APL Deconstruction: Arcane

Guides and discussions of all things specific to the Mana Adepts of Azeroth.
User avatar
Curnivore
Posts: 827
Joined: Mon Sep 28, 2015 9:26 am

Re: APL Deconstruction: Arcane

Unread postby Curnivore Sat Jan 09, 2016 8:13 am

I wonder if we need an alternative basic rotation script. One that may be suboptimal on top DPS but optimal for testing the core of the rotation on Patchwerk. I do not know though if that is technically possible without getting erroneous assumptions as results.

Alternatively the complex scripts could have alternatives for testing, e.g. a drop down menu giving options "Default (Optimized) APL, burn at ACX3, Start burning at 0 Arcane Charges when crystal is used or ring is used on pull" etc.
skiz
Posts: 182
Joined: Fri Aug 22, 2014 12:27 pm

Re: APL Deconstruction: Arcane

Unread postby skiz Sat Jan 09, 2016 10:35 pm

how can we reliably edit the script to test for fewer stacks of Arcane Charge before the opener
Impossible. Well, generally the APL is a priority list with a loot of boolean expressions. You can build an APL that does (mostly) exactly what you want to do. However, trying to look into results of what actually happens will show some problems. Looking into that specific issue I do get:

- the target of the precast is always the boss. (You'll see that in the Sample Sequence table)
=> Solution: delete precast from every profile I do compare. actions.precombat should just contain buffs.

- changing burn initialisation and checking that you'll go into burn on the pull isnt that hard.

- PoF probably can just be procced by Arcane Missile and Arcane blast on a UM profile. (You'll get the idea if you run the sim with debug=1 and look into the log). This cant be fixed inside APL. Basically, once PC is up robomage is just able to procc PoF on the Crystall.

- RPPM procc is incredibly high on the first cast with ~80% proccchance. So PC 0 is probably overvalued while PC 1 up to PC 4 are undervalued. (Same here with debug=1). No one knows how RPPM is actually working. This method of faking is okayish, but doesnt work for analyzing PC and PoF. This will mess up your results even more.

To get a correct answer from simcraft you need to change the code AND the APL.
to avoid using the Crystal for Arcane Missiles?
If I get that right, you want to know how to change the APL that you do not cast Arcane missiles DURING PC. Go for it. But I doubt it is worth the time. The braindump from Komma about Tome of shifting words is quite decent. And a correct answer to that question is a little bit more complex. It depends on:

- gear: Itemlevel of tome.
- Glyphed / Unglyphed AP
- Overpower / PC (well, in this case it is just PC, but generally speaking)
- PC is just another amplifier, if a solution of that problem works without PC, it will be the same with PC.
- Fighttime : on an i.e. 25 sec reaver pull you'll probably prioritize AM way higher
- Mana: Robomage can run oom and he likes to do nothing if oom.
- Relevance and RNG: You'll probably not generate many AM charges during PC

Edit: It is actually quite funny what happens in there. Worth to look at it. :)

Well, maybe you're talking on weaving missiles. There are examples of that somewhere else.

P.S. If you want to change a APL, I do recommend the copy-option. Look at "Copy=" on the simcraft wiki. Furthermore turn on deterministic RNG. Samplesequence table and the samplesequence on the html report are often enough data to see if an APL works. But basically, just decent boolean constructions will ensure your APL is working. Going deeper into debug=1 log is usually necessary on any kind of proccing stuff.

Return to “Arcane”

Who is online

Users browsing this forum: No registered users and 26 guests