This site makes extensive use of JavaScript.
Please enable JavaScript in your browser.
Live
PTR
10.2.7
PTR
10.2.6
Beta
Seal of Wisdom and you.
Post Reply
Return to board index
Post by
pezz
I looked around Wowhead, Wowwiki, and Elitist Jerks for some comprehensive math on the mana returns on this seal, but didn't find very much. I've done just a couple two minute tests to get a rough idea of how it works (i.e., ICD, PPM, etc.).
I had five different set ups. All of them were two minute tests and included 100% uptime on JotP. After that there were two ways they were broken up, with one modification to one additional test.
I first broke things up by weapon. Happily, I have a
pretty speedy weapon
and a
much slower one
just from my own gear sets, so two tests each used these two weapons.
The second way I broke up testing was by haste. Combing my best +haste gear from my holy and ret sets gives me 515 haste rating, also known as 20.42% melee haste rating. I wore this set for the speedy tests, and was naked except for my weapons (which unfortunately both have some haste on them) for my slow tests.
One other little quirk: On my speedy weapon and max haste test I also judged on CD.
The numbers:
For my fully hasted, judging on CD, fast weapon test I had:
36 SoW procs, 101 melee attacks with 7 failures (i.e., dodges and misses*), and 12 successful judgments. Base attack speed of 1.60, eventually modified to 1.18 actual attack speed. Also note that my holy spec does NOT include Imp. Judgments, which is why it's 12 judgments and not whatever it would be otherwise.
36 procs x 4% total mana = 144% total mana refunded over two minutes. 120/5 = 24 five second increments in two minutes. 144%/24 = 6%tmp5 (six percent total mana per five seconds).
For my fully hasted fast weapon, judging just to keep JotP up:
38 SoW procs, 101 melee attacks with no failures and two judgments with no failures. Base attack speed of 1.60, eventually modified to 1.18 actual attack speed.
38 x 4 = 152/24 = 6.3 repeating%tmp5
Next we have the low haste, fast weapon test, judging only as often as necessary to keep JotP up.
34 procs, 87 melee attacks with 9 failures, two successful judgments. 1.60 base attack speed, eventually modified to 1.37 actual attack speed.
34 x 4 = 136, 136/24 = 5.6 repeating%tmp5
Now fully hasted slow weapon, only judging for JotP
41 procs, 48 melee attacks with 5 failures, two successful judgments. 3.50 base attack speed, eventually modified to 2.52 actual attack speed.
TL;DR: 6.83 with a repeating three % tmp5
This seemed a bit high, so I ran another two minute test on this one.
36 procs, 48 melee attacks with 8 failures, two successful judgments.
TL;DR: 6%tmp5
And finally my slow weapon with as little haste as possible, again, only barely keeping JotP up
34 procs, 41 melee attacks with four failures, two successful judgments. 3.50 base attack speed, eventually modified to 2.94 actual attack speed.
TL;DR: 5.6%tmp5
*Note: I included glancing blows as successes. I'm reasonably sure for my purposes that they are.
So, what do we make of it.
For one, I'm reasonably sure there's no ICD. Judgments and melee attacks that landed just about simultaneously both procced SoW, and in the comments section for the spell itself here on Wowhead there was a claim of multiple hits of the same DS proccing multiple occasions of SoW.
Secondly, I am just about positive that there's a PPM. What it is exactly is hard to say. I'd eye ball it around 36 to 38 roughly. I don't feel like the exact number is terribly important. I was only interested to see if it exists because I wanted to see whether a faster weapon would be worth an amount of mp5. Interestingly, it seems to work on both melee and judgments combined, since judging on CD did not increase the PPM over judging twice to keep JotP up.
So why, then, did you get 41 procs on your slow weapon low haste test?
On a spell with a PPM but no ICD, slower weapons have a much higher chance on each individual hit to proc an effect. My guess is I got an extra hit while running in before I started the stopwatch and an extra hit running out after I'd stopped it. I was having to somehow hit the in game stop watch, judgment, and run in all at the same time, so such an error is entirely plausible and would account for the discrepancy observed.
So there you go! SoW can be a valuable tool to keep your mana up, but it seems like healers still don't have to worry about weapon speed.
Post by
svirve
I'm sorry to say this but, 2 minute tests are about 8 minutes too short for anything remotely accurate, and 58 minutes short of something definite.
That you're judging as well kinda screws up any PPM numbers since you don't know the proc chance of judgement.
That there's no ICD is a fact I thought to be well known, in your Holy guide i led a discussion on SoW/JoW proccing up to 8 times (4 times each) on a single judgement.
The PPM is ~15 which adds up with your numbers. You had an average PPM of 18.5 throughout your tests, and as i mentioned, they were very short. P.S. There used to be a thread on EJ about the PPM of SoW. It's gone now though.
Glancing blows can proc effects so you were right to assume them as succesfull hits =)
You know I respect you and honestly I'm not trying to bash you here.
I get what you're trying to do here, but this is only really applicable when you're farming or doing heroics. Cause in raids, and especially progression raids, you wont be hitting anything for any time worth mentioning.
What i would like to see on almost the same subject though is how weapon speed affects judgement proc chance. Cause that's something you will do no matter what difficulty raid.
Seeing as how judgement has a 10(8) sec CD any test would be 2+ hours demanding that you sit by the computer and judge every 10(8) sec.
That's something at least I would find extremely applicable and useful.
Once again, I'm not trying to bash you here.
Post by
Squishalot
Judgement is classified as a melee attack and would also proc SoW in line with melee attacks.
Personally, I'd just use a haste set /w slow weapon and haste set /w fast weapon, and ignore judgements. To test ICD, use Ret with judgements, CS and DS on cooldowns, but again, it doesn't tell you anything unless you trawl through the combat log and look at the minimum gap between procs.
Post by
335643
This post was from a user who has deleted their account.
Post by
Quest
You can make an autoit script that presses a button over and over and leave it for an house at a test dummy.
BUT THEN YOUD GET SUED BY BLIZZARD FOR COPYWRITE ISSUES BECAUSE ITS LOADED INTO RAM ETC DONT DO IT MAN TURN BACK WHILE YOURE CLEANqnzs.
Post by
svirve
Im Going to have to disagree with you here Svirve. Though I havent tried Professor or the Blood Council yet, I've been able to stand and melee swing on almost every boss for the entire fight in ICC while healing.
Even if you can stand at melee range, bet is you're not going to get away that many swings, spare GCDs are few and even with a 1.7 speed weapon each spellcast will reset the swing timer.
The % total mana return that pezz listed is when no melee swing gets left out.
If you use recount or record combat logs for WWS (or similiar) look at how many swings you can average per minute.
My bet is you'll have a hard time reaching even 15 swings per minute.
Post by
25071
This post was from a user who has deleted their account.
Post by
Squishalot
My bet is you'll have a hard time reaching even 15 swings per minute.
You do realise that 15 swings per minute = greater mana recovery than Divine Plea?
At 6%tmp5, that's 6% mana (or 2400 mana on 40k) for 5 seconds of non-casttime work. So every time you cast Beacon, Judgement, Holy Shock, and infused Flash of Light, you'll get a melee swing or two.
Potentially, you'll get better mana returns from a faster swinging weapon, because it's less likely to be cut-off before you get your melee swing in.
Post by
pezz
I'm sorry to say this but, 2 minute tests are about 8 minutes too short for anything remotely accurate, and 58 minutes short of something definite.
Oh I'm aware of that. I wasn't going for anything like exact numbers. I just wanted to set my mind at rest about very rough estimates on PPM and ICD, and I couldn't find such numbers with an easy search. Based on what else I've seen and what I've found there isn't an ICD, or at least not one that is going to affect anything a holy paladin can manage to do, and while I wouldn't hazard a definitive guess at the PPM, I think the very rough numbers I've found combined with common sense and a knowledge of how these mechanics usually work is enough to satisfy me that I can ignore weapon speed when selecting my weapon.
I'm also not at all trying to get an exact amount of %tmp5, that was more just a curiosity. As long as it's more than zero and doesn't affect my healing I'm happy with it.
I'm also not offended at all, I do value your input, and you make a valid point.
I get what you're trying to do here, but this is only really applicable when you're farming or doing heroics. Cause in raids, and especially progression raids, you wont be hitting anything for any time worth mentioning.
I'm not sure I agree with this part though. The key to this being a useful source of mana is the fact that instant casts do not affect the swing timer. Hands of Freedom, Salvation, Protection, and Sacrifice, Aura Mastery, DiSac, Cleanse, Bubble, Lay on Hands, SS, Beacon, possible aura dancing, Divine Plea, Judgment, Divine Illumination, Hammer, and GotN if you're a Draenei are all useful spells in a paladin's arsenal, none of which are going to reset your swing timer. Further, any of these could be followed up immediately by an HS cast, which won't reset the swing timer, and may give an FoL proc which also won't reset the swing timer. Considering haste effects your melee swing timer on a steeper curve than it effects the GCD, and considering you'd never actually use a two handed weapon as a holy paladin, there's definitely room in there to get a couple melee attacks off every once in a while. You're not going to make 100% use of the potential PPM, I realize, but you can definitely find some times when you can (and should) string a couple instant casts together.
Post by
svirve
Even if you can stand at melee range, bet is you're not going to get away that many swings, spare GCDs are few and even with a 1.7 speed weapon each spellcast will reset the swing timer.
At 40k mana raid buffed a single swing is 1600 mana, which is almost 2 holy lights. It's nothing to be sneezed at.
It might be added that no holy paladin is running without at least a fair amount of haste on their gear, bringing down the swing timer considerably.
I've even gotten a few melee swings off at Anub heroic, keeping my mana full until p3 hits, although that was more the exception than the rule. I'm just saying that standing in melee "attacking" the boss while healing is certainly not anything to be neglected as a regen tool for paladins.
Just to be picky, 1600 mana is not even 1.5 HLs, so no it's not almost 2 holy lights. Though I agree it's nothing to ignore.
The poster i was replying to mentioned meleeing a whole fight.
What i said earlier was:
I get what you're trying to do here, but this is only really applicable when you're farming or doing heroics. Cause in raids, and especially progression raids,
you wont be hitting anything for any time worth mentioning
.
Which is basically what you're saying as well, you can melee swing sometimes but you do not do it much at all.
And seeing as the test performed was with swinging constantly, it's not applicable to a raid were you can swing very very little.
My bet is you'll have a hard time reaching even 15 swings per minute.
You do realise that 15 swings per minute = greater mana recovery than Divine Plea?
At 6%tmp5, that's 6% mana (or 2400 mana on 40k) for 5 seconds of non-casttime work. So every time you cast Beacon, Judgement, Holy Shock, and infused Flash of Light, you'll get a melee swing or two.
Potentially, you'll get better mana returns from a faster swinging weapon, because it's less likely to be cut-off before you get your melee swing in.
That was actually a shot from the hip, 15 sounded like a fairly reasonable number with a ~1.7 speed weapon and accounting for the haste as well.
The 6%tmp5 number is when swinging constantly yes, 15 hits over a fight would land you 5.6 returns for a total gain of 22.4% of your total mana, so it's slightly less than DP. That was using 1.5 speed because of haste and all.
This is why i'm saying this is not applicable to raids, you will never, ever see 6%tmp5 during a raid unless either of the following applies: switch of phase and the boss is still vulnerable to attacks, a pause in a phase were there is little to none damage and you have something to hit or you're doing naxx. The first two won't last for more than ~15 seconds at the most, and the last one, well...
Oh I'm aware of that. I wasn't going for anything like exact numbers. I just wanted to set my mind at rest about very rough estimates on PPM and ICD, and I couldn't find such numbers with an easy search. Based on what else I've seen and what I've found there isn't an ICD, or at least not one that is going to affect anything a holy paladin can manage to do, and while I wouldn't hazard a definitive guess at the PPM, I think the very rough numbers I've found combined with common sense and a knowledge of how these mechanics usually work is enough to satisfy me that I can ignore weapon speed when selecting my weapon.
I'll accept that, well, i basically have to since that's what i did throughout the protribution tests. ;p
Once i get around to finishing my Advanced Holydin Compendium, i'll make sure to include the PPM and the fact that there is no ICD on SoW.
I'm also not at all trying to get an exact amount of %tmp5, that was more just a curiosity. As long as it's more than zero and doesn't affect my healing I'm happy with it.
Ah i see, misunderstanding on my part then :)
I'm also not offended at all, I do value your input, and you make a valid point.
Glad to hear =)
I'm not sure I agree with this part though. The key to this being a useful source of mana is the fact that instant casts do not affect the swing timer. Hands of Freedom, Salvation, Protection, and Sacrifice, Aura Mastery, DiSac, Cleanse, Bubble, Lay on Hands, SS, Beacon, possible aura dancing, Divine Plea, Judgment, Divine Illumination, Hammer, and GotN if you're a Draenei are all useful spells in a paladin's arsenal, none of which are going to reset your swing timer. Further, any of these could be followed up immediately by an HS cast, which won't reset the swing timer, and may give an FoL proc which also won't reset the swing timer. Considering haste effects your melee swing timer on a steeper curve than it effects the GCD, and considering you'd never actually use a two handed weapon as a holy paladin, there's definitely room in there to get a couple melee attacks off every once in a while. You're not going to make 100% use of the potential PPM, I realize, but you can definitely find some times when you can (and should) string a couple instant casts together.
I know it's kickass when casting instant spells but my point was that the tests you did and the facts you provided is not applicable in raids, raid numbers wont even come close to those you presented. SoW is an awesome extra source of mana regen, but it is
VERY
situational.
Just to make it clear, everyone should be meleeing for SoW returns when possible. It is awesome, it really is.
It is cool to see how much it can really return, the only reason i'm being critical is because it has no chance of becoming even remotely close to realistic.
Post by
pezz
I think we might have misunderstood each other. My point was not that you can come even close to 40 some hits per minute, and your point was not that you'll never melee. I can agree that your total melee swings aren't going to come close to what I got in my tests, but I do think a paladin that finds times to use all of his or her various utility spells is going to have a surprising number of chances to squeeze in a melee hit here and there. I think that's what we're both saying, just with different language.
Potentially, you'll get better mana returns from a faster swinging weapon, because it's less likely to be cut-off before you get your melee swing in.
I doubt it. Haste speeds up your melee swings more than it speeds up your spell casting. Since it's usually going to be reasonably appropriate to put HS in with whatever other instant cast you do, that's two GCDs+latency time minimum. Since you'll never equip a two hander your attack speed is probably going to top out at 2.70 completely unbuffed with haste. That's already faster than two GCDs unbuffed with haste, and the more of that you get the faster your melee swings are going to be compared to your GCD, especially if you're at the soft-cap.
Also, if your regular raid has Frost Death Knights or Enhancement Shamans, they are going to kind of help your swing timer blow your GCDs out of the water.
Edit: wait, never mind. I'm thinking of the fact that you're guaranteed a hit or two even with a slow weapon, but I temporarily forgot that you have a chance to
juuuust
maybe squeeze an extra one in with a faster weapon. This part of the discussion is pretty much academic though, I can't imagine two weapons being close enough for something like this to be the make-or-break factor in which one to use.
Edited for great dumbness.
Post by
svirve
I think we might have misunderstood each other. My point was not that you can come even close to 40 some hits per minute, and your point was not that you'll never melee. I can agree that your total melee swings aren't going to come close to what I got in my tests, but I do think a paladin that finds times to use all of his or her various utility spells is going to have a surprising number of chances to squeeze in a melee hit here and there. I think that's what we're both saying, just with different language.
Yea, I'm just a bit critical ;p
Potentially, you'll get better mana returns from a faster swinging weapon, because it's less likely to be cut-off before you get your melee swing in.
I doubt it. Haste speeds up your melee swings more than it speeds up your spell casting. Since it's usually going to be reasonably appropriate to put HS in with whatever other instant cast you do, that's two GCDs+latency time minimum. Since you'll never equip a two hander your attack speed is probably going to top out at 2.70 completely unbuffed with haste. That's already faster than two GCDs unbuffed with haste, and the more of that you get the faster your melee swings are going to be compared to your GCD, especially if you're at the soft-cap.
Also, if your regular raid has Frost Death Knights or Enhancement Shamans, they are going to kind of help your swing timer blow your GCDs out of the water.
Edit: wait, never mind. I'm thinking of the fact that you're guaranteed a hit or two even with a slow weapon, but I temporarily forgot that you have a chance to
juuuust
maybe squeeze an extra one in with a faster weapon. This part of the discussion is pretty much academic though, I can't imagine two weapons being close enough for something like this to be the make-or-break factor in which one to use.
Edited for great dumbness.
The thing with haste and procs is that haste wont work against it (i recently learned), if you have a 2.7 speed weapon it'll have the same proc chance if it is hasted to 1.5 sec speed as if it was unhasted.
With all the haste being tossed around, i'm betting a slower weapon would still be benifitial since it benefits more from haste as well as the higher proc chance.
Post by
pezz
I think this discussion is just conjecture until someone comes up with solid numbers for the PPM. At the moment it's 'potential extra hit or two from a fast weapon 'versus' higher proc chance on potentially less hits from a slow weapon.' Both points make sense for something where a solid PPM isn't known.
Post by
Squishalot
The thing with haste and procs is that haste wont work against it (i recently learned), if you have a 2.7 speed weapon it'll have the same proc chance if it is hasted to 1.5 sec speed as if it was unhasted.
With all the haste being tossed around, i'm betting a slower weapon would still be benifitial since it benefits more from haste as well as the higher proc chance.
Shouldn't make a difference. 10% melee haste = 10% more swings at 15ppm, irrespective of your weapon speed. All I'm advocating is that a faster weapon is less likely to have as much swing clipping due to casting, i.e. higher uptime.
Post by
svirve
I think this discussion is just conjecture until someone comes up with solid numbers for the PPM. At the moment it's 'potential extra hit or two from a fast weapon 'versus' higher proc chance on potentially less hits from a slow weapon.' Both points make sense for something where a solid PPM isn't known.
The PPM is ~15 which adds up with your numbers. You had an average PPM of 18.5 throughout your tests, and as i mentioned, they were very short. P.S. There used to be a thread on EJ about the PPM of SoW. It's gone now though.
Running a test now while doing my late night browsing to confirm that number is still accurate.
The thing with haste and procs is that haste wont work against it (i recently learned), if you have a 2.7 speed weapon it'll have the same proc chance if it is hasted to 1.5 sec speed as if it was unhasted.
With all the haste being tossed around, i'm betting a slower weapon would still be benifitial since it benefits more from haste as well as the higher proc chance.
Shouldn't make a difference. 10% melee haste = 10% more swings at 15ppm, irrespective of your weapon speed. All I'm advocating is that a faster weapon is less likely to have as much swing clipping due to casting, i.e. higher uptime.
I know what you're saying but my point is that if you have a 2.7 speed weapon it'll have a proc chance of ~70% while a weapon with a 1.5 speed will have a 37.5% proc chance.
At 1093 (33.3%) haste rating the 2.7 speed weapon will have a 2.025 speed.
The 1.5 speed weapon will have a 1.125 speed.
Not only will the slower weapon have gained an extra 0.3 speed over the faster one, it will still have a 53% higher proc chance.
Post by
frmorrison
SoW has a 15 proc per minute chance.
Post by
Squishalot
I know what you're saying but my point is that if you have a 2.7 speed weapon it'll have a proc chance of ~70% while a weapon with a 1.5 speed will have a 37.5% proc chance.
At 1093 (33.3%) haste rating the 2.7 speed weapon will have a 2.025 speed.
The 1.5 speed weapon will have a 1.125 speed.
Not only will the slower weapon have gained an extra 0.3 speed over the faster one, it will still have a 53% higher proc chance.
It won't have gained an extra 0.3 speed. The speed difference is 2.7/1.5 = 1.8x initially, and is 2.025 / 1.125 = 1.8x at the end too.
*happily puts svirve back in his place*
See? This is why you need me along in the Vent channel, so the carpets don't get wet from all the drooling...
Post by
svirve
I know what you're saying but my point is that if you have a 2.7 speed weapon it'll have a proc chance of ~70% while a weapon with a 1.5 speed will have a 37.5% proc chance.
At 1093 (33.3%) haste rating the 2.7 speed weapon will have a 2.025 speed.
The 1.5 speed weapon will have a 1.125 speed.
Not only will the slower weapon have gained an extra 0.3 speed over the faster one, it will still have a 53% higher proc chance.
It won't have gained an extra 0.3 speed. The speed difference is 2.7/1.5 = 1.8x initially, and is 2.025 / 1.125 = 1.8x at the end too.
*happily puts svirve back in his place*
See? This is why you need me along in the Vent channel, so the carpets don't get wet from all the drooling...
1.5-1.125 = 0.375
2.7-2.025 = 0.675
0.675-0.375 = 0.3
It will have gained an extra 0.3 speed.
And of course it's going to be the same ratio between the two, I'm increasing them both by the same "percentage", haste is not a fixed percentage so i hope you know what i mean.
And that extra 0.3 speed in this case is very very important. Seeing as we're trying to squeeze in swings during GCDs which are more oftenly than not shorter than swing speeds.
As for the test i mentioned, pic
one
and pic
two
.
I used my prot set since it has no haste, is hit capped, and has some expertise. The two latter aren't that important but made the test go faster.
I have the combat log as well but cba uploading it and if this doesn't suffice you can just go do something to yourself.
At a 5779 mana pool I'm getting back 231,16 mana on each proc, which gives us 285 procs during my test.
694 succesfull swings with a 1.7 speed weapon gives us 1195.1 seconds or 19.91833 minutes.
Which gives us a PPM of 14.5 (14,493982030852686896084082047807).
Once again, small scale tests are fairly inaccurate, and since blizz likes round numbers i can fairly certainly say that the PPM is either 14 or 15.
Post by
Squishalot
And that extra 0.3 speed in this case is very very important. Seeing as we're trying to squeeze in swings during GCDs which are more oftenly than not shorter than swing speeds.
I know, that's why I'm advocating using the faster weapon. If your swing times are 2.025 and 1.125 secs, then with 1.125sec GCDs, you're likely to get two hits (due to latency) with the first instant cast, and one more hit with every other instant cast you do. (First hit when you cast Beacon, second hit just before you start casting Judgement, third hit just before you cast HL or something.)
If you only use one instant cast, then you won't get any melee swings in using the 2.025sec (hasted) weapon from the first instant cast, and you'll only get a hit from the second, and every 2nd one from then on (and occasionally an extra one from an odd one).
End story - you'll get much, much higher uptime if you can possibly find a 1.5sec weapon. There are probably breakpoints, so a 1.6sec weapon will be no better than a 2sec weapon, for its ability to fit in between GCDs.
Post Reply
You are not logged in. Please
log in
to post a reply or
register
if you don't already have an account.