DW Builds 3.2 -Revenge of the Offhand - Page 53

Astalion · 08/03/09, 7:54 PM

@doc: Did you try the priority I listed above? Having PS and GF that high on the priority only seems to serve to clip BP ticks, risk using death runes for GF, and generally use a GCD for low damage/RP return - and the dps difference shows in my tests at least.

Krevak · 08/03/09, 8:19 PM

Using what Doc provided I was getting the same results that he is, sitting around 6600~ DPS with PS and DC sitting at 4%/16% respectively.

However, I changed the priority around a bit, moving DC up above IP and PS and this boosted the DPS to consistently 6900.

<?xml version="1.0"?>
<Priority>
	<FrostFever></FrostFever>
	<BloodPlague></BloodPlague>
    <GhoulFrenzy></GhoulFrenzy>
    <DeathCoil></DeathCoil>
    <IcyTouch></IcyTouch>
    <PlagueStrike></PlagueStrike>
    <BloodStrike></BloodStrike>
</Priority>

Ability	Total	%	Landed	Hit%	Crit%	Miss%	Average
Plague Strike	79283117	3	49513	50	49	0	1601
Icy Touch	284522983	11	118962	67	32	0	2391
Blood Strike	55992533	2	29406	57	42	0	1904
Death Coil	546307423	21	96902	59	40	0	5637
UB	109260347	4	96902	100	0	0	1127
Frost Fever	55680424	2	52898	100	0	0	1052
Blood Plague	98397305	3	93408	100	0	0	1053
Necrosis	114334950	4	345430	100	0	0	330
Blood Caked Blade	85223051	3	103410	100	0	0	824
Wandering Plague	64151794	2	60910	100	0	0	1053
Main Hand	362841107	14	169422	42	43	14	2141
Off Hand	208667527	8	176008	42	42	14	1185
Ghoul	292817443	11	399972	87	12	0	732
Gargoyle	137696400	5	41400	86	13	0	3326

DPS 6931
Total Damage 2495176404 in 100 h
Threat Per Second 4643
Generated in 210s
Template : DW Unholy.xml(C:\Documents and Settings\jthines\Desktop\Temp2\DKSimulator0.9.8\Templates\DW Unholy.xml)
Priority : DW Unholy 2.xml(C:\Documents and Settings\jthines\Desktop\Temp2\DKSimulator0.9.8\Priority\DW Unholy 2.xml)
Presence :Unholy
Sigil of :VengefulHeart
RuneForge :FallenCrusader / FallenCrusader
Pet Calculation :True

Not sure if this is the case, but it did provide a noticeable difference.

sweberry · 08/03/09, 8:24 PM

@Döc: That's the reason everyone is using the statset added in the simulator here for. It's not that weird that you get different numbers if you use a different statset. And obviously playing DC higher is not a DPS loss, it seems.

I'm not gonna try to simulate the 17/54 as it's not a spec I like anyway, but just wonder if anyone has tried adding the <DeathCoilMaxRp> into the priority?

Krevak · 08/03/09, 8:43 PM

Originally Posted by sweberry

the 17/54 as it's not a spec I like anyway, but just wonder if anyone has tried adding the <DeathCoilMaxRp> into the priority?

Interestingly enough, I was able to squeeze a little bit more out of the spec. Same stuff as before but I modified the priority a bit.

<?xml version="1.0"?>
<Priority>
	<FrostFever></FrostFever>
	<BloodPlague></BloodPlague>
    <GhoulFrenzy></GhoulFrenzy>
    <DeathCoilMaxRp></DeathCoilMaxRp>
    <IcyTouch></IcyTouch>
    <PlagueStrike></PlagueStrike>
    <BloodStrike></BloodStrike>
    <DeathCoil></DeathCoil>
</Priority>

Obviously with DC ranked normally above all the rune abilities save GF you'd not need to use that so by adding that in the old DC spot I dropped DC down to the bottom again.

Ability	Total	%	Landed	Hit%	Crit%	Miss%	Average
Plague Strike	83407848	3	53324	51	48	0	1564
Icy Touch	310053273	12	124963	67	32	0	2481
Blood Strike	62109071	2	33369	57	42	0	1861
Death Coil	483448927	19	83220	59	40	0	5809
UB	96699715	3	83220	100	0	0	1161
Frost Fever	53585488	2	48874	100	0	0	1096
Blood Plague	98768593	3	90011	100	0	0	1097
Necrosis	112021757	4	347603	100	0	0	322
Blood Caked Blade	84267758	3	104266	100	0	0	808
Wandering Plague	61966306	2	56488	100	0	0	1096
Main Hand	355450447	14	173950	43	41	14	2043
Off Hand	204256769	8	173653	43	42	14	1176
Ghoul	357906914	14	396828	87	12	0	901
Gargoyle	158001076	6	45586	86	13	0	3466

DPS 7005
Total Damage 2521943942 in 100 h
Threat Per Second 4516
Generated in 203s
Template : DW Unholy.xml(C:\Documents and Settings\jthines\Desktop\Temp2\DKSimulator0.9.8\Templates\DW Unholy.xml)
Priority : DCMaxRPTest.xml(C:\Documents and Settings\jthines\Desktop\Temp2\DKSimulator0.9.8\Priority\DCMaxRPTest.xml)
Presence :Unholy
Sigil of :VengefulHeart
RuneForge :FallenCrusader / FallenCrusader
Pet Calculation :True

I ran it a few times and this about where it averages out.

dr_AllCOM3 · 08/03/09, 8:50 PM

Originally Posted by Fargom

1. Kithus did a 550 hour sim, Doc did a 100 hour sim.
2. Kithus used 150 latency, Doc used 200.
3. Kithus's plague strike was 2% of total, DC was 20%. Doc's PS was 4%, DC was 17%.
4. Do you attribute this gain simply to the new priority you came up with?

1. Shouldn't make a difference.
2. I used 150ms, as I wrote in my post.
3.+4. Must be due to my slightly different priority. Also I don't know what char he used and I don't have the statset2 anymore.

Originally Posted by sweberry

@Döc: That's the reason everyone is using the statset added in the simulator here for.

I use a real char from the live server. Doesn't really matter, if you just look at the dps difference and not the total value. Theorycrafting is more than pressing a button

.

Originally Posted by Krevak

Interestingly enough, I was able to squeeze a little bit more out of the spec. Same stuff as before but I modified the priority a bit.

I get -40dps with such a priority.
Do you use the DWsetstat2 char? It's not that good actually. It has less Arp and more Str than mine. Might explain the DC/PS differences.

sweberry · 08/03/09, 9:38 PM

@Döc:

I agree that the stats on the setstat2 seems a bit odd, particularly the high strength, I think it's supposed to be "a bit into 3.2"-stats, but I'm not sure.

Regardless, you would get the point of the statset if you got off your high horse for a while. The point is to be able to share your results with a fellow theorycrafter without having to figure out if his dps is higher than yours because of his gear, and how much is because of the gear. Or if as in this case you sim it and simply state "this is correct" we can do nothing but assume this as we cannot duplicate it in your gear (we could, but it would be fairly annoying for everyone to do). Thus you must see that when you suddenly post something that is not simmed with your own stats everyone will wonder why your numbers are off.

Basically, to help it would be much preferable if you used the statset while we're still in the state of comparing specs, then you can personally sim it with your own gear for your own. It could always be discussed if to change the statset or add another one depending on spec, but I really don't think it's that important.

Krevak · 08/03/09, 9:58 PM

Originally Posted by dr_AllCOM3

I get -40dps with such a priority.
Do you use the DWsetstat2 char? It's not that good actually. It has less Arp and more Str than mine. Might explain the DC/PS differences.

Ahhh yeah that was it. DWsetstat1 and DWsetstat3 both were consistent around 6500 DPS. Oh well so much for that thought.

dr_AllCOM3 · 08/03/09, 10:00 PM

I'm not on any high horse. I'm well aware of the benefits of a common statset, but a random one with made up numbers it provides no benefit to me (or to anyone else). I'd rather use a statset I know well.
You can continue using setstat2, but it won't work for comparing different specs. The itemization for the three trees is too different. I actually have seperate statsets for each tree. Modifying my spreadsheet makes that an easy task.

Originally Posted by Krevak

Ahhh yeah that was it. DWsetstat1 and DWsetstat3 both were consistent around 6500 DPS. Oh well so much for that thought.

On further inspection I think set 2 and 3 are in the wrong order. It's 1700->1900->1800 Str right now.

Dauraane · 08/04/09, 12:25 AM

I was looking at the talents for 0/53/18 and was wondering. Why not get UB instead of IIT? Since most raids will probably have a windfury totem, does the 5% haste from IIT out do the 2min CD UB talent?

Nobble · 08/04/09, 3:36 AM

Originally Posted by Dauraane

I was looking at the talents for 0/53/18 and was wondering. Why not get UB instead of IIT? Since most raids will probably have a windfury totem, does the 5% haste from IIT out do the 2min CD UB talent?

I believe if we are not going to provide IIT it is better that we spec for the Glyph of disease (though personnally I will be sticking with IIT 10 man raiding means every buff anyone can provide to the raid is a useful buff..)

@Doc isn't awareness a better glyph for Unholy as Orlgin illustrated in post 760

Astalion · 08/04/09, 4:41 AM

Originally Posted by Nobble

@Doc isn't awareness a better glyph for Unholy as Orlgin illustrated in post 760

It depends on the spec - 0/18/53 uses blood presence and Obliterate, and benefits more from awareness, 0/17/54 uses unholy presence and is mainly about spamming death coil and single rune attacks, and doesn't even use any of the abilities on the awareness sigil.

fled · 08/04/09, 8:24 AM

Tried to back track a several pages however I cant get a solid answer here:

KM/Rime proc use the HB ability?
The RP dump is FS?

Seeing that today is patch day, I just wanted to be crystal. I know im risking hand holding but being new to frost, everyone is talking about using KM/Rime simultaneously but not stating what ability is to be used on the proc(S).

Astalion · 08/04/09, 8:43 AM

@fled: Yes, that's basically it - my testing suggest that you want to use your Rime procs even if KM hasn't procced yet just before the next Obliterate in order not to waste potential extra Rime procs, but that was using Glyph of Disease, I saw someone suggest just skipping Rime procs entirely if KM isn't up when you're using other Glyphs (well, HB glyph would be different), but the difference there should (I think) be more or less neglible.

phantazum · 08/04/09, 9:40 AM

Originally Posted by Astalion

@fled: Yes, that's basically it - my testing suggest that you want to use your Rime procs even if KM hasn't procced yet just before the next Obliterate in order not to waste potential extra Rime procs, but that was using Glyph of Disease, I saw someone suggest just skipping Rime procs entirely if KM isn't up when you're using other Glyphs (well, HB glyph would be different), but the difference there should (I think) be more or less neglible.

Only using Rime when KM procs is a rather large dps boost. You would be surprised at how much difference it makes. Basically its the only attack worth interrupting your rotation for. There is no point using rime if KM hasn't proced because frost strike and obliterate do more damage. The rotation is just right for this small interruption, more interruptions like normal Rimes will only lower your dps because you are delaying obliterates or frost strikes.

Astalion · 08/04/09, 9:44 AM

@Phantazum: I wasn't talking about actually interrupting the rotation, in a Disease rotation you'll have enough time for RP dumping to take one GCD "off" every now and then to use Rime, even without KM - my testing show's it's better than letting the Rime proc rot.

@Odii: I've been working on the priority/rotation system for HB, and it's looking very promising. I think I might be able to make it work a bit better though, I'll post my results when I'm done messing around with it :P