanical identity that I assumed any system that didn't centre it would be a non-starter.
I have every confidence in the Deadfire devs, and clearly what they accomplished already with the turn based mode is amazing, so I think they must have introduced rounds for what seemed like a good reason, but I just can't imagine what was so broken about roundless that it was worth nuking action speed (and to a lesser extent breaking Intellect and Resolve as well).
The one thing that does seem like a challenge with roundless is the turn queue UI. The one advantage (AFAICT) that rounds (and measuring durations in rounds) gives you is that a lot of events that would otherwise need their own initiative value can instead piggy-back on a character's turn (e.g. effects ending, DoTs pulsing). That limits the number of things that have to be in the turn queue UI, which prevents it from being too cluttered to be legible. For example, consider casting Relentless Storm with five enemies in range. Now each of the strikes for the storm need an entry in the queue, and the end of the stun for each of the five enemies needs an entry in the queue. That's eight or nine (not sure how many strikes the spell actually makes) new rows in the queue UI just because someone cast a single spell. In a decent sized fight, that could easily get out of hand.
That said, I don't think the problem is insurmountable. One possible approach would be to have a different type of row in the UI for events which aren't character's turns (so casts finishing, effect durations expiring, DoTs pulsing, etc) which would be more compact (though the existing row element is pretty compact) and more importantly batches all of the effects that are happening between two character's turns, so that there's only ever one row between any two character's turns. Instead of an icon and some text, you could just show a row ordered from left to right of icons where each icon represents an event. Then you can show the full details of what's happening and when if you mouse over the row.
Anyway, obviously they've received a lot of feedback about the action economy problems of TB mode, so hopefully they'll have a solution of some sort in place before TB makes it out of beta.
Edited to Add: Of course, there is also the obvious solution to these problems, which is to get rid of rounds, but measure durations in Turns instead (which is how FF X does things anyway, and is why it doesn't have to deal with these issues). Things would basically work the same way as in the current TB mode, with effects expiring and DoTs pulsing on character's turns. This means you don't really need any UI changes, and it's no more complicated to understand for the player than the current system. The downside is that it will affect balance, as character's taking faster actions will be affected by things for less "time", because they take turns more frequently, and character's taking slower actions will be affected for longer. You also need to come up with a way to prevent players from ending afflictions early by effectively skipping turns (I guess you wouldn't be able to end your turn without either taking a standard or cast action, or moving a sufficient distance?). It also doesn't do anything to help fix Intellect or Resolve, which would still be somewhat busted. On the whole, though, it would probably still be a big improvement.
-
There is a difference though. Say for instance a 6 in Dex gets you one attack, 12 gets you 2 and 18 gets you 3 attacks. Now any Dex from 7-11 and 13-17 is a waste of points since it gets you nothing tangible. Currently in RTwP every single point of Dex does something. With multiple attacks per action it creates the break points issue people have been mentioning. Plus multiple attacks in a single action begins to throw the action economy out of whack if you can exploit it.The balance would be the same as in RTwP mode, though (including movement, which would now cost initiative), so I don't think that's a major concern. On the other hand, having action speed only affect basic attacks would significantly alter the balance between casters and non-casters, for example.
I meant that in an FF X style roundless TB mode (which doesn't suffer from the problem you're describing) the balance would be the same as RTwP. I've edited the post for clarity.
-
Yeah, as others have said, the assumption is that once you get rid of rounds, everything to do with time is now measured in Initiative ticks. That includes effect durations (which means Intellect now works properly again) and also when DoTs go off (so they would be on their own Initiative, not bound to when the character takes actions, just like in RTwP).
This shouldn't cause any balance problems (again, because it's really just exactly how things worked in RTwP) but it may be a bit of a challenge communicate what's happening to the player.
-
First poster here. Made an account just to join in on feedback on turn based, as this is my favorite type of game.
Like the idea of FFX style speed/action system over turnbased. Multiple turns including possible repositioning could be overpowered though. In RTwP, usually the more actions are just continuing to attack, but at a faster rate.
How about doing it like the old Darksun D&D games. You still only get one action per turn, and don't get extra actions for being faster, however If you would normally get 2-3 attacks in that amount of time (because of attack speed and recovery time), when you make an attack, your character attacks multiple times in one action.
For example:
Warrior with plate mail and greatsword has one action per turn. He makes a standard attack with his action. The warrior makes one swing of the sword.
Monk with leather armor and lightning strikes and high dexterity. He makes a standard attack with his action. The monk attacks 3-4 times in succession before his turn ends.
I know all these suggestions have more complications but that is a general idea.
The balance in an FF X style roundless TB mode would be the same as in RTwP mode, though (including movement, which would now cost initiative), so I don't think that's a major concern. On the other hand, having action speed only affect basic attacks would significantly alter the balance between casters and non-casters, for example.
-
That said, I agree with people saying that a direct translation of the RTwP action economy to turn-based would also be awful. Even super efficient players in RTwP are going to waste some actions, but every action can be carefully considered in turn-based, meaning action economy advantages would be far more pronounced. Just look at how people are mentioning things like weapon swapping and free actions, actually not very meaningfully different from their RTwP implementations, as being problematic.
But that doesn't mean that discrete rounds with initiative is okay either. Who is ever going to choose to use a rapier and dagger under this system? Some way to get extra actions in exchange for investing in speed is something a huge amount of the game is built on.
Once you get rid of rounds, balancing the action economy is just a matter of tuning action speed bonuses, possibly at a global level. As long as there's one action per round action speed is too weak to be a useful balancing factor, so you need to start messing with damage values and who knows what else.
-
While I do agree that there are problems with the current Initiative system, a lot of this feedback IMO fails to take into account the downsides of having full new turns available to a character. Getting a new chance to act would be far more powerful than in RTwP, as you get effectively more Stride, more Cast or Standard Actions (you technically get a lot of those in RTwP, but it's very hard to micro them effectively so the action economy isn't as lopsided in practice), etc.
I still agree that this is at least something worth considering for the devs, but I expect that it would be toned down in some form from RTwP. Bit surprised that Dexterity doesn't give additional standard attacks during a turn tho'.
Dexterity granting additional standard attacks would violate one of the design goals for Attributes in PoE, which is to avoid hard breakpoints. If Dexterity granted additional attacks there would only be a few values of Dex where your number of attacks actually changed, and setting Dex to any other value would be a newb-trap.
-
1
-
-
After playing (and enjoying) a lot of the turn based combat, I agree that this would be the way to go. The current TB system is a lot of fun, but the damage to the game balance and build diversity is significant. I also think that action speed is too important to the mechanical identity of PoE to be devalued the way it is in the current TB mode. I think a FFX (or FFTA) style TB mode without rounds would feel more like PoE than the current system does.
What concerns me, though, is that it's such an obvious idea that I wonder if they didn't already try it and discard it. It's not clear to me what the fatal flaw would be in such a system, but it would be a little harder to grok for players in a number of ways. The biggest downside I can think of is that it would be harder to communicate effect durations to the player. A duration of N rounds is easy to understand, but if you measure the duration in Initiative ticks, it's less clear how long the effect lasts in practice. You'd need to add when an effect ends to the turn queue UI, which could result in a lot of clutter. One solution to the UI clutter, though, could be to add a more compact queue entry style that would be used for effect expiration and cast completion, while the existing style continues to be used for character turns. This would allow you to show more queue entries in less vertical space. You could even have the new entry style handle batches, so in the relatively common case where an ability applies more than one effect with the same duration, the expirations would be shown in a single entry (with details on mouseover). An upside, though, to measuring duration in Initiative ticks as opposed to rounds is that it would make Intellect work a lot more smoothly.
The other problem I can imagine is with movement. The current movement system is pretty simple (and generous compared to RTwP), but the OP has already suggested the obvious solution to movement, which is simply to charge Initiative for it (with some maximum movement per turn to prevent players from completely screwing themselves).
Even if they have already evaluated something similar, I hope they take another look. The current TB system is fun in the moment to moment play, but it does introduce a lot of problems, most of which just go away completely if you get rid of rounds. Given the pre-existing depth of the mechanics, I really think the TB system is this close to being one of the best combat systems to ever be in an RPG, and more than anything I think a FFX style turn system is what would put it over the top.
-
So, I've noticed that there's two different notations for damage bonuses. Some damage bonuses (like the bonus from Might, or Burning Lash) are written as "+N% Damage" (e.g. +20% Damage, +50% Damage) and others (like the bonus from weapon quality and many talents and abilities) are written as "xN Damage" (e.g. x1.2 Damage, x2.0 Damage).
The first notation implies the bonus is additive and the second implies it's multiplicative, which has a significant impact on what the bonus means. Are there really two different kinds of damage bonuses, one additive and the other multiplicative? If so, how are they applied (there's a few different possibilities, each of which would produce different final damage totals)? If not, I guess that means that all of the instances of the "xN Damage" are wrong, and should be "+((N-1)*100)% Damage"? (If that's the case, is there a bug open to have that corrected?)
-
To be fair, according to my model, all you have to do to give breastplates a viable niche is change the -40% recovery penalty to a -35% recovery penalty. It's not exactly a massive overhaul.
-
So, I built a much more complete and accurate model of when it's appropriate to wear different armors...
[snip]
Interesting. It looks like that model is assuming a 1:1 ratio of dps to tanks, does it change any if you assume that there are 2 or 3 dps'ers per tank?
Also I don't think it changes anything, but those deflection values seem a little odd. 46 seems a little optimistic for dps (the bb fighter has 38 at level 5) and 64 seems a little low for a tank (bb fighter does have 67 with a shield and defender activated, but he's missing 20 points of deflection he could have from better talents and another ~18 from stats).
And yeah, there's some really neat looking armors inbetween plate and padded. It'd be nice to have a reason to wear them.
The model assumes one dps and one tank, but changing that shouldn't have any affect on the results. All that matters is the ratio of attacks that go to the tanks vs the dps's. The deflection values were just read off the bb rogue and bb tank from one of my saves, but the bb builds are a bit weird. I played around with some other deflection values, and it shifts the ranges where you want clothing, padded, or plate armor around a bit, but it doesn't change any of the conclusions. Critically, there are no deflection values that make breastplates viable.
-
So, I built a much more complete and accurate model of when it's appropriate to wear different armors. It models a tank and a DPSer at around level 5 wearing Fine armor fighting lions. It doesn't do much to change the conclusions from my earlier calculation, though. Clothing is (obviously) the best armor for ranged attackers, plate is best for tanks, and there is a good argument for putting melee DPSers in padded armor. Unfortunately, the model also shows that breastplates (and presumably other medium armor) is completely useless. This saddens me, because I think it's the best looking armor in the game.
A chart can be found below:
The model can be viewed here:
https://docs.google.com/spreadsheets/d/1W5D9oTAMkiYpPAVwz6gnSxNCamI1bvQTBcXwJ8FnIKE/edit?usp=sharing
-
1
-
-
So, I did some quick calculations, and at least for 5th level characters fighting lions, if your tank is > 80% effective (that is, your tank is getting hit 80% of the time, and 20% of the time your DPSer is getting hit) then you're better off wearing clothing than armor. Between 80% and 40%, light armor beats medium armor, and under 40% tank effectiveness, you're better off in medium armor.
This is a very rough estimate, and the calculations are pretty sensitive to a number of variables (particularly how much damage your tank takes from attacks) but it seems to generally point to the idea that ranged characters really shouldn't be wearing armor (no great surprise there, I guess) but light armor may make sense for melee DPS characters.
-
I'm not sure the DPS penalty is that high. It's something like -12% for padded armor, and -25% for a breastplate. That's a fair amount, but it's not apocalyptic (for padded, anyway) and classes you're likely to use for melee DPS all have higher endurance than the wizard, so they benefit more from DR.
-
Without some numbers behind it, that's just not an argument. How minor a survivability boost? How large an amount of damage?It's an argument whether or not there are any numbers behind it. But I have this chart kicking around from the last version, and I like charts, so here, enjoy this chart. Paladins are assumed to have 5 deflection and 3 DR over fighters.
Slapping the heaviest armor in the game on a wizard increases the average attacks until they die from 2.3 to 3.2. Putting plate on a tanky fighter increases their lifespan from 23 swipes to 58 swipes.
Now, I'm sure you're going to say, but what about other creatures. And let me assure you, the shape doesn't change much for other creatures. But if there is a creature you would like me to put into the chart, I would be happy to do so.
No, that chart is pretty interesting; I don't think there's much value in measuring against different creatures, this seems like a reasonable benchmark. I am interested in seeing the difference in survivability between no armor, light armor, and medium armor for melee DPS builds on Ciphers, Rogues, Barbarian, Rangers, etc, and seeing the difference in DPS for the different armors.
-
Unless the fight starts out of a cutscene, the current AI makes it trivial to make sure that only your tank is attacked. You can try it yourself, have your tank walk forward and draw aggro, then let everybody else walk up and start shooting stuff. See how long it takes for anyone else to get damaged.
Sure, that is my usual strategy. I don't think the AI is nearly aggressive enough about attacking your back-line, but I have had beetles tunnel past my front-line, and enemy casters hit everyone with AoEs. In unfamiliar areas, you can also get adds coming in from unexpected directions. Melee DPS characters are attacked more often, and are more likely to be worth putting in armor.
Even if that were ever fixed, both deflection and DR give minor boosts to surviability when you have none and increasingly larger benefits as you stack them. Characters with low defense gain little if they choose to invest in defense but lose a large amount damage.
Without some numbers behind it, that's just not an argument. How minor a survivability boost? How large an amount of damage?
-
1
-
-
Everyone in this thread is taking it as given that light and medium armour are dominated by clothing and heavy armour, but no one has actually demonstrated that.
Combat can be viewed as an optimization problem where you are attempting to maximize the value of (damage done to enemies) / (per-rest resources consumed). Wearing light or medium armor reduces damage output, but it also reduces consumption of a per-rest resource (Health), so it's only obvious that light and medium armor are worse for non-tanks than clothing if you're assuming that only your tanks are ever damaged. In my experience with the beta, that assumption simply isn't true, and it's likely to be even less true in the full game.
To actually know one way or another, you'd have to produce a model that calculates (damage done to enemies) / (per-rest resources consumed) under various conditions. Then you could see whether the conditions where light or medium armor are optimal choices could plausibly occur in the game.
-
1
-
-
No they are not screwed. They are still better to move away and use all their attacks while melee guy only gets his opportunity attack. That is of course if the ranged guy is faster than melee guy. Just like in IE games, you could not kite 1v1 if you were not faster.
So does PoE have less attacks. If you are running back you are not getting most DPS as you have slower recovery times.In DnD you have to build for that mobility. It does not come standard. Look at Shot on the Run (3E). Even after a host of feats to get it, you only get to make a single attack while moving.
http://www.dandwiki.com/wiki/SRD:Shot_on_the_Run
Also, its kinda funny that you list DnD concerning this.
And in 5e you can move and do all attacks. And you could do so in 2e as well (which IE games are based on).
In neither 5e or 2e could you kite, however. In either system, the pursuer can move twice as fast as a character moving and shooting, and once the pursuer catches the kiting character, the kiting character is screwed. They can either run, or move and shoot, at which point they provoke an attack of opportunity (rules for that were in 2e, though under a different name and not implemented in IE) or they can safely disengage and move at normal speed. Either way, next turn the pursuer just catches up to them again.
That's true in 5e, assuming the the pursuer doesn't have the "Charger" feat, but not in previous editions, and especially not in 2e. In editions prior to 5e, charging was a thing anyone could do, so a kiter would have to be at least twice as fast as their pursuer, or they would suffer an attack of opportunity and a charge for every attack they made. I was just flipping through my 2e PHB, and it turns out in that edition things are even worse. The rules for withdrawing from melee are so harsh that unless you have another friendly character next to you to block your pursuer, there's almost no point in bothering. It's not even clear that you can move out of melee and attack in the same turn, but assuming it is permitted, your pursuer would first get a free attack on you, or multiple attacks if they have them, and then next turn they can just charge you. Charging lets you move 150% of your speed and then attack (with a bonus to hit), whereas you can only move 50% of your speed and make a ranged attack, so unless the kiter is more than three times faster than their pursuer, they are seriously screwed.
-
So does PoE have less attacks. If you are running back you are not getting most DPS as you have slower recovery times.In DnD you have to build for that mobility. It does not come standard. Look at Shot on the Run (3E). Even after a host of feats to get it, you only get to make a single attack while moving.
http://www.dandwiki.com/wiki/SRD:Shot_on_the_Run
Also, its kinda funny that you list DnD concerning this.
And in 5e you can move and do all attacks. And you could do so in 2e as well (which IE games are based on).
In neither 5e or 2e could you kite, however. In either system, the pursuer can move twice as fast as a character moving and shooting, and once the pursuer catches the kiting character, the kiting character is screwed. They can either run, or move and shoot, at which point they provoke an attack of opportunity (rules for that were in 2e, though under a different name and not implemented in IE) or they can safely disengage and move at normal speed. Either way, next turn the pursuer just catches up to them again.
-
I am actually not sure how engagement mechanics is a solution for kiting. can someone explain to me how that adds up? I have been hearing this thing for quite a while now.
It's not a full solution to kiting, but it does discourage kiting in some circumstances. Specifically, it discourages kiting with a character that has already been engaged, because they would need to suffer a disengagement attack (or take some action to prevent one) before they can begin kiting. It doesn't help a melee character catch a kiting character in the first place, though, which is why it's not a full solution.
I don't think there's any single mechanic that can solve kiting. Any solution has to be a combination of mechanics, AI, and encounter design.
-
That's right, you can. However since prone, stunned and paralyzed units can't fight back and get huge penalties to defenses ... why on earth would you stop attacking them?
Retreating isn't the only reason to disengage, though, you may also be trying to get to a squishier target.
-
1
-
-
On further thought, I think bonuses for attacks from behind are actually not great either conceptually or practically. On a conceptual level, presumably you move more slowly near melee characters because you're being forced to be concerned more with your personal safety than moving as quickly as possible. If that's the case, it doesn't make that much sense to both be moving slowly, and suffer defense penalties (i.e., moving slowly is the cost you're paying to not be extra vulnerable to attacks). Practically, it's probably best to only have one type penalty, instead of just stacking on more and more different penalties arbitrarily. With just slowed movement you already have two different tuning nobs to tweak (radius of the effect, and how much it slows your movement) adding on more is only going to make balance more difficult.
Not that slowed movement near melee characters is likely to be implemented at this point, but it is certainly an interesting idea.
-
Hm, yeah, facing would probably be a bit tricky.
-
Considering engagement main purpose is to prevent kiting and moving easily to back lines best way would be to make it so its penalty is not bonus attacks but slowed movement.
Try to run through melee? You move so slow enemy melee kicks your ass while you waste time just moving.
Try to kite someone that engaged you? You move slow and they can still attack you easily as their attacks are not slow.
Considering in PoE everyone does lots of damage and attacks fast, just slowing movement should be enough as enemies will get enough extra attacks on you while you move away/through but a good defensive character can still move away when badly damaged so to have someone else take his place without suffering mortal blows that make moving away pointless in the first place.
I really think OE should experiment with this instead of engagement attacks doing crazy damage.
I think that could work, and I would be interested in trying it out (maybe in combination with bonuses on attacks against enemies facing away from you).
-
Even the IWD:HoW AI you're presenting as a positive example very rarely meets my criteria for being smart, by your own admission.
Sure but it's what I would like to see in the game at a bare minimum. Encounter design can also make up for AI deficiency. The Encounter design in the beta isn't great, but at least Adra beetles have a couple of ranged attacks now.
Sure, but I have to assume that if all the enemies in IWD:HoW were actively trying to get to your backline characters (using the highest-AC targeting clause, lets say), that would dramatically change the feel of the game when compared to its current state, where most enemies are content to waste time attacking people in plate armor. I don't think IWD:HoW gives you the tools you need to deal with enemies that are that aggressive on a constant basis (which is where the passive, soft CC of engagement is, to my mind, beneficial).
Time Unit Based Combat System
in Beta Feedback for Turn-Based Mode
Posted
Yeah, this idea has been discussed several times on the board already. I agree with you that's it's probably the ideal way of implementing turn-based combat for Deadfire (certainly better than the existing system, and I think also better than an action point based system). It's actually so ideal that when I heard they were doing a turn-based mode I just assumed that's how they would do it, and was pretty surprised when it wasn't.
In any event, the next big patch is probably going to be out in a few days, and I expect that will include any major changes they're going to make to the turn-based mode, so we'll soon see if that's the direction they've chosen to go.
Korica, it's the system used in all of the Final Fantasy Tactics games (the original, Advance, and A2), and in Final Fantasy X. I think other people have mentioned a similar system was also used in the Heroes of Might and Magic games, and it's probably turned up in other places as well.