Min/Maxing: Are these extra 1000 DPS increases worth it?


One of my recent theorycrafting thoughts was on looking at APL’s and reducing the complexity to see how this alters DPS output. The idea is that a less complex rotation will be easier to use in-game, and maybe even counter-intuitively increase damage.

Brain Load

For any player there will be a maximum of things that you can pay attention to at one time, ie: how well you multi-task. To throw some arbitrary numbers out, if a rotation requires 75% of your concentration this means if the encounter requires more than 25% of the rest of your concentration you’re either going to mess up your rotation, the mechanics, or both. To help with both sides of that equation we have raid addons like DeadlyBossMods or BigWigs to deal with encounter mechanics, and a wide array of custom Weak Aura profiles to help with your rotation.

Maelstrom Management

If you can reduce the complexity around using a single ability then you can offload that into Weak Auras and just go “hit button X when thing is glowing”. One example of that is Earth Shock for Elemental, which has a couple of use lines with different conditions on each.

actions.single_asc+=/earth_shock,if=maelstrom>=117|!artifact.swelling_maelstrom.enabled&maelstrom>=92

actions.single_asc+=/earth_shock,if=maelstrom>=111|!artifact.swelling_maelstrom.enabled&maelstrom>=86

The 117/92 line is 5 rows higher than the 111/86 line, and both are designed around minimising maelstrom wastage before & after using Lava Burst. For these I substituted the following line in place of the first entry (higher priority than Lava Burst) and removed the second.

actions.single_asc+=/earth_shock,if=maelstrom=125|!artifact.swelling_maelstrom.enabled&maelstrom>=100

This change reduced the T20M (Non-gamble) profile from 1,286,283 DPS to 1,273,598, a decrease of 12,685 DPS or 0.986%.

This is part of the reason for looking at this: individuals will look for any and all small changes that can be made for small DPS increases, but now that players are passing one million DPS on a regular basis those few thousand adjustments are trivial by comparison. Sure, for the high end raiders who can manage both complex rotations and mythic raid encounters these small changes may be both practical and important for tight DPS checks, but for players with lower progression/skill these adjustments may not be necessary at all.

That 8,923 DPS loss from my tweak to Earth Shock gives roughly 2.7m extra damage over a 5 minute fight, vs the average damage per Earth Shock cast is 2.18m damage.

Arcane Mathematics

It’s a half joke that any APL for Arcane requires some literal arcane knowledge to put together. The current T20M profile can be found here and has 111 lines in the action list, although 38 are just comments. Rather than remove parts of this I’ve had to essentially start again with building a “basic” version at 35 lines (7 comments) which is reasonably close to the default: 127k vs 134k for a difference of ~5.22%. I have yet to add specific stuff to handle legendary effects, different talents or set bonuses but it’s still a good example of how simple APLs can be effective.

[Edit Sept 4th]

I’ve run some further sims looking at stat weights. The default profile has Vers (25.72) > Crit (23.81) > Haste (23.27) > Mastery (21.17) > Int (20.67) but my “simple” APL has Mastery (25.44) > Vers (23.13) > Crit (21.44) > Int (18.28) > Haste (16.48). Given that they’re using the same profile with 10824 Crit, 8868 Haste, 6016 Mastery and 4710 Versatility it suggests a more optimized gear set would narrow the difference between the two profiles.

Sims vs Logs

When comparing sims to logged fights it’s important to remember that the sim “actor” is able to consider a larger number of things at any one point in time. This means that a highly complex rotation is easily handled in any simulation but requires a lot of assistance to master in game, which includes weak aura setups, macros and muscle memory. It’s also important to remember that when comparing the two you shouldn’t just look at the DPS results. If you messed up somewhere but had a higher than expected proc or crit rate then the results could be similar, meaning you’re more likely to think you didn’t screw up (or screw up badly) than you actually did.

Cross class comparisons

Given the differences between these “simple” APLs and the defaults it’s possible that some of the perceived variation between classes and specs in those “stacked sim” charts is just down to how many of these small min/maxing differences have been found, and whether they’ve been applied to the default profiles or not.

Advertisements

About binkenstein
I'm a geek, living in Christchurch, New Zealand.

One Response to Min/Maxing: Are these extra 1000 DPS increases worth it?

  1. Munahe says:

    This is very interesting Binkenstein. Especially the fact that stat weights change so much if you simplify the rotation. I wonder how this relates to actual play.

    Would it be doable to parse combatlogs and map an APL back onto it? Or since thats probably overkill/impossible, calculate your actual stat weights by going over your own actual rotation? (of course, everyone tries to stick to the ‘best’ apl, but in practice it would be more like the simplified apl for most mortal players)

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: