 |
| Welcome to Elitist Jerks |
We're testing some new features on the site regarding OpenID registration and coordination with gamerDNA. If you experience any issues with registering an account, please take the time to fill out a report and send it to this e-mail address. We would appreciate any assistance you could provide in making sure everything is functioning as intended. Thanks!
If this is your first visit, please be sure to check out the FAQ and the forum rules. Users must register to post and new registrations are subject to a one day "mute" period to get acquainted with the community.
|
06/21/09, 5:51 PM
|
#801
|
|
Glass Joe
|
Endoscient is correct, the onus is on you to show the accuracy of your model. Just because the model includes some measure of HL, FoL, and HS does not mean it is anywhere near useful. For example, it seems your model weights HS about the same as HL. However HS is used very infrequently while HL is used extremely frequently. How can such a model come close to predicting gains in a real encounter?
On another note, regarding the regen nerf: It was posted earlier that a good strategy when healing with two holy paladins is to alternate breaks of 15 seconds to DP + SoW melee. It seems like it might be better to do 25-30 second rotations instead, using 15 for DP, and 10-15 for either FoL or HL/SoW. Reason being that you spread out the lower-healing phases while making sure to always have one paladin spamming their full HL. I was also wondering, what do you guys do if you have 1 paladin plus another class healing the tank. If you need to DP, it seems a lot more dangerous since other classes don't have as much throughput as a holy paladin spamming HL. It would be nice to match up your DP with tank cooldowns, but after the illumination nerf it doesn't seem like we will have that luxury. Thoughts?
|
|
|
|
|
|
06/22/09, 2:00 AM
|
#802
|
|
Don Flamenco
Human Paladin
Shadowsong (EU)
|
Hm, to elaborate more on previous post: if in some fight two paladins are now healing two different tanks by HL spam, in 3.2 with the smart use of beacon thay could do exactly the same healing by casting HL twice less often (doing HL in turns, for example each 30 sec in one min) WHILE using DP each CD without gimping the healing on the tank (in 'idle' periods). So with two paladins / two tank scenarious all the changes are a clear win.
Also, I can not think about any fight except maybe Algalon (haven't healed by myslef, but seen the parses) where you really have to spam HL from the first to the last second or tank will die. Thorim hard I spam HL only after 7-9 stacks, Hodir hard (3-4 healers) - only during frozen blows or at the end of the fight just to burn my mana, Steelbreaker last - only in phase 3, XT hard - only during the tantrum.
I solo healed hodir 10 hard and even then I didn't have to spam HL really non stop. I feel, without crunching any numbers, that 3.2 changes will make my role and playstyle more interesting. Yes, I couldn't spam HL like no tomorrow, but I don't HAVE to.
To explain my point - there is NO fight where HL-spam HPS is mandatory. If there would be such a fight no class could replace the paladin and people would be forced to bring 2 healers to cover the absence of one paladin. That would be a big flaw in the encounter design and people would QQ till the fight is nerfed. Of course there are a few fights where our monstrous raw HPS is really helpfull, but no fight where it is mandatory. And if it's not mandatory than we CAN HL less if we optimize our playstyle better.
There were a few people who posted their parses and showed that with LoH glyph and mana pot they won't be OOM while maintaining the same level of HL usage. I would encourage everyone who dislike 3.2 changes look at their own parses and calculate how much mana they would loose and compare that with their mana at the end of the fight and sources of mana they probably neglected to cover the loss in 3.2.
@ Endoscient and Braque
I don't defend Braque model, but I would like to note that imho there is one fight where it can be relevant - Vezax hard with all regen switched off. Then if your weightings represent the actual percentage of spells used in that fight it should give the correct estimate for the overall increase in effectiveness (basically it will just say how big would be this increase for set number x, y, z of HL, FoL, HS). Though even in this case int might not be rated correctly, just close to the optimal value.
In general model should give correct results for SP (since changing SP we don't change tha rate at which we will cast spells - it doesn't affect mana pool), but it is irrelevant since other ratings will be wrong.
@ Braque
What people are trying to say is that (mainly because Illumination mechanics that return mana based on its usage) you can not model dynamical healing by looking at sort of static cases and then mixing them with weightings that represent the spell usage in a specific fight. This may work only for SP, adding any other stat WILL change the weightings of possible spell usage and this change isn't reflected in your sheet. As Endo wrote earlier, your model doesn't take into account time and mana trade-offs - by adding more stats like int or crit or mp5 you could potentially cast more HL but those HLs will decrease the amount of FoL and HS you are able to cast.
A few examples that you could try in your model:
1. how will your spreadsheet rate mp5 in a fight where length is small enough to cast HL non stop? For example put 30sec fight length. Obviously it should give 0 value even with regen allowed (non-Vezax mode). Will it rate it as 0?
2. How will your model rate 1 mp5 for a very specific set of parameters where this 1 mp5 makes exactly 1 more HL possible if you cast 1 FoL less. Obviously perfect model would rate this 1 mp5 MUCH higher than 1 mp5 in the case where it is not enough to cast one extra HL (since optimum model assumes you use your full mana to achieve highest HPS WHILE casting close to non stop).
To make your model work you could introduce different fight lengths on top of weighted sum, but that would work just because you have 6 parameters to tune to receive 5 values...
Last edited by Palados : 06/22/09 at 3:43 AM.
|
|
|
|
|
|
06/22/09, 3:32 AM
|
#803
|
|
I am a nice guy
Blood Elf Paladin
Mal'Ganis
|
Originally Posted by DiamondTear
And with all the powerful raid healing tools the other classes have, I can't really see us as a raid healer.
|
I think this is one part where I have to disagree, but I can see your point of view. Ulduar itself has one too many raid-wide raid damage fights, and it's a very poor model and one I hope we do not see more of in 3.2 (Profound Darkness on Vezax, P2 fire nova on mimiron, tremor on Freya+3, high voltage on steelbreaker last, etc).
However, when it comes to deep damaging RSTS effects, Paladins are very strong at raid healing. I can concede that if we're dealt a lot of raid-wide damage scenarios, then yeah, we're probably going to have another Sunwell on our hands. Regardless, I'm sure we'll manage.
|
|
|
|
|
|
06/22/09, 3:47 AM
|
#804
|
|
Don Flamenco
Human Paladin
Shadowsong (EU)
|
I think that many paladins and Blizzard are thinking about 'raid healing' differently.
Blizzard: raid healing is healing anyone except tank.
Paladins (and healer community in general): raid healing is healing raid-wide AoE damage.
Thus paladins will agree that they are viable in raid healing only if they have some sort of AoE heal, while Blizzard thinks we are fine with 3.2 changes.
Let us look at Maexxna fight. There raid healing is, basically, healing the cocooned players and we excel in it as single target healers. Another example - Thorim non hard assuming people avoid stuff that can be avoided. There you will have to heal only occasional chain lightnings and paladins are fine. There are many fights where spells that heal a lot of people at the SAME time are not crucial. In all those fights palas are good raid healers. Actually I think that HL rotation on 5 targets is giving more HPS than CoH 'spam' for the same level geared priest and paladin. Thus we are 'bad' raid healers ONLY if mechanics requires topping many people at the SAME time. And I think I am fine with this role, assuming that no tier of raiding will have 100% fights with heavy raid damage.
Also think about this - can any class but paladin solo heal hodir10 hard in not top guild raid? I really doubt it unless raid avoids Icicles and doesn't let Biting Cold to stack while at the same time stay hugged for AoE healing (not really doable for average guild). AoE powerfull classes will fail simply because they couldn't keep tank and raid up at the same time - loads of people are spreaded too much for AoE to reach them and tank gets big spikes at times. Paladin can do it with BoL on tank and HLing the raid.
Or think about XT fight mechanics without tantrums - 1 person at a time gets a bombs and has to be actively healed. Paladins in 3.2 can do it while healing the tank at the same time. No other class can do it as effective as we can. In this situation we are perfect raid healers - heal a lot of damage on one person while keeping tank up at the same time. It will spare 1-2 raid healing spots in those type of fights.
Kologarn - only shammy can heal 3 gripped targets (by rotating CH) better than paladin (who rotates HL). Priest and druids can't produce such HPS because WG and CoH cooldowns. But unlike shammy we CAN heal the tank at the same time.
Auraya - I bet that 3 paladins in 3.2 can heal this fight in 25man, while 3 healers of ANY other class will have big problems trying do so.
Vezax hard - get one priest or shammy for inspiration and stack paladins since FoL is the most effective spell in game for that fight (plus you can rotate DS in phase2). If you have to ACTIVELY AoE heal in Vezax hard fight you are doing something wrong and will wipe anyway.
Yogg-Saron - if your kickers are good then a bunch of paladins will heal this fight without any problem.
To be honest, only T8 fights where stacking paladin healers wouldn't work in 3.2 will be Mimiron, Freya+3 and Algalon assuming semi-decent tactics execution. I think that HL rotations will suffice for Hodir frozen blows since there you don't have other serious stuff oneshotting people at low health as in Freya encounter (assuming they play reasonably well).
Don't get me wrong, I don't want to say that in 3.2 we will be the best raid healing class. However we will be able to cover raid healing role much better than we can now with the new BoL mechanics. And only thing to prevent us from spamming HL on raid is mana regeneration nerf. Either that or remove the CD off CoH and WG to let priests and druids excel in raw many-target HPS.
Last edited by Palados : 06/22/09 at 4:21 AM.
|
|
|
|
|
|
06/22/09, 6:21 AM
|
#805
|
|
Don Flamenco
Human Paladin
Bronze Dragonflight (EU)
|
Originally Posted by Braque
I really can't be bothered with you any more. Your argument is specious. A simulation can tell you one thing: Can you heal this fight. Yes or no. If you can, any "improvement" is irrelevant.
|
The one thing you're learning from a simulation is the minimum amount of power required to successfully heal the fight under the conditions you have chosen, which will most commonly be optimal conditions as they're easier to simulate. Improvement on that is far from irrelevant, as you won't always have the optimal conditions. If you can't step up to compensate when something goes awry, you're done.
|
|
|
|
|
|
06/22/09, 8:03 AM
|
#806
|
|
Piston Honda
Dwarf Paladin
Turalyon (EU)
|
Good morning fellow paladins and theory crafters.
I am at the office now, with access to a Windows PC, and have been able to see how the exported spreadsheet appears in Excel - that is to say extremely broken, unfortunately - so I have a better understanding of where some of the misunderstandings might be coming from.
First of all, I'd like to clear up, the model does not weight holy shock as of equal importance to Holy Light/Flash of Light (re: HolyCow). The sheet does not work as described by Endoscient.
As to 'proving it's relevance', I would have thought that people here would understand you can't prove a positive, and that falsification is the foundation of the scientific method, that said clearly the usefulness if the spreadhsheet has been cast into doubt - all I can think of as a reassurance is to point out that using the weightings "correctly" selects Breastplate of the Devoted from XT hard mode as the BiS paladin chest, as is generally accepted by the raiding community. All the other 'top picks' from the weighting scale it produces also match the community's better judgment as far as I can see:
Plate Armor - Items - World of Warcraft
Given the criticism from Endoscient I am still hoping for some concrete examples of incorrect picks from the weightings, that I can use as a basis for improving the model. I presume his reaction to the sheet is based on some genuine problem with the results it produces, and not just a question of him not liking the methodology.
In reply to Palados: The sheet supports any length fight you are interested in (though you do have to change the fight length in 2 tables which is kind of stupid, I'll change that later). Given a situation where the paladin doesn't need any more MP5 to spam HL for the duration the sheet will correctly weight MP5 at 0. It also correctly reduces the int weight down to the SP contribution it makes, and crit down to the value from bonus healing. Note the Vezax fight weightings I gave from the sheet correctly exclude any regen stats (but high value on int for mana pool).
|
|
|
|
|
|
06/22/09, 10:38 AM
|
#807
|
|
Don Flamenco
Human Paladin
Shadowsong (EU)
|
Originally Posted by Braque
Given the criticism from Endoscient I am still hoping for some concrete examples of incorrect picks from the weightings, that I can use as a basis for improving the model.
|
While in general the choise of items is OKish, there are some wrong choises. Razorscale shoulderguards are rated by you higher than T8 (10man). It means that you probably rate int too little and haste with mp5 a bit too high. Also your model doesn't seem to support gemming for int neglecting 'lame' socket bonuses, that would clearly give slightly better healing output. This 'low' int rating has the reason described above - adding int not just allows you to cast more of each spell (that your model will reflect) but it also allows you to increase the HL/FoL ratio (thus increasing the fight HPS even more), that isn't reflected.
Also if you look at your model results and RAWR results closer, you can see that the correlation between them is good only if you look at 'burst healing' while neglecting the 'fight healing' (and even then the model may be off in a few cases giving one wrong item on top like those shoulders).
To sum it up: you could use the weightings produced by your sheet to rate the items, but at times it will give wrong results and rawr will do it better. Applying Occam's razor I would say that one would better use rawr.
Last edited by Palados : 06/22/09 at 10:54 AM.
|
|
|
|
|
|
06/22/09, 11:05 AM
|
#808
|
|
Piston Honda
Dwarf Paladin
Turalyon (EU)
|

Originally Posted by Palados
While in general the choise of items is OKish, there are some wrong choises. Razorscale shoulderguards are rated by you higher than T8 (10man). It means that you probably rate int too little and haste with mp5 a bit too high. Also your model doesn't seem to support gemming for int neglecting 'lame' socket bonuses, that would clearly give slightly better healing output. This 'low' int rating has the reason described above - adding int not just allows you to cast more of each spell (that your model will reflect) but it also allows you to increase the HL/FoL ratio (thus increasing the fight HPS even more), that isn't reflected.
Also if you look at your model results and RAWR results closer, you can see that the correlation between them is good only if you look at 'burst healing' while neglecting the 'fight healing' (and even then the model may be off in a few cases giving one wrong item on top like those shoulders).
To sum it up: you could use the weightings produced by your sheet to rate the items, but at times it will give wrong results and rawr will do it better. Applying Occam's razor I would say that one would better use rawr.
|
Looking at the order that the spreadsheet ratings give sholder items, I have to say, I agree with it. I don't think T8 (10 man) is worth taking over the razorscale ones (unless you need them for the 4 part set bonus), which in turn I don't think are worth taking over elevated lair pauldrons. At this point I'd say that I believe that many paladins underrate mp5 and overrate intellect, especially in the light of recent changes (the link i gave is weighted for the 3.2 nerfs). That said, I have the sheet set up to assume DP being used about half of the maximum possible - some support chaining it, which would would give a higher weighting to int, but I find I rarely need to chain DP on cooldown, so YMMV on that.
Your comment about HL/FoL ratios not being considered by my spreadsheet is incorrect, the spreadsheet weighting table automatically adjusts the weighting scale towards HL the longer the base scenario allows you to keep casting HL.
EDIT: PS - one of my reasons for building a spreadsheet, rather then depending on Rawr (other than the perverse pleasure I get out of doing stuff like that) is that I don't really like "scouting" gear and then building an ideal set in an external tool, I prefer to be surprised by drops when I explore new content, but it's useful to have a mod (ie Pawn) that can give me a quick idea of where an item stands relative to what I have. I don't expect a weighting scale to be more accurate than a simulation, nor do I try to claim my spreadsheet will be "better" than Rawr. That'd be silly. Different tools for different jobs. 
|
|
|
|
|
|
06/22/09, 11:29 AM
|
#809
|
|
Von Kaiser
A
Gnome Hunter
No WoW Account
|
After seeing the changes and skimming posts since then I've seen a lot of people praising the changes and a lot of people really down about them, but nothing that's particularly addressed my concerns as a raid leader. I haven't really healed on my paladin all that much since Sunwell other than a handful of normal mode Ulduar bosses that we only needed one tank so I thought I'd check here for some perspective.
Before Paladins were amazing tank healers. I'd direct our Holy Paladin to spam Holy Lights on the tank and the HPS he was able to put up was incredible. With the nerf to both Illumination and overall mana, I'm relatively certain that this spam will no longer be viable (and many people have agreed with that suspicion throughout this thread). If Paladins can no longer maintain the high hps they could before, wouldn't Discipline Priests become the dominant tank healer (depending on how hard the Penance nerf hits them). The absorbs and continued longevity of a Discipline Priest seems like it'd be better than a Paladin throwing a mix of Holy Lights and Flash of Lights.
The other major aspect of encounters has always been raid healing. The Beacon change, while definitely helping with Paladin raid healing ability, still comes no where close to the mass raid healing power that Shamans, Holy Priests, and Resto Druids (lol no nerfs) can do, especially since Glyph of Holy Light won't get nearly as much use since you'll be using plenty of Flash of Light. I can definitely think of some situations where having a healer pick off low raiders while doing other healing could be useful, but in the long run, large amounts of large coverage raid healing is more effective.
So, if Discipline Priests are now the dominant tank healer and Restoration Druids, Restoration Shamans and Holy Priests shadow a Paladins raid healing, why do I continue bringing a Holy Paladin to my raids? Am I underestimating the power of Beacon? Am I overestimating the nerf to mana regen? Am I wrong to completely exclude the Flash of Light HoT since heals are likely to be cast on a target without Sacred Shield?
|
|
|
|
|
|
06/22/09, 11:40 AM
|
#810
|
|
Von Kaiser
|
Originally Posted by Palados
Also think about this - can any class but paladin solo heal hodir10 hard in not top guild raid? I really doubt it unless raid avoids Icicles and doesn't let Biting Cold to stack while at the same time stay hugged for AoE healing (not really doable for average guild). AoE powerfull classes will fail simply because they couldn't keep tank and raid up at the same time - loads of people are spreaded too much for AoE to reach them and tank gets big spikes at times. Paladin can do it with BoL on tank and HLing the raid.
|
In my guild every class has solo healed hodir10 hard mode. Paladins probably have the hardest time of it, tbh. Without a DSac to reduce the frozen blows damage we fall too far behind during frozen blows and people die.
|
|
|
|
|
|
06/22/09, 12:06 PM
|
#811
|
|
Piston Honda
Blood Elf Paladin
Zuluhed
|
Originally Posted by Toppazz
So, if Discipline Priests are now the dominant tank healer and Restoration Druids, Restoration Shamans and Holy Priests shadow a Paladins raid healing, why do I continue bringing a Holy Paladin to my raids? Am I underestimating the power of Beacon? Am I overestimating the nerf to mana regen? Am I wrong to completely exclude the Flash of Light HoT since heals are likely to be cast on a target without Sacred Shield?
|
Because paladins are the only class that can heal both the raid and the tank effectively at the same time. The mindset you're demonstrating is stuck in current, old-fashioned ideals: one-to-two tank healers, five raid healers. Usually no less than 6 healers. No one heals outside their target.
This Beacon change means that you can assign 2 paladins to heal the raid, with beacon on the tank, and 3 other healers to help with raid healing, for a total of the same healing output of 6 healers, using only 5. In other words, Beacon itself has become the new "tank healer." It eliminates the need for a dedicated raid slot for a tank healer, since two holy paladins (or even one, of adequate gear and skill level) can cover a tank simply by healing the raid. It is an incredibly powerful tool, provided we can resolve the mana issues incoming with it.
|
|
|
|
|
|
06/22/09, 1:05 PM
|
#812
|
|
King Hippo
Ermad
Human Paladin
No WoW Account
|

Originally Posted by Braque
Good morning fellow paladins and theory crafters.
I am at the office now, with access to a Windows PC, and have been able to see how the exported spreadsheet appears in Excel - that is to say extremely broken, unfortunately - so I have a better understanding of where some of the misunderstandings might be coming from.
First of all, I'd like to clear up, the model does not weight holy shock as of equal importance to Holy Light/Flash of Light (re: HolyCow). The sheet does not work as described by Endoscient.
As to 'proving it's relevance', I would have thought that people here would understand you can't prove a positive, and that falsification is the foundation of the scientific method, that said clearly the usefulness if the spreadhsheet has been cast into doubt - all I can think of as a reassurance is to point out that using the weightings "correctly" selects Breastplate of the Devoted from XT hard mode as the BiS paladin chest, as is generally accepted by the raiding community. All the other 'top picks' from the weighting scale it produces also match the community's better judgment as far as I can see:
Plate Armor - Items - World of Warcraft
Given the criticism from Endoscient I am still hoping for some concrete examples of incorrect picks from the weightings, that I can use as a basis for improving the model. I presume his reaction to the sheet is based on some genuine problem with the results it produces, and not just a question of him not liking the methodology.
In reply to Palados: The sheet supports any length fight you are interested in (though you do have to change the fight length in 2 tables which is kind of stupid, I'll change that later). Given a situation where the paladin doesn't need any more MP5 to spam HL for the duration the sheet will correctly weight MP5 at 0. It also correctly reduces the int weight down to the SP contribution it makes, and crit down to the value from bonus healing. Note the Vezax fight weightings I gave from the sheet correctly exclude any regen stats (but high value on int for mana pool).
|
It is insanely difficult to have weightings that don't pick Breastplate of the Devoid as BIS, it is by far the best in that slot because of its high item level and 3 gem slots. Your value of crit is too high for 3.2 changes. It rates items like [Nebula Band] much higher then I think it should. That isn't the only incorrect weighting, but the most obvious. You shouldn't confuse the fact that items with higher item level are picked as meaning it is producing correct results.
I described the sheet as simulating casting Holy Shock every 6 seconds for 6 minutes and averages that in equally with FoL/HL, which is what it does. Whether or not that is equal depends on how you define equal, equal time spent, mana spent, healing done, etc.
You still refuse to defend the reasoning for your results. You keep saying they are true until proven otherwise, which is not the scientific method you claim it to be. You cannot prove something because there is no found counterexample, you can only disprove it because there is one. It is also quite possible to prove stuff positive with induction, contradiction, and similar methods.
[e] I am using a Mac with Numbers, so no misunderstanding is coming from that.
Last edited by Endoscient : 06/22/09 at 1:25 PM.
|
|
|
|
|
06/22/09, 1:32 PM
|
#813
|
|
Von Kaiser
A
Gnome Hunter
No WoW Account
|
|
This Beacon change means that you can assign 2 paladins to heal the raid, with beacon on the tank, and 3 other healers to help with raid healing, for a total of the same healing output of 6 healers, using only 5.
|
Or, I could bring a Discipline Priest and four Restoration Druids (or Holy Priests or maybe Restoration Shamans at patch) and have safer tank healing (the primary tank healer is looking at the tank's HP after all) with some solid support tank healing and probably do double the effective raid healing.
And, I'm not sure if we're looking at the same content or not, but I can't think of many encounters where a Paladin can solo heal a tank if they aren't able to use 100% HL. FoL's HPS is way too low to deal with my average incoming damage across hard modes thus far.
|
|
|
|
|
|
06/22/09, 1:35 PM
|
#814
|
|
Piston Honda
Dwarf Paladin
Turalyon (EU)
|
I'm pretty sure I did try to explain/defend the reasoning, but apparently, based on the incorrect descriptions of the sheet repeated here, people are having trouble understanding it. Is it worth me repeating it? Probably not, sadly the board seems overwhelmingly hostile to new models, and it's hard work to keep a positive attitude. A possible reason that crit might be 'overvalued' would be my treatment of throughput and 'full duration' values as being equal, but given I've not yet had any mana problems that seemed fair. I could add a slider for that. What weightings do you get from your model, given similar start values as I'm using? Do you think there is any genuine error in the mechanics modeling? Or is it just the system I use to go from 'results' to 'stat weightings'? Seeing as all criticism has been of the weighting model can I assume the mechanics appear solid?
|
|
|
|
|
|
06/22/09, 1:51 PM
|
#815
|
|
Glass Joe
Blood Elf Paladin
Shattered Hand
|
Originally Posted by Braque
I'm pretty sure I did try to explain/defend the reasoning, but apparently, based on the incorrect descriptions of the sheet repeated here, people are having trouble understanding it. Is it worth me repeating it? Probably not, sadly the board seems overwhelmingly hostile to new models, and it's hard work to keep a positive attitude. A possible reason that crit might be 'overvalued' would be my treatment of throughput and 'full duration' values as being equal, but given I've not yet had any mana problems that seemed fair. I could add a slider for that. What weightings do you get from your model, given similar start values as I'm using? Do you think there is any genuine error in the mechanics modeling? Or is it just the system I use to go from 'results' to 'stat weightings'? Seeing as all criticism has been of the weighting model can I assume the mechanics appear solid?
|
There are a couple of things I thought of that I don't see your model taking into consideration. Replenishment will normally be up. Did the model have replenishment on it? If not, taking it into consideration will raise the value of int. The other thing is that haste does more than raise your HPS; it allows you to heal targets before raid smart heals do. I think this would be a hard thing to include in some sort of model, but it is still important.
|
|
|
|
|
|
06/22/09, 2:05 PM
|
#816
|
|
Piston Honda
Dwarf Paladin
Turalyon (EU)
|
Originally Posted by ClayMask
There are a couple of things I thought of that I don't see your model taking into consideration. Replenishment will normally be up. Did the model have replenishment on it? If not, taking it into consideration will raise the value of int. The other thing is that haste does more than raise your HPS; it allows you to heal targets before raid smart heals do. I think this would be a hard thing to include in some sort of model, but it is still important.
|
Replenishment is included, yes, I have it set to 90% uptime by default on the sheet. I totally agree with you about haste, thats just one of those things where you look at the numbers and say - what the hell, I don't care if I get more healing over all with (ie) elevated lair pauldrons, I want the extra cast speed I get get from (ie) razorscale pauldrons. The theorycrafting is there to lets us make better decisions about those kinds of trade offs, but it'll never make the decision for you.
|
|
|
|
|
|
06/22/09, 2:07 PM
|
#817
|
|
King Hippo
Ermad
Human Paladin
No WoW Account
|
Originally Posted by Braque
I'm pretty sure I did try to explain/defend the reasoning, but apparently, based on the incorrect descriptions of the sheet repeated here, people are having trouble understanding it. Is it worth me repeating it? Probably not, sadly the board seems overwhelmingly hostile to new models, and it's hard work to keep a positive attitude. A possible reason that crit might be 'overvalued' would be my treatment of throughput and 'full duration' values as being equal, but given I've not yet had any mana problems that seemed fair. I could add a slider for that. What weightings do you get from your model, given similar start values as I'm using? Do you think there is any genuine error in the mechanics modeling? Or is it just the system I use to go from 'results' to 'stat weightings'? Seeing as all criticism has been of the weighting model can I assume the mechanics appear solid?
|
You need to demonstrate why the results produced by the arbitrary situation modeled in your spreadsheet is relevant to real fights. Which you have not done, even when asked many times. Just because it models spells and throws them together in some formula doesn't mean it produces correct results.
There might still be errors in mechanic calculation, I don't agree with the base methodology so I have been focusing on that.
|
|
|
|
|
06/22/09, 3:35 PM
|
#818
|
|
Don Flamenco
Human Paladin
Shadowsong (EU)
|
Originally Posted by Toppazz
Or, I could bring a Discipline Priest and four Restoration Druids (or Holy Priests or maybe Restoration Shamans at patch) and have safer tank healing (the primary tank healer is looking at the tank's HP after all) with some solid support tank healing and probably do double the effective raid healing.
And, I'm not sure if we're looking at the same content or not, but I can't think of many encounters where a Paladin can solo heal a tank if they aren't able to use 100% HL. FoL's HPS is way too low to deal with my average incoming damage across hard modes thus far.
|
I am not sure the nerfed DS priests will excel paladins in MT healing. First of all it was shown that for holy/prot spec SS could absord as much damage as priest shields over one minute, though in a weird way (think of them as 'block rating shield': small but often, while priest shield is 'avoidance shield': big but rare).
Secondly, as me and a few other people have written before, you still will be able to provide CURRENT level of HL/fol ratio with mana pot/loh glyph/mana tide (that most pallies just ignore now). You say that "I can't think of many encounters where a Paladin can solo heal a tank if they aren't able to use 100% HL" but if you actually look at the parses of any guild you will see that paladins do not use HL really non stop. My point is that paladins in 3.2 can spam HL if needed just not as long as they can now, HL spam won't be over, it just will not be mindless. If in any encounter paladin is forced to spam HL really non stop or tank will die I guarantee than no disc priest of similar gear level would solo keep the tank up in this situation. Just look at this link:
WoW Meter Online - Combatlog Replay
Do you think that any disc priest could provide over 10k HPS (healing+absorbs) for the whole fight that lasts longer than 2-3min? Or even 8k?
Another thing that was already mentioned - two paladins can provide currently achievable HL spam by rotating their divine pleas. This means that 2 paladins with the new beacon will be equivalent to 3.5 pre-3.2 paladins that have plea without MS debuff (each pala has double output 75% of the time and usual output 25% of the time). Thus bringing two palas you actually bring 3.5 MT/raid healers of current paladin power (without nerfs it would be like 4 current paladins).
Originally Posted by Braque
At this point I'd say that I believe that many paladins underrate mp5 and overrate intellect, especially in the light of recent changes (the link i gave is weighted for the 3.2 nerfs).
|
You believe in something that is not supported by real simulations (rawr). 1 point of mp5 will give the same mana in 3.2 and one int will give almost the same amount of hps/effectiveness. Patch will just change the value of crit drastically.
Rawr simulations show that t8 shoulders will give slightly better overall fight raw healing done and marginally better burst HPS than razorscale shoulderguards. Even in 3.2 crit HPS won't be changed (only effectiveness) so I expect them being slightly better even with the patch changes.
Last edited by Palados : 06/22/09 at 3:53 PM.
|
|
|
|
|
|
06/22/09, 4:48 PM
|
#819
|
|
Von Kaiser
A
Gnome Hunter
No WoW Account
|
Palados - Is the post regarding Sacred Shield in this thread? I've only been skimming recently since I don't heal as a Paladin much anymore, but I didn't come across it in my readings. I definitely don't doubt it, but would love to see some numbers on it if they're available. I guess I've been almost completely disregarding the Sacred Shield buff and assuming Discipline Priests bring all the absorbs to the raid, but that certainly isn't true.
You answered what was mostly my primary concern (Paladins being able to maintain the same type of HPS that they can currently put out in high tank damage situations like Thorim 7+ stacks or Steelbreaker last). I still feel like relying on a Beacon for all of your tank healing is pretty dangerous, but that's more of a matter of opinion than anything else.
|
|
|
|
|
|
06/22/09, 4:55 PM
|
#820
|
|
MELF Master Race (also, better then pigtail orcs)
Blood Elf Paladin
Mal'Ganis
|
Originally Posted by Palados
Another thing that was already mentioned - two paladins can provide currently achievable HL spam by rotating their divine pleas. This means that 2 paladins with the new beacon will be equivalent to 3.5 pre-3.2 paladins that have plea without MS debuff (each pala has double output 75% of the time and usual output 25% of the time). Thus bringing two palas you actually bring 3.5 MT/raid healers of current paladin power (without nerfs it would be like 4 current paladins).
|
To be fair, thats pretty inflated as far as number of healers. Say what you will, but paladins are getting no additional raid-healing boost. After 3.2, we will be just as good at raid healing as we are today (except with less mana).
Experiment: Take a paladin today, put him on raid healing and have him keep beacon 100% uptime on the MT (make sure to have a normal amount of tank healers so that you dont wipe). Compare his raid healing numbers to your normal core of Holy Priests and Resto Shaman/Druid. Small power flashes and non-smart healing HL bombs will keep said paladin WAY behind the other 3.
Translate this to 3.2:
Holy Paladins REMAIN primarily a tank healer thru beacon, but are able to add supplemental healing to raid. Even if the raid is taking 0 damage, the paladin will still be chaincasting something, either HL or Flash. But now in 3.2, our ability to be a tank healer is diminished due to mana (some fights mana isn't an issue, others like hardmodes, its definately an issue). However, to accommodate for this nerf, we gain the ability to add additional raid healing.
Edit: @Gtt, Paladin HPS on tanks can remain the nearly the same (for certain durations). Our total throughput is not being changed, for the most part. However, there are exceptions.
1) Regearing/gemming/specing from Crit rating/Crit talents to protection will happen by many. We will pick up bonuses to Sacred Shield and Divine Guardian for raid damage absorption. This means that there will be a subtle drop in HPS due to the loss of some crits, but not a ton.
2) Duration. As it stands now, Paladins can put out unsurpassed tank HPS for a very long duration. However, in 3.2, that will go WAY down. The 50% cut to illumination returns, smaller mana pools, and different gear/talents will mean that we can only put out that HPS for a short 'burst', otherwise we run a significant risk of going OOM (without being able to innervate back up).
Last edited by tiberion02 : 06/22/09 at 5:00 PM.
|
Borderlands: tiberion02, add me.
|
|
|
|
06/22/09, 5:01 PM
|
#821
|
|
Don Flamenco
Human Paladin
Shadowsong (EU)
|
Originally Posted by Toppazz
Palados - Is the post regarding Sacred Shield in this thread?
You answered what was mostly my primary concern (Paladins being able to maintain the same type of HPS that they can currently put out in high tank damage situations like Thorim 7+ stacks or Steelbreaker last). I still feel like relying on a Beacon for all of your tank healing is pretty dangerous, but that's more of a matter of opinion than anything else.
|
post is here The Holy Paladin Guide for 3.1
Important part:
|
Assuming Divine Guardian + 4PT8, this translates to roughly 3,222 damage absorbed every 4 seconds.
|
that will translate to 48330 maximum possible absorb per minute. This number doesn't take into acount 'hidden' 10% buff to SS from 4pT8 that will raise the threshold to something around 50k absorb per minute.
Also while relying on beacon is dangerous now, it will be absolutely safe with 3.2 changes (60y is a huge distance by the way). In the worst case, when paladin is unable to spend GCD to renew BoL, he will just spam MT directly.
|
|
|
|
|
|
06/22/09, 5:04 PM
|
#822
|
|
Von Kaiser
|
Tiberion: I raid heal pretty frequently and I always come in a solid second // third on EHPS meters. Obviously it depends on the fight and damage pattern but on most fights I feel that I am a quite competitive raid healer.
|
|
|
|
|
|
06/22/09, 5:06 PM
|
#823
|
|
Divine Protector
Blood Elf Paladin
Mal'Ganis
|
Originally Posted by tiberion02
To be fair, thats pretty inflated as far as number of healers. Say what you will, but paladins are getting no additional raid-healing boost. After 3.2, we will be just as good at raid healing as we are today (except with less mana).
|
Not so much. In 3.2, from 60 yards away you can Beacon the MT (I don't know an example of what another 20 yards gives you as a bonus, but more range is nice) and spam heal someone that has a increased healing bonus talent (Paladin with Divinity or Assassin Rogue, there are others).
The fact that Beacon takes into all healing done is a huge boon, since smart heals done in 3.1 may eat up your Beacon healing, making Beacon have little value for raid healing in 3.1, since you can't count on Beacon to do anything.
Another thing to take into account is 3.2 is not even on the PTR. I can't think of one PTR that the initial changes went to live without tweaks.
The biggest thing for me is 3.2 is having a "bigger number". The healing done may not matter much, but it feels good for me to have the most healing done (which I have in Ulduar a few times, but never as Holy).
|
DK - Ashbane Failure is the condiment that gives success its flavor.
|
|
|
|
06/22/09, 5:19 PM
|
#824
|
|
Von Kaiser
A
Gnome Hunter
No WoW Account
|
Originally Posted by Palados
Also while relying on beacon is dangerous now, it will be absolutely safe with 3.2 changes (60y is a huge distance by the way). In the worst case, when paladin is unable to spend GCD to renew BoL, he will just spam MT directly.
|
I'm actually more concerned with the paladin selecting the correct heal for their tank assignment while actually watching the and landing that heal on the raid.
|
|
|
|
|
|
06/22/09, 5:21 PM
|
#825
|
|
MELF Master Race (also, better then pigtail orcs)
Blood Elf Paladin
Mal'Ganis
|
edit^^^: Or the 'I was casting Holy Light on someone who was near death, but they died before the cast went off, which lead to my healing gap of >3 seconds' and the tank died. This will happen, it'll be rare, but you will hate everything in the world when it does.
Originally Posted by frmorrison
Not so much. In 3.2, from 60 yards away you can Beacon the MT (I don't know an example of what another 20 yards gives you as a bonus, but more range is nice) and spam heal someone that has a increased healing bonus talent (Paladin with Divinity or Assassin Rogue, there are others).
The fact that Beacon takes into all healing done is a huge boon, since smart heals done in 3.1 may eat up your Beacon healing, making Beacon have little value for raid healing in 3.1, since you can't count on Beacon to do anything.
Another thing to take into account is 3.2 is not even on the PTR. I can't think of one PTR that the initial changes went to live without tweaks.
|
My point is, that in 3.2, you will be doing the same amount of raid healing as you can now (if your assigned to raid heal now). Beacon isn't going to help you keep the other 24 people in the raid who aren't the MT alive. I was saying, mostly directed to Toppazz, that if you want to see what to expect for raid healing from a paladin (compared to your normal raid healers), give your holy paladin a raid or 2 to only be a raid healer and not cast on the MT at all.
Now, I do wonder if healing bonuses on other characters will effect beacon heals. I personally doubt it very much, and expect the 3.2 beacon to heal for the unmodded heal amount. However, if I can cast on a Warlock with the +20% healing bonus and that transfers to the tank, I will have a very different opinion on this patch. Then it will just be a matter of convincing (forcing) the Warlock to gimp their DPS for a big healing buff.
And sure, 3.2 hasnt even hit the PTR, and I do expect tweaks, but there is not point in speculating on unannounced changes at this point. I'd rather just focus on what we know so far.
|
Borderlands: tiberion02, add me.
|
|
|
|
|