sorry, but "durability line" is not a good idea, imo
x is HP (linear)
f(y) is a defence function, where y is defence tooltip
Zmax = x/(1-f(y)/100), (maximum damage taken) is utility function (durability)
now you build map (x,y) for Zmax (map is the set of indifference curves) with dx = 400, dy = 100 (steps, GS-wise)
I assume that for each indifference curve you take the point that lies closer to zero... Bad idea:
1) gear for different toon-types is different (GF can easily stack 5k defence, what to do with that?)
Simple, whenever you have a choice between HP and Defense you choose HP since it will help more. As I stated in the first post, this is a tool to help you see whether HP or Defense will help you more. It doesn't mean a GF has to reduce their defense in order to be on the line. The line just represents the spot where it doesn't matter whether you add HP or defense because they will help an equivalent amount.
I don't see how that is relevant. A character is still going to fall somewhere on this chart and will need to decide what to put in the defense slots in their equipment.
Which would be relevant if I were making a guide on building the best character possible for a given amount of AD. I'm not. This is strictly a "what's best?" analysis. Not a "what is most cost effective" analysis. As an aside: you know calculus, but forget the number 3? Come on, man! Bring your A-game to this math brawl!
Better idea: build map, find curve on map that better suits your goals (max damage you want to survive), twink your gear to make it optimal
e.g. for 35000 HP 1500 defence (which is optimal, according to graph) the values: 33800 HP 1849 defence will be on the same indifference curve, which can be easily achieved by swapping radiant r10 with azure r10 and adding 49 defence from somewhere, but even without this additional 49 defence result will be almost the same.
The key word there is "almost." Yes, it would be "almost" as good. But it would not be "as good." If you have 35000 HP and 1500 defense and swap out a Rank 10 Radiant for a Rank 10 Azure it will be "almost" as good, but you will be decreasing the amount of damage your character can withstand. Hence, why the line is where it is.
If you would like to expand on this by creating your own "cost optimization" charts or creating your own "indifference curves" feel free. Criticize my work if it is inaccurate, but please don't criticize it for not being something it was never intended to be.
Sure. But here are the problems that do matter and why I will use your chart as a baseline but not in overall decissions about my build:
1. PVP most semi-serious PVP players are going to be around 30% ArPen. At which point 1600 defense and 0 defense have the same effect. Deflect ignores ArPen and is a lot more effective under these conditions. Basically unless you are a tank or playing low-end PVP you can forget defense. If you are tanky subtract about 1500 from your defense before using the chart. But even that won't work because of diminishing returns. Love to see a version assuming 30% ArPen on attackers.
2. PVE between pots LS and regen your real concern is not 'what is my EHP?' your real concern is how much spike damage you can take and how fast you can recover from it. So how much of the spike you can mitigate weighs heavier in the calculation because mitigated damage does not decrease your HP pool and having a bigger pool does not improve LS (it will have an effect on regen). At this point high end content spikes can be in the 30K range. Once you are over 30K HP the real concern is in how much of that you can mitigate and how fast you can recover it.
3. In most cases this is a good formalism to prove the point that radiants are usually best in defense slots. Actually making build decissions based on anything other than enchants is likely a mistake. If you end up swapping a radiant in for a azure due to this data then you have made the right choice. If you change gear set to raise your HP or lower your defense you have been mesmerized by numbers and are making a mistake.
with all the new artifact and jewelcraft accessories, its hard to stack tenacity
if you mix all 3 (HP, tenacity, def) I wonder what will be the result
*tenacity only exist on equip and it also give control and cirt severity debuff
0
katbozejziemiMember, NW M9 PlaytestPosts: 856Arc User
edited January 2015
They should move the HP bonus from Radiant enchantments to Silvery or something. It's getting worse and worse as we get better gear and radiants already have the lowest drop rate AND the highest desirability.
The line just represents the spot where it doesn't matter whether you add HP or defense because they will help an equivalent amount ... If you would like to expand on this by creating your own "cost optimization" charts or creating your own "indifference curves" feel free. Criticize my work if it is inaccurate, but please don't criticize it for not being something it was never intended to be.
:sigh: Well, i expected answer like this. I never said that you graph is inaccurate. Problem here that this graph is less informative that indifference map (it's not my term, don't need to put it in quotes). http://en.wikipedia.org/wiki/Indifference_curve . Each indifference curve on map is made of spots where "it doesn't matter whether you add HP or defense because they will help an equivalent amount". Map gives you the possibility to choose whether you stay at the point of your graph or you follow the curve to less effective (in terms of GS) but more effective for your goals point (e.g saving AD (?)). Also you forget about that you are making the graph for a game, not some abstract graph for 2 functions, and this "almost" word really matters.;-)
Again: i'm not saying that your graph is bad, i'm just saying that this line is not a panacea.
Also, if you build a full map, like http://s10.postimg.org/hckt7wlnt/Untitled_1.jpg you will see that effectiveness of increasing HP vs defence greatly decreases as HP increases.
My point is that's you want azures in all slots for PVE (if you are not a GF, ofc, and if you have >=30k HP). :-) Why? Because durability increase from using radiants is not that high (not to say about AD).
As for creating my own topic: look, my mate sent me a link to this thread, I did some calculations and posted them on my guild's site to prevent people for paying lots of money for even more radiants, as my guild is 99% PVE. This thread only increases "radiant rush", which is absolutely useless, if spec is bad or guys don't know how to play. I also think that we both know that PVE content in this game is so unchallenging now with this all new gear, that it doesn't really matter what to put in these defence slots, unless you are going to solo what? Tiamat? :-)
Again: i'm not saying that your graph is bad, i'm just saying that this line is not a panacea.
For a general guide is more than enough, if pvp tournaments would be involved then we could go further and look at the potential burst/recovery time and burst recovery time. For pve the same, if there would have been pve that would present any kind of challenge...
400 hp can mean the difference between surviving a burst or not, i cant tell you how many times someone escaped with less than 1000 hp, or forced me to overkill only to pop SF and get away ... But that hardly matters in casual pvp.
If i can withstand the burst with both, as a warrior class i would take def stones to increase my dmg.
The problem is that a fair number of people are going to look at the charts you made and not know how to interpret them. I was setting out to make this easily understandable to whomever is playing this game.
People should absolutely use their discretion in determining whether or not they have the disposable currency to go for the optimal build (with Radiants) or whether they should settle for a still good, but slightly sub-optimal build (Azures). This guide is about "here's a tool to determine what is optimal." Given the uniquness of each class (GF's with their Shield, GWF's with Unstoppable (and the fact that more defense gives them power), SW's with crazy healing capabilities, whether or not one has high DPS and Lifesteal), and also given whether one is playing PvE or PvP (where Tenacity, high Armor Penetration, and other considerations come into play), and different economic considerations, people can and should engage their brains and deviate from the line as they determine necessary.
*Edit*
One other thing. I decided to investigate this after attempting to solo Epic Temple of Spiders. I needed every stinking drop of survivability I had to make it through there alive. Afterward I decided it was worthwhile to look into whether I had the right distribution of defensive stats. I didn't. If I had it may have gone a bit smoother.
[SIGPIC][/SIGPIC]
0
thefabricantMember, NW M9 PlaytestPosts: 5,248Arc User
:sigh: Well, i expected answer like this. I never said that you graph is inaccurate. Problem here that this graph is less informative that indifference map (it's not my term, don't need to put it in quotes). http://en.wikipedia.org/wiki/Indifference_curve . Each indifference curve on map is made of spots where "it doesn't matter whether you add HP or defense because they will help an equivalent amount". Map gives you the possibility to choose whether you stay at the point of your graph or you follow the curve to less effective (in terms of GS) but more effective for your goals point (e.g saving AD (?)). Also you forget about that you are making the graph for a game, not some abstract graph for 2 functions, and this "almost" word really matters.;-)
Again: i'm not saying that your graph is bad, i'm just saying that this line is not a panacea.
Also, if you build a full map, like http://s10.postimg.org/hckt7wlnt/Untitled_1.jpg you will see that effectiveness of increasing HP vs defence greatly decreases as HP increases.
My point is that's you want azures in all slots for PVE (if you are not a GF, ofc, and if you have >=30k HP). :-) Why? Because durability increase from using radiants is not that high (not to say about AD).
As for creating my own topic: look, my mate sent me a link to this thread, I did some calculations and posted them on my guild's site to prevent people for paying lots of money for even more radiants, as my guild is 99% PVE. This thread only increases "radiant rush", which is absolutely useless, if spec is bad or guys don't know how to play. I also think that we both know that PVE content in this game is so unchallenging now with this all new gear, that it doesn't really matter what to put in these defence slots, unless you are going to solo what? Tiamat? :-)
Hey ther3n, its me sharp I did ask you if you would map this, so thanks for doing all this work
The problem is that a fair number of people are going to look at the charts you made and not know how to interpret them. I was setting out to make this easily understandable to whomever is playing this game.
People should absolutely use their discretion in determining whether or not they have the disposable currency to go for the optimal build (with Radiants) or whether they should settle for a still good, but slightly sub-optimal build (Azures). This guide is about "here's a tool to determine what is optimal." Given the uniquness of each class (GF's with their Shield, GWF's with Unstoppable (and the fact that more defense gives them power), SW's with crazy healing capabilities, whether or not one has high DPS and Lifesteal), and also given whether one is playing PvE or PvP (where Tenacity, high Armor Penetration, and other considerations come into play), and different economic considerations, people can and should engage their brains and deviate from the line as they determine necessary.
*Edit*
One other thing. I decided to investigate this after attempting to solo Epic Temple of Spiders. I needed every stinking drop of survivability I had to make it through there alive. Afterward I decided it was worthwhile to look into whether I had the right distribution of defensive stats. I didn't. If I had it may have gone a bit smoother.
Would you mind attempting to solo it again...purely for the sake of comparison?
well, Sharp, i see what you are doing here, mate, xD
Abaddon, about interpreting the graphs. Don't understand how it can be hard to interpret, sorry. But i think i will post my "method".
I always choose HP over Defense as it buffs regeneration and other type of heals that are based on % of health or max hit points. Also certain skills and items bypass defence.
Comments
Simple, whenever you have a choice between HP and Defense you choose HP since it will help more. As I stated in the first post, this is a tool to help you see whether HP or Defense will help you more. It doesn't mean a GF has to reduce their defense in order to be on the line. The line just represents the spot where it doesn't matter whether you add HP or defense because they will help an equivalent amount.
I don't see how that is relevant. A character is still going to fall somewhere on this chart and will need to decide what to put in the defense slots in their equipment.
Which would be relevant if I were making a guide on building the best character possible for a given amount of AD. I'm not. This is strictly a "what's best?" analysis. Not a "what is most cost effective" analysis. As an aside: you know calculus, but forget the number 3? Come on, man! Bring your A-game to this math brawl!
The key word there is "almost." Yes, it would be "almost" as good. But it would not be "as good." If you have 35000 HP and 1500 defense and swap out a Rank 10 Radiant for a Rank 10 Azure it will be "almost" as good, but you will be decreasing the amount of damage your character can withstand. Hence, why the line is where it is.
If you would like to expand on this by creating your own "cost optimization" charts or creating your own "indifference curves" feel free. Criticize my work if it is inaccurate, but please don't criticize it for not being something it was never intended to be.
1. PVP most semi-serious PVP players are going to be around 30% ArPen. At which point 1600 defense and 0 defense have the same effect. Deflect ignores ArPen and is a lot more effective under these conditions. Basically unless you are a tank or playing low-end PVP you can forget defense. If you are tanky subtract about 1500 from your defense before using the chart. But even that won't work because of diminishing returns. Love to see a version assuming 30% ArPen on attackers.
2. PVE between pots LS and regen your real concern is not 'what is my EHP?' your real concern is how much spike damage you can take and how fast you can recover from it. So how much of the spike you can mitigate weighs heavier in the calculation because mitigated damage does not decrease your HP pool and having a bigger pool does not improve LS (it will have an effect on regen). At this point high end content spikes can be in the 30K range. Once you are over 30K HP the real concern is in how much of that you can mitigate and how fast you can recover it.
3. In most cases this is a good formalism to prove the point that radiants are usually best in defense slots. Actually making build decissions based on anything other than enchants is likely a mistake. If you end up swapping a radiant in for a azure due to this data then you have made the right choice. If you change gear set to raise your HP or lower your defense you have been mesmerized by numbers and are making a mistake.
if you mix all 3 (HP, tenacity, def) I wonder what will be the result
*tenacity only exist on equip and it also give control and cirt severity debuff
example of building a map (3 radiants vs 3 azures):
http://www.wolframalpha.com/input/?i=ContourPlot[++y%2F%281+-+%2849.99*x%2F%281643.6+%2B+x%29%29%2F100%29%2C+{x%2C+1000%2C+1900}%2C+{y%2C+30000%2C+33600}]+
yepp, HP is more effective, as we see, now let's look at effectiveness:
33600/(1 - (49.99*1000/(1643.6 + 1000))/100) = 41435.4
30000/(1 - (49.99*1900/(1643.6 + 1900))/100) = 40985.6
you can survive 449.765 (!) more damage, that's really a great deal, right?;-)
Again: i'm not saying that your graph is bad, i'm just saying that this line is not a panacea.
Also, if you build a full map, like http://s10.postimg.org/hckt7wlnt/Untitled_1.jpg you will see that effectiveness of increasing HP vs defence greatly decreases as HP increases.
My point is that's you want azures in all slots for PVE (if you are not a GF, ofc, and if you have >=30k HP). :-) Why? Because durability increase from using radiants is not that high (not to say about AD).
As for creating my own topic: look, my mate sent me a link to this thread, I did some calculations and posted them on my guild's site to prevent people for paying lots of money for even more radiants, as my guild is 99% PVE. This thread only increases "radiant rush", which is absolutely useless, if spec is bad or guys don't know how to play. I also think that we both know that PVE content in this game is so unchallenging now with this all new gear, that it doesn't really matter what to put in these defence slots, unless you are going to solo what? Tiamat? :-)
For a general guide is more than enough, if pvp tournaments would be involved then we could go further and look at the potential burst/recovery time and burst recovery time. For pve the same, if there would have been pve that would present any kind of challenge...
400 hp can mean the difference between surviving a burst or not, i cant tell you how many times someone escaped with less than 1000 hp, or forced me to overkill only to pop SF and get away ... But that hardly matters in casual pvp.
If i can withstand the burst with both, as a warrior class i would take def stones to increase my dmg.
People should absolutely use their discretion in determining whether or not they have the disposable currency to go for the optimal build (with Radiants) or whether they should settle for a still good, but slightly sub-optimal build (Azures). This guide is about "here's a tool to determine what is optimal." Given the uniquness of each class (GF's with their Shield, GWF's with Unstoppable (and the fact that more defense gives them power), SW's with crazy healing capabilities, whether or not one has high DPS and Lifesteal), and also given whether one is playing PvE or PvP (where Tenacity, high Armor Penetration, and other considerations come into play), and different economic considerations, people can and should engage their brains and deviate from the line as they determine necessary.
*Edit*
One other thing. I decided to investigate this after attempting to solo Epic Temple of Spiders. I needed every stinking drop of survivability I had to make it through there alive. Afterward I decided it was worthwhile to look into whether I had the right distribution of defensive stats. I didn't. If I had it may have gone a bit smoother.
Hey ther3n, its me sharp
Would you mind attempting to solo it again...purely for the sake of comparison?
Abaddon, about interpreting the graphs. Don't understand how it can be hard to interpret, sorry. But i think i will post my "method".