This site makes extensive use of JavaScript.
Please enable JavaScript in your browser.
Live
PTR
10.2.7
PTR
10.2.6
Beta
The basics - hit, crit, haste, and spell damage
Post Reply
Return to board index
Post by
irharry
It looks like its been edited for clarity now. The first page has some incorrect and misleading info (posted by me) that I've left in there in case people are interested in the one roll/two roll system. Its an interesting concept, and I believe melee still uses the one roll system, which is what I thought spell damage was using as well.
Post by
iranou
Hi, I've just hit 80 and found this as a useful guide to get started with gearing my
demo lock
(feel free to criticise) for raids, but I have a few questions. Sorry if these have been covered - I skim-read this thread and couldn't find the answer.
Firstly, does SP = dmg? If not, how are they related? The OP uses both terms.
First rule of warlocking
- +hit is your friend. Until you're capped, +Hit is the single biggest dps boost you will get, in terms of damage returned for stats invested. Put simply, 1 +hit rating is worth more to you than 1 of any other stat.
You've emphasised the importance of +hit, but is it necessary to prioritise +hit above all other stats until hit capped? I.e. if I find a piece with some hit, should I choose it over another piece with high SP, haste etc but 0 hit?
you generally want to stack stats in this order;
+hit (till capped) > +dmg > +haste > +crit > spirit.
The relative worth of stats looks something like this,
crit_rating=0.66
haste_rating=0.98
hit_rating=1.56
intellect=0.32
spell_power=1.46
spirit=0.47
Would you recommend using these numbers in wowhead's item comparison tool (
example: chest pieces
) in place of the
preset weightings
(which I've been using so far)?
Thanks.
Post by
asakawa
spellpower is transferred into actual damage via each spell's coefficient. depending on their cast time (or doT length) each spell receives a different amount of your spell power. long casts get more benefit otherwise long cast spells would always be inferior to quick spells.
in basic terms, though, raising your spellpower will raise how much damage each spell hits or ticks for. other stats won't raise the base number.
as you see on those weightings hit gives more benefit (point for point) than any other stat. this is why it should be prioritised. it also has a cap at which point none of your spells miss any more. since it does the most and has a clear point at which you're done with it it's very good to stack it first and cap your hit before moving on to other stats like haste or crit.
all items you look at should have spellpower so it's less of a concern. never take a drop in spellpower when changing gear unless there's a considerable amount of hit on it.
yes use those weightings on wowhead's item comparison or other sites with similar functions. if in doubt about 2 pieces of gear these tools are very good at giving you an idea which item will bring the most dps.
Post by
sasexw
i don't really understand sasexw's post (i reread it a few times but can't make head-nor-tail of it)
however, crit rate is lowered by low hit rates.
a roll is done to see if it's a hit or miss then a separate roll is done to see if the resulting hit is a crit. it's not 25% of spells cast, it's 25% of hits.
to find out your adjusted crit rate (after misses) you multiply your crit and hit percentages.
using our example:
25% x 83% = 20.75%
I just wanted to say that the critical hit , is a hit . What is a critical hit?A spell that hits the target and makes damage more than normal.Your spell must hit the target but this counts as a critical hit
I just tried to clear up that it wont be the hit diffrent than crit its almost the same thing
100 spells damage the target 25% of these are critical strikes so the critical hits are 25% of the total(100%) hits
i hope that clears it
Post by
asakawa
i hope that clears it
not even a little. at least for me - i may just be being a little dense today.
there's a roll to see if it hits or misses.
▼
if it hits there's another roll to see if it crits.
▼
your crit percentage is the percentage of the second roll so if you're missing much it will lower your overall crit rate.
if you're agreeing with that then great. if not then you'll have to word it a lot better because i have no idea what your point is ^_^
Post by
sasexw
totaly agree
Post by
asakawa
hehe ^_^
Post by
irharry
Again, the confusion here is largely due to incorrect information in the comments beneath the original post. I'm seriously considering removing all the info..
Post by
asakawa
aaah, i see. sorry, i never read past the main post >.<
well, this has raised a point. it would take a tremendous amount of testing to get a clear answer and i have to admit that i'd taken on faith that somebody had done that testing. the only numbers i've seen have been pretty amateurish and lacklustre so perhaps it's an assumption that's taken hold.
i'll endeavour to find this out for sure. i (like scooby doo and the gang) hate an unsolved mystery ^_^
what've you seen to make you doubt the 2 roll system irharry?
Post by
irharry
aaah, i see. sorry, i never read past the main post >.<
well, this has raised a point. it would take a tremendous amount of testing to get a clear answer and i have to admit that i'd taken on faith that somebody had done that testing. the only numbers i've seen have been pretty amateurish and lacklustre so perhaps it's an assumption that's taken hold.
i'll endeavour to find this out for sure. i (like scooby doo and the gang) hate an unsolved mystery ^_^
what've you seen to make you doubt the 2 roll system irharry?
I spent several hours testing with a large amount of crit gear, and no hit gear soon after wrath was released, and recorded higher crit ratios than I should of under the two roll system, which led me to believe that they had rolled not only the stat, but its effects into the one, melee based, one roll system. My numbers did not precisely match what I expected under the one roll, but they were much, much closer than what the two roll should have produced. I don't have my data anymore unfortunately, but I did a decent amount of testing, three tests each of 500 searing pains at three different crit rates.
I am the only person that thinks this however, so I eventually retracted all my comments. Personally I'm still not sure whats going on, but I figured that I had to be realistic and conclude that the chances I'm right and the rest of the warlock community is wrong was pretty small.
I simply can't find anyone else (here, at wowwiki, or at EJ) who agrees with me. Personally I'm not convinced that anyone has done enough testing to be sure, but again, for the sake of avoiding spreading misinformation I've shut up :)
Post by
asakawa
lol i see.
without any testing, thinking only logically, my problem with the single roll idea is, what happens in an extreme example of an 80% miss rate but 25% crit rate. eg, a level 40 lock SBing Dr boom (i'm not sure what the miss rate is at 30 levels but you see what i'm getting at).
does crit push a normal hit off the table? that would be absurd.
using an opposite example. it would be entirely possible for a lock to get 83% chance to crit on loatheb with 0 +hit.
with a single roll there would be no normal hits again. this may be the easiest way to test the theory.
i sense that i'm dragging up something you've already dealt with (possibly got flamed for) and moved on from here so apologies for that but i'm not willing to dismiss your idea out of hand. as you say, the rigorous testing that gave us the 2 roll idea may well have been done in vanilla wow and the idea stuck with nobody considering revisiting it.
2 rolls makes the most logical sense to me but wow's background spreadsheet rarely obeys logic.
Post by
irharry
i sense that i'm dragging up something you've already dealt with (possibly got flamed for) and moved on from here so apologies for that but i'm not willing to dismiss your idea out of hand. as you say, the rigorous testing that gave us the 2 roll idea may well have been done in vanilla wow and the idea stuck with nobody considering revisiting it.
2 rolls makes the most logical sense to me but wow's background spreadsheet rarely obeys logic.
I didn't strictly get flamed for it - but I suspect a few people only held their tongues because I had done a little more posting than they had :)
I have not been able to find any conclusive testing either way, other than my own, which while i found strongly indicative, was not 100% conclusive. I bowed to popular opinion and let it drop, but if you were able to find some way of proving it.... mint :)
Post by
WoWTheoryCraft
without any testing, thinking only logically, my problem with the single roll idea is, what happens in an extreme example of an 80% miss rate but 25% crit rate. eg, a level 40 lock SBing Dr boom (i'm not sure what the miss rate is at 30 levels but you see what i'm getting at).
does crit push a normal hit off the table? that would be absurd.
I like this idea, in fact you should be able to test it using an appropriate level test dummy.
The only thing that would complicate the results is level-based resists (and levelling a 'lock to the appropriate level to test), but this approach does sound promising.
If you choose the setup carefully it won't require a very large sample size at all to provide data with a good degree of confidence.
Post by
asakawa
have either of you ever seen what the miss rate looks like at much higher level differences? i'm so used to seeing that wowwiki table which goes to 5 levels in either direction.
Post by
WoWTheoryCraft
From what I've read it's an additional 11% miss rate for every level above +2 (for PvE), but I don't know how solid that info is.
Post by
asakawa
hmm, might be easier to do the loatheb test then. all the factors are known in that set up though you only get to do the test once unless you purposefully wipe before killing him so you can try again ^^
doubt i could convince my guild to do that.
although, one test should be enough. you'd have to be careful to never cast unless you have the 50% crit buff but if you can go in with 33% crit from gear/talents and make sure you're not getting any extra raid buffs like scorch...
in a 2 roll system with 0 hit rating you should see 68.89% crit rate (+17% misses = 14.11% normal hits) and in a one roll system you won't see any regular hits at all.
though this assumes that crit rate
can
push normal hits off the table. possibly an assumption too far.
Post by
midnightzak
Per the 1 roll VS 2 roll debate-
If there is an easier way to test (not wiping a guild raid) then I will definitely help out.
Cool if it will work on a test dummy, but am up for anything...
Post by
iranou
The relative worth of stats looks something like this,
crit_rating=0.66
haste_rating=0.98
hit_rating=1.56
intellect=0.32
spell_power=1.46
spirit=0.47
Just bear in mind that these numbers are always going to depend on your own ratios as well, and they're never ever going to be exact (credit to Asakawa for these numbers from simulationcraft).
Thought I'd google simulationcraft and it's awesome. Empirical data ftw! Check out
this page
for details of the top DPS casters - nice to see locks at the top :D
Also note further down the page are the DPS scale factors for each spec (including 7 different lock specs), and links to wowhead and lootrank listings for each one. I've found these links very useful for finding the best gear for my
spec
. Only been 80 for a few days, but thanks to this I've been able to select the best gear to prepare me for raids.
Post by
asakawa
interesting how different the values are for destro and affliction for hit and SP... makes me a little dubious. i'll try and run a few simcraft passes myself on those builds and see what i get.
Post by
325825
This post was from a user who has deleted their account.
Post Reply
You are not logged in. Please
log in
to post a reply or
register
if you don't already have an account.