Goblin Squad Member
|
| 1 person marked this as a favorite. |
Part of the problem of monotone damage output (without crits) is that it can make the ebb and flow of combat easy to predict.
The spread in damage around the mean (the standard deviation) in PFO looks to be around 20% or so, though it changes depending upon skill. (See here for some older analyses, specifically slides 9-12 of PFO-dmg-crit.pdf. Note that all plots on crits are now obsolete.)
How does this compare to a game where your normal damage falls into a tiny range and you have a certain chance of critting for double? For 10% crit chance the standard deviation is about 27% and for 20% crit chance it is around 33%. (For large N, the standard deviation divided by average damage is approximately sqrt[c(1-c)]/(1+c). )
So yes overall damage output is certainly less variable than in a typical game with double crits. However, this is an intentional design choice as far as I know. Non-crit damage has pretty large variance here compared to most MMOs, but if you take into account crits in other games it is smaller than most MMOs.
|
Goblin Squad Member
|
| 3 people marked this as a favorite. |
We need a wiki. I mean, we have a fantastic live wiki called Nihimon but I feel burdened by always having to look his way for clarification, details and links to the correct blog. :(
Edit: Or Nightdrifter. :)
Nihimon: words
Nightdrifter: numbers:-)
|
Goblin Squad Member
|
IMO wonky damage spikes don't add to the tactical side of the game. No crits makes damage easier to predict after you've been wailed on for a few seconds, but easier to predict does not always mean easier to deal with.
I'm mixed on this. Easy to predict damage means that a player will very quickly figure out if he can't win a fight. While Ryan thinks that getting killed by a lucky shot in PVP isn't fun, I think it is infinitely worse to be stuck in a PVP fight you know you CANNOT win. Players stuck in that situation will try to flee from the fight (depending on your objective with the fight, this maybe ok or it may be irritating) or simply give up on the fight to die quickly and do something else (I find killing a player that won't fight back to be the anathema of fun). Damage boosting crits can mean that a player always has a chance in a fight, keeping him or her invested in the fight. I've been on both sides of a lucky crit roll, and I'll take the risk of getting nailed by one if it means I've got a chance to dish one out.
To mitigate the effect of crit damage, most games give you options for builds that reduce your foe's crit chance or the magnitude of its effect (Fortification property in Pathfinder TT). Give the player an option to heavily defend himself against crits at the expense of other defensive options (DR, AC bonus, other special properties), just as creating a crit build for your character means sacrificing other things (to hit chance, normal damage, weapon variety, etc...) And if you are really against large damage spikes, you can always cap the max crit damage to x2 normal damage, with a heavy diminishing returns to achieve that.
|
Goblin Squad Member
|
While Ryan thinks that getting killed by a lucky shot in PVP isn't fun, I think it is infinitely worse to be stuck in a PVP fight you know you CANNOT win.
I think the keyword there is "stuck". You should be able to try different ways to disengage once you realize you're in over your head.
|
Goblin Squad Member
|
Try to have a practical exit strategy in mind before initiating hostilities.
Yet in a game where so much information (flags, names, affiliations, reputation counters) is displayed with trumpets, drums, bells, whistles, and neon signage for even the most inept onlooker to peruse at leisure that age-old truism is really moot, I suppose.
Since there will be no secrets about who is who, whole constellations of potentially meaningful interaction is simply reduced to fight or flight.
|
Goblin Squad Member
|
While Ryan thinks that getting killed by a lucky shot in PVP isn't fun, I think it is infinitely worse to be stuck in a PVP fight you know you CANNOT win.I think the keyword there is "stuck". You should be able to try different ways to disengage once you realize you're in over your head.
Theoretically, yes you would. However we don't know anything about possible ways of disengaging beyond running for it. For instance, what kind of crowd control is out there, and how effective will it be? We know there won't be mounts initially, and therefore escaping via mount can't happen.
I'm not opposed to the system GW is proposing for crits, it could be really well done. I just think their rationale exchanges one issue for another and is still going to leave a section of players thinking PVP is not fun. Since the devs control the keys to the system, it would seem to me that throttling the damage of crits down so their aren't ridiculous high crit spikes would be just as effective at preventing a crit from one-shotting someone.
|
Goblin Squad Member
|
While Ryan thinks that getting killed by a lucky shot in PVP isn't fun, I think it is infinitely worse to be stuck in a PVP fight you know you CANNOT win.
Well, that is a problem which is entirely divorced from whether or not crits do extra damage. If crits do additional damage, but you are 1 v 10, you know you cannot win the fight, regardless of how much extra damage your crits do. So I don't think making crits do additional damage has any effects on that.
In my opinion, unless your opponent has told you exactly what they have to bring to combat (or you've observed them fighting before) there will be uncertainty about whether you can win a fairly even fight, even without throwing in more RNG to the equation. Player skill and the minutiae of equipment choices, skill choices, etc., should make fights uncertain enough, IMO.
|
Goblin Squad Member
|
| 1 person marked this as a favorite. |
Well, that is a problem which is entirely divorced from whether or not crits do extra damage. If crits do additional damage, but you are 1 v 10, you know you cannot win the fight, regardless of how much extra damage your crits do. So I don't think making crits do additional damage has any effects on that.
Well, if they're all low rep characters stuck in T1 and you're in high end T2 or even T3 with appropriate resists against what they're using I can see you winning 1 v 10. If they're smart and mix up damage types and find your weakness then maybe not. Depends on the matchup, but it definitely seems possible to have corner cases where 1 v 10 is winnable.
In my opinion, unless your opponent has told you exactly what they have to bring to combat (or you've observed them fighting before) there will be uncertainty about whether you can win a fairly even fight, even without throwing in more RNG to the equation. Player skill and the minutiae of equipment choices, skill choices, etc., should make fights uncertain enough, IMO.
Very true. Stats are very much relative in PFO. You might match up well against one opponent of equal skill and comparable quality equipment and then get stomped by another. It might simply be a matter of what enchants against what resists are on your armor and what damage type you do. Or to a lesser degree which defense (ref, fort, will) you attack and how your attack bonus stacks up.
The fact that it can change so easily somewhat forces you to travel in groups with mixed damage types and mixed defenses you attack. That way you don't run into an opponent you just happen to stack up very poorly against. Not sure if this is intentional or not by the devs, but it very much fits Ryan's "maximize human interaction" mantra.
|
Goblin Squad Member
|
As for clerics that will choose negative energy, I understand you want healers to be able to heal injuries. Is the opposite true as well? Will negative energy clerics be able to inflict injuries? Or is the plan for negative energy just to be HP based damage during fights and all injury healing ability will be able to be cast by any cleric?
sort of odd that I am replying to myself but am still hoping that i can get an inkling as to what the devs have in mind for negative energy/ evil clerics..
|
Goblin Squad Member
|
I think it is slightly erroneous to determine the viability of tiered equipment vs equipment vs level vs number at this point xD
Why? Between the blogs and dev posts (mostly Stephen Cheney) we've been given a lot of info about how the combat system works. Certainly not everything, but it's enough to do basic modelling of the system.
Right now you can just model the system as people standing still hacking away at eachother. Movement, flanking and the like are just extra complications. (Relevant xckd.) For some cases you can already determine the outcome with reasonable certainty. Take a T3 fighter wading into a group of T1 fighters where everyone is using physical damage. Take the information that's been given and you can calculate how much everyone will be hurting eachother. That T3 fighter is going to be tearing the T1's to shreds while they're lucky to be doing more than a couple of damage per hit. Do you really need to wait until the game is out to know the outcome of that fight? Sure, not every fight is so easy to know the outcome of due to complications, but that doesn't mean you can't figure out certain cases.
|
Goblin Squad Member
|
well if we are assuming they are all fighters doing physical damage, or mages doing magic damage then sure, we can reasonably determine. I'm taking into account the amount of dev change, and the articles of the blogs I haven't read yet. So I guess it would be erroneous for me to say that. But you do have a good point.
I'm just hesitant until I see it actually happen.
My main point, however, is determining player skill before hand. A tier three level ten might do better vs ten tier one level ones on average, but that level ten player might be really b@d, or those level ones might be exceptionally skilled.
Further, the devs made a point saying a level one character will still be viable no matter what stage the game is in. It remains to be seen HOW viable.
|
Goblin Squad Member
|
The fact that it can change so easily somewhat forces you to travel in groups with mixed damage types and mixed defenses you attack. That way you don't run into an opponent you just happen to stack up very poorly against. Not sure if this is intentional or not by the devs, but it very much fits Ryan's "maximize human interaction" mantra.
Accessing an opponent's "level of threat" could be important before (or premeditating before) making the decision to run-away sooner than later, if need be. So noticing a particular atire or equipment and the quality of it will be partly or mainly responsible before actual combat numbers flash up?!
I really like the sound of uncertainty and how that could influence player to form groups. Thanks for the current interpretation.
|
Goblin Squad Member
|
Any stealth changes to the game mechanics would probably make most theorycrafting obsolete. So all I can do is hope nothing significant changes or that the devs keep us updated.
Player skill is probably the hardest thing to model. Right now I have no idea how to even begin dealing with it properly. It might be possible to deal with it in a limited way by saying "player X always makes this choice, player Y does random stuff, and player Z makes the optimal choice". That's possible to code up, but it's a lot of work.
The cases I deal with are purely character skill and gear based. If a fight is close based on that then player skill really matters. If gear differences mean that a player is ridiculously outmatched then it'll take an equally ridiculous amount of player skill to tip the odds. Eg. your opponent's attacks kill you 40 times faster than your attacks kill him. Damn near impossible to win. If it was say 1.5 instead of 40 then sure player skill could potentially tip the balance.
If all theorycrafting was reliant upon knowing player skill for every case being looked at then theorycrafting would be damn near impossible. So admittedly it's all about what looks best on paper. (It's actually code, not paper, but you get the idea ;) )
As for new player viability:
The numbers we've seen from the escalation encounters in the videos appear to be fights where a new character could certainly contribute. We were even given the stats for a Goblin Scout. A new combat PC should definitely be able to solo one of those. So if your settlement needs to deal with escalations similar to the ones in the video a fresh character is certainly worth bringing.
|
Goblin Squad Member
|
Well, that is a problem which is entirely divorced from whether or not crits do extra damage. If crits do additional damage, but you are 1 v 10, you know you cannot win the fight, regardless of how much extra damage your crits do. So I don't think making crits do additional damage has any effects on that.Well, if they're all low rep characters stuck in T1 and you're in high end T2 or even T3 with appropriate resists against what they're using I can see you winning 1 v 10. If they're smart and mix up damage types and find your weakness then maybe not. Depends on the matchup, but it definitely seems possible to have corner cases where 1 v 10 is winnable.
Just curious - if that 1 v 10 is against 10 rogues who have sneak attacks when they aren't the target, how does it work out, or do we not know enough at this point on sneak attack?
|
Goblin Squad Member
|
Just curious - if that 1 v 10 is against 10 rogues who have sneak attacks when they aren't the target, how does it work out, or do we not know enough at this point on sneak attack?Well, that is a problem which is entirely divorced from whether or not crits do extra damage. If crits do additional damage, but you are 1 v 10, you know you cannot win the fight, regardless of how much extra damage your crits do. So I don't think making crits do additional damage has any effects on that.Well, if they're all low rep characters stuck in T1 and you're in high end T2 or even T3 with appropriate resists against what they're using I can see you winning 1 v 10. If they're smart and mix up damage types and find your weakness then maybe not. Depends on the matchup, but it definitely seems possible to have corner cases where 1 v 10 is winnable.
Depends on how much bonus base damage sneak attack adds. A couple of the hits being 71 in the Q4 video lead me to believe it can be rather large. If so, that T3 guy better hope he threaded his expensive gear!
The 1 v 10 example I gave is just an extreme case where facing 10 clones who you are good against is winnable. I expect 1 v 10 random lowbies is a different story.
|
Goblin Squad Member
|
@Nightdrifter
You are a boon to the community friend
I love how we can tell which one is the liberal sciences theory crafter type and which is the applied sciences theory tester.
If we are basing on the character skill and gear alone, I guess there isn't much to argue about other than semantics. Obviously things are not for sure, but then to argue that makes it a pointless discussion, cause that would be arguing the entire setup of the game, which is just silly.
;) tally-ho tally-ho
|
Goblin Squad Member
|
| 1 person marked this as a favorite. |
Thanks :)
Pretty much all the info I have ultimately comes from Stephen Cheney (a dev) in some way, so only fair he gets credit!
I always think of the old Flintlocke cartoon where a character named "the Wuuf" points out that if you're going to spend a ton of time playing a game you might as well spend some time learning the mechanics. Since we can't actually play yet ...
|
Goblin Squad Member
|
Well, that is a problem which is entirely divorced from whether or not crits do extra damage. If crits do additional damage, but you are 1 v 10, you know you cannot win the fight, regardless of how much extra damage your crits do. So I don't think making crits do additional damage has any effects on that.Well, if they're all low rep characters stuck in T1 and you're in high end T2 or even T3 with appropriate resists against what they're using I can see you winning 1 v 10. If they're smart and mix up damage types and find your weakness then maybe not. Depends on the matchup, but it definitely seems possible to have corner cases where 1 v 10 is winnable.
Okay Mr. technical, some 1 v 10's will be winnable. :P My point was supposed to be that if crits do damage instead of debuff, your enemies get whatever advantages you get out of that so it doesn't really tip any odds or even up any imbalances that there would be.
|
Goblin Squad Member
|
| 2 people marked this as a favorite. |
What we're not going to try to do is create well-defined role templates out of those combos. Players may do so on their own initiative, but we're not going to be held to them - what you do in terms of how you build out your character is your own choice, it's not a blueprint endorsed by us.
This is key, imo. So many games have "Best in Slot" charts, "Best Talent Tree" set-ups, skills or abilities that must be known (Bio-med in SWTOR comes to mind). Creating a system where every character is not only different, but creating duplicates or twins is almost impossible, is getting close to an ideal. Each player's play style should be available, even if it does not meet a "best" of anything. We will ultimately be paying to have fun. When the fun ends, so does the money.
|
Goblin Squad Member
|
| 1 person marked this as a favorite. |
One character may be in a hopeless fight. But what are the bigger dimensions? How many company members/allies to help? Can a cleric keep character from "death"? And characters are marked, so death is not permanent. If a character 'dies' but the team wins, they can protect the site until the spirit returns.
|
Goblin Squad Member
|
| 2 people marked this as a favorite. |
Am I the only one that's a little disappointed by the change to a non-secure protocol FROM a secure protocol (HTTP is transmitted clear text, HTTPS transmitted encrypted)? This will only generate future security problems for the website if all commands are sent in clear text. It's a step back.
|
Goblin Squad Member
|
Am I the only one that's a little disappointed by the change to a non-secure protocol FROM a secure protocol (HTTP is transmitted clear text, HTTPS transmitted encrypted)? This will only generate future security problems for the website if all commands are sent in clear text. It's a step back.
They're working on the http -> https SSL issue now.
|
Goblin Squad Member
|
sort of odd that I am replying to myself...
Oh, I dunno. It doesn't seem a terribly uncommon occurrence around these parts. Stick around and you'll see the same crop up here and there in the most surprising quarters.
It gets even more interesting when the same author posts a diatribe against the very argument he made a few months ago.
|
Goblin Squad Member
|
| 2 people marked this as a favorite. |
Am I the only one that's a little disappointed by the change to a non-secure protocol FROM a secure protocol (HTTP is transmitted clear text, HTTPS transmitted encrypted)? This will only generate future security problems for the website if all commands are sent in clear text. It's a step back.
I am glad someone besides me pays attention to such details.
|
Goblin Squad Member
|
sort of odd that I am replying to myself...Oh, I dunno. It doesn't seem a terribly uncommon occurrence around these parts. Stick around and you'll see the same crop up here and there in the most surprising quarters.
It gets even more interesting when the same author posts a diatribe against the very argument he made a few months ago.
I once took part in a PbP game and created two characters for it. Yet kept it secret I was playing both. I spawned an intense IC and OOC rivalry between both characters that often got the GM and other players to jump in and play peacekeepers. :P
|
Goblin Squad Member
|
| 1 person marked this as a favorite. |
@Nightdrifter
You are a boon to the community friendI love how we can tell which one is the liberal sciences theory crafter type and which is the applied sciences theory tester. (....)
+1
Pretty much all the info I have ultimately comes from Stephen Cheney (a dev) in some way, so only fair he gets credit!
That's a very well formulated disclaimer!
Nightdrifter's conclusions are obviously only as good as his information (ie assumptions and hard data), and are strictly limited to the cases he models. However the underlying methods appear very sound (robust, even) and can easily accommodate new assumptions and data.
In short: Nightdrifter knows what he's doing and has given us a very good vantage point for discussing certain game aspects.
However (over)interpreting his data should be done with a bit of caution as there are plenty of explicit and implicit assumptions that affect the answers, the corner cases are often far apart, and it is not always clear if we are discussing potential outcomes, likely outcomes, the validity of model or assumptions, etc. etc.
The "10 v 1" discussion is a prime example here. If you ask Nightdrifter "in a duel, would you rather have higher tier equipment or a friend to help you?", you may get a different picture than with the current default question which is "will normally a T3 fighter beat two T2 fighters if they stand still and autoattack each other?"
|
Goblin Squad Member
|
| 1 person marked this as a favorite. |
If you ask Nightdrifter "in a duel, would you rather have higher tier equipment or a friend to help you?"
How much better equipment do I get? And how does my friend's gear/skill match up against my opponent's? :p
I'd consider movement and combos to be the biggest parts missing from my analysis for small scale combat. Movement is hard to deal with in current formats. Combos are still a wildcard. We've only seen one example set of slotted attacks and two potential combos. How effective those specific combos are is very highly dependent upon how often 'opportunity' comes up ... which means the combos are movement dependent.
I'm torn on how important movement will be. Casters and archer style characters will obviously want to keep out of melee range. However, I tend to think melee will be less running around eachother than some other games. Since facing isn't really important there's no point in trying to get behind your opponent so he can't attack you and things like that (no circle strafing). Also running circles around a fighter will make you open to opportunity and grant him all sorts of bonuses to his combos. So there's no advantage to running around your opponent and against fighters there's a distinct drawback to doing it. That favors just standing there.
One possible advantage to movement in melee combat is blowing all your stamina, backing out of melee range to avoid taking any hits while you can't attack, then returning to melee once you have stamina back. However, that may provoke opportunity and your opponent can probably follow you easily. So it might be advantageous or it may do more harm than good. Intuitively I expect both opponents to blow their stamina at roughly equal rates, meaning that when you're out with roughly coincide with him being out. So all backing out does is make you open to opportunity. Again that favors stationary combat.
There's some Ryan Dancey quote somewhere *casts Summon Nihimon* about combat being largely about conditions and less about movement. That might favor stationary combat over very heavy movement based combat.
In all likelihood it's too early to say how important movement will be. How much and when to move likely falls into the hard to model category of player skill.
Goblinworks Executive Founder
|
I'd say the biggest factor influencing multiple combatants is that multiple combatants won't be polite and fight to death. When the one gets close to taking down one of the ten, the victim will withdraw, forcing the one to suffer attacks from the nine and chase down his target, or switch targets, allowing his first target to swing back in unnoticed, or heal up for a second round.
|
Goblin Squad Member
|
| 1 person marked this as a favorite. |
There's some Ryan Dancey quote somewhere *casts Summon Nihimon* about combat being largely about conditions and less about movement. That might favor stationary combat over very heavy movement based combat.
I think this is probably what you're remembering.
The key thing about the combat system as it is currently envisioned is that your goal is to maximize the combos not worry about position and facing. You can see various indicators blinking on and off in the UI as the fights progress. What we want players to do is think about the various attack options they have to inflict a condition on a target or benefit from a condition on a target.
Instead of having a pre-programmed order that you use your attacks you should be choosing moment to moment what the most effective option is based on the conditions affecting you and your target, and your knowledge of what your allies can do based on what conditions you can inflict on a target.
So its more strategic and less tactical than WoW style "run the macro again" combat.
(emphasis in original)
| 2 people marked this as a favorite. |
Meaning we can use it at will?
Give me that spell! I need to troll the Nihimonster!
|
Goblin Squad Member
|
While I can really appreciate the intellectual contortions involved, are we over thinking combat? It is my opinion that, overall, quality of weapon/armor will fall victim to superior tactics and facility with "choosing moment to moment what the most effective option is based on the conditions affecting you and your target, and your knowledge of what your allies can do based on what conditions you can inflict on a target" (as Ryan has said) more often than one might think. In the end I believe that superior social organization and planning will be the true measure of successful PvP.
|
Goblin Squad Member
|
While I can really appreciate the intellectual contortions (....) I believe that superior social organization and planning will be the true measure of successful PvP.
Agree on both counts. The key variable for pvp remains: "you and what army?"
(hint: if you don't know the answer, you can call a friend)
PS: But that's not diminishing the value of what Nightdrifter is doing. For evaluating builds and equipment - and even more so for evaluating proposed mechanics - the analyses are valuable.
|
Goblin Squad Member
|
Should point out that not only has the blog been updated, but it appears much of the goblinworks site has been as well. They've included some nice new screenshots.
