If I change two 5dmg/4crit gems into two 6dmg/5hit gems (the smallest possible change to reach 16% hit), my dps rises from 1115 to 1121. My total damage drops from 563565 to 555714.
That's a half a percent DPS gain to but a 1.4% damage loss.
If I change two 5dmg/4crit gems into two 6dmg/5hit gems (the smallest possible change to reach 16% hit), my dps rises from 1115 to 1121. My total damage drops from 563565 to 555714.
That's a half a percent DPS gain to but a 1.4% damage loss.
Uhm.... wow? Of course your total damage drops, when the hell was the last time you nuked straight Lbolts for 8min straight, no pauses, and accidentally deleted all your mana potions. The total damage statistic is meaningless, Elemental shaman should never run out of mana, which is why the Total damage figure is lower. The spreadsheet says you run out of mana after 505seconds because it is not a simulation, it is not real, it is just using Water shield, talented Mp5, set bonuses, etc to give an estimate. The DPS is all that matters, or that should matter, otherwise you are bad. No offense. You are looking at a 6dps gain from ONE gem change, in my book thats freakin huge.
If I change two 5dmg/4crit gems into two 6dmg/5hit gems (the smallest possible change to reach 16% hit), my dps rises from 1115 to 1121. My total damage drops from 563565 to 555714.
That's a half a percent DPS gain to but a 1.4% damage loss.
That's because crit reduces the mana use, increasing your sustainability, and thus increasing the overall damage done.
Do a comparison between a pure stam gem (with the stam value set to 0), then add a pure hit gem. Then see what the change is.
At the moment you are comparing 8 crit with 2 dmg/10 hit, which is not a good comparison to make when looking at +hit.
Yes, I'm aware of what crit does. Now if we actually go back to my first post, you'll see where I stated that _I_ can't increase hit or haste because every ounce of damage I output is necessary in my raiding situation and those things lower my total damage.
And I made that statement in response to Daidalos suggesting to raptorgsus that increasing hit to the cap would increase his dps. He then went on to talk about mana returns.
My whole point was that when mana is an issue, socketing hit instead of crit is not always the best answer.
Yes, I'm aware of what crit does. Now if we actually go back to my first post, you'll see where I stated that _I_ can't increase hit or haste because every ounce of damage I output is necessary in my raiding situation and those things lower my total damage.
And I made that statement in response to Daidalos suggesting to raptorgsus that increasing hit to the cap would increase his dps. He then went on to talk about mana returns.
My whole point was that when mana is an issue, socketing hit instead of crit is not always the best answer.
1 point to the contrary is that if you don't max hit, and you get resisted then you have 0% chance to crit on that cast (obviously). This makes you miss clearcasting opportunities, and therefore spend MORE mana on casts following a resisted lbolt.
The answer is to socket +hit +spelldmg instead of straight +spelldmg gems to get to the cap on hit (hit is better than +spelldmg when not capped), and keep the necessary crit in order to keep your mana up, IF MANA is an issue. I honestly don't know why mana is an issue above T4, and if you have mana troubles the best solution is swapping in Totem of Pulsing Earth during the fight (~63mp5 if spamming Lbolts) or switching to Adepts + Major mageblood instead of flasks. If you have mana troubles stay away from spell haste. If you do those things and still have issues, with a spriest in your group, then the spriest needs a swift gkick. If you don't get a spriest in raids, which must be the only explanation, then you would need to look into the +int and mana return meta socket as well as tide walker trinket, or if you can't get one, pendant of the violet eye. Of course all of this kills Elemental dps but having mana is more important, and obviously if you are in a raid and don't get a spriest, then your guild is using you for your totems and doesn't care about your DPS.
To be frank there are tons of items available starting at the Kara level to keep your spell hit capped as you progress, there is simply no excuse, and regemming should never be necessary. Arena weapons have +spell hit, they are excellent options as weapons when not hit capped. Honor reward necklaces have a yellow socket you can gem with +hit +spell dmg, making it one of the best necks in the game if you arent hit capped. Just mentioning two of a plenitude of options.
My whole point was that when mana is an issue, socketing hit instead of crit is not always the best answer.
You have to hit before you can crit. Spells are a 2 roll system.
Let's look at it a different way. Let's suppose you traded off the 8 crit for the 2 damage and 10 hit via 2 gem swaps (shoulders would be a good place to put them).
Currently, you sit at 36.35% crit and 15.35% hit. This gives you an effective crit rate of (thanks to the 2 roll system) .3635 * (.83+0.1535) = 35.75% crit, losing an effective 0.6% crit to resists.
The trade off would put you at 35.99% crit and 16.04% hit, capped at 16% of course. This would give you an effective crit rate of .3599 * (.83+.16) = 35.63%, losing an effective 0.36% crit to resists.
The effect of spell hit is best seen through DPM, and especially in low mana situations, where a resist is the most painful. So let's roughly simulate that on the spreadsheet.
So save a second copy of ShamStats, one with your current gear, one with manually added 2 damage, subtract 8 crit rating, and add 10 hit rating to the shoulder slot.
Now, remove all mana/sustainability buffs from the Main page (AI, BoK, BoW, Shadow Priest, Super Mana Pots) and change the starting mana to half your mana pool (I used 5k since this is a rough demonstration). You'll notice that on the spreadsheet with the gem swap in place, you're doing 7 more DPS and more total damage as well.
That effect is what is gained from spell hit: more efficient mana usage. So when mana's an issue, disregarding spell hit for spell crit will actually yield a negative effect. If mana's not an issue, stacking other stats can sometimes yield better results in theory. The biggest problem is that, just like the new SR, you'll experience a larger standard deviation from your DPS output then you will with capped spell hit.
Originally Posted by DeeNogger
I look forward to seeing these "numbers". Notice that I put the word numbers in quotations. Thats sassy type for "you're full of shit".
Originally Posted by Florrie
Nothing spells out attraction quite like being given books about the slaughter of your people.
Yes, I'm aware of what crit does. Now if we actually go back to my first post, you'll see where I stated that _I_ can't increase hit or haste because every ounce of damage I output is necessary in my raiding situation and those things lower my total damage.
And I made that statement in response to Daidalos suggesting to raptorgsus that increasing hit to the cap would increase his dps. He then went on to talk about mana returns.
My whole point was that when mana is an issue, socketing hit instead of crit is not always the best answer.
Theorycrafting, simulations, and in game testing via Dr. Boom, WWS, etc. all show that for every caster class +hit rating is always the most valuable stat until hit capped. There aren't any if and, or but's about it.
Milkk, Diadalos, Daler, Bink etc. have all pretty much covered your flawed reasoning with respect to shamstats and its calculation of time until oom. But try thinking about it in this light. On a boss fight. It will be a certain length. If you are doing 6 more dps and the fight is 5 minutes you will be doing 1800 more damage with the hit/damage gem over the crit/damage gem
Now, if the fight is longer and you don't do the extra 6 dps then you will still end up doing more damage in the end. Great. What does that tell you about hit and crit? Nothing. You are comparing apples to oranges with the fight length.
I don't really see any more point in discussing this, though, as you won't be swayed by opinions and analysis of people that I have grown to have a lot of respect for upon reading through thie whole thread.
Last edited by Durakghar : 03/26/08 at 7:16 PM.
Reason: Daler's contributions were valuable in trying to help understanding as well.
If you need more mana for longer figths, whine at your shadow priests if that is your argument to not get hit capped, totall dmg from just base stats is completely pointless.
I mean shamans need a really lousy amount of hit anyway it's not like it will eat into your stats like a raving rabbit.
Wodi, I've been getting 1 in 10.6 avarage after I made a mod to count the refreshes, I do have a substantially more haste than you do. 1.8 sec cast before mods
While it is a lot better than my previous post, forgive me for being WWS stupid ^^, I am still getting no were near the afore mentioned uptime.
Couple of things. I replaced two +8 crit gems with 2 +8hit gems, and a +4crit/5dmg gem with a +4hit/5dmg gem. I am now at 53 hit. I went to do our guilds usual TK/SSC runs this week and got #1 damage on Al'ar and #2 on Lurker. Only the Destro locks in my group can outdps me now unless aoe is involved. I have no math, but experience in this raiding group has shown me to usually be 3-5th overall per fight, and now I am fighting for 1-4th just because of 2% extra hit. Best 100g I ever spent. Guildmates are boggled that a shaman can challenge for top dps - I just say "go read elitist jerks and L2P"
On the haste trinket, it goes back to my earlier question about weightings. I know 40 haste is pretty hot, but one thing I personally think about haste is that it's only good if you have enough of it. If you only have 40 overall I don't think it is going to make a big difference, but if you get 100 or more, then you starting 'feeling it'. I'm personally going to buy that trinket but I don't think I will use it until I get a couple more haste items to complement it.
For trinkets I'm currently using Xiri's Gift and Icon with good success, though I'm considering the 40 crit trinket just for the passive increase. Despite the math I don't have much confidence in Quag's Eye, TLC, or other 'maybe it will fire, maybe it won't' items.
long time reading here but up to now there was no need to post as someone else always was there to post the correct question ;-)
But as 2.4 is live on server we have a new possible trinket - 40spellhaste + "mini" healthstone
According to the dps-spreadsheets 40spellhaste is worth about 60 spelldamage - and therefore a possible alternative trinket to use?
Depending on your stat weights a little under 60 but yes its a decent trinket from a dps standpoint and quite nice if its a fight where you can take advantage of the temp health buff however its still not as good as [Icon of the Silver Crescent] which requires fewer badges and works well in a lust without any increased mana useage. If you are really bad at clicky trinkets or only have crescent and no other options its certainly not a bad trinket to have.
Originally Posted by raptorgsus
Couple of things. I replaced two +8 crit gems with 2 +8hit gems, and a +4crit/5dmg gem with a +4hit/5dmg gem. I am now at 53 hit. I went to do our guilds usual TK/SSC runs this week and got #1 damage on Al'ar and #2 on Lurker. Only the Destro locks in my group can outdps me now unless aoe is involved. I have no math, but experience in this raiding group has shown me to usually be 3-5th overall per fight, and now I am fighting for 1-4th just because of 2% extra hit. Best 100g I ever spent. Guildmates are boggled that a shaman can challenge for top dps - I just say "go read elitist jerks and L2P"
To sum up the last 2 pages.
"Hit cap good"
At least one person got it ;-)
On the haste trinket, it goes back to my earlier question about weightings. I know 40 haste is pretty hot, but one thing I personally think about haste is that it's only good if you have enough of it. If you only have 40 overall I don't think it is going to make a big difference, but if you get 100 or more, then you starting 'feeling it'. I'm personally going to buy that trinket but I don't think I will use it until I get a couple more haste items to complement it.
I don't really buy this argument. Its like saying .5% crit isn't worth much since you don't notice it. I also don't buy the argument of you need x num of mins before you can get an extra cast in. This assumes everyfight without haste it ends precisely as your last LB finishes. Think about it how many times have you been 90% through a cast and a boss or mob died. Haste increases your chance of completeing that last spell regardless if you have "saved" 2s of cast time yet.
For trinkets I'm currently using Xiri's Gift and Icon with good success, though I'm considering the 40 crit trinket just for the passive increase. Despite the math I don't have much confidence in Quag's Eye, TLC, or other 'maybe it will fire, maybe it won't' items.
Quags eye hasn't been shown to be that strong of a trinket. I think all the math that tried to show it was ended up disproved. Its a very good blue but it is not one you should really be using once you have access to the badge trinkets. TLC is a good trinket there is no way I'd use xi'ri's over TLC.
For trinkets I'm currently using Xiri's Gift and Icon with good success, though I'm considering the 40 crit trinket just for the passive increase. Despite the math I don't have much confidence in Quag's Eye, TLC, or other 'maybe it will fire, maybe it won't' items.
"Despite the math" is not something you want to throw around in a place like this.
Situations like quick battles, arena, etc. are where proc trinkets lose their attraction, but the reason why they are competitive (and it's the same reason why hit caused such a major difference in your performance, I'd bet) is that they follow pretty predictable proc rates and give measurable effects over time. PvE is a much larger sample size, so while they may not "feel" as dependable or powerful as a passive stat trinket, they give bonuses that can be weighted the same as any other stat.
Oh now you've gone and done it - you've dredged up the relic debate in me. I currently use Totem of the Void because again, I like reliable passive benefits or things I can activate on demand. The 100 haste proc relic has gotten a LOT of posts recently. I don't use it.
I know this place is all about the math, but even the math is in dispute on that damn relic. Do you guys really think 100 haste is up consistently enough to overpower the 55 damage? Math doesn't do much for lag, totem dropping, or running away. That 55 goes with you no matter what and the 100 is a crapshoot even if you're standing still.
PS lowering the GCD on totems to 1 second also greatly helped my DPS on Al'ar - I drop ToW, WoA, and MS every time I have to chase him to a new platform in phase 1. Saves me a full CL cast per relocate. I still think totems should drop all at once or with no GCD, but its better than it was. Damn Boomkin aura just follows him around. No fair.
Going back to the new haste trinket and comparing it to the other pvp trinkets, the math would indicate that the haste one is > the crit one > the dmg one correct?
If 100 haste screws up your LB/CL rotation, or doesn't change the number of LB's you cast before casting CL, don't use it.
So if 100 haste changes (screws up) the number of LB in your LB/CL rotation or if it doesn't don't use it? Is that a clever way of saying don't use it? :-p
In regards to forgetting to click on use trinkets, I have my trinkets bound to commonly used skills on my characters to ensure that I receive a reliable amount of use out of them. On my shaman, I have a macro that I use when I get in to my rotation:
#showtooltip Lightning Bolt
/use Icon of the Silver Crescent
/cast Lightning Bolt
/console Sound_EnableSFX 1
/run UIErrorsFrame:Clear()
This will activate the trinket every time it's off cooldown when you cast a Lightning Bolt. The console and run lines at the end stop the error text/sound spam about items not being ready and such.
Last edited by Gaise : 03/27/08 at 3:14 PM.
Reason: Me fail English? That's unpossible.
Has anybody experimented with TLC since the patch?
Item Cast Spells: Many spells cast by items were being cast as if the caster were the same level as the item. Most of those will now be cast at the player’s level, giving them a reasonable chance to miss, be dispelled, or be resisted.
That implies that TLC now has the standard 17% chance to miss against bosses that we have, but I haven't been able to find confirmation yet. The patch notes say "most" but don't specify which items are affected.
If 100 haste screws up your LB/CL rotation, or doesn't change the number of LB's you cast before casting CL, don't use it.
Eh, I have to respectfully disagree with the latter. The extra 100 haste may not push me away from a 3:1 rotation, but it does increase the number of successful casts I finish before mob/boss X dies, for the reason Daidalos pointed out above:
I also don't buy the argument of you need x num of mins before you can get an extra cast in. This assumes everyfight without haste it ends precisely as your last LB finishes. Think about it how many times have you been 90% through a cast and a boss or mob died. Haste increases your chance of completeing that last spell regardless if you have "saved" 2s of cast time yet.
I'm personally quite satisfied with the skycall totem during MH trash clears. It seems to lessen the impact of moving to redrop FN/Magma totems as needed (and god bless the 1.0 sec totem GCD )
Originally Posted by DeeNogger
I look forward to seeing these "numbers". Notice that I put the word numbers in quotations. Thats sassy type for "you're full of shit".
Originally Posted by Florrie
Nothing spells out attraction quite like being given books about the slaughter of your people.
In regards to forgetting to click on use trinkets, I have my trinkets bound to commonly used skills on my characters to ensure that I receive a reliable amount of use out of them. On my shaman, I have a macro that I use when I get in to my rotation:
#showtooltip Lightning Bolt
/use Icon of the Silver Crescent
/cast Lightning Bolt
/console Sound_EnableSFX 1
/run UIErrorsFrame:Clear()
This will activate the trinket every time it's off cooldown when you cast a Lightning Bolt. The console and run lines at the end stop the error text/sound spam about items not being ready and such.
OMFG thanks for that I've been using macros like that for months and just ignored the constant error text and sounds. It wasn't so bad because I bound it to CL instead of LB, but still ever 6 seconds is still annoying.
#showtooltip Chain Lightning
/cast Elemental Mastery
/use Icon of the Silver Crescent
/use Xiri's Gift
/cast Blood Fury (racial)
/cast Chain Lightning (rankx)
/console Sound_EnableSFX 1
/run UIErrorsFrame:Clear()
That's hot. I find if I keep the activations on my CL, they stay up pretty reliably. The only thing that would optimize them better is if I saved all the cooldowns for Bloodlust and then activated them all before and during Bloodlust. Pretty hard to time all that without wasting possible uptime on the abilities.
The X > Y math is pretty hard to apply when each fight has different requirements. I think if you have to move around at all or replace totems more often than normal the 100 haste proc fails compared to the 55 dmg even though on paper it might win - has anyone got math that includes variables for movement time and rate of totem drops? I don't know how to prove it mathematically, but I'm pretty sure the haste isn't worth it on most boss fights and certainly not on trash.
I have a question my The World of Warcraft Armory says I have 436 int unbuffed. When you hold over int it says i have 7.65% crit from int. That works out to be 436 / 7.65 = 56.99 int to equal 1% crit. I thought its supose to be 80 int = 1% crit at 70.I was trying to figure out my crit % but not sure. Also if you hold over my spell crit it says I have a 386 crit rating from gear witch is 17.47% crit and a crit % of 25.14%. So if you minus 17.47% from 25.14% you get 7.67% witch makes me asume that int is 56.99int = to 1% crit instead of the 80 int = 1% crit. That or our character screens are bugged in game and in Armory.
So based on if the screens are correct then 1 int is worth approx. .0175 instead of .0125.
I figured it out from reading some other sites. It is "(Intellect/80) + Class Specific Constant". The Class constant for Shaman is "2.2"
Source : Spell critical strike - WoWWiki - Your guide to the World of Warcraft
Tested it and it calculates perfect to what character sheet says on armory for spell crit hit under intelect.
Add (Spell Critical Strike Rating/22.08) to that number and thats where the crit chance % comes from on armory.
The X > Y math is pretty hard to apply when each fight has different requirements. I think if you have to move around at all or replace totems more often than normal the 100 haste proc fails compared to the 55 dmg even though on paper it might win - has anyone got math that includes variables for movement time and rate of totem drops? I don't know how to prove it mathematically, but I'm pretty sure the haste isn't worth it on most boss fights and certainly not on trash.
Well, that's going to be difficult to quantify (the time spent moving, that is).
Would a WWS active DPS time help? My thought is to take the active DPS time from WWS, and multiply that by the projected uptime (since WWS and most mods are coming up short on the uptime due to refreshes) to get an effective uptime %, then run the DEP values. I think it'd be napkin math at best, so if someone's got a better idea, I'm all ears.
Originally Posted by DeeNogger
I look forward to seeing these "numbers". Notice that I put the word numbers in quotations. Thats sassy type for "you're full of shit".
Originally Posted by Florrie
Nothing spells out attraction quite like being given books about the slaughter of your people.
So if 100 haste changes (screws up) the number of LB in your LB/CL rotation or if it doesn't don't use it? Is that a clever way of saying don't use it? :-p
What did you mean by screw up?
The big problem with haste is that we want to be using CL as soon as the cooldown is up.
So if we have zero haste, and can get in a perfect 3 LB 1 CL rotation in 7.5 seconds, having the totem proc will mean we get the 3 bolts cast in 5.6 seconds, and end up waiting 0.4 seconds for CL to be up to cast it again, which means we only really save 0.09 seconds overall.
If the proc takes us passed the point where a 4/1 rotation results in higher dps, then we'd be changing from a 3/1 rotation to a 4/1 rotation and back multiple times during a fight, and the static haste you have is only useful when the totem procs.
I would only use it if I can get passed that "minimum haste" value.
This is the graph from the main post, assuming hit capped and 40% crit
So when I get 4pc T6, I'll look at getting about 6% haste from items before I even consider swapping them in.
You have to hit before you can crit. Spells are a 2 roll system.
Hi, could you give me the source that supports this statement, since wowwiki says otherwise. they might be wrong, so I would be glad to see your source.
Hi, could you give me the source that supports this statement, since wowwiki says otherwise. they might be wrong, so I would be glad to see your source.
Actually, WoWWiki says precisely that you must first hit before you can crit. Go back and reread the section titled "The effect of Spell Hit chance on Spell Critical Hit chance."
03/26/08, 5:21 PM
)
Similar Threads
