Update 23 Second Look

[BigErkyKid]

Now, I have 200 PRR

200 * (100/100 + 200) = 200*1/3 = 200/3 = 66.66 damage taken, the game will round this to 67.

Now, I have 300 PRR

200 * (100/100+300) = 200/4 = 75% mitigation, 25% points of damage taken. I take 50 damage

.

If we really must discuss this, which is self explanatory and incredibly obvious, the best way to look at it is to ask how many DAMAGE points I mitigate with an extra PRR.

Or if we can just compare these two options, 200 mitigation give me (200-67)/200 = 0.65 damage mitigation per 1 PRR.

And the second option, 300 PRR gives me (200-50)/300 = 0.5 damage mitigation per 1 PRR.

It is obvious there are decreasing returns to PRR. It follows from the mathematical definition of "decreasing returns" which just means that the second derivative of the PRR function is negative. Damage mitigation is a concave function in PRR.

There is no need to discuss over these details. Any player can look at a table and see:

Oh look, going from 200 to 300 PRR makes me take this % of extra mitigation. Is it worth it?

The discussion is whether the proposed change is working as intended and to that I have to give a very firm NO.

[poltt48]

Just wondering how hard was it to get 250 PRR? I mean, would this be considered the norm at high levels or only if you built specifically for max PRR? I'm just kind of wondering what the percentage of people would actually have that high of a PRR. If you really have to build and invest to reach that number then I would think there would be a low amount of people actually obtaining 250 PRR so it really wouldn't be much of an issue. But if getting that 250 PRR was an easy thing to do then I could see a real problem.

With tank getting 250 prr I not the hard on all you really need is the bracers and the 5 insightful prr ring and rest is feats, enhancements, and ED.

[BigErkyKid]

Did some more testing. On a bard splash (12 bard, 6 fighter, 2 rogue) I can get to 140 PRR (possibly more) while blitzing.

At the same time, I would have 40 MRR (reduction of around 30%?) and over 20% dodge.

This together with evasion and quick high saves (without optimizing, around 60).

This is on whether now heavy armor will catch up with other styles.

Right now I'd say no.

[The_Human_Cypher]

Three attempts on Epic Normal for the new Epic Abbot raid ('Mark of Death') with Kookie and three wipes with a full raid party. The raid may be overpowered at this point but a challenge for the veteran players would be a good thing.

[Throwdown]

Devs you're doing PRR backwards.

More PRR should have increasing returns, while lower amounts should be LESS effective.

This incentivizes people to really invest it in to be invincible. It also means you can't rely on it in pajamas, which is part of the point of this exercise.

[Zasral]

Ok so I will be lr'ing my paladin into a Harper intelligence based paladin, and taking none of kotc. Drop extend cause I get it for ap now and take insightful reflexes instead. The Harper tree should have been the new kotc tree except exchange the int hit/damage to charisma lol.

[MangLord]

I really like the Harper tree, but I'm generally stretched pretty thin in enhancement trees as it is. Not that I don't like it, but is it really necessary? More often than not, I have no problem capping out my favorite tree and dipping into the racial and low hanging fruit in a second class tree. I tend to pick races that synergize well with the class (dwarf fighter, elf ranger, drow pale master), so the racial tree usually has a lot to offer, and I'll usually put about 20pts into it, mostly racial weapon bonuses.

Right now I'm playing an Elf 16cleric/4fighter, and Warpriest is a pretty heavy investment. Looking at bard, Swashbuckler also seems to be a heavy investment. That's good, because it means the tree has a lot of great stuff, but I already feel like I need more AP without the addition of Harper. I understand that I don't need to invest in the Harper tree at all, but I hate being torn between 2-3 great choices and agonizing over it until I weigh all the options and then second and third guess myself. It's nice to have more options, but I feel like I need another 10AP as it is.

My plan for Bard life is a drow rogue2/bard18 going with Swashbuckler and racial rapier/short sword bonuses. I'd thought to focus on Dex for damage, reflex saves, etc, with a lighter investment in INT and CHA, but if i can do INT for damage much cheaper and earlier, then it seems like a no brainer to dump DEX and use INT. Unfortunately, now i have to think about another feat in a feat starved build, assuming i want empower heal, quicken, insightful reflexes and all three SWF enhancements. Arrrgh!

I'd like to see att/dam options other than INT. Given the nature of the Harpers organization, Dex and Cha options don't seem out of line. The classes that will benefit most from Harpers enhancements don't have a lot of breathing room as far as feats go, (bards especially) so adding Insightful Reflexes to a rogue's must have feat list is going to make me lose out on something I already need for a tricky class to build correctly, IMO. Cha for attack/damage would greatly help Paladin, as well.

[???]

I know that it is somewhat useless, but how about fixing DDO User Manual in the test folder? It is still in the 10-level era.

[BigErkyKid]

OK, took the time to crunch some DPS numbers to go beyond impressions.

First let me briefly clarify what I am doing. This is simply a pure raw DPS test, no other considerations are taken. I do not want to mix equipment much into the numbers too much, as it can greatly vary. What I will do is to get a multiplier to your base damage (that includes equipment, enhancements and what not) and compare those multipliers across destinies.

To see how this works. Suppose I am peasant with a weapon that has a crit profile x2. This is just to see how I do it, so forget about destinies real classes and real weapons. My damage will be:

BASE*1 will happen 19/20 times and BASE*2 for 1/20 times. This can be rewriten as BASE* (1*(19/20)+2*(1/20)). Or in another words, my damage will be whatever I can stack in my BASE*1.05.

This exercise will assume a paladin with holy sword, full line of SWF, overwhelming critical, IC:slash, critical multiplier from LD and doublestrike from DC.

The old system

DC (including fully stacked doublestrike): BASE*2.53
LD (including fully stacked blitz): BASE*5.25

In other words, under the other system, blitz was roughly double the damage of DC.

The new system

DC (including fully stacked doublestrike): BASE*4.3516
LD (including fully stacked blitz): BASE*4.662

So LD is only slightly ahead in terms of damage multiplier.

However, even the best raw power destiny is .88 of the old blitz

There are two CRUCIAL issues: 1) since you can take more hits because of the extreme change in PRR, you don't need to back out to heal as frequently and 2) the new paladin holy sword is UBER

Let's examine the PRR side a bit more. Take for instance an old build, say the dwarven parody that had ~100 PRR. Before it mitigated around 40%, now out of the box it mitigates 50% and in blitz around 60%. This means roughly speaking means it can stay 20% more in combat.

Just for the sake of the example, look at the dwarven parody (without double strikes, don't use this numbers to compare to the paladin):

Old system

Damage multiplier in fully stacked blitz: 4.275

New system

Damage multiplier in fully stacked blitz: 4.028
Combat time adjusted damage multiplier: 4.028*1.2= 4.83

So the the dwarven parody sees a 13% increase in melee power.

The final issue is power creep. Some people are saying that overall melee is a lot easier. They are right! Let's take the top DPS option that the game now provides. This is probably a paladin with a khopesh. This build is a significant power increase over previous possibilities. I will compare a dwarven parody with double strikes included in the old system to a paladin in the new system. Perhaps that dwarf was not the peak in power, but surely was close to it. Feel free to provide the numbers for another build. I am aware this is not exactly the same exercise I did before with the paladin since that was not available pre changes.

Old system top: Dwarven parody (35% double strike)

Damage multiplier in fully stacked blitz: 5.77
Combat time adjustment: assume 100PRR as a base, which provided 40% mitigation.

New system top: paladin with khopesh (30% double strike)

Damage multiplier in fully stacked blitz: 5.78

Blitzed PRR 170: provides 63% mitigation. This is 23% more than the parody before, meaning 23% more time of combat time.

Combat time adjusted damage multiplier: 5.78*1.2= 6.936

So 6.936/5.78=1.2. So top melees builds are now 20% more powerful than before in terms of adjusted raw power.

Conclusions

1. DC is now very close in raw DPS to LD.

2. There is a massive buff to damage mitigation.

3. Paladins are incredibly powerful this update. This is because previous competence bonus were limited to weak weapons, but holy sword can take the best one.

4. Damage mitigation adjusted numbers are what provide a feeling of generalized increased power.

5. There is significant power creep. The top melee build is now 20% more powerful than the previous one. In addition, there is a feeling of more power because melee combat is much more forgiving.

Remarks

Careful bringing down melee power even more. Melees are fighting alonside with very high burst damage builds, even if they are more survivable than others (PRR changes) if melee power keeps diminishing they won't be able to contribute because their relative DPS will be too low. If mobs are dead, it does not matter if you are very survivable.

[Portalcat]

Devs you're doing PRR backwards.

More PRR should have increasing returns, while lower amounts should be LESS effective.

This incentivizes people to really invest it in to be invincible. It also means you can't rely on it in pajamas, which is part of the point of this exercise.

PRR has always had diminishing returns.

[Nagantor]

Diminishing returns are fine for PRR, yes. Current system has drawbacks for sure.

Problem 1: You have a fast decrease in the effective use. Main goal is already achieved when damage is reduced enough to prevent instant death from heavy hitters and you get enough time to heal when fighting groups of trash (without doing nothing but healing). Everything beyond is far less important.
Optimally we'd start with low gain, reach the important treshhold with moderate effort and then it starts to diminish like now.

Problem 2: It's far too easy to reach the sweet spot with Robe/Light Armor. Did you consider adding another scaling factor depending on armor type? Applied to the current formula we'd end with "damage * 100 / (100 + PRR) * modifier". Modifier could be 0.50/0.75/1.0/1.25 for cloth/light/medium/heavy and the cap values for PRR/MRR adjusted. (Values are for the principle, actual implementation would require some testing for good values.)
I'm aware that sheltering items for example will then give less if they don't have a good armor to work upon but that's fine with me. Without (strong) armor it's very hard to get into highly effective amounts. Still possible, but you need a definite effort while heavy armor starts at a noticeable effect by itself. If you want to stay mobile, make use of it and evade by feat or plain movement, stay out of the thick combat... and don't consider yourself a tank.

Death of cloth/caster tanks is nothing I'd mind to see. Those character types have different roles in my eyes. Monks might stay the exception having some tricks allowing them to tank without armor - requiring some feat or ability which should require Monk 12 (or higher) in my opinion. A real Monk special, not a splash feat.

[zerit2002]

Can we please stop arguing about the semantics of how the PRR was stated and focus on how TERRIBLE this new idea for blitz is?

I just popped on Lamma, stood in a room full of mobs, 10 or so, popped a blitz, killed everything...sounds good, right? 0 CHARGES. 0. 0? What's with this? Why can't I get a charge from killing TEN enemies?

And since i haven't effectively gained a blitz, yet, I don't know how terrible the reduction in damage will be when its fully stacked. 50...doesn't seem like anywhere close to the appropriate number for a 10 stack of a top tier Epic Moment.

But, seriously, someone needs to go through this with an extremely fine toothed comb, preferably someone that gets on and tries to maintain a blitz, and fix why the chance to gain a charge is so ridiculously minuscule.

More. Epic. Sadfaces.

No, that's not what I'm saying at all. I wasn't paying attention to how much damage I was doing per swing, but, yes, everything was dead before I gained one charge.

When I went for my second attempt, 5 minutes later, I zerged thru to get around 25 mobs to surround me. Threw a bunch of cleaves, lay wastes, momentum swings, stunning blows, etc. I gained four charges. By the time I ran to the next room full I was at two, threw some attacks in the group, made it to six. Kept zerging as hard as I could to try to reach the mini-boss at the middle of the quest. only there was I able to get my stack to 8... Kept zerging to the end fight. Kill all the trash and the semi-boss, wind up with 7 stacks for the main boss. Start on him, work it up to a ten stack. At this point, im doing ~1200-1500 on big crits, 200-300 on hits. WITH A TEN STACK. My build is not tweaked out with all the best of the best, but, with an 80str and power of the forge going, with a ten stack of blitz, these numbers are absolutely abominable.

If the goal of changing blitz was to keep people in a party competitive but still be able to hold their blitz, e.g. not stacking on kills, this was a complete and utter fail.

Must zerg to keep it alive, things dont live long enough to stack.

Hey devs! Give this man some attention!
He is absolutely right!

In the present implementation of blitz charges are too hard to get and too easy to lose, even SOLO, with full control over the quest.
Trying to keep blitz up in a group like this would be a nightmare. Things are gona die long b4 we get charges.

And yeah, the dmg increase for a 10 stack is too low.

[XodousRoC]

Can we please stop arguing about the semantics of how the PRR was stated and focus on how TERRIBLE this new idea for blitz is?

I just popped on Lamma, stood in a room full of mobs, 10 or so, popped a blitz, killed everything...sounds good, right? 0 CHARGES. 0. 0? What's with this? Why can't I get a charge from killing TEN enemies?

And since i haven't effectively gained a blitz, yet, I don't know how terrible the reduction in damage will be when its fully stacked. 50...doesn't seem like anywhere close to the appropriate number for a 10 stack of a top tier Epic Moment.

But, seriously, someone needs to go through this with an extremely fine toothed comb, preferably someone that gets on and tries to maintain a blitz, and fix why the chance to gain a charge is so ridiculously minuscule.

More. Epic. Sadfaces.

LD also does NOT (for those that weren't aware) play nicely with the new KoTC cleaves either. They don't qualify u for Momentum swing, which is a non-starter. The stated goal was to help a feat starved class by implementing enhancements to replace the cleave/gr cleave feats (with whom the enhancements share cooldowns). Testing on Lammy shows the enhancements do NOT qualify as prereqs for momentum swing. For all those (including myself) who've been asking, "Yes, but do the enhancements play nice with LD?", the answer was silence. Now I answer it, the answer means they've not saved me any feats at all.

Even. More. Epic. Sadfaces.

[Oliphant]

Will the PRR and MRR changes affect enemies' abilities?

[Altrelo]

I LR'ed my Halfling Assassin rogue in order to better make use of Harper Agent PrE. I found that my base damage in Legendary with 10 stacks of Blitz was about ~30 while my SA damage didn't that much of an increase, ~2-6. While I did more damage on rolls of 19-20 in Legendary, the utility that Shadowdancer has makes it so that I won't switch my Assassin to Legendary, even though what stopped me was the tactical feat charge up. Turbine has at least succeed in keeping the other destinies competitive, for these builds at least.

Though Dex will only be raised to 17, it means that Assassins can no longer take Improved SA. Shadowdancer could also boost rogue levels for our abilities, spending half of my points in Intelligence upgrades to even have a viable Assassinate DC preposterous. I can't even make use of all the utility abilities that Shadowdancer has, because I don't have the points for it. Getting Steakthy for the +6 DC to Assassinate, getting Lithe to tier and getting Shadow Form takes up all of 12 extra points. Without Lithe, I can't make use of the extra dodge I'll get for my Shadow Charges that won't be used due to not having an ability that can use them after I use Shadow Form.

[Silverleafeon]

Will the PRR and MRR changes affect enemies' abilities?

No, monsters are custom made.

[Severlin]

If we are looking at PRR, dissecting the raw numbers and calling them "diminishing returns" is misleading.

In terms of functionality, we look at survivability as "time until dead" or, more accurately, time until I have to interrupt my DPS and use resources to heal. This is important because, barring random factors, if your survival time > your kill time then you win. More accurately for DDO, the less resources you have to use per unit of time, the more likely you will reach the next rest shrine alive. If you look at its effect on time until heal, PRR gives linear returns.

As an example, lets say I have 1200 hit points and decide that I always have to heal when I hit 200. That means I have 1000 hit points to chew through before I have to both interrupt my DPS and use resources to recover health.

If incoming damage is 100 DPS, I think it's easy to see that with no mitigation I will have to heal in 10 seconds.

Now let's look at how PRR affects that time until heal:

At 0 PRR I can fight for 10 seconds before I need a heal.
At 100 PRR I can fight for 20 seconds before I need a heal.
At 200 PRR I can fight for 30 seconds before I need a heal.
At 300 PRR I can fight for 40 seconds before I need a heal.

The pattern follows a linear progression; assuming constant damage each 100 PRR increases my time until heal by 10 seconds.


Now, the issue I have with the language being thrown around in this thread that describes PRR as diminishing returns is that it examines the raw mitigation numbers, but it doesn't take into account that fact that as PRR increases there is less incoming damage to mitigate. This exagerrates the effect of each PRR point, which leads to a linear progression of survival time.

It's survival time we examine when looking at this mitigation.

As for whether "PJ" builds can leverage too much PRR, it was a concern when we increased the PRR formula and also removed the PRR limit on no armor and light armor builds. I also wonder if that's more of an issue with Earth stance than the current system on Lamannia. When I get in next week we will look at the numbers.

Sev~

[Ganak]

As an example, lets say I have 1200 hit points and decide that I always have to heal when I hit 200. That means I have 1000 hit points to chew through before I have to both interrupt my DPS and use resources to recover health.

I suggest that a different logic be considered.

1.) If you are at 200 HP and have agro in EE, you are dead.

2.) More importantly, how do you recover the HP, and how fast?

Sure I can try a pure 20 fighter or barb and start chugging pots for a few minutes to get back up, but be left behind.

[Severlin]

I suggest that a different logic be considered.

1.) If you are at 200 HP and have agro in EE, you are dead.

2.) More importantly, how do you recover the HP, and how fast?

Sure I can try a pure 20 fighter or barb and start chugging pots for a few minutes to get back up, but be left behind.

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~

[Qhualor]

If we are looking at PRR, dissecting the raw numbers and calling them "diminishing returns" is misleading.

In terms of functionality, we look at survivability as "time until dead" or, more accurately, time until I have to interrupt my DPS and use resources to heal. This is important because, barring random factors, if your survival time > your kill time then you win. More accurately for DDO, the less resources you have to use per unit of time, the more likely you will reach the next rest shrine alive. If you look at its effect on time until heal, PRR gives linear returns.

As an example, lets say I have 1200 hit points and decide that I always have to heal when I hit 200. That means I have 1000 hit points to chew through before I have to both interrupt my DPS and use resources to recover health.

If incoming damage is 100 DPS, I think it's easy to see that with no mitigation I will have to heal in 10 seconds.

Now let's look at how PRR affects that time until heal:

At 0 PRR I can fight for 10 seconds before I need a heal.
At 100 PRR I can fight for 20 seconds before I need a heal.
At 200 PRR I can fight for 30 seconds before I need a heal.
At 300 PRR I can fight for 40 seconds before I need a heal.

The pattern follows a linear progression; assuming constant damage each 100 PRR increases my time until heal by 10 seconds.


Now, the issue I have with the language being thrown around in this thread that describes PRR as diminishing returns is that it examines the raw mitigation numbers, but it doesn't take into account that fact that as PRR increases there is less incoming damage to mitigate. This exagerrates the effect of each PRR point, which leads to a linear progression of survival time.

It's survival time we examine when looking at this mitigation.

As for whether "PJ" builds can leverage too much PRR, it was a concern when we increased the PRR formula and also removed the PRR limit on no armor and light armor builds. I also wonder if that's more of an issue with Earth stance than the current system on Lamannia. When I get in next week we will look at the numbers.

Sev~

thank you for explaining this so someone like me who doesn't understand DDO math very well actually can understand how this works. im also glad to see some of you guys jumping into threads to correct us when forumites explain things incorrectly.