Options

Item bonus stacking - how do they work?

tangoraventangoraven Registered Users 42 Posts
Between Chrilion, Rewand and Crazyaddict, we probably have most of the important the items stats and ability bonuses clearly discovered.

How about item bonuses and how they stack up? My Oda is currently wearing the Darkness Runic Leather, Lvl 7 (Hit Rate plus attack speed bonus) with Eternal Hood and Boots, both at Lvl 10 (attack speed bonus only).

I'm thinking of upgrading the Eternal armours to Darkness but if the attack speed and hit rate does not stack, it doesn't seem that worthwhile to spend the runes just to get more armour points (my Oda has close to 1k defense).

In fact, if the attack speed doesn't stack, I might downgrade Oda to other items with bonuses like movement speed which can be useful as Oda is now my main "tank".

Comments

  • Options
    rewandrewand Registered Users 588 Posts
    Several of us have done experiments which suggest that most of the capabilities do not stack. Here's how the tests work.

    Attack/cast speed:
    1. Test with one hero at a time
    2. Load up on as many attack speed bonuses as possible
    3. Find a suitable mission. One in which the first battle faces someone difficult enough to not die quickly. If needed throw in a healer to make the test run longer.
    4. Attack for a set amount of time (30 seconds, 1 minute etc) and count the number of attacks. Not just hits but attempts.
    5. Record results
    6. remove all bonus items and repeat
    7. try with just the single highest bonus.

    The results we've found is that only the highest bonus counts.

    For dodge put someone with slow foot speed against a single ranged unit. Run your hero to the far corner and count the attacks vs dodges but don't count misses. Tip, only heal when necessary.

    For stun, fear, critical hit it's more tricky. Like the dodge test you have to count the ratio of hits which trigger the skill to normal hits. Headhunter events can often be a good test scenario. Long tests are best for accurate results. What we are looking for is a statistically significant difference. To properly test requires a good deal of expertise but my tests (dodge, chaos) resulted in the score with two items being worse than the score with one item leading to the conclusion that the items do not stack. I haven't tested the others.
  • Options
    EziekialEziekial Registered Users 166 Posts
    Here was the post we did on the first few tests back in july

    http://ggnbb.glu.com/showthread.php?17631-Accessory-Items-STACK
Sign In or Register to comment.