Your items, your corrections
Start with manual entry. Screenshot recognition will come later, with every extracted value editable before comparison.
A little clarity for your next adventure
Two items. One equipment decision. Bring your stats together, see what changes, and keep the unknowns in view.
The workbench is open. Compare raw stats today. Damage recommendations and screenshot recognition are still being developed.
What’s available01 / Your equipment
Copy the stat lines from each item. Leave unknown values blank; enter 0 only when you know a stat is absent. Use the same units for both items.
Choose a PNG, JPEG, or WebP image up to 10 MB. It stays in this tab. Automatic text recognition is not available yet; enter its values below.
02 / The comparison
Enter a matching stat for both items to see the difference here.
02 / The comparison
Your inputs have changed. Compare again to update these results.
| Stat | Equipped | Candidate | Change |
|---|
Confidence: Low. Stat differences alone cannot establish which item deals more damage. Scaling, caps, accuracy, skill effects, and activity modifiers still need verified evidence.
Read the current assumptions →Built with the unknowns in view
Start with manual entry. Screenshot recognition will come later, with every extracted value editable before comparison.
Today’s result is candidate minus equipped, in matching units. Percentage stats use percentage-point changes. Blank values stay unknown.
Future recommendations will include High / Medium / Low confidence, key factors, and source-linked assumptions. CP is not used as a damage verdict.
No damage formula has been validated for this preview. No Attack, Main Stat, Critical Rate, Critical Damage, Accuracy, or skill-level scaling is assumed. Activity is recorded as context and does not change raw differences.
This first form covers six stat lines, not every possible item effect. A higher number in the table is not an equip recommendation.
View the mechanics research register ↗