Going over the past 10 pages or so, I see this topic brought up several times, but the only real "discussion" I see are statements alot the lines of "many mechanics have changed, and that's the result Rawr spits out now". No offense, but that is not a valid explanation. It doesn't do anyone any good to use a program if we can't understand WHY it gives the results that it does. So let's think about it, what mechanics specifically have changed that would make Str/AP and Crit/Agi so much more valuable per point than Hit/Expertise?
For Str/AP, the obvious changes are Divine Strength increased to 15%, the addition of Sheath of Light, and the changes to how seals/judgements scale with AP/SP. These are some pretty massive buffs to the scaling of the Str stat, so I think it's easily believable that 1 Str could be more DPS than 1 Hit, especially since the relationship between Str and Hit was never an obvious one to begin with, since Str is not a %DPS increase like Hit/Crit.
For Crit/Agi, the only major changes I can think of are the increased critical damage bonuses on Art of War and Righteous Vengeance. I really don't see how those talents could be nearly enough to make 1 Crit rating better than 1 Hit rating. Even if we overexaggerate the Crit bonus and pretend like it multiplies all crits by 2.5, then 1% Crit would only be an approximate 0.9% DPS increase (assuming 40% crit). 1% Hit should give more than 1% DPS increase, since Seals cannot proc if the original autoattack/CS/DS misses. Now throw in the fact that 1% Hit is only 15.7 rating while 1% Crit is 22.1 rating, and I really don't see how 1 Crit rating can beat 1 Hit rating.
Are there any other major mechanic changes that I'm forgetting, that could explain the results Rawr gives?
I can't find the discussion at the moment, but in either this thread or the generic pally WotLK thread Avitus has a two post gemming guide that covers much of the reasoning. Key points as I remember them are as follows. Much of the value of hit/expertise centered around keeping up other pallies' judgements. With that need removed they are of less value. Divine Strength is now +15% Str so it is more powerful again than it was. IIRC Str > Hit > Crit but those values can change based on other factors. At 80, Agi is a lot less powerful in comparison to crit than it was at 70. These conclusions have been posted based on calculations by Bellator and also by Avitus using Redcape's spreadsheet.
EDIT: My bad memory. Here is the link to the first gemming post (the second is on the next page):
Seal of Blood can crit, but I don't think it can miss.
Hm, there's something I wasn't aware of (as someone who never got to use Blood until recently). In essence, SoB damage should still be subject to Hit rating, since it "misses" if the original autoattack/CS/DS misses. But that would eliminate the extra scaling of Hit rating that you get with SoC effectively having 2 chances to miss.
What about JoB, can it miss?
Originally Posted by JettJaguar
I can't find the discussion at the moment, but in either this thread or the generic pally WotLK thread Avitus has a two post gemming guide that covers much of the reasoning. Key points as I remember them are as follows. Much of the value of hit/expertise centered around keeping up other pallies' judgements. With that need removed they are of less value. Divine Strength is now +15% Str so it is more powerful again than it was. IIRC Str > Hit > Crit but those values can change based on other factors. At 80, Agi is a lot less powerful in comparison to crit than it was at 70. These conclusions have been posted based on calculations by Bellator and also by Avitus using Redcape's spreadsheet.
Actually Avitus' posts about gemming (last post of p. 237 and first post of p. 238) were the first things I went back to check, since I'd immediately trust his in depth summary over my back of the napkin analysis. But what he had posted (which was before Rawr came out), was to always gem Str over Hit if there aren't juicy socket bonuses involved, but in cases where yellow sockets are beneficial, to gem Hit over Crit if you are under the hit cap. So this is consistent with what I had thought - that there are significant mechanic changes to explain Str being more valuable than Hit, but that I don't see how Crit can be more valuable than Hit. According to Rawr, Avitus' gemming summary posts are dead wrong, Rawr says that you should never gem hit, and just gem crit instead.
Going over the past 10 pages or so, I see this topic brought up several times, but the only real "discussion" I see are statements alot the lines of "many mechanics have changed, and that's the result Rawr spits out now". No offense, but that is not a valid explanation. It doesn't do anyone any good to use a program if we can't understand WHY it gives the results that it does. So let's think about it, what mechanics specifically have changed that would make Str/AP and Crit/Agi so much more valuable per point than Hit/Expertise?
For Str/AP, the obvious changes are Divine Strength increased to 15%, the addition of Sheath of Light, and the changes to how seals/judgements scale with AP/SP. These are some pretty massive buffs to the scaling of the Str stat, so I think it's easily believable that 1 Str could be more DPS than 1 Hit, especially since the relationship between Str and Hit was never an obvious one to begin with, since Str is not a %DPS increase like Hit/Crit.
For Crit/Agi, the only major changes I can think of are the increased critical damage bonuses on Art of War and Righteous Vengeance. I really don't see how those talents could be nearly enough to make 1 Crit rating better than 1 Hit rating. Even if we overexaggerate the Crit bonus and pretend like it multiplies all crits by 2.5, then 1% Crit would only be an approximate 0.9% DPS increase (assuming 40% crit). 1% Hit should give more than 1% DPS increase, since Seals cannot proc if the original autoattack/CS/DS misses. Now throw in the fact that 1% Hit is only 15.7 rating while 1% Crit is 22.1 rating, and I really don't see how 1 Crit rating can beat 1 Hit rating.
Are there any other major mechanic changes that I'm forgetting, that could explain the results Rawr gives?
This is exactly what I've been saying. Rawr is giving me a lot of different numbers to which I don't agree, specifically the ones concerning hit. I'm certain there are some numbers that are completely off somewhere in there, because hit should always be a bigger increase in damage than crit if you're not capped yet.
It's also worth noting that in your arguments you mentioned that crit might benefit from the damage bonuses on Art of War and Righteous Vengeance, and both of them are being changed to not affect crits anymore.
Originally Posted by JettJaguar
I can't find the discussion at the moment, but in either this thread or the generic pally WotLK thread Avitus has a two post gemming guide that covers much of the reasoning. Key points as I remember them are as follows. Much of the value of hit/expertise centered around keeping up other pallies' judgements. With that need removed they are of less value.
[snip]
This bit is actually wrong. Hit value wasn't really about keeping other pallies' judgements (although it certainly helped). Hit (and for that matter, expertise) is important because you cannot land a seal if you miss your auto attack. Avituus posted the gemming guide in which it clearly describes that it's mostly better when you gem for Strength and Hit instead of Strength and Crit if the socket bonus is a strength or hit bonus. There's also exceptions, but those are rare and go on a case by case basis.
In my opinion, your rankings for stats were right (I'll just add expertise there):
Str > Hit/Expertise >> Crit
Here's Avituus's post. This one and the previous post explain it thoroughly.
It's also worth noting that in your arguments you mentioned that crit might benefit from the damage bonuses on Art of War and Righteous Vengeance, and both of them are being changed to not affect crits anymore.
This is true of course, but I think that the main goal for now should be checking whether Rawr is making accurate calculations given the assumptions that it is using (which for now, is AoW/RV as crit damage bonuses). Also, even after they do get changed, with RV being a deep wounds type ability that relies heavily on Crits, we may still see around the same relative scaling value for Crit.
By the way, while I'm thinking of it, does anyone know if the newest version of Rawr has Heroic Presence (1% Hit Shaman racial)? I don't remember seeing it on there, but I'm at work now and I'm not sure if I had the most recent Rawr version at home. If it's not on there, can it be added?
On a slightly different note, how does moving DS from holy to physical damage effect ArP as a pally stat? It would seem to be a pretty major shift and I wonder how it now plays into the mix of stats.
EDIT: Would this also effect rotation priorities since individual CS attacks now deal 10% more damage than DS attacks?
According to Rawr, Avitus' gemming summary posts are dead wrong, Rawr says that you should never gem hit, and just gem crit instead.
Unfortunately the case right now is Spreadsheet vs Rawr (vs reality) and it's impossible to know which is more accurate. We need everyone on the case trying to zero in on the actual differences and trying to figure out what's being calculated differently.
I do agree with you however, it's hard to logically explain why crit would outdo hit in Rawr, which is why the spreadsheet seems more accurate. One thing I do know is that the Spreadsheet uses mechanics from 1 build prior to the build Rawr uses, meaning they never were going to give the same results anyway. That said, a few things have changed again since the release of 3.0 (i.e. including JotM doing less than JoB for some reason).
I would strongly suggest that both Rawr and the Spreadsheet plan to make new releases as soon the next build hits (where righteous vengeance changes), this time with unified assumptions and coefficients, at which point it will finally be possible to conclude that the discrepancies are due to some bug in one (or both) of the tools, rather than not knowing whether it's due to using different wow builds to base the math off.
Originally Posted by bellator
When all our abilities settle down, i'll try to get it updated, but takes a while and not a lot of time on my hands atm.
I know exactly where you're coming from Part of the reason why I've not really started analyzing numbers during the first part of beta. However it seems we're getting closer to that day where we can sit down and actually calculate things without having to worry whether they're going to change everything the next day.
At the moment, the outstanding issues I can think of are:
-Sheath, what exactly is it supposed to affect and what not? Is that intended?
-Revising multipliers after RV/AoW changes.
-Alliance vs Horde JoB ("JotM"), which is the bug?
-JoC judgement coefficients intended or a bug?
Originally Posted by JettJaguar
On a slightly different note, how does moving DS from holy to physical damage effect ArP as a pally stat? It would seem to be a pretty major shift and I wonder how it now plays into the mix of stats.
EDIT: Would this also effect rotation priorities since individual CS attacks now deal 10% more damage than DS attacks?
DS was always firmly below CS in a higher DPS first clash priority. At 10 sec cooldown DS was already barely above Consecration even and becoming physical damage does not help its case.
It's unknown what exactly will apply the new Righteous Vengeance DoT, I'm assuming any white, cs, ds crit, but it could also be judgement crits. Judgement priority will depend on this fact.
ArP will obviously become more valuable now that it works with one additional attack, how valuable in comparison to other stats is yet to be determined.
Going back to the the whole Hit vs Crit issue, the AoW change should push down the worth of Crit somewhat, while the changes due to the new RV mechanic are unknown. It will depend one whether the DoT is applied to Judgements and how much damage it will do, whether it has a rolling effect, before we can conclude whether the loss of the old RV "+crit damage" multiplier is a gain in overall worth of the crit stat, a loss or no change. Too many unknowns. Have I mentioned how much I hate unknowns? :/
On a slightly different note, how does moving DS from holy to physical damage effect ArP as a pally stat? It would seem to be a pretty major shift and I wonder how it now plays into the mix of stats.
EDIT: Would this also effect rotation priorities since individual CS attacks now deal 10% more damage than DS attacks?
From all the logs I've been accumulating, 50% of our damage (CS, DS, Swing) benefits from ArP and approx 40% (Swing, most seals) from haste. I've found Command to be the best seal over the last few days but it also makes it harder to estimate how many are proc'ing of which abilities.
I do agree with you however, it's hard to logically explain why crit would outdo hit in Rawr, which is why the spreadsheet seems more accurate. One thing I do know is that the Spreadsheet uses mechanics from 1 build prior to the build Rawr uses, meaning they never were going to give the same results anyway. That said, a few things have changed again since the release of 3.0 (i.e. including JotM doing less than JoB for some reason).
I would strongly suggest that both Rawr and the Spreadsheet plan to make new releases as soon the next build hits (where righteous vengeance changes), this time with unified assumptions and coefficients, at which point it will finally be possible to conclude that the discrepancies are due to some bug in one (or both) of the tools, rather than not knowing whether it's due to using different wow builds to base the math off.
A next version of the rawr module is being worked on. Until I get solid numbers, however, I can't do much.
The hit versus crit thing is a different story. A lot of classes are noticing now that maxing hit is no longer the best course of action, and that secondary stats are doing better. I have no clue why the spreadsheet isn't showing what Rawr is, but the beauty of Wowhead is that it actually gives the calculations for almost all of the ret paladin abilities... just plug in values and VOILA! The answer, as I see it, will likely lie in how rotations are modeled, and the fact that I ignore HoW use in my model atm. And when it comes to rotations, due to the way ret pallies are now, I need numbers from you guys. Something like..."In this 3 minute 45 second brutallus I used W hammer of wraths, X crusader strikes, Y divine storms, and Z judgements."
Also keep in mind on live, white and seal damage was easily 65% of your total damage. Now that they are a lot lower, maybe 45% (if you count only the seal damage due to white hits), hit has a lower value than crit. Also, judgement, which hits like an absolute truck when it crits, get nothing from hit. It is by far and away our most powerful ability now. Now CS and DS also get more crit damage. Crit was also always good for us, but now it just obliterates all other stats other than STR/AP. The fact that the spreadsheet still values hit higher has me thinking that IT, in fact, is the inaccurate option. Of course, I may be biased since I code the Rawr module... but these reasons alone seem like a logical reason for them to work this way. It seems to me that people are trying to fit a square peg into a round hole; dump your preconceived notions of mechanics, WOTLK changed EVERYTHING. I had to re-write no less than 90% of the main calculation page of the Rawr module for the 3.0 patch, so just keep that in mind.
The only thing on my plate atm is to activate LOTP in buffs and fix up the racials. And also make DS physical damage.
Last edited by Zurm : 10/17/08 at 2:50 PM.
Formally Xyrm/Zurm, the Ret Pally. Now playing my rogue, Zyrm, more casually with RL friends.
I understand that maxing hit is not an absolute necessity anymore, but it's still a very valuable stat. How valuable is the question: More than crit or less?
Anyway, I'm leaning towards ignoring any differences in results until the next versions of both spreadsheet and rawr are released and actually go off the same build, then we can start narrowing down on what's correct or wrong. It's getting too messy to keep up with and try to explain everything, when some of the core coefficients used are different.
Compared to the other changes that are happening to Ret, this is a fairly minor issue, but does anyone else understand why the glyphs for some of our primary damage abilities deal with mana efficiency? [Glyph of Holy Wrath], and now we are told that [Glyph of Crusader Strike] is being changed to reduced mana cost. Honestly, it was worthless for PvE before this change, but the new version is just as useless. Perhaps I'm missing something, but I can't seem to run out of mana even when using Exorcism, Holy Wrath, etc. Why the emphasis on efficiency that, as best as I can tell, isn't necessary?
-Alliance vs Horde JoB ("JotM"), which is the bug?
-JoC judgement coefficients intended or a bug?
Going back to the the whole Hit vs Crit issue, the AoW change should push down the worth of Crit somewhat, while the changes due to the new RV mechanic are unknown. It will depend one whether the DoT is applied to Judgements and how much damage it will do, whether it has a rolling effect, before we can conclude whether the loss of the old RV "+crit damage" multiplier is a gain in overall worth of the crit stat, a loss or no change. Too many unknowns. Have I mentioned how much I hate unknowns? :/
The answer to the first (which JoB is correct, 25% or 45%) will help answer the second, since both Seals should be close.
Crit will go down in value once RV is in. I assume RV will work like Deep Wounds, except it will do Holy damage instead of Bleed damage.
Originally Posted by Siawn
Why the emphasis on efficiency that, as best as I can tell, isn't necessary?
My guess is the devs like some things that are just bad. Hey, it is another way to tell if a player isn't very good at their class.
Millions of words are written annually purporting to tell how to beat the races, whereas the best possible advice on the subject is found in the three monosyllables: 'Do not try.'
Compared to the other changes that are happening to Ret, this is a fairly minor issue, but does anyone else understand why the glyphs for some of our primary damage abilities deal with mana efficiency? [Glyph of Holy Wrath], and now we are told that [Glyph of Crusader Strike] is being changed to reduced mana cost. Honestly, it was worthless for PvE before this change, but the new version is just as useless. Perhaps I'm missing something, but I can't seem to run out of mana even when using Exorcism, Holy Wrath, etc. Why the emphasis on efficiency that, as best as I can tell, isn't necessary?
This is a shot in the dark, but the only time I can think of these being useful is if blizzard becomes really fond of mana drain/burn mechanics in the raids in WoLK. You can also look at it in the light that there are also very valuable enchants and some very wtf enchants currently.
I think Blizzard said that they aren't done making glyphs yet. Glyphs have already gotten changed, I'm sure they will be making more changes in the future as well.
To be honest, I am really hoping they put the judgment refreshing mechanic into a glyph of crusader strike glyph.
Last edited by Saltycracker : 10/17/08 at 3:23 PM.
Crit will go down in value once RV is in. I assume RV will work like Deep Wounds, except it will do Holy damage instead of Bleed damage.
Deep Wounds is triggered by crits however. If this is the case for the RV dot, it's impossible to say that the worth of Crit will go down without knowing all details. We can keep assuming things, but we're really going no where.
AoW is definitely a reduction in the worth of Crit however.
<snip>
the beauty of Wowhead is that it actually gives the calculations for almost all of the ret paladin abilities... just plug in values and VOILA!
<snip>
I agree, that's wonderful from Wowhead and I'm using it myself to crunch numbers. But Wowhead is either multiple builds behind or not updating this - for instance it has the same stats for Blood and Martyr, which we know don't match. Having the correct numbers at our fingertips will make it easier to generate correct results.
We're mostly waiting for numbers to stabilize (they keep changing on us) and new Righteous Vengeance with bated breath.
I agree, that's wonderful from Wowhead and I'm using it myself to crunch numbers. But Wowhead is either multiple builds behind or not updating this - for instance it has the same stats for Blood and Martyr, which we know don't match. Having the correct numbers at our fingertips will make it easier to generate correct results.
I checked Wowhead yesterday and it had the difference in formulas for judgements for Blood and Martyr (perhaps it was the wotlk site, but it was clear).
Millions of words are written annually purporting to tell how to beat the races, whereas the best possible advice on the subject is found in the three monosyllables: 'Do not try.'
I'm clearly seeing .28 on both of them. I just tried clearing my cache and viewing again - no change.
Edit: And that's a big whoops for me. I had been misreading the thread and thought the seal not the judgement changed, so was totally focused on wrong numbers. Thanks frmorrison!
I am referring to Judgements, the Seal damage is the same and testing confirms. I see .45 MW for Blood and .25 MW for Martyr, when the spells are supposed to be the same.
I did test them (on pre-mades) and got different judgement numbers as well.
Millions of words are written annually purporting to tell how to beat the races, whereas the best possible advice on the subject is found in the three monosyllables: 'Do not try.'
A next version of the rawr module is being worked on. Until I get solid numbers, however, I can't do much.
The hit versus crit thing is a different story. A lot of classes are noticing now that maxing hit is no longer the best course of action, and that secondary stats are doing better. I have no clue why the spreadsheet isn't showing what Rawr is, but the beauty of Wowhead is that it actually gives the calculations for almost all of the ret paladin abilities... just plug in values and VOILA! The answer, as I see it, will likely lie in how rotations are modeled, and the fact that I ignore HoW use in my model atm. And when it comes to rotations, due to the way ret pallies are now, I need numbers from you guys. Something like..."In this 3 minute 45 second brutallus I used W hammer of wraths, X crusader strikes, Y divine storms, and Z judgements."
Also keep in mind on live, white and seal damage was easily 65% of your total damage. Now that they are a lot lower, maybe 45% (if you count only the seal damage due to white hits), hit has a lower value than crit. Also, judgement, which hits like an absolute truck when it crits, get nothing from hit. It is by far and away our most powerful ability now. Now CS and DS also get more crit damage. Crit was also always good for us, but now it just obliterates all other stats other than STR/AP. The fact that the spreadsheet still values hit higher has me thinking that IT, in fact, is the inaccurate option. Of course, I may be biased since I code the Rawr module... but these reasons alone seem like a logical reason for them to work this way. It seems to me that people are trying to fit a square peg into a round hole; dump your preconceived notions of mechanics, WOTLK changed EVERYTHING. I had to re-write no less than 90% of the main calculation page of the Rawr module for the 3.0 patch, so just keep that in mind.
The only thing on my plate atm is to activate LOTP in buffs and fix up the racials. And also make DS physical damage.
On Judgement damage... Judgement (I think) can miss, and does get benefit from hit. Judgements are based off the ranged attack table and hence cannot be blocked/dodged or parried but can definitely miss. Please correct me if I'm wrong. This would definitely make it more susceptible to hit.
Some stats are better than hit. This is certainly the case, as Strength is now much better than hit. It, however, does not explain why Rawr considers crit as a better choice than hit. It doesn't make any sense. Tarja made some good arguments (16 hit vs 21 crit for 1% for instance) and all you say is that you trust Rawr. I understand that you're confident in your programming skills, I think we all are, but there might be a number that's wrong somewhere. I just think that the changes were so big that it'd be worth it to go line by line through the code and see what's going wrong. I can help if you need an extra hand.
I've never seen a judgement miss or get fully resisted... maybe it's because I've been at hit cap forever but I'm pretty sure judgements are guaranteed to land.
I also explain WHY I feel crit should be better:
Also keep in mind on live, white and seal damage was easily 65% of your total damage. Now that they are a lot lower, maybe 45% (if you count only the seal damage due to white hits), hit has a lower value than crit. Also, judgement, which hits like an absolute truck when it crits, get nothing from hit. It is by far and away our most powerful ability now. Now CS and DS also get more crit damage.
I'm not tooting my own horn, Avitus even mentioned that Rawr was based off a more recent version than the spreadsheet.
Formally Xyrm/Zurm, the Ret Pally. Now playing my rogue, Zyrm, more casually with RL friends.
Pre-3.0, Command and Righteous Judgements could miss, however Judgement of Blood, Light, Wisdom, Justice never could miss.
Post-3.0, Judgements cannot miss, since you are using Judgement of Light/Wisdom/Justice, which since the Pally patch cannot miss.
Millions of words are written annually purporting to tell how to beat the races, whereas the best possible advice on the subject is found in the three monosyllables: 'Do not try.'
I've never seen a judgement miss or get fully resisted... maybe it's because I've been at hit cap forever but I'm pretty sure judgements are guaranteed to land.
I also explain WHY I feel crit should be better:
I'm not tooting my own horn, Avitus even mentioned that Rawr was based off a more recent version than the spreadsheet.
Judgements definitely miss. I am under hit cap and have seen it on multiple occasions.
According to Rawr, Avitus' gemming summary posts are dead wrong, Rawr says that you should never gem hit, and just gem crit instead.
That is the big thing that was throwing me for a loop, even in my completely sleep deprived state. Despite being under hitcap by huge amounts, Rawr kept reporting to me that changing the 5str 5crit gem in my shoulders to 10 hit was a loss of dps, it also told me that changing the 5str 5crit to 10 hit in my legs (which I did anyways) was also a loss of dps, despite the fact the gear I was experimenting with I was 3% off of hit cap. The gear was upgrades, but not substantial enough upgrades (compared to Rawr pre-3.0) to make that much of a dps difference as it does now.
Basically, I know hit is of less value then it was before and I knew that while messing around with Rawr, but is the extent that Rawr is reporting it to be less important correct?
Thank you for any and all discussion :) I have read over the past few pages just brushing up on the new mechanics but this one in particular I didn't see an explanation through enough to make me stop worrying about it. A lot of the new conversations sparked from it have been helping that a lot, so I really appreciate it! The question is getting answered now.
Judgements definitely miss. I am under hit cap and have seen it on multiple occasions.
Using alliance seals... pre-3.0.
Either way, this weekend Avitus will look over my math in Rawr. I think everyone on this board can agree he is well enough knowledged in the mechanics to give a good evaluation of my model.
Formally Xyrm/Zurm, the Ret Pally. Now playing my rogue, Zyrm, more casually with RL friends.
10/17/08, 12:15 PM
Part of the reason why I've not really started analyzing numbers during the first part of beta. However it seems we're getting closer to that day where we can sit down and actually calculate things without having to worry whether they're going to change everything the next day.
