Weapon Enchant

Hi everyone,

I was recently wondering which weapon enchant should I go for. Let me explain a bit.

I'm currently raiding in 10 Man Heroic Throne of Thunder (1/13 Killed, though). As I always felt very comfortable with my survivability, I decided to go for a high DPS optimization a while ago. Therefore, I used to play with Dancing Steel instead of Windsong during 5.1 raids. But now, I'm asking myself if dancing steel is still a relevant choice. Here's why :
- Our Vengeance grew a lot with the new raid's encounters.
- I am also currently raiding partly in heroic, and monotanking normal bosses whenever it's possible.
- I suppose it's useless to remind anyone here that the more vengeance we have, the less strength is valuable.


So here's the question :
Considering how both enchants work (i.e. with the possibility of having multiple procs of windsong at a time), which weapon enchant will provide me the best DPS output ?

For very large values of vengeance, would an additional 200 haste make enough of a difference to make the Living Steel Weapon Chain an option? (The haste would obviously come from not having to get 200 expertise from other gear, allowing you to forge/gem it to haste instead)

Koatanga wrote:For very large values of vengeance, would an additional 200 haste make enough of a difference to make the Living Steel Weapon Chain an option? (The haste would obviously come from not having to get 200 expertise from other gear, allowing you to forge/gem it to haste instead)

Pretty sure near 50% uptime on almost 4k AP is more dps than 200 haste rating (probably slightly less because of how reforging works).

At current Vengeance levels, I don't think Windsong will pull ahead. Windsong is 2 RPPM, Dancing Steel is 2.3 RPPM. As a rough estimate, that means you get 13% uptime on the crit and haste buffs (8 seconds each every minute). That same rough estimate would put Dancing Steel at around 46% uptime (2.3*12/60), though that's an over-estimate due to munching. A more reasonable estimate would be about 37% uptime (1-(1-p)^N, with p=2.3*1/60 and N=12). The same calculation for Windsong drops it to 12.55%, but let's be generous and round up.

So 37% uptime on 1650 Strength vs. 13% uptime on 1500 crit and 1500 haste. Using the DPS stat weights for 150k Vengeance:
STR = 1.466
haste = 2.517
crit = 1.645

0.37*1650*1.466= 895 DPS

0.13*1500*2.517= 491 DPS
0.13*1500*1.645= 320 DPS

So Dancing Steel grants about 895 DPS, while Windsong grants around 811 DPS. That's at 150k Vengeance; below that point Dancing Steel will continue to dominate. At higher Vengeance values Windsong will eventually surpass Dancing Steel.

To estimate that crossover, let's assume that Windsong scales linearly with AP (pretty good assumption). We start with 40k AP from gear, so the base Windsong DPS value is 811*40/190= 171 DPS. The remaining DPS is due to the 150k Vengeance AP, giving (811-171)/150= 4.27 DPS per 1k vengeance. We can check that this is correct: 171+4.27*150=811.5, which is within rounding error (I rounded several values to the nearest DPS along the way). Thus, we want to know at what Vengeance value X we reach 895 DPS, or 171+4.27*X=895. Solving that gives X=170.

So the Vengeance breakpoint at which Windsong becomes higher DPS than Dancing Steel is around 170k Vengeance AP. Note that if you want the most average DPS over a fight, you also need to average your Vengeance value (meaning including periods of off-tanking, where you generally have less Vengeance).

Does that STR stat weight include the 10% AP buff?

Fetzie wrote:Does that STR stat weight include the 10% AP buff?

Yes. On second thought, it also includes Kings, so I actually need to correct that slightly. Fixed.

Well, thanks a lot for all the answers. Since the breakpoint is that high, I guess I'm going to stay with Dancing Steel for Throne of Thunder.

i guess that makes windsong dps enchant of choice for 25m hc tot

theckhd wrote:At current Vengeance levels, I don't think Windsong will pull ahead. Windsong is 2 RPPM, Dancing Steel is 2.3 RPPM. As a rough estimate, that means you get 13% uptime on the crit and haste buffs (8 seconds each every minute). That same rough estimate would put Dancing Steel at around 46% uptime (2.3*12/60), though that's an over-estimate due to munching. A more reasonable estimate would be about 37% uptime (1-(1-p)^N, with p=2.3*1/60 and N=12). The same calculation for Windsong drops it to 12.55%, but let's be generous and round up.

So 37% uptime on 1650 Strength vs. 13% uptime on 1500 crit and 1500 haste. Using the DPS stat weights for 150k Vengeance:
STR = 1.466
haste = 2.517
crit = 1.645

0.37*1650*1.466= 895 DPS

0.13*1500*2.517= 491 DPS
0.13*1500*1.645= 320 DPS

So Dancing Steel grants about 895 DPS, while Windsong grants around 811 DPS. That's at 150k Vengeance; below that point Dancing Steel will continue to dominate. At higher Vengeance values Windsong will eventually surpass Dancing Steel.

To estimate that crossover, let's assume that Windsong scales linearly with AP (pretty good assumption). We start with 40k AP from gear, so the base Windsong DPS value is 811*40/190= 171 DPS. The remaining DPS is due to the 150k Vengeance AP, giving (811-171)/150= 4.27 DPS per 1k vengeance. We can check that this is correct: 171+4.27*150=811.5, which is within rounding error (I rounded several values to the nearest DPS along the way). Thus, we want to know at what Vengeance value X we reach 895 DPS, or 171+4.27*X=895. Solving that gives X=170.

So the Vengeance breakpoint at which Windsong becomes higher DPS than Dancing Steel is around 170k Vengeance AP. Note that if you want the most average DPS over a fight, you also need to average your Vengeance value (meaning including periods of off-tanking, where you generally have less Vengeance).

I'm pretty sure you need to include weapon speed in PPM calculations? or it only for trinkets
Also I believe mastery proc increases dmg through alabaster shield glyph?

Now regarding uptime... I'm looking at logs and just don't get the same numbers
http://www.worldoflogs.com/reports/85zs ... #tab-auras
http://www.worldoflogs.com/reports/85zs ... #tab-auras

Yeh it depends which of 3 is going to proc more, but you always get 1 on at least 11%+, the other two on 15-20, so that's roughly 50-55% combined.
If lucky crit and haste will be around 15%, which I think is a fair increase

ps. don't forget windsong can proc 1-3 at a time, which means you can get a huge dps boost at a time...BL+all cds at start yes yes:)

Weapon speed has no effect on RPPM calculations. It has an effect on old-school PPM effects, but both enchants are RPPM.

The mastery proc's effect on Alabaster Shield is not going to be very large, ~2% higher chance to block is not going to give you a lot more AS charges. Not to mention that you probably won't have AS glyphed for every encounter.

Your logs show ~15%-17% average uptime on any given proc, which is consistent with this calculation. Remember that the 13% here is without any haste, and RPPM effects scale with haste. Both enchants will scale about equally with haste though, which is why I've ignored it.

And sure, they can proc 1-3 at a time during BL+cooldowns... or they could not. It's great if they do line up, not as great when they don't. And all told, Dancing Steel is far more likely to be up during BL because of its higher uptime.

theckhd wrote:Weapon speed has no effect on RPPM calculations. It has an effect on old-school PPM effects, but both enchants are RPPM.

The mastery proc's effect on Alabaster Shield is not going to be very large, ~2% higher chance to block is not going to give you a lot more AS charges. Not to mention that you probably won't have AS glyphed for every encounter.

Your logs show ~15%-17% average uptime on any given proc, which is consistent with this calculation. Remember that the 13% here is without any haste, and RPPM effects scale with haste. Both enchants will scale about equally with haste though, which is why I've ignored it.

And sure, they can proc 1-3 at a time during BL+cooldowns... or they could not. It's great if they do line up, not as great when they don't. And all told, Dancing Steel is far more likely to be up during BL because of its higher uptime.

Yeh, however if both Crit+Haste proc at the same time they contribute to each other and can give higher DPS increase.
It's pure RNG, but I've seen it a lot(mostly at start ofc, because of proc downite). Might switch to WS and configure WeakAuras to show whwn both of them are up to test the chances of it:)

On average, those effects should cancel out. I don't think "but they can line up during BL" is a valid argument since it isn't guaranteed. Plus, while the multiplicative effect is non-trivial, it's also not as big an increase as I think you're imagining it to be.

The best argument for WS is that in heroic modes, 200k+ vengeance is not uncommon. I may switch to it soon myself.

mh.... would be interesting seeing the DPS comparison of WS and DS vs vengeance. Along with the average uptimes at specific haste points in order to really get an impression. I'm closing in on 16k haste rating now so it is starting to get really interesting here.

theckhd wrote:On average, those effects should cancel out. I don't think "but they can line up during BL" is a valid argument since it isn't guaranteed. Plus, while the multiplicative effect is non-trivial, it's also not as big an increase as I think you're imagining it to be.

The best argument for WS is that in heroic modes, 200k+ vengeance is not uncommon. I may switch to it soon myself.

Fair enough. What about WS as def enchant?

It's not a bad choice. I'm not sure any of them really stand out as strong for defense anymore. I liked Colossus, but it also doesn't scale with boss damage, so by heroic T15 content it's falling behind. May still be a better smoothing enchant than Windsong, but probably not by much if at all. And the dps trade from going to Dancing Steel or Windsong is pretty huge.