Archived

This topic is now archived and is closed to further replies.

Squeakster

On the Value of Haste

60 posts in this topic

One can imagine a website or downloadable program where you could upload your combat log and have it spit out specific stat weightings for your exact performance.

Before you go too far down this path you might check out shaman_hep and the accompanying EJ thread, It appears to be fairly well regarded and may even serve as a decent starting point for a priest tool.

Share this post


Link to post
Share on other sites

- Holy concentration uptime :

This is basicaly the same idea but for holy and regen, haste will also increase your chance to keep holy concentration up (depending on how you manage your prom/coh/filler use of course). And it scales with crit for that matter as well. So it's not a straigth mana consumption increase there.Good point again, and I will add this to my original post.

I really don't like the kind of argument saying "Haste increases the number of heals you cast, therefore your holy concentration uptime".

Technically, this is really true.

But the effect is marginal, and people shall not count on it. If you want to increase your holy concentration uptime, you gear for crit, and/or you change your spells selections (with short spells that can trigger it). You don't gem for haste.

The gain in holy concentration uptime is far from the loss of the increased number of cast.

Just for a quick numerical evaluation :

HC is an 8s buff that procs after any crit.

I'll take the best situation for haste : the charater currently has no haste, 30% crit, and can choose between 10% haste and 10% crit.

He maintains a cycle of 1 POM - 1 COH - Renew - Renew - SOL-FH, that takes roughly 8s (including lag).

In 8s, 2 spells can triggers HC.

The current downtime is 0.7^2 = 0.49. Uptime is 51%.

With 10% more crit, downtime is 0.6^2 = 0.36. Uptime is 64%. The gain is 13% uptime.

With 10% haste, the character can know cast 2.2 HC-friendly spells per 8s. Downtime is 0.7^2.2=0.456. Uptime is 54.4%. The gain is 3.4% uptime for 10% haste, in a really favorable scenario.

Share this post


Link to post
Share on other sites

I'll take the best situation for haste : the charater currently has no haste, 30% crit, and can choose between 10% haste and 10% crit..

Only error I see in your statement is the 10% crit vs 10% haste. Haste and crit have different rating numbers for 1% gain. Therefore we should be comparing say 200 crit rating to 200 haste rating, as items are weighted based upon rating numbers not percentages.

I do believe that crit is very strong for disc, but this should be vetted by more people than me before it becomes gospel.

I'm following this discussion mostly to try and verify my belief in this as well. :) Think I might have to finally resocket crit for int for the LK.... tearing my mana pools apart.

Share this post


Link to post
Share on other sites

It feels like the thread has drifted off abit.

Remember, You "can only" heal as much as there is need for which is why I'd consider discussions as this one to be pretty useless.

Healing has too many factors involved to say "this is the best" for example, what healing setup, what encounter, how good is your group (bad tank/people tanking fires). Discussions as this one is really only viable for one thing which is heavy aoe encounters which of we only have two bosses (in ICC) of excluding a few phases such as "phase 1" of Festergut etc.

As I've told every priest who've asked me for tips so far including guildies asking for tips, the best way is to try and balance things, Holy wise a decent balance of Int, Spi and Haste (in my opinion) is the stats to try and balance whilst crit and spellpower fall abit behind in the current content. When you reach Hardmodes I'd consider Haste to be slightly better until you reach around a 1000 and the "DR" really starts to kick in for real. For example my Penance is currently at 1.5 sec and casting up a PW:S for Borrowed Time which is 25% haste only decreases it with 0.2 sec or so. Getting "GCD capped" would be considered valueable, but is it really worth sacrificing so much to get there, you're talking hundreds of spellpower, int, spirit you'd have to sacrifice to gain those 0.1 seconds which you, more than 50% of the time don't really need?

Share this post


Link to post
Share on other sites
The spellpower calculation was very back of the envelope and I'd like to do it "correctly" at some point. I thought I made that clear in the post, but re-reading it now I see I should have been more explicit about this. I made two assumptions in coming up with the 33 SP -> 0.48% ratio. The first one, which is probably not very big, is that I ignored the boundary cases where a non-overheal gets turned into an overheal. The second is I ignored the base heals of spells, so I simply said that 1% increase in spellpower would increase all non-overheals by 1%. 52% of the direct heals in the parse had some overhealing, and I looked up the disc priest from the parse on the armory and he had ~3300 sp unbuffed.

That makes sense, thanks for the explanation.

-----------

Before you go too far down this path you might check out shaman_hep and the accompanying EJ thread, It appears to be fairly well regarded and may even serve as a decent starting point for a priest tool.

It appears that the Shamans have beaten us to the punch. This is exactly the kind of thing I was thinking about, thank you for calling our attention to it Mutagen.

The page for shaman_hep states, "This program is written in Perl and needs a Perl interpreter," so I guess we need someone knowledgeable in Perl to adapt it to Priest healing. I wonder how difficult it would be? I would imagine it wouldn't be too bad, just changing spell names and some of the variables. Although I bet parsing absorb effects would require some work.

Share this post


Link to post
Share on other sites
I really don't like the kind of argument saying "Haste increases the number of heals you cast, therefore your holy concentration uptime".

Technically, this is really true.

But the effect is marginal, and people shall not count on it. If you want to increase your holy concentration uptime, you gear for crit, and/or you change your spells selections (with short spells that can trigger it). You don't gem for haste.

The gain in holy concentration uptime is far from the loss of the increased number of cast.

Just for a quick numerical evaluation :

HC is an 8s buff that procs after any crit.

I'll take the best situation for haste : the charater currently has no haste, 30% crit, and can choose between 10% haste and 10% crit.

He maintains a cycle of 1 POM - 1 COH - Renew - Renew - SOL-FH, that takes roughly 8s (including lag).

In 8s, 2 spells can triggers HC.

The current downtime is 0.7^2 = 0.49. Uptime is 51%.

With 10% more crit, downtime is 0.6^2 = 0.36. Uptime is 64%. The gain is 13% uptime.

With 10% haste, the character can know cast 2.2 HC-friendly spells per 8s. Downtime is 0.7^2.2=0.456. Uptime is 54.4%. The gain is 3.4% uptime for 10% haste, in a really favorable scenario.

I agree that the effect of haste on Holy Concentration uptime is significantly less than the effect of crit and that the cost of that extra two-tenths of a spell will far outweigh the regen gained from higher Holy Concentration uptime, but that doesn't make it irrelevant. The goal of this thread is to develop an understanding of the effects of haste as fully as possible, which means accounting for all of the effects, even though some are smaller than others.

Going along with your numbers, which look like a sound approximation to me, and assuming 1800 Int and 1500 Spirit, spirit based regen provides 0.016725*1500*(1800)^(.5) = 1064 mp5 outside the five second rule or when Holy Concentration is up, and 1064/2 = 532 mp5 inside the five second rule. Assuming 90% of time is spent inside the five second rule and 51% HC uptime, one's overall regen would be 1064*0.1 + 532*0.9*0.49 + 1064*0.9*0.51 = 829 mp5. Increasing HC uptime to 54.4% makes that calculation 846 mp5, a 17 mp5 increase for 10% haste. Yes it is minor, but I want this to be comprehensive.

---------

My responses in bold:

It feels like the thread has drifted off abit.

Remember, You "can only" heal as much as there is need for which is why I'd consider discussions as this one to be pretty useless. If you would prefer to avoid such a "useless" thread and continue healing however you see fit, then please do. But if you would like to contribute to the discussion, then by all means, post something constructive.

The idea that "you can only heal as much as there is need for" is broached in the original post in the context of mana usage. It would appear from your statement that you are trying to argue that adding haste provides no benefit whatsoever because the amount of damage needed to be healed over the course of the encounter does not change as well. I will assume you are not trying to say that, because obviously that is absurd.

Healing has too many factors involved to say "this is the best" for example, what healing setup, what encounter, how good is your group (bad tank/people tanking fires). Discussions as this one is really only viable for one thing which is heavy aoe encounters which of we only have two bosses (in ICC) of excluding a few phases such as "phase 1" of Festergut etc. Yes, the value of haste in healing is an extremely complicated and difficult nut to crack, but that does not make it unworthy of our efforts. My goal for this thread is not to boil the entire debate down to one sentence proclaiming that one stat or technique is superior to all others. The goal is to more fully develop our understanding of the way haste affects how we play and our success in raiding. I fail to see how that only applies to situations with lots of AoE damage. Haste affects almost all aspects of our gameplay, regardless of the way damage is dealt in the particular encounter.

As I've told every priest who've asked me for tips so far including guildies asking for tips, the best way is to try and balance things, Holy wise a decent balance of Int, Spi and Haste (in my opinion) is the stats to try and balance whilst crit and spellpower fall abit behind in the current content. Do you have any math, data, or logical reasoning behind these statements? I'm not saying you are wrong, I'm saying that we are here to discuss these things, and if you make unsuported statements like this it gets us nowhere. Additionally, the goal of this thread is not to eventually print out some unequivocal, never-changing set of stat weightings for every stat that appears on our gear. Balancing longevity stats like spirit and intellect with throughput stats is an entirely different and complicated question.

When you reach Hardmodes I'd consider Haste to be slightly better until you reach around a 1000 and the "DR" really starts to kick in for real. What? First of all, you are making the fallacious argument that somehow hard modes have a special requirement for haste that does not exist for other stats or for normal modes. This is something I explicitly talk about in the original post. There is nothing special about haste in regards to hard modes. And the diminishing returns only start to "kick in for real" after 1000 haste? Where is your math for that? There is nothing special about 1000 haste, the diminishing returns are always present.

For example my Penance is currently at 1.5 sec and casting up a PW:S for Borrowed Time which is 25% haste only decreases it with 0.2 sec or so. You need to check your math here. If your Penance is 1.5 second cast time you are running with a total of 33% haste. Borrowed Time will give you an additional 25% haste, making the new cast time 2/(1.33*1.25) = 1.2 seconds, a savings of 0.3 seconds.

Getting "GCD capped" would be considered valueable, but is it really worth sacrificing so much to get there, you're talking hundreds of spellpower, int, spirit you'd have to sacrifice to gain those 0.1 seconds which you, more than 50% of the time don't really need? First of all, nobody here is saying that every priest needs to 1 second GCDs at all costs. We are not even saying that haste is better than other throughput stats. Secondly, you say that you would rather have spell power, intellect and spirit rather than enough haste to reach 1 second GCD's, but you provide absolutely no reasoning, math, or data to back this up. You want us to just take your word on it? And how did you decide that 50% of the time you do not need to take 0.1 seconds off your GCD?

Share this post


Link to post
Share on other sites
The page for shaman_hep states, "This program is written in Perl and needs a Perl interpreter," so I guess we need someone knowledgeable in Perl to adapt it to Priest healing. I wonder how difficult it would be? I would imagine it wouldn't be too bad, just changing spell names and some of the variables. Although I bet parsing absorb effects would require some work.

It is possible with some work to achieve a similar result for Rawr using the Custom Role option. I still don't have T10 2pc and 4pc added, but from the logs I've seen so far, 2pc is more or less a 11% increase on FH and 4pc is more or less useless due to limitations from circumstances.

Share this post


Link to post
Share on other sites

Just out of sheer curiousity, Squeakster. No offense.

How are you supposed to crack this "nut" which changes equation every second you're pulling, your guildmates dps, your tanks do an attack, the encounter mechanics. The amount of scenarios you'd have to take into account would be too much for any human being to handle. The only thing you can "assume" is how much haste will give you on a few handful of encounters in ICC.

I'm not against this topic, I like it. It's a summary of what me and firstaidspec was discussing not to long ago in the compendium. I'm just saying, all we need is a estimate of how haste balances out on;

1) "PoM, Renew, CoH encounters"

2) "PoM, CoH, PoH encounters"

3) "FH, GH, CoH, PoM encounters"

4) "Everything in your arsenal encounters"

5) "Shield spam encounters"

6) "FH, Penance, Shield, GH encounters"

All we really can discuss on the subject is how much are we going to get out of each spell in an optimal encounter, what I listed is as we all know some of the most common ones.

Share this post


Link to post
Share on other sites
Just out of sheer curiousity, Squeakster. No offense.

How are you supposed to crack this "nut" which changes equation every second you're pulling, your guildmates dps, your tanks do an attack, the encounter mechanics. The amount of scenarios you'd have to take into account would be too much for any human being to handle. The only thing you can "assume" is how much haste will give you on a few handful of encounters in ICC.

Same way we do for DPS. theorycraft, sim and estimate (the comparative static estimation a la shamanHEP is most promising). The goal isn't to gather a complete accounting but to get a little beyond the "hand wave" explanation for haste valuation.

Share this post


Link to post
Share on other sites
The page for shaman_hep states, "This program is written in Perl and needs a Perl interpreter," so I guess we need someone knowledgeable in Perl to adapt it to Priest healing. I wonder how difficult it would be? I would imagine it wouldn't be too bad, just changing spell names and some of the variables. Although I bet parsing absorb effects would require some work.

God damn it, Perl. The entire thing's in one single file that's almost 30,000 lines. A quarter of that is probably comments, but still. Yecchhhhh. You'd be better off starting from scratch.

Share this post


Link to post
Share on other sites
Same way we do for DPS. theorycraft, sim and estimate (the comparative static estimation a la shamanHEP is most promising). The goal isn't to gather a complete accounting but to get a little beyond the "hand wave" explanation for haste valuation.

We can definitely say things.

But compared to dps, which have a cycle and tend to spam it in most encounters (with some movement, or adds focus sometimes), we don't have it, and every encounter is a new theorycraft for us.

That doesn't mean that theorycraft has no value at all, but we shall be really careful with its assumptions before using it.

For example, Bobturkey's stats weighting and gear ranking is interesting, but because of its assumptions, I really prefer the "Take any iLVL increase" rule as holy priest, as long as the new gear doesn't have any hit (and if possible, MP5). I can find a fair use of all other stats. The distortion made by the strong assumptions (flash heal spam, MP5<->spellpower conversion) is in my mind greater than the difference in the final weighting.

Only exception is trinkets, where I use the difference to adapt to my current gear options and encounters. and where ilvl is sometimes meaningless (hello Solace).

Share this post


Link to post
Share on other sites

The page for shaman_hep states, "This program is written in Perl and needs a Perl interpreter," so I guess we need someone knowledgeable in Perl to adapt it to Priest healing. I wonder how difficult it would be? I would imagine it wouldn't be too bad, just changing spell names and some of the variables. Although I bet parsing absorb effects would require some work.

Pivoting off the complaint above (not quoted), the file is gargantuan. Most of it is declarations for the various spell weights, talents, and items which might show up in the combat log. You should be able to read it with any bare bones text editor. Depending on your OS, you might even be able to run the program (e.g. if you have a mac or run lunix, you've got perl installed already).

Share this post


Link to post
Share on other sites

Spell power is a little trickier because different spells have different coefficients. A full listing of Priest spell coefficients can be found at Spell power coefficient - WoWWiki - Your guide to the World of Warcraft. Perhaps the best I can do is give an example of a spell with a coefficient of 80.68% (PWS, Flash Heal, PoM and others have this coefficient):

Example
(Flash Heal)

((1887+2193)/2 + .8068*0) = 2040.0

((1887+2193)/2 + .8068*100) = 2120.7

We see that an increase in 100 Spell Power yields a 2120.7/2040 = 1.0396, or a 3.96% increase in healing. This is also a 3.96% increase in HPS regardless of cast times because the cast times between the two spells are equal and would cancel out.

This is quite misleading in my view; you have based your percentage increase upon the assumption that you begin with zero spellpower. Using an equally arbitrary baseline of 4,000sp, the extra 100sp would only add 1.53% to your healing done/HPS.

But we cannot neglect the itemization cost of the three stats. Haste rating and Critical Strike rating have equal 1.00 StatMods when it comes to item budgets. However, the conversion between rating and percentage are different between the two. Specifically, 32.79 haste rating equals 1% haste, while 45.91 crit rating equals 1% crit. This means that crit rating is 40% (45.91/32.79) more expensive than haste rating. So when considering gems/gear, note that the choice is not between 1% haste or 1% crit, but rather between 1.4% haste and 1.0% crit.

Spell power has a StatMod rating of 0.86, meaning that for every 1 haste or crit rating a piece of gear has, it could have 1.16 spell power (1/0.86). In the example I used above a 100 spell power increase resulted in 3.96% higher HPS. That 100 spell power could instead be 100/1.16 = 86.2 haste rating, which is 86.2/32.79 = 2.63% haste. Under the 1%:1% rule, 2.63% haste is a 2.63% HPS increase. So it would appear that gemming for spell power results in a larger increase in healing that haste.

I agree with your conclusion, but (as above) the reasoning seem faulty to me. The most important thing to keep in mind is that spellpower, crit, and haste all increase the value of each other as you gain more. Adding more spellpower makes haste more valuable. Adding haste makes spellpower more valuable - at least for a holy priest - as has been mentioned, disc priests are limited in the value they can extract from haste. There are definitely points at which a point of haste will be more valuable in terms of HPS than a point of spellpower for a holy priest.

Even with this in mind, I still support your conclusion that gemming for spellpower should be the preferred option. We only really get to have true freedom in choosing stats when we choose gems, since all available gear has a certain balance of stamina/intellect/spellpower, and then any two of spirit/crit/haste. When gemming, you can choose regen (i.e. intellect, possibly spirit), or throughput - so crit, haste, or spellpower. Of the three, only one increases the effectiveness of every part of every spell a priest casts - and thats spellpower.

Share this post


Link to post
Share on other sites
This is quite misleading in my view; you have based your percentage increase upon the assumption that you begin with zero spellpower. Using an equally arbitrary baseline of 4,000sp, the extra 100sp would only add 1.53% to your healing done/HPS.

I agree with your conclusion, but (as above) the reasoning seem faulty to me. The most important thing to keep in mind is that spellpower, crit, and haste all increase the value of each other as you gain more. Adding more spellpower makes haste more valuable. Adding haste makes spellpower more valuable - at least for a holy priest - as has been mentioned, disc priests are limited in the value they can extract from haste. There are definitely points at which a point of haste will be more valuable in terms of HPS than a point of spellpower for a holy priest.

Even with this in mind, I still support your conclusion that gemming for spellpower should be the preferred option. We only really get to have true freedom in choosing stats when we choose gems, since all available gear has a certain balance of stamina/intellect/spellpower, and then any two of spirit/crit/haste. When gemming, you can choose regen (i.e. intellect, possibly spirit), or throughput - so crit, haste, or spellpower. Of the three, only one increases the effectiveness of every part of every spell a priest casts - and thats spellpower.

You know what, you are right. I think I will remove that part of my original post and temporarily put something more vague there. I feel like this discussion really does need a comparison of haste and spell power so I really wanted to include at least something, even if flawed. However, I think what I have there now is just too flawed to really be of use, and is, like you say, misleading. I would ideally love to have something mathematical in there about the relative value of Spell Power, but I'm having trouble thinking of a way to come up with a more general comparison.

The idea that the three throughput stats are interdependent is something I should have been more explicit about.

And for your last point, perhaps the haste vs crit vs SP section should include a subsection specifically on gearing/gemming, stating the 5 possible gear combinations that Nicene posted earlier, as well as something on throughput versus longevity, and also your last point about the broad effectiveness of SP compared to crit and haste.

-------------

God damn it, Perl. The entire thing's in one single file that's almost 30,000 lines. A quarter of that is probably comments, but still. Yecchhhhh. You'd be better off starting from scratch.

Yeah, now that I look at the code maybe it would not be so simple to convert it for our uses. At least we know that the premise is sound and we can be justified in thinking this is a worthwhile approach.

-------------

We can definitely say things.

But compared to dps, which have a cycle and tend to spam it in most encounters (with some movement, or adds focus sometimes), we don't have it, and every encounter is a new theorycraft for us.

That doesn't mean that theorycraft has no value at all, but we shall be really careful with its assumptions before using it.

For example, Bobturkey's stats weighting and gear ranking is interesting, but because of its assumptions, I really prefer the "Take any iLVL increase" rule as holy priest, as long as the new gear doesn't have any hit (and if possible, MP5). I can find a fair use of all other stats. The distortion made by the strong assumptions (flash heal spam, MP5<->spellpower conversion) is in my mind greater than the difference in the final weighting.

Only exception is trinkets, where I use the difference to adapt to my current gear options and encounters. and where ilvl is sometimes meaningless (hello Solace).

I think we all agree that theorycrafting for healers is neither as simple nor as conclusive as theorycrafting for damage dealers. Like you said, everything varies by fight and assumptions like FH spam and mp5<->SP conversions lead to very questionable results, but to me that seems like the exact argument for a tool like shaman_hep; it will provide values based on the your exact spell usage for a specific fight, not based on a series of questionable assumptions like FH spamming.

Like a Richelieu said previously, a tool like shaman_hep would not help very much with deciding what is the best healing strategy or spell usage for a particular encounter, but it would, somewhat conclusively, say exactly how to improve your gear to optimize your performance for that encounter and how you play during it. And that, despite its limitations, is very valuable.

------------------

It is possible with some work to achieve a similar result for Rawr using the Custom Role option.

I had never taken that close of a look at the custom role option, but now that I do it does appear to be somewhat of a halfway step between purely theoretical, assumption based theorycrafting and the combat log parser I am imagining. Being able to specify the exact fight length and how many of each heal you cast is nice, but it does miss some of the things shaman_hep does.

For instance, the crit rates for each type of spell (how often did Improved Flash Heal result in an extra crit?); overheal percentages, both crit and non-crit; how many spells that were cast were haste capped (1 sec GCD); how many times did your target die while you were in the act of casting a heal on him/her;

One could imagine a priest version of shaman_hep to also include things like: what percentage of SoL procs were used, expired, or overwritten; what percentage of FH casts were effected by BT haste (and PWS casts, and GH casts, etc.);

Those are only the things I could think of as I write this, I'm sure there are many more interesting things we could glean from such a combat log parser. The custom role aspect of Rawr is a nice stepping stone, but I think the possibilities are much larger.

Share this post


Link to post
Share on other sites
I had never taken that close of a look at the custom role option, but now that I do it does appear to be somewhat of a halfway step between purely theoretical, assumption based theorycrafting and the combat log parser I am imagining. Being able to specify the exact fight length and how many of each heal you cast is nice, but it does miss some of the things shaman_hep does.

For instance, the crit rates for each type of spell (how often did Improved Flash Heal result in an extra crit?); overheal percentages, both crit and non-crit; how many spells that were cast were haste capped (1 sec GCD); how many times did your target die while you were in the act of casting a heal on him/her;

One could imagine a priest version of shaman_hep to also include things like: what percentage of SoL procs were used, expired, or overwritten; what percentage of FH casts were effected by BT haste (and PWS casts, and GH casts, etc.);

Those are only the things I could think of as I write this, I'm sure there are many more interesting things we could glean from such a combat log parser. The custom role aspect of Rawr is a nice stepping stone, but I think the possibilities are much larger.

How useful is it to know how many more crits imp. FH gave you? I also believe that Overhealing as long as you do not go OOM is not harmful, which means knowing overhealing is not that vital. Spells cast while Haste Capped could be interesting to know, but for the most time, they only happen during Heroism. How many people die when you are trying to heal them is not relevant, as thats pretty generic.

Overwritten SoL procs, BT hasted casts etc could be displayed easily. Don't know if it will change your gearing choices however. And this is perhaps better for a log parser than a theory tool.

The Custom Role is designed to let you know how to improve your Throughput and Sustainability given a specific fight you found challenging. Since we all know the game is highly random based, every fight will not end up being exactly the same, so I did try to avoid having you be too detailed.

Share this post


Link to post
Share on other sites
How useful is it to know how many more crits imp. FH gave you? I also believe that Overhealing as long as you do not go OOM is not harmful, which means knowing overhealing is not that vital. Spells cast while Haste Capped could be interesting to know, but for the most time, they only happen during Heroism. How many people die when you are trying to heal them is not relevant, as thats pretty generic.

Overwritten SoL procs, BT hasted casts etc could be displayed easily. Don't know if it will change your gearing choices however. And this is perhaps better for a log parser than a theory tool.

The Custom Role is designed to let you know how to improve your Throughput and Sustainability given a specific fight you found challenging. Since we all know the game is highly random based, every fight will not end up being exactly the same, so I did try to avoid having you be too detailed.

First of all, I apologize if my post came across as an attack on your work or anything like that, because that was not my intention. I think you have done a great job with the priest module and it's something that I personally use all the time.

Knowing the number of crits Improved Flash Heal adds is necessary to understanding the value of the talent. What if I find out that it's getting me only an extra couple of crits each night? I guess I can't really see myself dropping the talent, but I still like to know exactly what I'm getting.

I agree that mana is not a huge concern for most players right now, and therefore overhealing probably isn't either. However, it's still useful data. What if 75% of your heals are either partially or completely overhealing? Or 90% have at least some overhealing? That has big implications in the value of spell power; any heal that was overhealing to begin with will not benefit from more SP (or crit, ignoring DA).

Number of GCD spells cast while at < or = 1 sec GCD is probably only useful for holy priests during heroism, but is a pretty big concern for disc priests. The value of haste because of Borrowed Time haste is often debated here, and data like that would be very important in answering that question.

The number of times your target died while you were in the process of casting a heal seems pretty important and relevant to me. Many people like to make the argument that stacking haste is the best approach to gearing because it lessens the chance that your target will die just before your heal lands. The validity of that could be investigated by seeing just how often that kind of thing occurs, and comparing between players with different levels of haste.

Again, I did not mean any of that to be an indictment of your work on Rawr. I still find that program to be a powerful tool and one that I use frequently. I just think there is the potential for a lot more.

Share this post


Link to post
Share on other sites

On the topic of Haste preventing deaths. If 10% of deaths happen between 0.001 and 0.2s of you completing a cast, another 20% happens between 0.0201 and 0.04, and remaining happens with > 0.04 or no heal incoming at all, you'd think that getting your haste up enough to 0.2s less would be a good thing? Well, once you get your haste up that 0.2s, you suddenly see that between 0.001 and 0.2 suddenly 22% of deaths happen before you finish your heal. That just means your statistics for deaths caused by too late heals went worse.

Yes, hasted heals may keep someone alive from time to time, I'm just not sure its possible or feasible to model it.

And on the topic of overheals. If 75% of your heals are overhealing, that just means 25% were not. If you are worried about players going dead, those non-overhealing heals might have saved him by having him topped off if you had more spellpower to toss it up to an overheal. Same goes for crit. This is why I don't consider overhealing a problem at all as long as it doesn't hurt your ability to sustain healing. If you toss a heal, its because someone needed it.

And regarding imp. FH. There is a slider in Rawr that allows you to tell Rawr how many % of your heals land on people <=50%. You'd need to have a parser that knew peoples hitpoints and then give you the %. I cannot guesstimate this in a model, you do need to figure it out yourself and apply.

And Disc and GCD capping. Personally I feel its more important to look at your haste while not buffed with Heroism nor Power Infusion nor Borrowed Time. Thats the heals that may be sluggish and need attention. Being capped will diminish the value of haste for Instant spells. Spells with cast time however, will still land faster.

I did not take your post as an attack on my work on Rawr, I'm actually happy to try to explain my point of view on why I do things the way I do, and I'm open to suggestions on how to improve it. And you do not need to agree with me.

Share this post


Link to post
Share on other sites

Going back to what Squeakster and Enqi briefly hit on: "If your Penance is 1.5 second cast time you are running with a total of 33% haste. Borrowed Time will give you an additional 25% haste, making the new cast time 2/(1.33*1.25) = 1.2 seconds, a savings of 0.3 seconds.

Penance is a very powerful heal, and to cut off 0.3 seconds of its channeled cast time is invaluable. Especially when you start to consider lag, and the fact that even though the first tick is "instant' it still has a travel time similar to POM. Even if it takes 0.1-0.3 seconds for the first tick to occur with lag, which very well may be a large over exaggeration but for discussion sake, the first tick is still hitting before the GCD is up, making this the fastest and largest spell we have.

Like The Not So Evil had said, I'm slightly more concerned with haste in regards to not having BT up, especially while tank healing. With assumed raid buffs of Wrath of Air totem and Swift Retribution/Imp Moonkin, and Enlightenment, this brings penance down to 1.454 seconds with assumed gear having 20% haste. 2/(1.06)(1.05)(1.03)(1.20). At 30% haste from gear its further reduced to 1.342 seconds.

In regards to 'hasted heals keeping someone alive from time to time', the basis of healing is to keep people alive, but if we're thinking of haste with regards to crit/regen/sp, then buffs such as Grace, Inspiration, and DA also should be considered. Having the highest uptime we can on these spells directly improves throughput and mitigation.

Share this post


Link to post
Share on other sites

hello,

firstly, I'will grant you to be patient with my english (I'm no native speaker) and with my reflexions, this is my first post.

I think we need the math parts, in order to help us choose gears, gems and so on. I read with interest the haste topic, I had long discussions with a friend of mine playing discipline too and we do not agree about it (I'm currently 700 and he was tending to a lower cap).

But currently I'm a bit confused, part of the topic seems to me very far away from my real life in the game. Let me explain it : I raid quasi strictly 10-men raids, I'm the main healer of the guild, mostly I work hand-in-hand with a palaheal, and we can have some help as 3° healer by a shaman (when she's there) or a drood (the two of them dps in first specialisation). My guild, it's part of the real life, is mostly made of armor lovers (full of wars, and paladins) and we have a lot of dwarf hunters, so for example, we have down saurcroc with 3 wars (1 prot + 2 dps), 2 pal (1 prot + 1 heal), 1 dk, 1 mag, 2 hunters and me, what this means is that most of the time I do not have any of the usual raids buffs ... so discussing about BT with heroism is no help for me.

So I'm not the disci spamming shields on the raid in a 25-men, I'm not staying just focused on the tank, and I do not benefit from part of the "normal" raid buffs ...

I was tending to a 800 haste cap (cap decides from the reading of previous post), and now, after all of those maths, I don't know what I can or must do ! (perhaps it's due to my poor english and miscomprehension?)

For me this discussion is important, and as I'm not able to do the math part and if we - as priest- need further data I volunteer to implement a data saver in my log and to propose this to other priests in my realm.

Share this post


Link to post
Share on other sites

If you're not spamming shield, your haste cap will be a 50% haste, with raids buffs and enlightment. Recall that haste is multiplicative with gear and different effects (for example, 6% enlightment, 5% wrath of the air totem, and 10% from gear would be 1.06*1.05*1.1 = 1.2243 = 22.43% haste).

You don't need absolutely to be at that cap. But haste is one of the best throughput stats if you have enough mana (which you should).

Share this post


Link to post
Share on other sites
Overhealing as long as you do not go OOM is not harmful, which means knowing overhealing is not that vital.

...

The Custom Role is designed to let you know how to improve your Throughput and Sustainability given a specific fight you found challenging. Since we all know the game is highly random based, every fight will not end up being exactly the same, so I did try to avoid having you be too detailed.

While I agree, "Overhealing as long as you do not go OOM is not harmful," your conclusion that "knowing overhealing is not that vital," does not follow. Overhealing reduces the effectiveness of spellpower, so if you're trying to decide how to itemize, knowing your overhealing rate is vital. Consider the limit of very high spellpower. If every FH you cast heals for 100k, then getting enough spellpower to bump that up to 101k wouldn't make any difference in your throughput at all. This is a problem with Rawr, and indeed most healing theory crafting, that it tends to make the rather unrealistic assumption that you have an unlimited bucket of health to heal into.

Share this post


Link to post
Share on other sites

However problematic you think theorycrafting assumption about unlimited bucket of health is, your assumption is equally problematic. Right now no healer can fill up a tank with 1 heal. Thus there is possibility for more extra bonus healing to have its effect. You can't fight a non-extreme with an extreme.

Share this post


Link to post
Share on other sites
However problematic you think theorycrafting assumption about unlimited bucket of health is, your assumption is equally problematic. Right now no healer can fill up a tank with 1 heal. Thus there is possibility for more extra bonus healing to have its effect. You can't fight a non-extreme with an extreme.

Obviously not 100% of heals cast result in an overheal. I used the extreme example only to illustrate the problem with the assumption, not because I considered it a realistic situation. From my empirical observation, at least in ICC-10, the number is closer to 50%. I would welcome anyone else who can provided numbers to do so, since there are obviously methodological issues with mine, but it's the best I have at the moment. Note that the key number here is the percent of heals that result in an overheal (however small) and not the percent of overhealing.

With the no-overhealing assumption you are over-estimating the value of spellpower considerably. I'm not sure how much of a code change it would require, but it would be very nice if Rawr had a parameter like this.

Share this post


Link to post
Share on other sites

No matter how we deal with this we're going to end down the path of Heal and Damage granularity, which I want to avoid because there are too many variables to deal with it.

The reason I find that Haste and Crit is *less* worth than Spell Power even in your scenario with 50% overheal, is that this extra Haste may just cause another healer (who earlier landed the heal in front of you causing your overheal) to overheal.

And when it all comes down to individual stats, its Regen stats (Int/Spi/MP5) vs Throughput stats (SP/Hst/Crt). If you can sustain your current healing, more Regen stats is not needed. Crit is generally unreliable, which can be an issue. That leaves us with Haste and SP. More haste can lead to less overhealing or more overhealing, but it is a way to make sure you land a heal faster. SPP makes your heals hit harder.

So what is the value of haste?

(This is where I let the reader realize that it has to be up to ones individual style to make this final decision.)

Share this post


Link to post
Share on other sites
I really don't like the kind of argument saying "Haste increases the number of heals you cast, therefore your holy concentration uptime".

Technically, this is really true.

But the effect is marginal, and people shall not count on it. If you want to increase your holy concentration uptime, you gear for crit, and/or you change your spells selections (with short spells that can trigger it). You don't gem for haste.

The gain in holy concentration uptime is far from the loss of the increased number of cast.

Just for a quick numerical evaluation :

HC is an 8s buff that procs after any crit.

I'll take the best situation for haste : the charater currently has no haste, 30% crit, and can choose between 10% haste and 10% crit.

He maintains a cycle of 1 POM - 1 COH - Renew - Renew - SOL-FH, that takes roughly 8s (including lag).

In 8s, 2 spells can triggers HC.

The current downtime is 0.7^2 = 0.49. Uptime is 51%.

With 10% more crit, downtime is 0.6^2 = 0.36. Uptime is 64%. The gain is 13% uptime.

With 10% haste, the character can know cast 2.2 HC-friendly spells per 8s. Downtime is 0.7^2.2=0.456. Uptime is 54.4%. The gain is 3.4% uptime for 10% haste, in a really favorable scenario.

I'm not sure but I think I see something wrong with this. You can't apply the 10% haste to the amount of spells you cast but you have to apply it to the GCD and casting times then look at how many spells you can cast.

The problem is casting 2 HC procing spells in 8 seconds with 0 haste and then only casting 2.2 with 10% haste.

If you have 0 haste your GCD is 1.5. Let's say you cast a cycle of PoM, CoH, Renew, Renew, Flash Heal. It takes about 8 seconds to spam those 5 spells. (With your cycle, it should only take about 6 seconds since the last spell is instant, even with some lag)

If you gain 10% haste, your GCD is now 1.36. In that same 8 second window, you can cast 5.9 spells, we'll say 6 spells though since it really takes slightly less than a full 8 seconds to cast the original 5 spell cycle.

Adding 10% haste allowed a whole extra spell to be cast in the same window of time.

So adding 10% haste really should look like .7^3 = 0.343 or 34.3% downtime and a 65.7% uptime. That is a 14.7% increase over 0 haste.

If I'm completely off let me know, maybe you could do a few more examples and explain your math more in depth for me?

Share this post


Link to post
Share on other sites

  • Recently Browsing   0 members

    No registered users viewing this page.