Update 23 Second Look

[FCofKhatovar]

are we still arguing semantics of phrasing and POV when looking at PRR?

WOW.

More. Of. The. Most. Epic. Sadfaces.

Please folks, we know how the PRR system works. We all see it a particular way.

THE NEW WAY IS STILL BETTER THAN THE OLD.

How bout how the best destiny for any melee is going to be near useless?

[FCofKhatovar]

Sounds to me a cheap solution like it was for SWF double strength bonus that got just removed.

I couldn't agree more with this.

Tho, I do have to give them credit for the work they are doing to improve this system.

I don't want people to misunderstand what I've been trying to say here...I DO think LD was overpowered. I DO think it needs a reduction.

I just think that reduction needs to be less severe. ~60% as powerful as it was before with a more easily sustained blitz.

This calculation should NOT take into consideration how much longer a player can stay alive and in DPS mode, as EVERY build is getting this very same increase in survivability.

[EDIT]: It also shouldn't take into consideration any melee power gained outside the destiny when comparing the numbers to what blitz was, as, all other builds are also getting these increases.

[BigErkyKid]

PRR has diminishing returns: two examples

Take Sev's example. A mob hits you for 100 DPS, you have 1000HP

Now suppose for whatever reason devs gave away a cosmetic item that granted 50 PRR.

In this case, you can fight for 15 seconds before healing.

Now you ask yourself the following question, by how much would I have to increase my PRR to double my combat time?

That is, I want to be able to fight 30 seconds before healing.

The answer is that you would need a total PRR of 200.

To double your current combat time you have needed to get 4 times your current PRR.

Let's bring a comparison between two builds.

Now suppose you are comparing two builds. Build A has 100 PRR and build B has 200PRR.

That is, build B has double the amount of PRR than build B. However, build B only has 1.5 more combat time than build A.

Now as it has been pointed out the crucial issue here is that it is easy for certain builds to reach high values of PRR but heavy armor wearers do not have much more PRR.

[the_one_dwarfforged]

Sounds to me a cheap solution like it was for SWF double strength bonus that got just removed.

i dont think giving those destinies a small amount more melee power is going to make a single bit of difference in how people view those destinies.

[BigErkyKid]

i dont think giving those destinies a small amount more melee power is going to make a single bit of difference in how people view those destinies.

Right on. They are too far behind with out without the buff. It will help those who insist in playing them in terms of their power vs content, though.

[-Avalon-]

PRR:
The issues I have noticed, as have others, is that it gives very strong power to low PRR builds while not favoring higher PRR-Based builds enough... After thinking about this, it came to me that the devs continue to attempt to push a somewhat bad formula that heavily benefits those with lower PRR's, instead of trying methods that give proper amounts to people.

The qualities PRR Formula needs to have:

1- Small % for lower PRR builds
2- Extreme Diminishing Returns at extremely high amounts (I believe around 350-400 is extreme, but given that I have not seen the current PRR handouts on Lamannia, that could be smaller, or larger... what is considered extremely high but reachable?)
3- Last, enough gain for those in the 'bubble' region between low PRR and extreme PRR, to entice people to get more of it.

Those qualities are almost the exact definition of an "S" curve, or Sigmoidal Function. I played around with some different equations, and finally made one that if you plug it into any graphing calculator will give you a decent setup where you get the following approximations:

PRR : % Protection
0 : 0
100 : 10
200 : 30
300 : 70
400 : 90
500 : 90
and so on...

Now, I am not saying use this particular formula, but if tweaked, could give exactly what people are looking for, which is a low % for those who are not interested in building PRR (Pajama Builds, Casters, etc), giving an upper boundary that makes the Dev's happy, and a fast climb rate that makes every PRR-point between 150 and 350 worth getting.

By the way, the formula I used (it may look tricky, but it works imo, cannot type it perfectly in basic text editors either, so will use ^ to denote an exponent)

90/(1+80e^(-1/50x))^(9/10)

The e is euler's number and there really is not a good way to substitute it from what I have found/seen in my experience, despite that it is considered rough equivalent to 2.7 (etc)... the 90 at the beginning gives the upper boundary of 90% also... now, I have never taken calculus or any higher math courses, so there may be ways to simplify that equation as well... but, to give an S shaped curve that fits what is needed for PRR/MRR to work properly, you need something like this I believe.

[Miightyy]

The PPR logic(formula) for the update 23 is good, thank you for looking into it and improving it. Just wish composite plating would get some more attention, maybe some PPR as well. Even if PPR can't be added, maybe something else can.

***A little off topic but still, maybe you guys can come up with a way for us melee to protect ourselves from those TR'd Magic missles and Force Missles that casters like to use in pvp...jeweled cloak with 25 charges goes down to 8 after being hit with "1" spell... how is that fair? A sorcerer per say clicks the spell once and it takes a chunk of my jeweled cloak charges ... I thought it was suppose to absorb spells (1 to 1)... Please guys, make it fair for pvp, do something about it or fix/change the wording on Jeweled Cloak, or come up with some way for us to be able to obtain protection from those TR'd missles.

[Pescha]

I don't care about prr BUFF Masters blitz.

[Chi_Ryu]

Now you ask yourself the following question, by how much would I have to increase my PRR to double my combat time?

That is, I want to be able to fight 30 seconds before healing.

The answer is that you would need a total PRR of 200.

I am not sure why you keep flogging this argument. In absolute terms of raw damage prevented, PRR is diminishing - no-one has disagreed so far. However, Sev's point is that in terms of extension to combat time it is linear - and this is indeed the case.

Your base starting point is 10 seconds combat time.
PRR 100 makes this 20 seconds.
PRR 200 makes this 30 seconds.

How much theoretical PRR would you need for 1 minute? I can tell you at a glance it is 500.
45 seconds? PRR350.
16.4 seconds? PRR 64.
60 minutes? PRR 35,900.

The relationship - in that respect - is entirely linear:
PRR needed to fight until time t = (t - 10) x 10.

That you cannot directly compare PRR raw values for different characters (as in your example #2) to make easy comparisons does not make the relationship non-linear.

[Blackheartox]

Sidetracking a bit from whole thread.

Sev, could you add to barb cores, like the lv 18 one or similiar a ability to negate silver flame potion debuffs and to be able to stack more then only 10 at once.

I think this would solve many issues regarding melle incoming damage since the ones left most behind are barbs since they really are the least survivable melle atm.

Also anoither question, why were silver flame pots made in the first place with the debuff option? Could you not like remake them since we are far beyond the point where they could be considerd overpowerd when taking into account ammounts of selfhealing most builds get from spells/innate abilities/slas

(since talk was about prr/ time needed to heal, why dont you simply add something like a cast limiter to silver pots, example after drinking silver flame potion you become silenced for 30 seconds unable to cast spells, while removing the debuff part --- just a simple suggestion that would be simpler to implement then whole tree remake or some new numbers to confuse players even more, also why no prr/mrr/mell power/ranged power potions in store or in game , there are many ideas you could use to earn money/help community)

[lyrecono]

i took some time to build a pure fighter tank
prr:
with a tower shield it has a 251 prr (71,5% damage reduced)
same set up but without shield: 200 prr (66,7% damage reduced)
mrr:
with a tower shield it has a 120 mrr (54,5% damage reduced)
same set up but without shield: 105 mrr (51,2% damage reduced)
a quick calculation tells me that it should reduce damage by 71% (100/100+240) mrrX2 due to towershield. the character sheet doesn't display it or the mechanic doesn't work yet?

both instances i spend 2 feats on shield mastery and i had twisted in leg. shield mastery.
only gear from the new update, the pdk shield and a Thunderholme weapon was used.
using a citw weaponset and having epic/iconic pastives would make this case even worse.
reaching these numbers was easy and i bet others can easely top that

if these prr changes were meant to help "sword and board tanks", why do i only reduce melee damage by 5%?
doesn't that defeat the purpose?
why wouldn't i opt for single weapon fighting feats and equip a shield only when getting into trouble during a raid?

[Douglas21]

if these prr changes were meant to help "sword and board tanks", why do i only reduce melee damage by 5%?
doesn't that defeat the purpose?

Let's try showing why that logic is wrong by taking it to the obvious extreme. By your logic, if the shield increased your damage mitigation from 66.7% to 100%, it would "only reduce damage by 33%". This hardly seems an appropriate description for a difference that makes you absolutely invincible.

The damage you take with the shield is about 85% of the damage you take without the shield. The shield thus offers a 15% relative improvement.

mrr:
with a tower shield it has a 120 mrr (54,5% damage reduced)
same set up but without shield: 105 mrr (51,2% damage reduced)

This, on the other hand is a rather tiny difference. There are a huge number of things that add only to PRR - various feats, destiny abilities, maybe some non-defender enhancements. It was my impression that different values for PRR and MRR were being reserved for future development, with everything supposed to count to both on initial release. Was that incorrect?

a quick calculation tells me that it should reduce damage by 71% (100/100+240) mrrX2 due to towershield. the character sheet doesn't display it or the mechanic doesn't work yet?

The shield doubles MRR only against evasionable attacks, not for everything. The character sheet shows general purpose works-against-everything MRR. To see the shield doubling effect, you'll need to go in a quest or brawling pit and get hit.

[Nascoe]

Right on. They are too far behind with out without the buff. It will help those who insist in playing them in terms of their power vs content, though.

Yeah, it helps them not feel as a horrible choice for melee. Also should make it a bit less painfull to get karma in off destinies to get from one to another sphere.

[Nascoe]

i took some time to build a pure fighter tank
prr:
with a tower shield it has a 251 prr (71,5% damage reduced)
same set up but without shield: 200 prr (66,7% damage reduced)
mrr:
with a tower shield it has a 120 mrr (54,5% damage reduced)
same set up but without shield: 105 mrr (51,2% damage reduced)
a quick calculation tells me that it should reduce damage by 71% (100/100+240) mrrX2 due to towershield. the character sheet doesn't display it or the mechanic doesn't work yet?

both instances i spend 2 feats on shield mastery and i had twisted in leg. shield mastery.
only gear from the new update, the pdk shield and a Thunderholme weapon was used.
using a citw weaponset and having epic/iconic pastives would make this case even worse.
reaching these numbers was easy and i bet others can easely top that

if these prr changes were meant to help "sword and board tanks", why do i only reduce melee damage by 5%?
doesn't that defeat the purpose?
why wouldn't i opt for single weapon fighting feats and equip a shield only when getting into trouble during a raid?

I think its the character sheet not correctly displaying (several other people remarked it was off), but I hope the devs look at that to make sure its working as intended and fix what is shown

[BigErkyKid]

I am not sure why you keep flogging this argument. In absolute terms of raw damage prevented, PRR is diminishing - no-one has disagreed so far. However, Sev's point is that in terms of extension to combat time it is linear - and this is indeed the case.

Your base starting point is 10 seconds combat time.
PRR 100 makes this 20 seconds.
PRR 200 makes this 30 seconds.

How much theoretical PRR would you need for 1 minute? I can tell you at a glance it is 500.
45 seconds? PRR350.
16.4 seconds? PRR 64.
60 minutes? PRR 35,900.

The relationship - in that respect - is entirely linear:
PRR needed to fight until time t = (t - 10) x 10.

That you cannot directly compare PRR raw values for different characters (as in your example #2) to make easy comparisons does not make the relationship non-linear.

It is not linear, it has never been and it will never be, regardless of whether I can represent it in a different way so that it looks like it. I would not be too quick at calling my numbers invalid or at doubting my ability to "grasp simple concepts". I might be rather absent minded and clumpsy, but I do work with mathematical models professionally.

This is why it looks like it is linear.

The basic unflawed reasoning is the following:

Fact:

If I decrease incoming damage by 10%, my combat time increases by 10%. Hence, since the damage reductions of PRR are decreasing, the increases in combat time must also be decreasing.

And exactly this is what is happening. The percentual increase of combat time is diminishing as I increase PRR, but in absolute terms it is not. This is why you keep seeing +10 seconds, but 10 extra seconds are not the same when I have 10 seconds of combat time or when I already have 30 seconds.That is why when I talk about "doubling" PRR it shows diminishing returns, but when I talk about absolute time in increase on its own it does not.

Example:

Take the example with PRR 100. This has a damage taken of 50%, meaning that my combat time doubles from no PRR. From 10 to 20 seconds.

Now increasing PRR to 200 places you at 33% damage taken, which is 50/33= 1.5 times better. This means that my combat time will be 1.5 the one for PRR 100. And it is: 30/20=1.5

Now going to PRR 300 places you at 25% damage taken. Compared to PRR 200, this is 33/25=1.32 times better. And combat time increases by 1.32 times the one of PRR 200. That is, 40/30.

So to sum it up, the first 100 PRR multiplied my combat time by 2, the next 100 PRR by 1.5 and the third 100 PRR by 1.32. Each additional 100 PRR has brought diminishing returns in combat time.

Now one could wonder, why does it matter? If every 100 PRR brings 10 seconds, then it means that the first 100PRR as as valuable as any extra 100 PRR. Is this true? If I have 10 seconds of combat time, an additional 10 seconds seem pretty handy. But when I already have say 50 seconds, do 10 seconds more look as appealing?

The issue with the current proposal

For the current proposal, because they bumped the value in damage reduction per PRR point, together with increasing the amount each armor gives, it means that light armor / pajamas are reaching high values of PRR. The difference between my DPS toons bard and paladin in lama is around 30-40 PRR. Hence, as my bard already reaches around 140 PRR, in terms of actual combat time benefits there isn't much of a difference (remember that PRR has diminishing returns).

Once we factor in dodge, my bard is actually better at damage reduction as I have demonstrated in this post:
https://www.ddo.com/forums/showthread.php/447627-Update-23-Second-Look?p=5418759&viewfull=1#post5418759


Not to mention that evasion, when combined with high reflexes, is always better than MRR.

In another post I showed how actually PJ are even better nowadays with respect to heavy armor even in terms of just pRR.

https://www.ddo.com/forums/showthread.php/447627-Update-23-Second-Look?p=5417341&viewfull=1#post5417341


The reason is exposed in the post above.

A simple way to change this would be to make damage taken non monotonic in PRR. At first, PRR could provide increasing returns. Then, to avoid it getting out of hand, it could provide decreasing returns after a certain treshold. This would give an edge to heavy armor.

[Chi_Ryu]

However, Sev's point is that in terms of extension to combat time it is linear - and this is indeed the case.

It is not linear, it has never been and it will never be

I don't think there is anywhere further we can go with this debate then.

I do work with mathematical models professionally.

That's nice for you, but irrelevant.

Not to mention that evasion, when combined with high reflexes, is always better than MRR.

In another post I showed how actually PJ are even better nowadays

This wasn't what I was arguing. You have good points here, but need to stop telling people that things they are saying are wrong when they are not, and when you are repeatedly proving something entirely different .

(Also, and this is rather off topic but just to be pedantic, a decrease in damage received by 10% leads to an increase in time-to-death of 10/9ths - i.e. 11.1%)

[BigErkyKid]

The extension in combat time is linear

So? Why do I care about the extension and not the actual gains in relation to previous thresholds?

If I offer a rich man 10 bucks for a job he will laugh at me. If I offer him 10% of his wealth, he might do it.

The same applies here. 10 seconds of extra combat time are not the same when I had originally 10 seconds or when I have 40 seconds.

The reason why we engaged in this debate to start with is because Sev claimed that the diminishing returns of PRR did not matter because it did not offer diminishing returns in combat time.Then he offered his example. However, as logic dictates, if PRR has dimishing returns in damage mitigation, it must have dimishing returns in combat time. And it does, as I have shown you in my previous example.

As other people have already said before me, trying to present PRR in a way that hides this decreasing returns is misleading.

Why is it misleading? Because one of the big issues is that some non heavy armor builds are reaching high levels of PRR. Between high levels of PRR the difference in damage mitigation is less important. And in order to create big differences between heavy armor and light armor in terms of actual mitigation you need to pump up your PRR a lot. Because the opportunity cost of PRR is increasing while the help it provides has diminishing returns this means that it is "unfair" for heavy armor.

So this is why I insisted so much on this issue. I honestly don't care much otherwise regarding the way you want to present PRR. It is what it is.

[Zerkul]

I couldn't agree more with this.

Tho, I do have to give them credit for the work they are doing to improve this system...

I think devs are doing great job so far, though tests are not finished yet and this new update is not ready yet. Though the way seems to be right.

[Ausdoerrt]

This is actually the same logic. If you are chugging potions, and you have a finite number of potions, then time until you must heal is extremely important and probably the most useful way to looking at PRR. The longer you can go without a heal, and further those potions will take you and the less DPS you will lose by chugging.

The actual numbers can vary, but the math is the same. You could change the hit points to 1600 and the healing threshold to 600, or 1800 and 1000, or any numbers really and the technique of measuring time until a heal still works.

Sev~

So how about we get some cure crit/heal potions?

[lyrecono]

Let's try showing why that logic is wrong by taking it to the obvious extreme. By your logic, if the shield increased your damage mitigation from 66.7% to 100%, it would "only reduce damage by 33%". This hardly seems an appropriate description for a difference that makes you absolutely invincible.

The damage you take with the shield is about 85% of the damage you take without the shield. The shield thus offers a 15% relative improvement.


This, on the other hand is a rather tiny difference. There are a huge number of things that add only to PRR - various feats, destiny abilities, maybe some non-defender enhancements. It was my impression that different values for PRR and MRR were being reserved for future development, with everything supposed to count to both on initial release. Was that incorrect?


The shield doubles MRR only against evasionable attacks, not for everything. The character sheet shows general purpose works-against-everything MRR. To see the shield doubling effect, you'll need to go in a quest or brawling pit and get hit.

what i was trying to say that the added bonus of having a shield is too little, and this is with 2 feats and a twist slot to boost it, for an attempt to boost plate and shield wearing toons (usually but not exlusively fighters, paladins and clerics) it seems more beneficial to say, a barbarian who sprang for a few levels of fighter.
15 % is weak and a fighter is still better of with swf.
or tr-ing into a functional class.

either the prr curve doesn't work for all the sources they put in or they don't want us to wear shields