Top War: Battle Game – How to Test Heroes

Testing Heroes (from The Perspective of Testing The New Hero Binary)

By Alan.

Main thing for testing most heroes is using sandtable, so I will first explain how to use sandtable. Some heroes don’t work well in sandtable, but most heroes do, and sandtable is very good tool for assessing heroes. It also helps establish trigger rate of heroes.

We can evaluate damage part of heroes, but we can’t evaluate protection part of heroes, as it’s not reflected in sandtable. Protection part (HP, indestructible, silence, etc) has to be tested in other ways. Main thing to know is, we always assess relative damage to damage done by regular attacks.

On this example, I’m testing current heroes Nelle, Quily and Binary, and their relative damage stays mostly the same:

Damage from regular attacks (which is some sort of benchmark) is base with which we measure damage from heroes. In this example, you also should note that I have both Quily and Nelle with es5, which can affect relative damage produced by them. Number of stars has much lower impact to relative damage (but does affect absolute damage).

So, we can see here that regular damage is 11.7aa, and Nelle does double that. Here, we mainly test new hero Binary, that does 8.49aa, which is about 72% of regular damage. Activation rate is also easy to see, it’s 166/450 (for Binary), ie about 37%. We do need to make more attacks to find average activation rate.

Also, we can see that Quily does 62.2aa damage, ie 5.3x more damage than regular attacks. When I was testing Quily-Nelle combo, I have noticed how Quily is boosted with Nelle. And, as Binary is hero similar to Nelle, it’s good to test how they alone influence Quily:

Notice that absolute damage is lower, but ratio of damage from Binary is similar (3.58/4.46=80%). Damage from Quily is now 3.45x regular damage (15.4/4.46), which is drop from 5.3x damage when Nelle helps boost Quily’s damage.

When using Nelle-Quily combo:

Quily’s ratio is 3x regular damage (25/8.29), and even Nelle did lower damage ratio 1.65x from 2x.

To test first findings, one more sandtable test of all 3 heroes:

We now have Quily 5x (from 5.3x first time), Nelle 1.94x (from 2x first time).

One more test (5th and last sandtable for a day):

Quily is at 5.1x, Nelle is at 1.96x.

Conclusion:

Even if Binary is at 1*, no es, nothing, there’s some real synergy between those three heroes that is reflected not only in boosting Quily, but also boosting Nelle’s damage. Those three heroes work great together, much better than just using them in pairs.