Behind the result - No mystery in the method
For each option, the gold-only quantity is floor(budget ÷ entered cost), the population-only quantity is floor(allowance ÷ entered population), and simplified capacity is the smaller value. The entire allowance is assumed to be dedicated to that single option, not divided between both. If an optional training time is supplied, quantity × time gives a serial-only illustration; parallel queues and delays are not modelled. Fixed editorial role prompts explain questions to consider. There are no claimed damage, health, DPS, exact counter, current timing or tier-list values.
Blank optional fields are omitted from the entered-input summary. Text is displayed literally, not interpreted as HTML.
Editing inputs clears the previous result. Build again before exporting. Download text creates a .txt copy; Print / Save PDF uses your browser. Review optional notes before sharing.
Before you go
What does the comparison show?
It puts two roles and your entered constraints side by side. Gold capacity and population capacity are calculated separately; the smaller quantity is the simplified capacity for that option.
Are the training-time numbers read from the game?
No. Optional training time is entered by you. The output multiplies it by the simplified stand-alone quantity for a serial-only illustration. It does not inspect queues, skins, upgrades, speed changes or current game balance.
Why does the same unit give two different plans?
Choosing the same label is allowed so you can compare two hypothetical cost or population assumptions. Differences then come from those inputs, not a hidden stat change or official variant database.
Can I print or download the comparison?
Yes. The text download and browser print include the selected units, your entered values, role prompts and capacity calculations.