"Max Health" debuff - Is this a valid test?

[Nevao]

<p><em>Please note that I'm posting this in Spells/Abilities since it's around the the Max Health and Stamina debuff mechanics, not just necessarily about rogue specific abilities. </em></p><p>So a guild member and I were having a discussion about if the "Change of Engagement" AA (Direct Damage and Max Health debuff) was worth taking over the "Sunder" AA (Direct Damage with a proc chance). I've always been rather leery of "Max Health" debuffs since you can not see the the actual effect. So after kicking it around a bit I decided to try to see if there was a way to parse the Max Health.</p><p>Since it's never been fully explained I am assuming the following about "Max Health" debuffs:</p><ul><li>If your max health is decreased while at 100% health  it will also decrease the "actual" health to the new max by the corresponding amount. You therefore would still have 100% health, just X% less. </li></ul><p>To test how well this actually works I copied over to Test Copy and grabbed a large number of heroric test dummies and just started wailing away, only using auto attack and procs, to try to establish a baseline health for the mob (since we can't see number only the %). I used a Heroric instead of Epic since I want to leave the house sometime today. The range of damage taken was between 903K and 907K with an average of 905K. The variable amount of damage taken makes me think one of two things is happening:</p><ul><li>Training dummies have a random amount of health, or</li><li>You can do more damage to a mob than it has health and the parser does not pick up.</li></ul><p>I'm guessing it's the latter, but regardless it means each scenario has to be run multiple times.</p><p>After I established my baseline I parsed out the following scenarios against a heroic dummy:</p><ul><li>Start with "Change of Engagement" and continue with Auto Attack, recasting CoE as it came up</li><li>Start with "Will to Survive" (a similar ability) and continue with Auto Attack, recasting WtS as it came up</li><li>Start with Debilitate (with the Stamina debuff), and continue with Auto Attack, recasting Deb as it came up. (I ran this assuming that Stamina debuffs would have the same net effect just on a different scale)</li></ul><p>I ran each scenario 5 times just to get a feel for effectiveness and decided to post those results to validate before going further. Here's the average damage taken for each test:</p><ul><li>Auto Attack Only (baseline): 905,096</li><li>Change of Engagement: 900,579 - <span style="color: #ff9900;">0.5% difference</span></li><li>Will to Survive: 903,923 - <span style="color: #ff9900;">0.1% difference</span></li><li>Debilitate: 798,925 - <span style="color: #ff9900;">12% difference</span></li></ul><p>Initially it looks like CoE did have a very slight effect, WtS was too close to be called different and Deb had the greatest effect of all. The concerning part about this though is that CoE should have had an 8% decrease in Max Health and WtS should have had a 5% decrease. Unless of course these dummies are internally tagged as "Epic" but that would mean 3% and 2% decreases. Obviously neither is happening. The only explanations I can think of is that Max Health debuffing is either not working or it's only debuffing the mob's "base health" and other factors are being applied after the fact. I'm assuming Debilitate, since not a % debuff, had an easier time in this test but there's a very good chance it won't scale up depending on mob design (is health controlled through base number, stamina, points of health per stamina, % health increase, + health amounts or some other mechanism).</p><p>Regardless something seemed very off so I pulled out a Test Copy buffed char on my wife's account, mentored my char down to 80 and then tried casting CoE and WtS on the other char in a duel:</p><ul><li>Max Health pre Duel: 14,175</li><li>Max Health after casting CoE: 13,206<span style="color: #ff9900;"> - 6.8% difference</span></li><li>Max Health after casting WtS: 13,569 <span style="color: #ff9900;">- 4.2% difference</span></li></ul><p>I did not run a Debilitate test.</p><p>It looks like the abilities are behaving if not quite as advertised in design with how they are described. So this leads me to the following questions:</p><ul><li>Is my testing strategy flawed?</li><li>If not are Heoroic dummies just not a good test subject?</li><li>If they are then does this just confirm that mob design renders an interesting ability near useless?</li></ul><p>Any thoughts would be appreciated.</p>

[Shotneedle]

<p>Heroic training dummies have 900,000 hp.</p>

[Nevao]

<p>Thanks Buffrat, that's good to know. So it's like Rift, we have "Overkill" in the parser it's just not being stated in any of the text.</p><p>That means that if the math were easy with those buffs the damage needed to kill a heroic training dummy should be:</p><ul><li>CoE cast: 882,000 (900K * 0.92)</li><li>WtS cast: 885,000 (900K * 0.95)</li></ul><p>Yeah, something is not right or not yet fully understood.</p>

[Geothe]

<p>And THIS is exactly the reason why Brigands have always hated Will to Survive and were even angrier in beta when we saw we were getting Will to Survive 2.0.</p><p>And Xelgad's basic reply to the entire thing was: deal with it.</p><p>And yet, your data here just further illustrates how completely useless/broken those abilities are.  Yet Xelgad, for some innane reason, thinks they are awesome!</p>

[Glenolas]

<p><cite>Nevao wrote</cite></p><blockquote><ul><li>You can do more damage to a mob than it has health and the parser does not pick up.</li></ul></blockquote><p>You practically always do more damage than it has health,and the  parser does pick it up.</p><p>The parser totals what SOE puts in the logs.    If you have a mob down to 8HP left, and hit it with an ice comet,  SOE will calculate the damage from the IC, and report it (for instance 100K).   It will also report the wizard killed the mob.   They will NOT calculate that the mob only had 8 health left and report the IC damage as 8.</p><p>Therefore, you can only use parsed damage done as an approximate measure of the mob's health, with the error being the "overage" of the damage of the hit that killed it vs it's remaining HP.   A big kill shot produces a bigger error in health.</p><p>In the case of debuffs,  you have to consider two potential benefits.</p><p>1.   In the case of ones that debuff the mob's actual health, you'll measure a lower value for the damage it took to kill it than it's known starting health.  (i.e. 800K killed the known 900K training dummy).   Remember, you don't know the actual HP the dummy had at death, just that it was less than 800K.</p><p>2.  In the case that the debuff makes the damage more effective...i.e. debuffed mob against cold makes the IC hit harder.</p><p>In this case,  the mob will still need 900HP to be killed, but  will take less spells/CA's/autoswings to total that amount.  (which will show up as FASTER time delivery of the 900HP vs a non debuffed mob.)    You'll still parse over 900HP to kill it, but measureably faster.  </p><p>When testing damage enhancing debuffs, you can fool yourself with only a few tests, as sometimes the mob dies faster due to hitting on the high side of the rather wide spread of all damage mechanisms  than anything tied to debuffing.</p><p>3.  If you think you are debuffing health (and not damage effectiveness) and you are measuring the same HP to kill,  it's not working as intended or you don't have the right take on the debuff.</p>

[Nevao]

<p><cite>Glenolas wrote:</cite></p><blockquote><p><cite>Nevao wrote</cite></p><blockquote><ul><li>You can do more damage to a mob than it has health and the parser does not pick up.</li></ul></blockquote><p>You practically always do more damage than it has health,and the  parser does pick it up.</p><p>The parser totals what SOE puts in the logs.    If you have a mob down to 8HP left, and hit it with an ice comet,  SOE will calculate the damage from the IC, and report it (for instance 100K).   It will also report the wizard killed the mob.   They will NOT calculate that the mob only had 8 health left and report the IC damage as 8.</p><p>Therefore, you can only use parsed damage done as an approximate measure of the mob's health, with the error being the "overage" of the damage of the hit that killed it vs it's remaining HP.   A big kill shot produces a bigger error in health.</p><p><span style="color: #ff9900;">Yeah, I figured that might be the case which is why I stuck just to Auto Attack (and the resulting gear procs)</span></p><p>In the case of debuffs,  you have to consider two potential benefits.</p><p>1.   In the case of ones that debuff the mob's actual health, you'll measure a lower value for the damage it took to kill it than it's known starting health.  (i.e. 800K killed the known 900K training dummy).   Remember, you don't know the actual HP the dummy had at death, just that it was less than 800K.</p><p><span style="color: #ff9900;">This is what I'm thinking is supposed to happen, but again there's no way to tell which is always going to cause a bit of suspicion on the ability.</span></p><p>2.  In the case that the debuff makes the damage more effective...i.e. debuffed mob against cold makes the IC hit harder.</p><p>In this case,  the mob will still need 900HP to be killed, but  will take less spells/CA's/autoswings to total that amount.  (which will show up as FASTER time delivery of the 900HP vs a non debuffed mob.)    You'll still parse over 900HP to kill it, but measureably faster.  </p><p>When testing damage enhancing debuffs, you can fool yourself with only a few tests, as sometimes the mob dies faster due to hitting on the high side of the rather wide spread of all damage mechanisms  than anything tied to debuffing.</p><p><span style="color: #ff9900;">I definitely undestand this part, mit/resist debuffs have never been in question</span></p><p>3.  If you think you are debuffing health (and not damage effectiveness) and you are measuring the same HP to kill,  it's not working as intended or you don't have the right take on the debuff.</p><p><span style="color: #ff9900;">And that's the million dollar question and point of this post.</span></p></blockquote>

[Nevao]

<p>Hopefully a Dev will come in and shed some light as we still do not have a real definite answer. We were told in Beta that the Dev's could not understand why we don't like this style of debuff. They obviously think it something special. I would like to be able to prove that either it is, or it's broken/not as effective as intended. Especially now that it's been used for two expansion end lines.</p>

[LardLord]

<p><cite>Nevao wrote:</cite></p><blockquote><ul><li>Auto Attack Only (baseline): 905,096</li><li>Change of Engagement: 900,579 - <span style="color: #ff9900;">0.5% difference</span></li><li>Will to Survive: 903,923 - <span style="color: #ff9900;">0.1% difference</span></li><li>Debilitate: 798,925 - <span style="color: #ff9900;">12% difference</span></li></ul></blockquote><p>My guess is that the training dummies have a <strong>massive </strong>hidden<strong> </strong>percentage-based health buff, which causes %-max health debuffs to be less valuable than intended and STA debuffs to be extremely powerful.</p><p>I wonder if the same holds true for normal mobs?</p>

[Wilin]

<p>I was involved in the discussion about this in beta. We never did get an explanation of the mechanics except to say that it's awesome so it should be attractive to us all.</p>

[Nevao]

<p><cite>Quabi@Antonia Bayle wrote:</cite></p><blockquote><p><cite>Nevao wrote:</cite></p><blockquote><ul><li>Auto Attack Only (baseline): 905,096</li><li>Change of Engagement: 900,579 - <span style="color: #ff9900;">0.5% difference</span></li><li>Will to Survive: 903,923 - <span style="color: #ff9900;">0.1% difference</span></li><li>Debilitate: 798,925 - <span style="color: #ff9900;">12% difference</span></li></ul></blockquote><p>My guess is that the training dummies have a <strong>massive </strong>hidden<strong> </strong>percentage-based health buff, which causes %-max health debuffs to be less valuable than intended and STA debuffs to be extremely powerful.</p><p>I wonder if the same holds true for normal mobs?</p></blockquote><p>I'm afraid that this is going to turn out to be the case. We have what is a decent ability in theory rendered useless by mob/encounter design. Traumatic Swipe anyone? Seriously though, this needs to be addressed if it's the case or the abilities that use this need to be modified. But before we can if it's the case or not we need confirmation on what's really happening and why.</p>

[Nevao]

<p>Any chance we  can get a Dev to weigh in here? This type of debuff constitutes end lines in two expansions and it's fair to ask what's going on and why it's not testing as advertised.</p>

[Darkor]

<p><cite>Nevao wrote:</cite></p><blockquote><p>Any chance we  can get a Dev to weigh in here? This type of debuff constitutes end lines in two expansions and it's fair to ask what's going on and why it's not testing as advertised.</p></blockquote><p>Not any dev, i want Xelgad to co me here and explain himself to this whole situation.</p>

[Silzin]

I am not sure about where or if your method of testing works, but i have an idea for testing it. Make a brawler, and make sure it has 0 Crit chance and 0 potince. then use the ability 'Fist of Devastation' on a solo training dummy. this should give you a 110% (I think at base) of max health. I know it it more then max health, but if it is a reliable 110% of max health then that can be charted more reliably then the attack damage method. Just an idea.

[Geothe]

<p><cite>Darkor@Nagafen wrote:</cite></p><blockquote><p><cite>Nevao wrote:</cite></p><blockquote><p>Any chance we  can get a Dev to weigh in here? This type of debuff constitutes end lines in two expansions and it's fair to ask what's going on and why it's not testing as advertised.</p></blockquote><p>Not any dev, i want Xelgad to co me here and explain himself to this whole situation.</p></blockquote><p>lol that'd take a miracle.Xelgad's reasoning as to why these are good debuffs is: "They are good!" heh.When in fact, the tests done in this very thread have shown the reality on how utterly worthless they are against NPCs.</p>

[Nevao]

<p>It looks like it's broken, but it may work. My personal guess is that it does work but that the levers used to raise mob hitpoints render it useless. But regardless until a Dev comes in and explains what is really happening and how we can tell that it's working/not working, we're not going to trust it. For a lot of things that may not be a big deal, but if we're going to be given this for Endlines we shouldn't need to bust out a oujia board to verify it.</p><p>I will admit, some of this is my bad for not having/taking the time to verify this during Beta but we still need some help here.</p>

[Xelgad]

<p>You guys are correct that this is not working as intended. We're already working on a fix that will merge stamina and percent-based max health debuffs and ensure that they function as intended regardless of how mobs are designed.</p>

[Wilin]

<p><cite>Xelgad wrote:</cite></p><blockquote><p>You guys are correct that this is not working as intended. We're already working on a fix that will merge stamina and percent-based max health debuffs and ensure that they function as intended regardless of how mobs are designed.</p></blockquote><p>YAY!</p>

[Nevao]

<p><cite>Xelgad wrote:</cite></p><blockquote><p>You guys are correct that this is not working as intended. We're already working on a fix that will merge stamina and percent-based max health debuffs and ensure that they function as intended regardless of how mobs are designed.</p></blockquote><p>Gracias Xelgad</p>

[Geothe]

<p>So its been several weeks.Do you plan on actually fixing these broken abilities anytime soon at all?</p>

[Yimway]

<p>not Soon(tm)</p><p>Its not worked for years, I wouldn't expect a fix in a few weeks, maybe by GU61.</p>

[Nevao]

<p>I don't know how many abilities there are that have one of those two effects but I would expect it to be a decent number. Since it will have to be a mechanics change and abilities change I agrree with Atan, it's going to take awhile. Personally I'm hoping to see something with LU60, but I wouldn't be suprised if it was LU61.</p>

[Geothe]

<p>Personally, I'd think that a Dev that preached and preached about how unbelievably awesome percentage health debuffs were and about how he cant comprehend how the player base would have such a negative reaction towards them, would actually want to fix said debuffs ASAP so he isn't caught just spewing BS.</p><p>But then again, this is SoE.</p>

[Xelgad]

<p>These changes have been made internally and should go to Test with GU60.</p>

[Nevao]

<p>Thank you Xelgad. Are there any details you can give on how they were adjusted (since you mentioned they were going to be merged)?</p>

[Xelgad]

<div>Mobs are designed with a relatively small amount of HP from Stamina, and then they get a massive "Max Health" modifier in order to raise their HP pool to the value the designer intends. "Max Health" debuffs have been working on that modifier, but when you debuff a 1000% "Max Health" modifier to 998%, you're obviously not going to see much effect. Fortunately, our coders have created a new stat that we will be using to apply a percent-based debuff to the mob's final HP pool value, instead of just debuffing the "Max Health" modifier.</div> <div> </div><div>While that has caused "Max Health" debuffs to be much weaker than intended, it has simultaneously caused Stamina debuffs to be much more powerful than intended. As such, we're converting all stamina debuffs to the new stat as well.  For example, the Sentinel's Fate AA that currently adds a Stamina debuff of 192.2 to Debilitate will now reduce the mob's HP pool by 2.5%. Other Stamina debuffs were converted at roughly the same rate. </div> <div> </div><div>As part of this conversion, we rebalanced the values of the "Max Health" debuffs to better mesh with the newly-converted Stamina debuffs and the intent of the content designers. For example, Will to Survive will reduce the HP pool on all mobs by 3%, instead of the old 5% value for heroics and 2% value for epics. In total, with current AAs and abilities, you will be able to reduce the HP pool of mobs by around 30% if you maximize debuffs.</div> <div> </div><div>I'm hopeful you guys will run these changes through the wringer once they're on Test, and I'll gladly clarify any points if needed.</div>

[Gaige]

<p>I personally think 30% is still too much, but hopefully it'll be enough to force guilds to kill raid mobs by their intended scripts instead of ignoring them completely and just burning them before anything happens.</p>

[Nevao]

<p>I look foward to testing this. Thank you for the explanation and the help.</p>

[Crabshack]

<p>When you can flawless a mob on your 2nd pull ever you know something isn't set up right. This fight looks pretty tough. Don't get me wrong, I'll take my loot and move along. But it's a shame that there is a way to trivialize the fight so bad. I'm sure when things get fixed we'll all be wishing we'd learned how to do the mob lol.</p>

[slippery]

<p><cite>Crabshack@Nagafen wrote:</cite></p><blockquote><p>When you can flawless a mob on your 2nd pull ever you know something isn't set up right. This fight looks pretty tough. Don't get me wrong, I'll take my loot and move along. But it's a shame that there is a way to trivialize the fight so bad. I'm sure when things get fixed we'll all be wishing we'd learned how to do the mob lol.</p></blockquote><p>That isn't a mechanic's problem, that is an encounter design problem. Don't confuse the two. The mechanic has been this way forever, blaming it on something that changed with DoV is mistaken. The fact is stamina debuffs a % of health right now, not a certain amount per point (which does make sense, because then it scales appropriately). If anything, other then changing it so % based debuffs actually work, I wouldn't change this at all. I might just institute a cap.</p><p>The problem you are describing is two fold. First encounters are designed with the idea that so much time is spent on adds. If you where to kill Kolskeggr by killing all the adds what would the effective amount of hp the encounter has be? Quite substantial I'd imagine. We had pulls where we did 50 million damage to kolskeggr and 300+ million damage to adds. When you design an encounter like that the better your gear gets and the more dps you can get on the named the less time you are going to have to spend killing adds. The higher your dps the exponentially shorter the encounter gets. Combine that with the fact that the gear boost was way too much and we are effectively wearing all current hard gear plus gear that should have come from Drunder and it isn't really surprising at all. Once you are killing harder mobs in Drunder you'd expect the current stuff to be farmed by the top guilds in that fashion.</p><p>P.S. Most guilds still won't be able to just burn most mobs.</p>

[Gaige]

<p>If this game is going to turn into debuff the mob's HP and burn it before it does its script I wonder who is going to want to play it.</p>

[Shotneedle]

<p><cite>Gaige wrote:</cite></p><blockquote><p>If this game is going to turn into debuff the mob's HP and burn it before it does its script I wonder who is going to want to play it.</p></blockquote><p>Oh man. Guess we should go back to SF and change Haraakat's script to auto port every 25% and Azara/Ernax's red text aoes to trigger every 25% since after guilds farmed loot they were bypassing the scripts.</p><p>Like Arabel said, this is the exact same thing that is happening now. Except instead of farming loot it all got upgraded massively, allowing us to bypass scripts.</p>

[Laiina]

<p><cite>Xelgad wrote:</cite></p><blockquote><p>You guys are correct that this is not working as intended. We're already working on a fix that will merge stamina and percent-based max health debuffs and ensure that they function as intended regardless of how mobs are designed.</p></blockquote><p>If this is fixed, it MAY also fix some other problems, such as:</p><p>Raw number type debuffs do NOT scale well to DoV mobs. All debuffs, that can be, should be percentage based so they scale up correctly. If my troub casts a debuff that reduces STR by 43, that might be a fair amount on a non-heroic SF mob, on a heroic DoV mob it is barely worh casting.</p><p>There is apparently no way to tell how what the actual stats are on mobs, perhaps by charming one you could see some of them, but since charmed mobs are somewhat nerfed it would only be an estimate.</p>

[Gaige]

<p><cite>Buffratx@Antonia Bayle wrote:</cite></p><blockquote><p>Oh man. Guess we should go back to SF and change Haraakat's script to auto port every 25% and Azara/Ernax's red text aoes to trigger every 25% <strong>since after guilds farmed loot they were bypassing the scripts.</strong></p></blockquote><p>Difference.</p><p>We're using STA debuffs to currently kill mobs WE ARE UNABLE TO KILL BY DOING THEIR SCRIPT.</p><p>That is my issue with it.  That is why I'd like to see mobs have a high enough HP to render bypassing the script, even with gear upgrades, impossible.</p>