Skill points 15 · Equipped items 0 · Selected talents 0
BATTLE SIMULATOR BETA
Build. Battle. Compare.
Pick heroes and builds. Compare combat and DPS under the same conditions.
This beta is under development. Supported heroes, items and combat calculations are still being improved.
RULESET PATCHPATCH 7.41f
Switch between duels and target measurements.
01Heroes · quick setup
Skill points 15 · Equipped items 0 · Selected talents 0
Skills, items & talents
Battle setup
Change settings before a battle or after resetting.
Left hero
Points 15/18
Items · talents · behavior
Equipment · 6 slots
0 / 6Select a slot to replace it. Components and held Moon Shards support duplicates; effect items currently support one per type.
Chance effects use seeded independent rolls rather than Valve PRD. Vision, illusions and extra targets are excluded. Lightning measures the first target; Desolator starts without kill/assist stacks.
Talents
Behavior
Auto-use · priority
The first learned spell or equipped active item that meets every condition is used, from top to bottom. Other actions, attacks and movement continue while a condition is unmet.
Use conditionsDefault conditions
Use conditionsDefault conditions
Use conditionsDefault conditions
Health 100% and mana reserve 0% add no restriction. Example: BKB at 60% own health; ultimate after 3 seconds. Conditions are checked when starting a cast; later health changes do not cancel it.
Range, mana costs, cooldowns, silence, taunts and existing buff checks still apply. These are user-defined experiment settings, not strategic game AI.
Exclude from Octarine reduction
- Attack damage
- 121.0
- Armor
- 10.3
- Attack interval
- 0.99s
- Magic resist
- 34.6%
Damage dealt 0Attacks 0Casts 0
Right hero
Points 15/18
Items · talents · behavior
Equipment · 6 slots
0 / 6Select a slot to replace it. Components and held Moon Shards support duplicates; effect items currently support one per type.
Chance effects use seeded independent rolls rather than Valve PRD. Vision, illusions and extra targets are excluded. Lightning measures the first target; Desolator starts without kill/assist stacks.
Talents
Behavior
Auto-use · priority
The first learned spell or equipped active item that meets every condition is used, from top to bottom. Other actions, attacks and movement continue while a condition is unmet.
Use conditionsDefault conditions
Use conditionsDefault conditions
Use conditionsDefault conditions
Health 100% and mana reserve 0% add no restriction. Example: BKB at 60% own health; ultimate after 3 seconds. Conditions are checked when starting a cast; later health changes do not cancel it.
Range, mana costs, cooldowns, silence, taunts and existing buff checks still apply. These are user-defined experiment settings, not strategic game AI.
Exclude from Octarine reduction
- Attack damage
- 107.0
- Armor
- 7.8
- Attack interval
- 1.16s
- Magic resist
- 29.5%
Damage dealt 0Attacks 0Casts 0
Random effects are seeded. Verified classifications use an increasing-chance PRD model after failures; this does not certify identical internal Valve randomness for every effect. Repeated runs describe this model.
Lina Lv.18
Idle4443
Axe Lv.18
Idle4443Helix 0/4
Loading original 3D models and animations0%
Ready0.0s
BETA · RULES 9
Repeated runs · equipment A/B
Repeat the same setup with different seeds. Win shares describe this model, not real match win rates. Computation stays on this device without external API requests.
Baseline A
Complete a battle or DPS measurement to remember the first result as A. This baseline resets when you leave the page.
Verification & calculation detailsNative stat/spell observations, source JSON and hero-specific limits.
Patch 7.41f has no selectable facets.
Valve removed facets in 7.41. Official data and all 23 retired facet definitions across these fifteen heroes were checked. This ruleset uses current innates, Scepter, Shard and talents. Official Valve 7.41 patch ↗
15 supported heroes · verification scope
Simulates these fifteen heroes’ spells, ultimates, innates, talents and supported equipment. Allies, creeps, illusions, vision and terrain outside the 1v1 are excluded. Hero-specific limits appear below.
Lina
Existing models, abilities and native observations are retained. A 1v1 has no secondary cleave or allied-aura targets.
Axe
Existing models, abilities and native observations are retained. A 1v1 has no secondary cleave or allied-aura targets.
Sven
Existing models, abilities and native observations are retained. A 1v1 has no secondary cleave or allied-aura targets.
Sniper
Existing models, abilities and native observations are retained. A 1v1 has no secondary cleave or allied-aura targets.
Zeus
Current-health damage, Lightning Hands and Nimbus. Arc hits the single opponent; Nimbus is a fixed, indestructible area. Vision and invisibility checks are excluded.
Lion
Channeling, Hex, Finger and its fist buff. Each fight starts without earlier hero kills/deaths: permanent kill stacks and respawn amplification start at zero.
Lich
Shield, channeling and Chain Frost bounces via Ice Spire. Sacrifice needs an allied creep and is unavailable. The spire uses simple geometry. Enemy AI attacks/destruction, the death Chain Frost talent and other spells during Scepter channeling are modeled; not every interaction is natively certified.
Phantom Assassin
Strike charges, evasion, Deadly Focus and the active innate. Rolls use a seeded PRD model, without certification of identical Valve internal randomness. Blur vision/invisibility is not reproduced.
Viper
Poison autocast, resistance reduction, attack slows and Break. Against the fixed-health target, Predator follows its configured health percentage. Extra targets and buildings are excluded.
Bristleback
Goo, Quill stacks, Warpath and rear mitigation. Linear combat models front/back, not side angles or 2D turning. Scepter turns the hero away while releasing quills.
Juggernaut
Blade Fury, slashes, crits and Healing Ward in a1v1. Extra-enemy bounce paths, freely controlled ward movement, terrain and vision are not reproduced.
Ursa
Current-HP damage, Fury Swipes stacks, Overpower attack-attempt charges, Enrage damage reduction and status resistance. The opponent and Roshan follow distinct rules.
Slardar
Bash uses an attack counter; Haze uses signed armor reduction. Water bonuses apply only inside created puddles/trails, not across the arena.
Drow Ranger
Frost Arrows is a mana-costing attack modifier. Gust, Multishot and Marksmanship use the single opponent; no extra enemies are invented.
Lifestealer
Rage, max-HP-based healing, Open Wounds and Scepter Infest are distinct. Allied-creep/hero hosts are unavailable in this1v1; Scepter enemy-hero Infest has separate conditions.
The five additions use checked Valve7.41f data, game files, original models, equipment and animations. Expand the native observations below for isolated stat, spell and interaction measurements and remaining differences. Matching selected cases does not certify complete fights.
Valve · 5 Heroes · 7.41f · JSON ↘
Observations verify only the recorded conditions, not all loadout DPS, Valve PRD or every interaction. Earlier ruleset records are retained and are not directly compared with this ruleset.
Compare with the actual game
Native engine observations availableCompare unbuffed values with the same patch, level, items and talents. Manually entered observations stay in this page.
PATCH 7.41f · Game client version 6969 / 11175759
| Metric | Web calculation | Actual game observation | Difference |
|---|---|---|---|
| Max health | 1440.00 | 0.000 ✓ | |
| Max mana | 1227.00 | 0.000 ✓ | |
| Average attack damage | 121.00 | 0.000 ✓ | |
| Armor | 10.30 | 0.000 ✓ | |
| Magic resistance % | 34.60 | 0.000 ✓ | |
| Attack interval (s) | 0.989 | 0.000 ✓ |
This condition has saved native observations. Editing a value switches that field to your input.
Isolated spell observations · 11
Level 18, no talents or attribute bonuses, max spell rank, against Sven. Damage includes burn. Only Flame Cloak uses Scepter. Damage is rounded up per hit; Lina burn uses the observed 1-second interval. Helix uses the health delta after four forced attacks. Timing samples the cast phase about every33ms. △ indicates a remaining difference.
| Ability | Damage web / native | Mana web / native | Cast point web / native |
|---|---|---|---|
| Dragon Slave | 286.0 / 286.0 ✓ | 120 / 120.0 | 0.35 / 0.367s |
| Light Strike Array | 253.0 / 253.0 ✓ | 130 / 130.0 | 0.45 / 0.467s |
| Laguna Blade | 887.0 / 887.0 ✓ | 450 / 450.0 | 0.30 / 0.300s |
| Flame Cloak | 0.0 / 0.0 ✓ | 50 / 50.0 | 0.00 / 0.000s |
| Battle Hunger | 300.0 / 300.0 ✓ | 80 / 80.0 | 0.30 / 0.300s |
| Berserker's Call | 0.0 / 0.0 ✓ | 120 / 120.0 | 0.30 / 0.267s |
| Culling Blade | 475.0 / 475.0 ✓ | 150 / 150.0 | 0.30 / 0.300s |
| Counter Helix | 160.0 / 160.0 ✓ | 0 / 0.0 | — |
| Storm Hammer | 227.0 / 227.0 ✓ | 110 / 110.0 | 0.20 / 0.200s |
| Warcry | 0.0 / 0.0 ✓ | 45 / 45.0 | 0.00 / 0.000s |
| God's Strength | 0.0 / 0.0 ✓ | 150 / 150.0 | 0.30 / 0.300s |
Additional observations · movement and item interactions · 14
PATCH 7.41f · Game client version 6969 / 11175759 · 10/29/2026, 3:21:07 PM KST
Measured at level 18 without talents in a separate local practice map. Sven’s actual defenses were recorded, with regeneration and outside damage blocked. All 14 conditions kept both units alive. Movement values use Sven; the two movement cases ran separately after previous Shrapnel zones expired.
| Check | Web / native or applied values |
|---|---|
| Three isolated Lina damage spells | 286 / 286 · 253 / 253 · 887 / 887 |
| Movement speed during 100% Headshot slow | 100 / 100 |
| Speed after Headshot expires inside Shrapnel | 227.5 / 227.5 |
| Speed after leaving Shrapnel | 325 / 325 |
| Assassinate cast point · base / Scepter | 2.0s / 0.5s |
| Mask · Satanic passive / active lifesteal | 22% · 25% / 170% |
| Desolator · Assault armor reduction | −6 · −5 |
| Daedalus critical multiplier | 225% |
| Monkey King Bar Pierce bonus | 70 magic · rounded with the attack |
| Maelstrom lightning | 110 magic → 78 after observed resistance |
Headshot and Shrapnel slows expire independently. Damage is rounded up per hit; lifesteal uses pre-rounding damage and preserves fractional precision. Native health displays integers, so accumulated healing was compared within 1.1 HP.
Pierce’s magic bonus uses magic resistance while the physical attack uses armor; their total is rounded together. Native events group both into one attack, so the event type alone does not identify every component. Crit and lightning combinations were also observed with normal attacks. This does not certify Valve PRD, full DPS across all loadouts, or innate melee damage block.
Download raw observations for all 14 conditionsMethod and remaining differences
Base stats, items, mana, cooldowns and cast points are checked against Valve data. Damage, recovery and facet interactions without native observations are not marked as measured.
Download the native measurement LuaAdditional observations · movement, items & level-30 talents
PATCH 7.41f · 6969 / SourceRevision 11175759 · 10/29/2026, 7:03:32 PM KST
| Condition | Health · web / native | Damage · web / native | Move speed · web / native |
|---|---|---|---|
| Sniper · Phase Boots | 1286 / 1286 | 109 / 109 | 335.0 / 335.0 |
| Sven · Phase Boots | 1946 / 1946 | 175 / 175 | 375.0 / 375.0 |
| Sniper · both boot types | 1506 / 1506 | 109 / 109 | 335.0 / 335.0 |
| Sniper · Phase active | 1286 / 1286 | 109 / 109 | 368.5 / 368.5 |
| Sven · Phase active | 1946 / 1946 | 175 / 175 | 450.0 / 450.0 |
| Sven · level 30, both talents | 2870 / 2870 | 248 / 248 | 325.0 / 325.0 |
Only the higher boot movement bonus applies. Melee/ranged Phase effects and all eight level-30 talents were compared. Headshot pushed outward despite reversed facing and caused no displacement during BKB. This does not certify exact native displacement or every other item combination.
Download raw observations for the eight additional conditionsSix added heroes · native verification
Local Workshop observations use the same build and patch. Target health, armor, resistance, suppressed regeneration and damage-event totals are recorded together. Counts are experiments, not actual-match samples or win rates.
- Stat configurations
- 36
- Compared stat metrics
- 432
- Spell / status fixtures
- 37
PATCH 7.41f · 6969 / 11175759 · 10/30/2026, 10:38:54 AM KST
Download native observations ↘Observations verify only the recorded conditions, not all loadout DPS, Valve PRD or every interaction. Earlier ruleset records are retained and are not directly compared with this ruleset.
How to read cast animation and channeling
- Cast animation
- The preparation before the spell begins. Moving into range, turning and command delivery latency are excluded.
- Channeling
- The time spent maintaining the spell after its cast animation. Movement or interrupts may end it; movement permissions and interruption effects vary by ability.
- Total cast time
- The sum of cast animation and channeling for a full cast. Abilities do not necessarily wait until the end to start taking effect.
Example: Elder Titan’s Echo Stomp has a 0.4s cast animation followed by 1.3s channeling, totaling 1.7s. The stomp takes effect when the cast completes without interruption.
Some abilities deal damage or apply effects while channeling. Projectile travel, separate effect delays, buff durations and stun durations are different from this total. Check the ability description for its effect timing.
A channeling time of 0s means no channeling. Not provided means the value is unverified. Passives show Not applicable.
These are base timings for the selected patch and ability rank. Octarine cooldown reduction does not shorten casting time. Talent or item changes to casting speed are excluded unless explicitly shown.
Attack animation is the time until a hit or projectile launch. Base attack interval is the cycle between attack starts. These values are before attack-speed modifiers.
Rules and sources
Simulates these fifteen heroes’ spells, ultimates, innates, talents and supported equipment. Allies, creeps, illusions, vision and terrain outside the 1v1 are excluded. Hero-specific limits appear below.
Combat uses a 0.02s step and may differ from animation frames. Terrain, vision, extra targets, unlisted items and Valve PRD are outside scope. Melee block subtracts16 from physical attacks with a seeded50% PRD model; enable it separately for the DPS target.
Random effects are seeded. Verified classifications use an increasing-chance PRD model after failures; this does not certify identical internal Valve randomness for every effect. Repeated runs describe this model.
Observations verify only the recorded conditions, not all loadout DPS, Valve PRD or every interaction. Earlier ruleset records are retained and are not directly compared with this ruleset.
Spells release after their cast animation without an extra forced recovery delay. Repeated attacks follow their attack interval. This does not predict real match win rates.
Patch, capture and source details
Hero data · PATCH 7.41f · 10/28/2026, 2:34:41 PM KST
Expanded equipment captured: 10/29/2026, 12:53:12 PM KST
This fixed ruleset is independent of the atlas patch selector. New patches require validation.
- Valve · Lina
- Valve · Axe
- Valve · Sven
- Valve · Sniper
- Valve · Zeus
- Valve · Lion
- Valve · Lich
- Valve · Phantom Assassin
- Valve · Viper
- Valve · Bristleback
- Valve · Juggernaut
- Valve · Ursa
- Valve · Slardar
- Valve · Drow Ranger
- Valve · Lifestealer
- Valve · Archived game files
Models and animations: Valve game files. Battle logic and effects: implemented by this site.
Recent battle events
- Start a fight to see casts and hits.