Sanctuary field notes

Diablo IV Field Guide

Researched, versioned notes on Diablo IV systems and decisions.

Theorycrafting a build: model the fight you want to win

Scope: Season 15, patch 3.2.1. The method is reusable; item interactions and numbers are patch-specific.
Checked: September 27, 2026.
Status: Working model. Its equations organize comparisons; they are not a claim to reproduce Diablo IV’s hidden damage calculation.

A build decision is only meaningful for an encounter. For an Arbiter Paladin, the target might be a dangerous high-tier pack or elite that can kill the character, with a boss check as a second case. Decide which of these is failing before you optimize: surviving the first burst, maintaining the rotation, hitting the important enemy, or dealing enough damage before the next dangerous attack.

Start with a fight model, not a character-sheet number

For each damage source (Hammers, Wing Strikes, a proc), use this sketch:

Useful damage over the fight ≈ attacks produced × hits per attack that connect with priority enemies × effective damage per hit during the actual buffs.

Then add the sources that matter. For a mixed Hammer/Wing Strike build, do not assume an aspect that improves one source improves the other. “Effective damage per hit” is a measured or estimated quantity with enemy state, Critical Strike Chance, skill ranks, and applicable bonuses included. These factors can change during a fight. The sketch is a comparison framework, not an exact damage formula or a promise that each factor is independent.

Lever The question to ask Arbiter example
Production How many attacks can the build produce in combat? Argent Veil can generate equipped Core skills during Arbiter’s Evade; manual Hammers cost casting time and Faith.
Connection How often do those attacks hit the enemy that matters, and how many hits result? Orbiting Hammers may miss if an evade carries you out of range; seeking Wing Strikes change targeting.
Hit strength How much does a connected hit deal under buffs actually present? A ring aspect can strengthen manual Hammers; some bonuses only apply to Holy or Disciple hits.
Uptime When do the relevant form, set, aspect, Vulnerable, and Crit conditions hold? Light’s Epiphany’s five-piece effect depends on an Arbiter of Justice cast; permanent form alone is not the same trigger.
Cost What do the gear, skill, resource, and defensive slots displace? Argent Veil occupies a ring; Aegis and Rally compete for a skill slot.

For the current Argent Veil tooltip, the evade trigger occurs per distance traveled and picks a random equipped non-channeled Core skill. If you equip another eligible Core skill alongside Blessed Hammer, you cannot count every trigger as a Hammer. The Season 15 patch notes are the version anchor for this page; confirm the item’s live behavior in game.

A break-even example, with made-up numbers

Suppose a comparable hard pack receives 12 connecting Hammer hits with Argent Veil and 9 connecting hits with a replacement damage ring. Ignore other damage and assume the Veil setup’s effective damage is 100 per hit. It deals 1,200 in this illustrative window. The ring setup needs more than 133.3 per hit (1,200 ÷ 9) to come out ahead. That is a greater than 33.3% improvement in effective damage per connected hit, relative to the entire Veil setup, including Veil’s own skill ranks, temporary damage bonus, and the replacement ring’s affixes and uptime. It is not the required number printed on one aspect. If the evade Hammers miss, or manual casting causes deaths or Faith gaps, change the hit counts. These are invented teaching numbers, not measured Season 15 values.

For a choice between two damage affixes, first check eligibility (which attack tags and targets qualify), then the bonus type and existing bonuses, then uptime. Blizzard’s damage-bucket explanation explains why stacking similarly worded additive bonuses cannot be valued as if each were a new independent multiplier. It is an older systems explanation, so use current patch notes and live tooltips for current item-specific behavior. If an interaction is uncertain, measure it rather than assign a precise multiplier.

Example: model a conditional target debuff

Vulnerable has two separate decisions: how to apply it, then whether to buy bonuses that require it. For the Arbiter build, Mother applies it on character contact, Seraph’s Wings applies it through Wing Strike hits, Exploit applies it on damage with a per-enemy restriction, and Falling Star has a chosen upgrade. Record Vulnerable uptime on the priority target during your damage window, not a generic percentage across a whole dungeon. A strong Vulnerable-only affix or an on-kill aspect loses value if the boss is not Vulnerable or you have no kills to ramp it. The comparison must include the socket, glyph, skill upgrade, and survival risk of the application method. Do not multiply every similarly named Vulnerable Damage roll as a separate multiplier.

Damage bonuses: compare the next increment

When one item says +40% damage and another says 40%[x] damage, the percentages do not describe the same gain. First ask whether the bonus applies to the attack and target you care about. Then ask what it is added to or multiplied by. Blizzard’s damage-bucket change made additional Critical Strike Damage, Vulnerable Damage and Overpower Damage additive with other applicable damage bonuses, while preserving baseline multipliers when their conditions are met. Blizzard’s explanation gives the baselines as 1.5× for a Critical Strike, 1.2× against a Vulnerable enemy and up to 1.5× for an Overpower, subject to its life condition. These are historical system explanations, not a promise that every Season 15 item follows a simple formula. The current Season 15 notes still distinguish [x] multipliers in item and skill changes. Read the specific live tooltip, especially effects that convert a stat into another bonus.

For a simplified comparison of one eligible hit, imagine base hit × (1 + applicable additive bonuses) × independent applicable multipliers. The additive term is a pool: if it is already +400%, its factor is 5.0. Another +40 percentage points changes that factor to 5.4, or 8% more damage, not 40% more. A separate 40%[x] factor, if it really applies to the same hit, would make it 40% more. These are invented numbers and an accounting model, not the game’s complete damage formula. A bonus that applies to only some hits must be weighted by those hits’ share of your damage.

Bonus to evaluate Practical comparison
+% Damage, +% Holy Damage, +% Damage to Vulnerable Count the displayed roll in the applicable additive pool only when the skill and enemy qualify. Compare new pool factor ÷ old pool factor; include uptime.
%[x] damage from an aspect, passive, set or glyph Check the exact wording, stack rule, cap and trigger. If it is an independent applicable multiplier, compare its new factor ÷ old factor, not its printed percentage in isolation.
Critical Strike Chance This determines how often the Crit baseline and Crit-only additive bonuses apply. A Crit-only bonus loses value on hits that cannot Crit; extra chance loses value near an effective 100% chance.
Vulnerable application The baseline 1.2× is available only while the priority target is Vulnerable. Additional Vulnerable Damage is an eligible additive bonus; applying Vulnerable more reliably can beat adding another roll.
Skill ranks, primary stat, attack speed and weapon damage Compare them with the same attack and actual rotation. They change the hit or the number of hits through different paths; do not automatically place them in the additive pool.

Example with uptime: suppose +40% Damage to Vulnerable would raise the additive factor from 5.0 to 5.4 on qualifying hits. It improves those hits by 8%. If half your relevant damage lands while the target is Vulnerable and all else stays equal, the whole-fight gain is roughly 4%. That estimate changes if Vulnerable windows contain your biggest attacks. A separate 20%[x] effect active on every relevant hit would instead improve them by 20%, provided it is genuinely independent and has no displaced cost. Compare the whole replacement: an item can also change skill ranks, Crit Chance, survival and attack production.

Cliptis’s damage walkthrough is a useful prompt to audit Paragon progress, glyph levels, main stat and conditional uptime when a copied build falls short. His earlier worked demonstration emphasizes that small omissions across those systems compound. That is build-author guidance, not a Season 15 breakpoint: measure your own damage sources, and prioritize the largest marginal improvement for a fight you can survive. For an uncertain tooltip interaction, compare repeated same-target training-dummy hits with one bonus changed, then check a real encounter for uptime and missed attacks.

Survival is a gate, not just another DPS stat

A damage gain has little value if you die before it works. Record Life and barrier available before the hit, mitigation for the damage type, whether the blow is blocked, and whether a defensive cooldown is active. Blizzard changed Armor and Resistances into rating systems and added a Toughness stat for damage types; old fixed armor caps and Torment penalty rules are poor inputs for a current calculation. Look at the combat state in the difficulty being tested.

A rough single-hit check is: does the damage that gets through exceed the Life plus barrier available at that instant? Treat healing and Fortify’s healing reservoir as sustain over time, not guaranteed protection from that instant’s hit. A build with high average survival may still fail the one large unblocked hit; compare the bad case as well as the average. For repeated damage, note healing, potion access, control effects, positioning, and how quickly enemies die. See Paladin survival for the defenses and boss-fight decisions in detail.

When comparing an offensive ring, aspect, or skill with a defensive choice, use a simple rule: first clear the survival and rotation gates at your target tier; then improve useful damage. More damage can itself reduce incoming attacks by shortening the fight, but only if you live long enough to deliver it.

Run a comparison that can change the conclusion

  1. Write the decision and the target encounter: for example, “Argent Veil versus a legendary ring against high-tier elite packs,” with a separate boss case if needed.
  2. Record the whole build difference: skill bar, aspects, affixes, resource solution, and which bonuses each variant gains or loses. Note patch and item rolls. A ring swap that forces another skill or aspect change is a build comparison, not a controlled single-variable test.
  3. Predict the mechanism and a break-even condition. For Veil: how many additional connecting hits must its evades provide to beat the ring’s effective hit-strength gain?
  4. Try several comparable engagements at the same tier. Record rough clear time, deaths or near-deaths, missed casts from Faith or movement, priority-target hits, and whether conditional buffs were present at the start and during the dangerous part. Compare cold starts as well as fully ramped packs.
  5. Keep a conclusion with its scope: “On this gear and in these packs, Veil wins because…,” or “The result is inconclusive because…”. Recheck after an item, aspect, or patch changes a key variable.

Tooltips alone cannot establish an exact simultaneous Hammer cap, proc interaction, enemy damage distribution, or every hidden multiplier. Keep those as open tests. A useful model names the observation that would reverse a recommendation, so the conclusion can change when new evidence arrives.