Archived

This topic is now archived and is closed to further replies.

Caltiom

SimulationCraft for ShadowPriests (PandaEdition)

11 posts in this topic

There now is a Shadow Priest report available at simulationcraft.org. It will be updated automatically within a few hours of code changes.

beta_chart.php?Priest_Shadow_T13H

beta_chart.php?Raid_T14H

The source code is available here: sc_priest.cpp , and the profiles here: Priest_Shadow_T13H.simc and Priest_Shadow_T14H.simc.

Watch the code changelog and profile changelogs: Priest_Shadow_T13H.simc, Priest_Shadow_T14H.simc for updates to the SimC priest model and the release notes for general changes to SimC.

Share this post


Link to post
Share on other sites

The current report is using the old 4.3 profile, slightly modified to now have 15% spell hit instead of 17% and reforged out of mastery as much as possible ( gear_mastery_rating=620 ).

I also want to give credit to Althor, who mainly developed the ShadowPriest module, double checked all the mechanics and fine tuned the action list and much else.

All the classes have undergone massive changes in MoP and there has only been limited development so far in SimC, so there are surely some bugs in there. We would greatly appreciate it if as many people as possible could go over the results presented here. If there is anything that looks wrong, please let us know.

Share this post


Link to post
Share on other sites

While discussing the exact behavior of Dragonwrath Tarecgosa's Rest on spells with both direct and tick damage components, Althor stumbled over a problem this causes on Devouring Plague, our biggest damage source. If the dd part of DP gets duplicated, it refreshes the dot, as one would expect. The problem now is that the DTR DP is executed with 0 Shadow Orbs, which results in a dot with 0 tick damage. The refreshed dot ticks for the expected 6seconds, each second, but just does no damage.

Implementing this in SimC resulted in a 850 dps loss, and especially reduces the benefit of DTR. But quickly comparing to normal weapons shows DTR still being clearly superior.

There is a new 501-3 release out there as well.

Edit: All damage relevant talents are implemented. 'Surging Darkness', 'Mindbender', 'Archangel', 'Twist of Faith', 'Power Infusion' and 'Divine Insight'. You can always check the source code to verify the implemented behavior.

The current release is focused at level 85, the level 90 talents are not yet implemented. They'll need a bit of testing when level 90 is available in Beta.

Share this post


Link to post
Share on other sites

So, is the health chart in the resources section accurate? Is, presumably, Shadow Word: Death backlash causing so much damage compared to Devouring Plague's self-heal that the simulated Priest is dying before the fight ends?

Share this post


Link to post
Share on other sites
So, is the health chart in the resources section accurate? Is, presumably, Shadow Word: Death backlash causing so much damage compared to Devouring Plague's self-heal that the simulated Priest is dying before the fight ends?

No, it is not yet accurate. We just started looking at the SW: D backlash implementation, and it was still using the old cataclysm one.

Our findings so far:

- Backlash isn't increased when you crit with SW: D

- There is always a backlash even if you miss with SW: D

- The backlash damage is slightly higher than the damage you do with SW: D. Apparently it gets calculated with a higher coefficient than the damage part.

- Tier13 2pc only reduces backlash damage by 1/3 and not by 95%.

The updated report, including a detailed "action" tracking the backlash, should be up in the next few hours.

Share this post


Link to post
Share on other sites
No, it is not yet accurate. We just started looking at the SW: D backlash implementation, and it was still using the old cataclysm one.

Our findings so far:

- Backlash isn't increased when you crit with SW: D

- There is always a backlash even if you miss with SW: D

- The backlash damage is slightly higher than the damage you do with SW: D. Apparently it gets calculated with a higher coefficient than the damage part.

- Tier13 2pc only reduces backlash damage by 1/3 and not by 95%.

The updated report, including a detailed "action" tracking the backlash, should be up in the next few hours.

Looking at the updated report, it seems like Shadow Priests run an HP deficit during the execute phase-- the self-damage from SW:D outstripping Devouring Plague's self-heal, even with 2PT13. Am I interpreting that correctly?

Share this post


Link to post
Share on other sites
Looking at the updated report, it seems like Shadow Priests run an HP deficit during the execute phase-- the self-damage from SW:D outstripping Devouring Plague's self-heal, even with 2PT13. Am I interpreting that correctly?

Yes you're interpreting that correctly.

Just looking roughly at the resource timeline, the priest ends the fight with 60-70% health, without T13_2pc that might come out at around 50%. I think that is pretty generous for a ~1.5min execute phase, compared to the pre-T13 era. The backlash is also much more stable, especially not punishing critical strikes smooths out the damage quite nicely. Taking the numbers from the simulation, the backlash hits between 9'900 and 11'800 damage.

Share this post


Link to post
Share on other sites

Work on the Level 90 talents is progressing. Even though Cascade probably is the most complicated action we ever put into SimulationCraft, it seems to work fine now. Ghostcrawler was so kind to publish the exact formulas for damage calculation for Cascade and Halo in respect to the distance of the target in Beta Class Balance Analysis - Forums - World of Warcraft . Those two spells are now working in SimC and are ready for testing. We hope to have more extended AoE profiles and reports at some point, even though the not-very good distance/position managment will make that complicated, especially for Cascade.

There are some Halo graphs at Beta Class Balance Analysis - Forums - World of Warcraft and Beta Class Balance Analysis - Forums - World of Warcraft .

Share this post


Link to post
Share on other sites

GC posted the details on how Shadow Word: Insanity works:

The damage scales based on the percentage of ticks that have already gone off from the spell (specifically, for each DoT, the damage is increased by (# ticks elapsed / (total ticks – 1)) * 100%, so for a base SW:P which would tick 6 times, SW:I gains 20% damage per tick elapsed)

I'm curious as to whether a sequence based around using SW:P and VT, in that order (to help their durations line up), and then Shadow Word: Insanity right before they expire will even prove to be a DPS gain over using your DOTs normally, let alone on par with Mindbender and From Darkness.

Share this post


Link to post
Share on other sites

Latest findings about Shadow Word: Insanity:

- Damage modifiers from multiple dots are additive to one another.

- It can proc Dragonwrath Tarecgosa's Rest. Similar to Devouring Plague, such a proc will then behave as if there are no dots on the target, and thus will not get the damage bonus from the SW: Insanity Mechanic.

- Dots get instantly removed now. ( For Mind Spike as well ). Previously there was a short delay on this.

According to Althor and some quick tests he did, it really isn't worth to use SW: Insanity ever, except possibly shortly before a monster dies. I'm hoping he can give us more detailed numbers on this.

I just released SimC-501-8, so you should be able to test any further action priority list ideas yourself.

Ps. 501-8 includes Holy and Discipline modules as well. We only have old T12 profiles, but the underlying simulation model should be complete and correct for MoP. So if anyone is interested, feel free to play around with it.

Edit:

Some Information about Devouring Plague behaving similar to Ignite:

After Ghostcrawler mentioned that DP behaves like Ignite, we just tested this a bit further to check if there is a possibility that no damage would ever be lost with it. Unfortunately this still is not the case.

As an example, let's assume a DP tick does 100 damage per tick per orb.

If you cast DP with 3 orbs, you will get 6 ticks a 300 damage. Now if this DP gets refreshed with 1 Orb let's say between the second and third tick, you'll get a refreshed dot with 7 ticks doing X damage. By the ignite-logic, X would be the remaining damage of the preivous dot ( 4 ticks * 300 damage ) plus the damage for the new dot ( 6 ticks * 100 damage ), divided by the remaining ticks ( 7 ). Or shorter: X = ( 4 * 300 + 6 * 100 ) / 7 = 257.

Unfortunately, that is not how it works. Even though it follows this ignite logic to some degree, the tick damage of the old previous amount is recalculated with the new amount of orbs. In our example, X will become ( 4 * 100 + 6 * 100 ) / 7 = 142. Those ( 300 - 100 ) * 4 = 800 damage are lost and gone.

For a DTR-proc this results in 0 damage, since ( 4 * ( 0 * 100 ) + 6 * ( 0 * 100 ) ) / 7 = 0. So it doesn't really matter if DP behaves Ignite-Like or not, a DTR proc completely destroys the previous DP tick damage pool.

In conclusion, refreshing a DP dot with less Orbs than the previous one still destroys damage, and DTR-procs negate it completely.

Edit2:

Ignite-Like mechanic for Devouring Plague was implemented into SimC. As expected it didn't really change that much, resulting in a 50 DPS increase for the T13H profile. This comes from the possibility/allowance to now recast a 3-Orb DP, which can occur in the execute phase with the additional Orbs from Shadow Word: Death.

Share this post


Link to post
Share on other sites

  • Recently Browsing   0 members

    No registered users viewing this page.