I don't get Rangers...

[dunehunter]

Rangers are effective vs enemy high level mages. They shines on the Fampyr island aganist those high deflection Fampyr mages with Pierce Sigil. Also if you wanna kill an Unbending Fighter, Ranger is a good idea.

[grasida]

I haven’t played one, but shouldn’t single class street fighter consistently be one of the best single target damage dealers? Especially at high levels with gambit it should compete with most multiclasses.

[Verde]

Not really. It's a gimmicky build for the Blunderbuss unless you multi. Way way too squishy to go head to head with flanking enemies AND be Bloodied as a single.

Edited June 23, 2018 by Verde

[Boeroer]

Sure you can use math to make a point that doesn't exist. If people want to think that way then I can't help that.

Of course it exists. Accuracy does affect average dps. Higher accuracy means higher chance for an higher attack roll quality. This can mean graze instead of a miss, hit instead of a graze or crit instead of a hit. All of those entail a significant damage increase that's hard to achieve with speed or dmg mods alone. Putting it at 2% is just an approximation over all cases. I don't know if this is a reasonable approximation or not. Unlike MIG or other dmg bonuses (or attack speed) accuracy's impact on dps is hard to measure because it highly depends on enemies' stats.

 

However there is NO doubt that more accuracy translates to increased dps (unless the difference between your accuracy and all enemies' defenses is so high that you ALWAYS crit. Only then it's not). And that leaves stuff aside that can happen when you crit (special procs like Enervating Blows, Blood Frenzy and whatnot).

 

So, claiming that accuracy has no effect on dps is plain wrong. And saying that you can't use math to determine the average dps gain per point of accuracy - because that doesn't exist - is ignorant. Of course you can. It's very complicated and you'd have to take every weapon,every ability and all enemies' defenses into account. But I guess if you'd do that then wouldn't be too far off (of 2%). Even if it's a very generalized approximation and might not help you in certain situations. Accuracy has a tremendous impact if it is lower than enemies' defenses and looses impact the higher it gets (compared to enemies' defenses). But it will add dps in nearly all cases.

 

Saying a hit roll is a hit roll and accuracy in this case doesn't help dps is like saying that a miss roll is a miss roll and all damage bonuses don't increase dps (in this case). You are not talking about dps in this case but simply about the damage per hit. That is not dps. Dps is an average value that you can only determine when looking at more than one attack (minimum two attacks). You can't determine dps with looking at one single attack resolution. Since damage per second is a value that's determined over time you have to look at a relevant number of attack resolutions to determine its value.

 

You can't say "oh the dps for this single attack is not influenced by accuracy" - because there is no dps (damage per second) when you only look at one single attack. Only the damage of this single hit. The time aspect (seconds) is simply missing. You need more attacks to determine dps. And when doing more attack resolutions you will get increased damage due to higher hit quality due to higher accuracy. Thus dps will rise.

 

Rangers can have the highest accuracy in this game - and that translates to higher dps. That is especially helpful on PotD (where defenses and AR are higher) and also against all bosses. If that's enough to balance out the shortcomings of the class I can't say. But I guess the complaints are hyperbole. I do think that single rangers have crappy high level abilities though and thus seem to be better for multiclassing.

Edited June 23, 2018 by Boeroer

[AeonsLegend]

 

Sure you can use math to make a point that doesn't exist. If people want to think that way then I can't help that.

Of course it exists. Accuracy does affect average dps.

 

Sorry for just quoting this tidbit, but dps and average dps imply different things. At the end of the game your dps is decided by three things: 1: average damage of your weapons with all buffs. 2: attack speed and recovery. 3. Chance to hit your opponent.

 

If you count all damage at the end of the game and look at accuracy values then yes you can calculate that a certain point of accuracy takes a percentage value of this damage.

 

But it doesn't work the other way around because it is an indirect reference based on theory and you shouldn't use it in a discussion to prove a point. There is no circumstance where accuracy directly affects dps. Only indirectly over a long time can it be measured. Saying 1 accuracy point is a 2% increase in dps is wrong. Saying that a certain value of accuracy contributes to a certain part of dps over the course of the game, yes sure. But again those are two completely different things.

Edited June 23, 2018 by AeonsLegend

[grasida]

 

But it doesn't work the other way around because it is an indirect reference based on theory and you shouldn't use it in a discussion to prove a point. There is no circumstance where accuracy directly affects dps. Only indirectly over a long time can it be measured. Saying 1 accuracy point is a 2% increase in dps is wrong. Saying that a certain value of accuracy contributes to a certain part of dps over the course of the game, yes sure. But again those are two completely different things.

 

You really seem to be confused about the seconds part of damage per second.

 

Not really. It's a gimmicky build for the Blunderbuss unless you multi. Way way too squishy to go head to head with flanking enemies AND be Bloodied as a single.

Again, I haven't played one, but you don't need both of those things. On another site, I read someone who described doing quite well with a single class streetfighter and largely ignoring flanked after the early game. The idea was to stand on the front line controlling positioning in the early stages of a fight, then jumping to the enemy's rear once bloodied and deleting the casters and ranged fighters.

 

Part of the reason that can work is that streetfighters gain comparatively little from might, since they already get overwhelmingly large amounts of +% damage. So I imagine they're durable enough to work as an off-tank or flanker if you go with low might and very high constitution.

[Verde]

I'm not sure why you reiterated you didn't play one, because that essentially makes your opinion complete conjecture based on someone else's playthrough.

 

I in fact played one, first as a single then multi with Fighter, to take advantage of being Flanked and Bloodied. With the new increase in difficulty it's just not as viable as single, because Rogues are squishy. The class itself offers no defensive benefits which is peculiar. As it is, it's a great multi with Fighter or Barb. But on it's own, it's not. It's not just about Constitution, it's about Deflection. So essentially you're building a Rogue like a tank and trying to crit all the time. Why not just multi? The entire pt of a Streetfighter is to get surrounded and hurt, which is the antithesis of a Rogue.

 

And if you did your due diligence you would know that a Blunderbuss is almost required esp for a single who, as said earlier, is not tough enough to stand toe to toe and get flanked. Take the already slow recovery system and add +50% to it, and that's what you have without Powdered Burns.

Edited June 23, 2018 by Verde

[AeonsLegend]

 

But it doesn't work the other way around because it is an indirect reference based on theory and you shouldn't use it in a discussion to prove a point. There is no circumstance where accuracy directly affects dps. Only indirectly over a long time can it be measured. Saying 1 accuracy point is a 2% increase in dps is wrong. Saying that a certain value of accuracy contributes to a certain part of dps over the course of the game, yes sure. But again those are two completely different things.

You really seem to be confused about the seconds part of damage per second.

 

Not really. It's a gimmicky build for the Blunderbuss unless you multi. Way way too squishy to go head to head with flanking enemies AND be Bloodied as a single.

Again, I haven't played one, but you don't need both of those things. On another site, I read someone who described doing quite well with a single class streetfighter and largely ignoring flanked after the early game. The idea was to stand on the front line controlling positioning in the early stages of a fight, then jumping to the enemy's rear once bloodied and deleting the casters and ranged fighters.

 

Part of the reason that can work is that streetfighters gain comparatively little from might, since they already get overwhelmingly large amounts of +% damage. So I imagine they're durable enough to work as an off-tank or flanker if you go with low might and very high constitution.

 

I think you are. Dps is calculated per second. That means you devide the damage you did so you get what you would deal in 1 second. But that's not what I'm contesting. I'm contesting the fact that people say 1 accurcy means 2% increase in dps. It doesn't work that way because it is an avarage, not a direct value. Edited June 23, 2018 by AeonsLegend

[Loren Tyr]

I think you are. Dps is calculated per second. That means you devide the damage you did so you get what you would deal in 1 second. But that's not what I'm contesting. I'm contesting the fact that people say 1 accurcy means 2% increase in dps. It doesn't work that way because it is an avarage, not a direct value.

 

As a rule I would imagine people are much more interested in DPS in the sense of the damage rate they will achieve, or could achieve in a specific scenario, not the damage they did previously (which, by the way, also requires a bunch of specific assumptions to get you to a meaningful number). This, by definition, will actually have a statistical distribution rather than a single value even if hits were guaranteed, unless the attack always does the same amount of damage rather than having a range (and technically it'd still qualify as a distribution even then). So even taking the need to actually hit out of the equation, the DPS numbers that people will cite are almost always not direct values either; they're the expected value of the damage distribution. 

 

You seem to be ignoring what DPS values are actually used for by people. One obvious and common application is to compare different weapons or weapon types with each other, compare abilities, compare boosts to different stats. For any of these comparisons, defining DPS just as (attack damage) / (duration of attack cycle) is not particularly useful. If attack 1 does 10% more base damage than attack 2, but has five fewer penetration and -20 ACC (with everything else identical), then clearly under the vast majority of circumstances that are likely to arise attack 2 is better. But the DPS as you propose to define it does not reflect this, so then what good is that definition? It's certainly of no use for this kind of comparison, and I struggle to think of applications where it would indeed suffice.

[Boeroer]

Exactly. That's like saying a dagger with 10 base damage and +5 ACC does the same DPS as a hatchet with 10 base damage and +5 to melee deflection. Which obviously is wrong as you migh admit.

 

The exact dps gain is different against different enemies, but there is one. It may be not that useful to approximate the dps gain for all situations and form an avarage number like 2% - but leaving it completely out surely is of even less use in any case.

 

Making it simple and using a value of whatever % (maybe 2%) can make sense. It will not be exact in most cases but will give a better estimation for dps comparison than leaving it out.

Edited June 23, 2018 by Boeroer

[AeonsLegend]

I don't understand why people confuse dps with average damage values. Average is average. dps is damage per second. If I gain 1 accuracy my damage per second does not increase. Because damage per second can also be calculated over 3 seconds. Sure I may get lucky and get a crit instead of a hit, but that is totally situation dependent and unreliable. Talk about potential dps instead of actual values and forego using exact numbers in a theoretical situation and we can talk.

 

I mean are you guys also going to complain that you get 0 crits with an ability that adds 25% crit chance? Or are you just assuming you're going to get 25% crits all the time?

Edited June 23, 2018 by AeonsLegend

[Boeroer]

ROFL - I give up.

 

Maybe Loren Tyr has more patience on this matter.

Edited June 23, 2018 by Boeroer

[grasida]

I'm really having trouble understanding what you're trying to say.

 

Let's assume the simplest possible situation. No crits or grazes or penetration or anything complicated. You attack once per second, deal 10 damage per hit and hit every time. What's your damage per second? It's 10. Very simple. Now assume exactly the same speed and damage, but not you hit only 50% of the time. Now what's your dps? It's 5. Again, very simple. In this case, 1% accuracy translates exactly to 1% dps, unless you simply can't miss at all, at which point further gains in accuracy do nothing.

 

Let's make it easier to visualize: Here is your string of attacks over 10 seconds. Each attack deals 10 damage and you attack every second:

 

100 accuracy: 10, 10, 10, 10, 10, 10, 10, 10, 10, 10. What's your dps? 10*10 = 100. 100/10 = 10.

50 accuracy: 10, 0, 0, 10, 0, 10, 10, 10, 0, 0. What's your dps? 10*5 = 50. 50/10 = 5.

 

This is your damage per second. You are dealing more damage each second on average because more of your attacks hit. Therefore Accuracy increases damage per second.

[Loren Tyr]

I don't understand why people confuse dps with average damage values. Average is average. dps is damage per second. If I gain 1 accuracy my damage per second does not increase. Because damage per second can also be calculated over 3 seconds. Sure I may get lucky and get a crit instead of a hit, but that is totally situation dependent and unreliable. Talk about potential dps instead of actual values and forego using exact numbers in a theoretical situation and we can talk.

 

I mean are you guys also going to complain that you get 0 crits with an ability that adds 25% crit chance? Or are you just assuming you're going to get 25% crits all the time?

 

Yes, damage per second. For which we need the following components:

- the rate at which we attack, the number of attacks per second

- the change that attack will actually connect and do damage

- the amount of damage dealt by the attack if it connects

 

Quite clearly, accuracy is going to affect the second component. Given that we have three different types of hit with different damage amounts, accuracy will also affect the third component. As, for that matter, will things like hit conversion effects and penetration. And yes, as should be quite obvious, the damage you will do per second is therefore dependent on the attributes of whatever you're trying to damage. That doesn't change the fact that you can just ignore factors like accuracy like you are intent on doing, and be left with a meaningful metric.

 

But maybe let's reverse this. How about you explain to us what use the DPS value you are proposing is to anyone? In terms of practical application, what can I actually do with it?

 

I'd also be quite interested in why you are so hung up on averages here. Almost invariably, even with your simplified DPS, you're going to end up using averages. Because let us suppose we're using an axe, just the standard 13-19 damage range with no penalties or bonuses. And suppose we are attacking with this axe at a rate of one attack per four seconds. What, according to you, is the DPS in this case; and how is that not an average? 

[XEternalXDreamsX]

Accuracy does help overall. I don't actually have a video for it (Or stats to back percent. Ect) but once your accuracy is well above the enemies' AR, you will likely hit steadily. Doesn't it have potentially diminishing returns once it is really high(excluding level scaling, POTD)?

 

Since Ranger has really good accuracy, wouldn't they be able to disperse stats differently and equipment?

[mosspit]

Accuracy does help overall. I don't actually have a video for it (Or stats to back percent. Ect) but once your accuracy is well above the enemies' AR, you will likely hit steadily. Doesn't it have potentially diminishing returns once it is really high(excluding level scaling, POTD)?

 

Since Ranger has really good accuracy, wouldn't they be able to disperse stats differently and equipment?

As mentioned, those scenarios can happen when the difference between Acc (Accuracy of your attacks) and the corresponding save for said attack (generally 1 of the 4 saves) is greater than 99. This will mean all instances of said attack will already crit and additional Acc will not change this.

 

Similarly, if the difference is too huge such that additional acc will still result in a difference of < -100, additional acc (which again assuming still keeps the acc low enough) will not help to make the attack qualify for graze.

 

In these fringe cases, acc will not do anything unless the respective threshold can be crossed.

 

But yes in a way, there is some diminishing returns if you compare the gains between a situation where an attack can still qualify for all 4 hit resolution states vs the extreme all crit or all miss situations. However, it is generally uncommon in a non-min/ax scenario.

Edited June 23, 2018 by mosspit

[mant2si]

Okay I will try to put some clarification about acc, at first you can check my build https://forums.obsidian.net/topic/102168-build-glanfathan-soul-hunter, one of main build feature is high acc, really high. Without consumable, with shield and pistol modal I can achieve ~130 acc (

), with queen hat, this value equal to 140.

If I used DW (Scordero's + Modwyr) without modal, my acc +150 and +160 with queen hat
If I used Single hand weapon, without modal and with hat I can hit ~172, If I used +perception drug, then acc will be ~180

But lest's stop at  ~130 and take as example Ukaizo fight, Ukazio with DD mod has 81 Deflection (huh pretty low), but I don't get +2% for DPS per acc (i.e 50 acc === 100% to dps), because when you overcome enemy deflection and get 100%  chance to hit, there no point rise acc above, because crit give you only +25% DM and some penetration (But of course I can't overpen Ukaizo armor without Alchemy)

At summary accuracy will give you dps only if you graze/miss to much otherwise there no point to rise accuracy so high. Much simpler will be take "Aware" affliction with Graze -> Hit conversion 

[Verde]

So essentially, accuracy has diminishing returns. A blanket % dps/acc is not correct.

[the streaker]

A ton of unique weapons have some wicked modifiers that only trigger on crits... and my Ranger/Rogue Maia was second in the group in damage dealt with minimal micro, so I'd say they're not too bad.

Edited June 24, 2018 by the streaker

[aimlessgun]

Okay I will try to put some clarification about acc, at first you can check my build https://forums.obsidian.net/topic/102168-build-glanfathan-soul-hunter, one of main build feature is high acc, really high. Without consumable, with shield and pistol modal I can achieve ~130 acc (

), with queen hat, this value equal to 140.

 

If I used DW (Scordero's + Modwyr) without modal, my acc +150 and +160 with queen hat

If I used Single hand weapon, without modal and with hat I can hit ~172, If I used +perception drug, then acc will be ~180

 

But lest's stop at  ~130 and take as example Ukaizo fight, Ukazio with DD mod has 81 Deflection (huh pretty low), but I don't get +2% for DPS per acc (i.e 50 acc === 100% to dps), because when you overcome enemy deflection and get 100%  chance to hit, there no point rise acc above, because crit give you only +25% DM and some penetration (But of course I can't overpen Ukaizo armor without Alchemy)

 

At summary accuracy will give you dps only if you graze/miss to much otherwise there no point to rise accuracy so high. Much simpler will be take "Aware" affliction with Graze -> Hit conversion 

 

Yes, in this extreme case of 160 acc vs 81 deflection (wow that is crazy low), the my spreadsheet says +1 acc is worth only 0.2% increase in DPS, so quite worthless to stack more at this point

 

When I said '2%' this is a rough estimate for say an average character on PotD where their acc is close to enemy defense or lower. For example right now in midgame my characters are around 80 acc and generally fighting enemies with 80-110 defenses. 

 

As the streaker says above lots of weapons have 'on crit' effects though, but that, like the penetration benefits, would be more work to model than I'm willing to invest!

 

Stacking almost anything in this game also has diminishing returns, if you already have +100% increased damage then one more point of might is not 3% more DPS, only 1.5% more (3/200 instead of 3/100). Same with dex if you already stacked a lot of action/recovery speed. So as in most RPGs you want a balance. 

Edited June 24, 2018 by aimlessgun

[Boeroer]

because when you overcome enemy deflection and get 100%  chance to hit, there no point rise acc above, because crit give you only +25% DM and some penetration

I think there's alsways is a point to achive crits instead of hits if you can. Not only does if give you +25% additive damage (which isn't so overhwhelmingly good but still useful) but there are a lot of abilites and items that do something on crit (Evervating Blows, Blood Frenzy, interrupts(!) and so on).

 

But that aside, a crit doesn't only give you "some" penetration, but multiplies your overall PEN with 1.5 - which leads to a better PEN quality (and dmg raise) in a lot of cases. The Guardian of Ukaizo has high AR, but if you yourself have stacked a lot of PEN (abilites, legendary equipment, Power Level boosts) you can improve your PEN quality with crits quite easily - even to the point of overpenetration.

 

But it's true that in these extreme cases additional points of accuracy don't do as much for you as in the more "normal" fights throughout the game. Actually this makes the ranger more valuable for PotD than for lower difficulties where you will reach the point of "more ACC is not adding that much anymore" a lot sooner.

 

Anyway: PoE's general rule of thumb "more ACC is always better" still applies.

Even more so in Deadfire because afflictions' impact on enemies' defenses is a lot lower than n PoE. Accuracy is very valuable and thus the acc buffs of the ranger make him valuable, too. And never forget the additional body - it's very, very useful to have a body who can't be injured but needs no resources to be there - even if it only works as cannon fodder. 

Edited June 24, 2018 by Boeroer

[Archaven]

Poe was about accuracy and deadfire is both accuracy and penetration. And what's really brilliant about pillars is that obsidian made gear (items/weapons/armors) as part of builds which makes the game really, really awesome.

Edited June 24, 2018 by Archaven

[knownastherat]

How is making gear as part of builds awesome? It's just stupid. The concept of PEN in Deadfire is stupid. 

Edited June 24, 2018 by knownastherat

[Boeroer]

Making gear part of builds is definitely not stupid. It adds a lot of replay value.

 

The PEN system is also not my favorite mechanic in Deadfire. It's just too tedious to keep track if your PEN is sufficient or not all the time. I think that's why Devoted, Berserker or Sharpshooter are popular subclasses: just make sure you pile up a ton of PEN and don't bother further with PEN vs. AR.

Edited June 24, 2018 by Boeroer

[knownastherat]

How does it add to replay value? A practical example would be nice. 

 

With multi-classing, there is so much replay value that only a fraction of players will take advantage of such value. Doing anything for a fraction of players is stupid  irrational.

 

edit: as of now PEN works like this: get a weapon with as much PEN as soon as possible (or armor with rating as high as possible), because else some encounters will not be possible or at least a chore independent of "build" or "skill", and be done with it for the rest of the game. wtf?

 

and if you are a caster, too bad, you have to wait till your PEN levels up .. lol ... or grab a weapon, right? 

Edited June 24, 2018 by knownastherat