[Prot] Dancing Steel and Windsong scaling

I've been seeing a few progression (and even some top-end) Prot Paladins turning to Windsong instead of Dancing Steel, and one idea that I've heard from folks is that because Dancing Steel refreshes its duration if it procs again while already active, it actually can lose ground as you gain more and more Haste rating. (Haste increases the proc rate of anything on the RPPM system - trinket procs, weapon enchants, even those lovely legendary metas.) Windsong procs one of three different secondary stats, so it doesn't have that effect - if it procs again immediately then you get two different procs of different stats. No problem!

I wound up testing this out in simcraft. I had to check it out as Retribution because the Prot one seems to default to some Cataclysm settings (I doubt a lot of tanks would be using SimCraft given all the variables involved in our damage output anyways), but this is what I found when comparing the weapon enchant uptimes while juggling Haste values up and down.

10671 haste rating

windsong - 19.46%, 19.33%, 19.35%
dancing steel - 53.87%

14174 haste rating

Windsong - 20.15%, 20.27%, 20.39%
dancing steel - 56.23%

16174 haste rating

windsong - 21.13%, 20.65%, 20.61%
dancing steel - 57.46%

21174 haste rating

windsong - 22.45%, 21.87%, 22.62%
dancing steel - 60.68%

The three percentages for Windsong track each of the individual buffs. Remember that the Crit one is pure damage (1500 Crit rating is a pretty substantial amount of DPS for us actually), and the Mastery/Haste ones are what we could care about for survival. I actually would think that a 12 second Haste proc would be worse than advertised for a Paladin since our benefit from Haste revolves around spell cooldowns and the global cooldown - it would still be quite good, but there would be a little weirdness around the beginning and end of the proc.

But anyways. I don't actually *see* a RPPM scaling issue here. The Windsong uptimes certainly are outpacing the Dancing Steel uptime if you directly compare the first and final figures, but it isn't particularly drastic. The combined Windsong uptimes went up by 8.8%, compared to the Dancing Steel uptime increasing by 6.87%.

21k Haste isn't achievable this tier (I think? maybe with item upgrades, full BiS haste gear, elixir, etc.) so that's really a good point to stop at imo.
Windsong actually can overwrite itself, as far as I know, but only if it procs the same stat that's already up.
Windsong actually can overwrite itself, as far as I know, but only if it procs the same stat that's already up.

Hmm. I figured it would default to proccing a different stat (shows you I did a WHOLE TON of research).

Still, it would have much much better odds of not doing this.
Yeah, I don't think the refreshing has any real impact on the scaling of either enchant. Theck did a comparison of the different enchants back in 5.1 (after they all went on the rppm system) and he basically found dancing steal to be the least attractive enchant for tankadins (even behind river's song). In comparison to River's Song, river has double the base proc rate of dancing, and it actually does have protection from overlapping procs because river's song can actually stack up to 2 (which basically means a second proc actually increases the duration of the first instead of decreasing it). And Windsong beat both of them in terms of TDR. And I don't really see any reason why increased haste levels would change their order (unless it got high enough that one of them had a near 100% uptime, which might favor river slightly).

The more interesting comparison was Windsong vs Colossus. Where he found that in Heroic T14 content Colossus was still typically far ahead of Windsong in TDR, and that even in peak damage intake periods they were roughly equal. And Colossus had the extra edge of being a more consistent/smooth source of damage prevention. Which basically made Colossus the hands-down best tanking enchant last tier.

Now I'm sure that boss dps has scaled enough in heroic 25 ToT that Windsong has pulled ahead in TDR (though I'd like to see by how much, it's haste-dependent and I'd attempt to rework through Theck's math, but I'd probably screw it up somewhere), it's just a matter of if it's ahead by enough TDR to outweight the smoothing aspect of Colossus. Theck himself appears to be using Windsong now, so I imagine it does; but I'd still like to see a revisit of the analysis given more current haste levels and boss damage to see by how much (in partcular I'm personally interested in whether the damage levels in 10n are enough to favor windsong now).
Honestly, the damage reduction provided by ANY of the enchants is pretty minor overall. If I was going to pick between minor damage reduction and minor damage reduction plus minor damage boost, I'd take the latter.
06/04/2013 09:22 AMPosted by Pancakê
Theck himself appears to be using Windsong now, so I imagine it does

Honestly? I haven't been paying too much attention to what Theck is or is not not saying/doing.

Apparently he's stacking huge amounts of Stamina gems (like previously) and then using the... DPS meta. Okie doke.
Apparently he's stacking huge amounts of Stamina gems (like previously) and then using the... DPS meta. Okie doke.

The legendary DPS meta increases damage output by 10% or more. Its contribution to fights is generally considered more significant than the tank meta.
I disagree (for 25H progression), and it also runs completely counter to the logic he was using to justify gemming Stamina over Haste.

But I mean, he can do his thing. :-P
And yes to back up Keten's earlier comment, Windsong can proc the same stat increase twice and refresh that buff's duration instead of giving a different stat.

Join the Conversation

Return to Forum