[Rogue/Warrior] Weapon Skill Adjustment Discussion - Page 22

Dontmindme · 09/07/07, 9:54 PM

Originally Posted by Lapp

Hello, I thought this would be a relevant question to ask here.

I am a 17/44 orc warrior, currently using a Drakefist Hammer in my MH, with almost enough nethers for Dragonmaw. I'll be getting the S2 OH axe soon (since it'll benefit from the +weapon skill racial where the mace, sword, etc won't). Considering the following mechanic changes, would it be worth it to respec axesmithing for the Planar Edge line?

1. Haste rating nerf
2. PPM nerf (uses current weapon speed, not base)
3. The 5 weapon skill I would be gaining since I am orc; apparently this gives 3% or so hit, some crit, and minus dodge?

Currently, the Mace BS line is hands-down better for on orc (or dwarf) because of 1 and 2 not being in effect despite the increased weapon skill with axe; however, with the reduction in potency of 1 and 2, will I see better results with the Axe BS line? I'm pretty sure that if I wasn't orc, the mace would still be the better choice... but will the weapon skill tip it into axe's favor? Thanks.

First off, I'm not a warrior and am not all that familiar with the weapons you are speaking about. Although I did look at them briefly. That said:
1) It is a substantial nerf. You will lose 1/3 of the haste effect.
2) This does not appear to be a nerf. This appears to already be the game mechanics that are currently in place. It's just that no one had noticed it. If something is proccing a lot now, it will still proc a lot. Nothing is changing here except spreadsheets that were modeled differently.
3) The question is whether you can get the 5 weapon skill some other way. 5 seems to be the magic number to get a substantial reduction in misses vs. bosses. Looks like 5 weapon skill saves you just over 47 hit rating vs. bosses.

Lolaan · 09/08/07, 1:48 AM

Originally Posted by Dontmindme

We are still trying to get to the bottom of the Dodge question. So far, it has proven to be more difficult to test (especially since there is no way to reach 0 Dodge with the available weapon skill which has proven to be the best way to really nail down the mechanic). The short answer is its being worked on along with "Parry" as this is of big interest to tanks.

As to your other questions, tests regarding hit rating have been done at those ranges. Each weapon skill reduced your "Miss" chance by 0.1% (which is a more accurate way of portraying the effect of hit rating). That part of the equation is solid. That is how much the miss chance is reduced; that's what it means to increase hit by 0.1%.

As far as Dodges are concerned, there is some thought that weapon skill may reduce these by 0.1% as well, but so far that is not nearly confirmed. We are still trying to figure out what the base dodge rate really is for bosses and whether the weapon skill effect to dodge is linear vs. bosses.

Has anyone considered testing dodge/parry rates with non-maxed level characters? You could get SIGNIFICANTLY higher level differences. So far anyhow, weapon skill seems to counteract all level based penalties. Looking back at that blue post and some of my WWS data (we have a feral tank not using earth warden, thus no weapon skill). Parry rates seem to be universally higher than dodge rates. Maybe his post wasn't so far fetched after all. This just wasn't the full picture.

Lets look at that post again:

Let's say you increase your weapon skill by 4.

Against a mob 3 levels higher than you, you get: 0.8% lower chance to miss,
0,4% lower chance to get dodged, 2,4% to be parried, and a 0,8% higher
chance to crit."

Lets also assume these are the first points of weapon skill you have received. For now, I will also assume that weapon skill is a direct counter to the penalties of the level difference (the mob's defense).

Thus!

1 point of weapon skill at defense-weapon skill >10 gives you:
0.2% -miss
0.1% -dodge
0.6% -parry
0.2% crit

As far as the hit goes, this looks very similar to the wowwiki theory. The current working theory here is that its 0.4% per point with a 1% bonus when you switch over to the second interval of the piecewise function. EDIT: Ignore this next part, it has been disproven Have we disproved with our testing that it is in fact 0.4% with a 1% bonus and not a 0.2% with a 2% bonus? Its been a bit since I read the entire thread so I just wanted to check. It seems like these would be extremely close and possibly within your testing parameters? I'll look at it more tomorrow.

But, onto the rest. Dodge at >10 is 0.1% per point and parry at >10 is 0.6% per point. Crit is 0.2% per point. But hey, didn't Blizzard say in their patch notes that they were making it 0.1%? I got a couple theories about this too, may or may not be right. The first one is that this is only the penalty for the >10 difference, which would be consistent with everything else. The second one (which is compatible with the first) is that blizzard gave the crit portion of weapon skill a bonus over its counteracting the mob's defense as its benefit instead of counteracting the enemies glancing blows.

For the moment, I will assume all these numbers (except hit) are correct. They may not be, obviously there's testing to be done. I still think the best option would be to use a twink where you can make an extreme level difference and thus get a larger change in the dodge and parry rates (assuming we can prove the progression is linear).

From the miss test we know a couple things about how mob levels behave.

There's a base value. (5% miss/dodge/parry (+19% for dw)
There's a modifier for skill points with a >10 difference (0.4% for hit is the currently believed value)
There's a modifier that can be different for skill points with <= 10 difference (0.1% for hit)
There is an additional penalty at the 10 point mark (1% for hit is the currently believed value)

If we assume that this is all true for all the stats weapon skill affects and that the information in the blue post is correct (except the hit which has been proven). We have the following.

Stat	Base	>10	<=10	+10
Miss	5(24)	0.4	0.1	1
Dodge	5	0.1	?	?
Parry	5	0.6	?	?
CritPen	0	0.2	?	?

Oh, and also one other thought on "eliminating" dodge or parry. You could go for mobs below your level to eliminate parry. This would find the <=10 value. This would obviously take SIGNIFICANTLY longer than hitting a boss for a long time since the mob would actually die if you can't find a high hp elite to use. To do this with 360 weapon skill:

0.1 per point, should see no dodges on a level 62 mob (molten core anyone?)
0.04 per point, should see no dodges on a level 47 mob (Archaedas in Uldaman has the highest hps for a 47 mob at 30k)

I also noticed that you do tested where you just skirt the cap. A rogue with mace spec could use the molten core mobs at 368. Double check my math here, didn't take statistics. If it is wrong, please provide an explanation so I'll know for next time, its appreciated.

0.1 per point, should see a dodge 1 in 500 hits
0.04 per point, should see a dodge in 1 in 1250 hits
Are these numbers too low to be useful? Would we need weapon skill to be 369? (1/1000 and 1/2500)

Anything higher would see a lot of dodges (or a good bit higher than 1/500)
Anything lower would see no dodges.

Now, none of this helps us determine what happens at 10 or above 10, but it gives us a good starting point to work with.

If this post seems a little discontinuous its because it was written over about 3 hours.

I might try double boxing this at some point but I don't have a whole lot of time lately so it'd be helpful if someone else could do the testing. I'll try to find time however.

EDIT3: Ignore the following two edits
EDIT: I just went back through and a 0.2 hit value seems near impossible based on sp00ns test. However, while the values in the blue post might not be accurate, I think that my theories are sound (granted they ARE theories!)

EDIT2: Hmm, Depending on Imp FF up time, I'm not sure how valid sp00n's data is, however my theory is still unlikely. I also think it was probably disproved somewhere earlier in the thread but I'm too tired to look it up right now.

Punscho · 09/08/07, 4:26 AM

Read through the whole thread again please.

There's already been testing vs mobs below your level, even at the first pages of this thread. With and without extra weapon skill. There's some conclusions about lowered avoidance rate among those posts.

Drop the blue post please, let's focus on our own tests. We havn't seen any reduction in parry as far as I know (see the paladin tests), only some deviations in dodge.

sp00n · 09/08/07, 8:16 AM

Jaaaaaaa!

After 4048 attacks I finally got the miss.
500 attacks with imp FF / dreanie aura seems quite possible, so that sucker absolutely pushed it to the limit.

I think we can now agree that the value 1 point of weapon skill gives is 0.4% for the first 5 points.

Captain Galvangar & Drek'Thar
354 weapon skill - 337 hit rating

Melee
Miss: 	     1   0.02%
Dodge: 	   185   4.57%
Parry: 	   111   2.74%
Glancing:  980  24.21%
Blocked:    38   0.94%
Hit: 	  2007  49.58%
Crit: 	   726  17.93%
Total: 	  4048    100%


Hemorrhage
Miss: 	     0      0%
Dodge: 	    53   5.16%
Parry: 	    26   2.53%
Blocked:     8   0.78%
Crit Block:  1   0.10%
Hit: 	   756  73.61%
Crit: 	   183  17.82%
Total: 	  1027    100%

Captain Galvagar is stunnable as mentioned before, explaining the lower dodge rates and therefore unfortunately this data set is not usable for the dodge discussion.
However, data only for Drek'Thar might be interesting, although I'm not sure if he cannot dodge while performing whirlwind as well.

Drek'Thar only:
354 weapon skill - 337 hit rating

Melee
Miss:        1   0.04%
Dodge:     136   5.74%
Parry:      78   3.29%
Glancing:  567  23.92%
Blocked:    25   1.05%
Hit:      1130  47.68%
Crit:      433  18.27%
Total:    2370    100%


Hemorrhage
Miss:        0      0%
Dodge:      35   6.14%
Parry:      19   3.33%
Blocked:     6   1.05%
Hit:       416  72.98%
Crit:       94  16.49%
Total:     570    100%

sp00n · 09/08/07, 8:46 AM

This are now how the tables look:

Target 3 levels higher, without any +Hit through talents

Skill    Hit   Miss    Total    Decrease
+ 0   441.54   28.0%   + 4.0%   0.0%
+ 1   435.23   27.6%   + 3.6%   0.4%
+ 2   428.92   27.2%   + 3.2%   0.4%
+ 3   422.62   26.8%   + 2.8%   0.4%
+ 4   416.31   26.4%   + 2.4%   0.4%
+ 5   394.23   25.0%   + 1.0%   1.4%  --> 1% jump
+ 6   392.65   24.9%   + 0.9%   0.1%
+ 7   391.08   24.8%   + 0.8%   0.1%
+ 8   389.5    24.7%   + 0.7%   0.1%
+ 9   387.92   24.6%   + 0.6%   0.1%
+10   386.35   24.5%   + 0.5%   0.1%
+11   384.77   24.4%   + 0.4%   0.1%
+12   383.19   24.3%   + 0.3%   0.1%
+13   381.62   24.2%   + 0.2%   0.1%
+14   380.04   24.1%   + 0.1%   0.1%
+15   378.46   24.0%   + 0.0%   0.1%

Target 3 levels higher, +3% Hit (e.g. Fury Warrior with Precision)

Skill    Hit   Miss    Total    Decrease
+ 0   394.23   25.0%   + 4.0%   0.0%
+ 1   387.92   24.6%   + 3.6%   0.4%
+ 2   381,62   24.2%   + 3.2%   0.4%
+ 3   375.31   23.8%   + 2.8%   0.4%
+ 4   369      23.4%   + 2.4%   0.4%
+ 5   346.92   22.0%   + 1.0%   1.4%  --> 1% jump
+ 6   345.35   21.9%   + 0.9%   0.1%
+ 7   343.77   21.8%   + 0.8%   0.1%
+ 8   342.19   21.7%   + 0.7%   0.1%
+ 9   340.62   21.6%   + 0.6%   0.1%
+10   339.04   21.5%   + 0.5%   0.1%
+11   337.46   21.4%   + 0.4%   0.1%
+12   335.89   21.3%   + 0.3%   0.1%
+13   334.31   21.2%   + 0.2%   0.1%
+14   332.73   21.1%   + 0.1%   0.1%
+15   331.15   21.0%   + 0.0%   0.1%

Target 3 levels higher, +5% Hit (e.g. Rogue with Precision)

Skill    Hit   Miss    Total    Decrease
+ 0   362.69   23.0%   + 4.0%   0.0%
+ 1   356.39   22.6%   + 3.6%   0.4%
+ 2   350.08   22.2%   + 3.2%   0.4%
+ 3   343.77   21.8%   + 2.8%   0.4%
+ 4   337.46   21.4%   + 2.4%   0.4%
+ 5   315.38   20.0%   + 1.0%   1.4%  --> 1% jump
+ 6   313.81   19.9%   + 0.9%   0.1%
+ 7   312.23   19.8%   + 0.8%   0.1%
+ 8   310.65   19.7%   + 0.7%   0.1%
+ 9   309.08   19.6%   + 0.6%   0.1%
+10   307.5    19.5%   + 0.5%   0.1%
+11   305.92   19.4%   + 0.4%   0.1%
+12   304.35   19.3%   + 0.3%   0.1%
+13   302.77   19.2%   + 0.2%   0.1%
+14   301.19   19.1%   + 0.1%   0.1%
+15   299.61   19.0%   + 0.0%   0.1%

Target 3 levels higher, one-hand or two handed weapon or ranged, no +Hit through talents

Skill    Hit   Miss    Total    Decrease
+ 0   141.92    9.0%   + 4.0%   0.0%
+ 1   135.62    8.6%   + 3.6%   0.4%
+ 2   129.31    8.2%   + 3.2%   0.4%
+ 3   123       7.8%   + 2.8%   0.4%
+ 4   116.69    7.4%   + 2.4%   0.4%
+ 5    94.62    6.0%   + 1.0%   1.4%  --> 1% jump
+ 6    93.04    5.9%   + 0.9%   0.1%
+ 7    91.46    5.8%   + 0.8%   0.1%
+ 8    89.89    5.7%   + 0.7%   0.1%
+ 9    88.31    5.6%   + 0.6%   0.1%
+10    86.73    5.5%   + 0.5%   0.1%
+11    85.15    5.4%   + 0.4%   0.1%
+12    83.58    5.3%   + 0.3%   0.1%
+13    82       5.2%   + 0.2%   0.1%
+14    80.42    5.1%   + 0.1%   0.1%
+15    73.85    5.0%   + 0.0%   0.1%

Target 3 levels higher, +3% Hit,
one-hand, two handed weapon or ranged (e.g. 2H Warrior with Precision, Hunter with Surefooted)

Skill    Hit   Miss    Total    Decrease
+ 0    94.62    6.0%   + 4.0%   0.0%
+ 1    88.31    5.6%   + 3.6%   0.4%
+ 2    82       5.2%   + 3.2%   0.4%
+ 3    75.69    4.8%   + 2.8%   0.4%
+ 4    69.39    4.4%   + 2.4%   0.4%
+ 5    47.31    3.0%   + 1.0%   1.4%  --> 1% jump
+ 6    45.73    2.9%   + 0.9%   0.1%
+ 7    44.15    2.8%   + 0.8%   0.1%
+ 8    42.58    2.7%   + 0.7%   0.1%
+ 9    41       2.6%   + 0.6%   0.1%
+10    39.42    2.5%   + 0.5%   0.1%
+11    37.85    2.4%   + 0.4%   0.1%
+12    36.27    2.3%   + 0.3%   0.1%
+13    34.69    2.2%   + 0.2%   0.1%
+14    33.12    2.1%   + 0.1%   0.1%
+15    31.54    2.0%   + 0.0%   0.1%

Lucke · 09/08/07, 11:24 AM

I understand the conventional wisdom that 308 hit rating was the cap; however these forums has lately shown that the current miss rating vs. L73 mobs is actually 28% base for dual wielders.

So, to make sure I am still on target:

1% HR = 17 hit?

5 to Weap Skill = 1% hit bonus vs. L73 Mob?

Thus the extra Weap skill = 71.4 HR (rounding down to 71) + the 5% from precision = 156 base hit rating.

28% base chance to miss on a 73 mob = 476 *base* HR needed to reach the "never miss" cap.

476-156 = 320, or 269 if a Boomkin is in the raid with Imp. FF.

I am looking for a math-check; if I understand the new finds correctly, we actually need the above+320 HR/269 w/Boomkin to reach the "never miss" target?

Any information on helping this information become clearer is appreciated; I would imagine any explanation on the above will help the lightbulb go off for more than just myself

Olgas · 09/08/07, 11:36 AM

Originally Posted by Lolaan

As far as the hit goes, this looks very similar to the wowwiki theory. The current working theory here is that its 0.4% per point with a 1% bonus when you switch over to the second interval of the piecewise function. Have we disproved with our testing that it is in fact 0.4% with a 1% bonus and not a 0.2% with a 2% bonus? Its been a bit since I read the entire thread so I just wanted to check.

Lolaan,

Yes, a .2% theory has been definitively disproved.

And, please read the whole thread before making a post like the one you did. Your post was not constructive to the analysis being done here.

Olgas · 09/08/07, 11:41 AM

Originally Posted by sp00n

Jaaaaaaa!

After 4048 attacks I finally got the miss.
500 attacks with imp FF / dreanie aura seems quite possible, so that sucker absolutely pushed it to the limit.

I think we can now agree that the value 1 point of weapon skill gives is 0.4% for the first 5 points.

Awesome news, Sp00n! You did it man!

We now have the complete formula for determining miss rate! Congratulations!

Your tables are a great way of expressing how it works. But, just for the record, I'll also express it using the formulas:

When Def Skill - Weapon Skill <= 10:
5% + (Def Skill - Weapon Skill)*.1%

When Def Skill - Weapon Skill > 10:
7% + (Def Skill - Weapon Skill - 10)*.4%

Olgas · 09/08/07, 11:47 AM

Originally Posted by Lucke

I understand the conventional wisdom that 308 hit rating was the cap; however these forums has lately shown that the current miss rating vs. L73 mobs is actually 28% base for dual wielders.

So, to make sure I am still on target:

1% HR = 17 hit?

No. At Level 70, +1% Hit is equal to 15.76923 Hit Rating. (You will usually see this shortened to 15.8)

5 to Weap Skill = 1% hit bonus vs. L73 Mob?

No. If you are a Level 70 player with 350 Weapon Skill, increasing your Weapon Skill to 355 will effectively give you a +3% hit bonus against Level 73 mobs/bosses.

Stock · 09/08/07, 11:49 AM

Originally Posted by sp00n

This are now how the tables look:

Target 3 levels higher, +5% Hit (e.g. through Precision)

Skill    Hit   Miss    Total    Decrease
+ 0   362.69   23.0%   + 4.0%   0.0%
+ 1   356.39   22.6%   + 3.6%   0.4%
+ 2   350.08   22.2%   + 3.2%   0.4%
+ 3   343.77   21.8%   + 2.8%   0.4%
+ 4   337.46   21.4%   + 2.4%   0.4%
+ 5   315.38   20.0%   + 1.0%   1.4%  --> 1% jump
+ 6   313.81   19.9%   + 0.9%   0.1%
+ 7   312.23   19.8%   + 0.8%   0.1%
+ 8   310.65   19.7%   + 0.7%   0.1%
+ 9   309.08   19.6%   + 0.6%   0.1%
+10   307.5    19.5%   + 0.5%   0.1%
+11   305.92   19.4%   + 0.4%   0.1%
+12   304.35   19.3%   + 0.3%   0.1%
+13   302.77   19.2%   + 0.2%   0.1%
+14   301.19   19.1%   + 0.1%   0.1%
+15   299.61   19.0%   + 0.0%   0.1%

Thank you very much, thats exactly what i was looking for. So with 365 Weapon Skill there would be a 300+hit Cap. So since i'm planning on having 366, im guessing it would be a 298 Hit cap. (0.1%x15.77 = 1.577, 299.61-1.577 = 298.033)

Anyways, im still debating on "is it worth it" to go that high. Weapon Expertise already gives +10. Going from 360-365 only means about a 7-8 Hit rating bonus.

Those 3 codes you posted is really helpful. Thank you once again.

• Vulajin · 09/08/07, 12:56 PM

At this point I have nothing to contribute except to say this: great job. It's always nice to be learning new things about this game, and you have all done a stellar job unraveling this particular mystery.

Lolaan · 09/08/07, 8:40 PM

Originally Posted by Olgas

Lolaan,

Yes, a .2% theory has been definitively disproved.

And, please read the whole thread before making a post like the one you did. Your post was not constructive to the analysis being done here.

*ahem* I did read the whole thread. In fact, if you had read the next sentence, I said that I read the entire thing. Granted I did not have the whole thing memorized so I probably forgot a few things, my bad. Did you read anything else I posted or are you just pointing out my errors?

How was it non-constructive? We're trying to figure out what weapon skill does. The hit seems nailed down now officially, I just wanted to double check that this theory was in fact disproven. I tried to break it down into from what I read weapon skill appeared to do as shown in my chart.

Now that hit has been nailed down, we need to move on to other things, namely crit, dodge and parry. I was proposing a way to get one section of the dodge and parry equation down. While it may not be perfect, is it not a step in the right direction? It might spark other ideas of how people can test things.

EDIT: P.S. If there is something else I missed and the entire post is truly invalid, I am fully prepared to simply remove it entirely, all I am interested in is finding the correct values for weapon skill, and if I am way off base and the post is not needed, it doesn't need to clutter up the thread. Note however, that there was a lot of information in the post, be sure it is all invalid before I do this please.

I can also shorten it to remove the incorrect data. Either way works.

seminarca · 09/10/07, 4:18 AM

Firstly, fantastic job to all the contributors in this thread. I've been following it for a very long time and the fact that you're all going through fairly tedious testing procedures in the name of enlightenment is very admirable. Cheers.

As the numbers for hit get nailed down, and focus shifts to dodge, parry and block, obviously things will get more difficult. It seems to me that part of the ease of testing the effects of weapon skill on hit has either been due to Rogue mace spec allowing steady increments of weapon skill, or being able to hypothesize and disprove by starting at a proposed "capped" hit rating and then working backwards while looking for misses. This can't be done with dodge, for example, as there is no "anti dodge rating" in the game.

Now I really don't know what might be an effective way to go about testing this, so I'll leave it to the experts. What I can offer the thread is tank data from combatlogs over the span of a few months (hope that's a decent sample size) covering Gruul, Mag and 1st half of SSC/TK. Aside from minimal to moderate differences in ap/crit/hit all of these logs will either be with +5 feral combat skill (+24 feral combat skill rating), or with +0. So I don't know how useful they will be for tracing down the effects on dodge, parry and block.

But if you guys think this data might be useful I can pester our WWS guy into sending me the logs and pull out the total number of hit/miss/crit/dodge/parry/block etc on a boss by boss basis (to filter out fights with adds etc).

sp00n · 09/10/07, 5:07 AM

Combat log from tanks would be helpful.
For dodge, you can basically take any fight if you know your weapon skill rating, but for parry, the fight has to be rather 'tank and spank', where you are almost 100% in front of the target.
Morogrim comes to my mind, as he only turns when selecting targets for his watery graves.

Let's see:
- Gruul: maybe good, depending on where the off tank stands
- Magtheridon: he can't dodge/parry while being banished, can he?
- Hydross: bad for parry due to his two phases
- Lurker: might be good
- Morogrim: could be the best fight
- Karathress: samples possible only on Karathress, the adds are level 71 (but otherwise good I suppose)
- Leotheras: bad due to his shadowform, but I think his name changes when he transforms, so data only for his human form should be ok (despite his whirlwind)
- Lady Vashj: good I suppose when you filter out the immune messages

- Voidreaver: depends on where the tanks stand and how often he turns
- Al'ar: again, depends on where the tanks stand during the debuff, ok for dodge

For other bosses I don't know, as that is our current progression.
In Karazhan, most of the fights there should be ok (despite Netherspite maybe, as there were some comments on his weird behavior).

koaschten · 09/10/07, 6:01 AM

If you are looking for Tank and Spank encounters to get data from tanks, Mother Shahraz and Teron Gorefiend are as static as it can become. Naj'entus perhaps, but he got this stupid shield and Gurtogg's Fel Rage screws data up.

Giske · 09/10/07, 9:06 AM

Anetheron is fairly static. Havent seen Kaz'rogal yet, but I imagine the war stomp ruins that one. Azgalor? Seems pretty tank and spank.

Gruntle · 09/10/07, 9:30 AM

Spoon, great work with nailing the 0.4%. Real dedication in getting that miss

When it comes to crit, dodge and parry, I'd just like to mention that the crit rate seen in the pala test by the paladin with 365 rating is still way too low. We discussed this a few pages ago but never really reached any conclusion. Naively one would expect the crit rate to match the tooltip value when you have 365 weapon skill (vs a lvl 73 mob), but in the paladin test the crit rate was about 3.3% (with a tooltip value of about 6%). Comparing with the results for the paladin with 350 weapon skil indicates that the crit per weapon skill is indeed close to 0.1 per skill (subject to fairly large uncertainties, but consistent anyway), but the fact that base crit seems to be lower than expected is still strange to me.

Nakhed · 09/10/07, 10:53 AM

Originally Posted by sp00n

Combat log from tanks would be helpful.
For dodge, you can basically take any fight if you know your weapon skill rating, but for parry, the fight has to be rather 'tank and spank', where you are almost 100% in front of the target.

What kind of stipulations would a test like this require on the tank, id like to help out but im not sure if my gear would meet the type of requirements. Ive got +50HR in my tanking gear with the base 350skill. The best i can do instance wise would be pretty much any boss in Kara. I DPS in Grulls so i wont have a tank log for that one. Let me know if i can help out.

sp00n · 09/10/07, 11:18 AM

Originally Posted by Nakhed

What kind of stipulations would a test like this require on the tank, id like to help out but im not sure if my gear would meet the type of requirements. Ive got +50HR in my tanking gear with the base 350skill. The best i can do instance wise would be pretty much any boss in Kara. I DPS in Grulls so i wont have a tank log for that one. Let me know if i can help out.

First we'd have to discover the 'true' dodge and parry ratings without any additional weapon skill.
For this, we need large sample sizes because of the randomness, preferably from mobs that do not turn around (e.g. secondary target ability), do not run away (Leotheras whirlwind) and don't have cases where they can't dodge or parry (casts?).

Furthermore we also should test out if there are any cases at all where dodge and/or parry is prevented. Maybe it is even dependent on the actual skill a mob uses, so not all channeled skills prevent dodge and parry.
[I know of no tests so far, only can remember of some discussions on whether players can dodge/parry while 'casting' (or drinking, or weapons not sheeted, etc)].

For dodge alone, I think Gruul would be ok. IF he can dodge while casting his shatter.

Crit is a completely different issue. You would need to be running with the exact same gear and the same group composition (i.e. either with or without a feral druid) and without any procs (Mongoose) for quite a few fights to gather enough data. The very low crit rate of the Paladins still worries me.

Charsi · 09/10/07, 12:37 PM

I could probably do it myself, but sp00n, would it be possible to include a table with the fury warrior 3% precision (as well as the rogue 5% precision).

sp00n · 09/10/07, 1:25 PM

Originally Posted by Charsi

I could probably do it myself, but sp00n, would it be possible to include a table with the fury warrior 3% precision (as well as the rogue 5% precision).

I've updated my posting (and also included 1h/2h/hunter with 3% hit through talents).

Nite_Moogle · 09/10/07, 1:28 PM

Originally Posted by sp00n

I've updated my posting (and also included 1h/2h/hunter with 3% hit through talents).

I'm swiping at least one of these and stuffing it in the upcoming EJ wiki.

XP-Dolphin · 09/10/07, 2:52 PM

Originally Posted by sp00n

Combat log from tanks would be helpful.
For dodge, you can basically take any fight if you know your weapon skill rating, but for parry, the fight has to be rather 'tank and spank', where you are almost 100% in front of the target.
Morogrim comes to my mind, as he only turns when selecting targets for his watery graves.

Let's see:
- Gruul: maybe good, depending on where the off tank stands
- Magtheridon: he can't dodge/parry while being banished, can he?
- Hydross: bad for parry due to his two phases
- Lurker: might be good
- Morogrim: could be the best fight
- Karathress: samples possible only on Karathress, the adds are level 71 (but otherwise good I suppose)
- Leotheras: bad due to his shadowform, but I think his name changes when he transforms, so data only for his human form should be ok (despite his whirlwind)
- Lady Vashj: good I suppose when you filter out the immune messages

- Voidreaver: depends on where the tanks stand and how often he turns
- Al'ar: again, depends on where the tanks stand during the debuff, ok for dodge

For other bosses I don't know, as that is our current progression.
In Karazhan, most of the fights there should be ok (despite Netherspite maybe, as there were some comments on his weird behavior).

Theras and I did parses before that showed that weapon skill seemingly had no effect on parry/block, but about double effect on dodge from expected. Give me some time to find that data, it might make the efforts on exactly nailing down values a bit easier. We just used a +15 weapon skill difference on our tests, though. No incremental increases.

Blah, same thread, lots of pages between. Sorry, I guess I should read up on the latest stuff. One of our rogues linked to a specific post here and I hadn't caught what has happened since.

sp00n · 09/10/07, 6:35 PM

Originally Posted by XP-Dolphin

Theras and I did parses before that showed that weapon skill seemingly had no effect on parry/block, but about double effect on dodge from expected. Give me some time to find that data, it might make the efforts on exactly nailing down values a bit easier. We just used a +15 weapon skill difference on our tests, though. No incremental increases.

Blah, same thread, lots of pages between. Sorry, I guess I should read up on the latest stuff. One of our rogues linked to a specific post here and I hadn't caught what has happened since.

Oh, good you're back. Do you have any information whether you were using specials as well in this fight or only white attacks, do you remember if you encountered dodges while she was casting and, by any chance, do you still have the combat log for that fight?

Originally Posted by Nite_Moogle

I'm swiping at least one of these and stuffing it in the upcoming EJ wiki.

There's going to be an EJ wiki? Wow.

Dontmindme · 09/10/07, 6:43 PM

If anyone has some Gruul combat logs laying around (with time stamps preferred), it shouldn't be too hard to locate the shatters and look for dodges and parries right before the casts.

I think the first step is trying to figure out why different bosses seem to be exhibiting different dodge and parry rates. If we find what's causing this, we might be able to figure out a way to work around the bad points (like ignoring everything X secs before a Shatter, for example).

I believe that Shatter is the only cast for Gruul, making him a very good candidate for such preliminary analysis.