|
| Thread: Initiative Weirdness |  This thread is pages long: 1 2 · «PREV |
|
Gidoza
Famous Hero
|
posted April 22, 2020 11:28 PM |
|
|
@frogo
Thanks for the analysis. I appreciate that you appreciate the situation. 
You're right that converging to 0 yields crazy results, but as you'll see from my Excel though - I make an effort not to render ridiculous values for certain things; e.g. I could make a level 3 archer unit with immense damage and 5 hit points, but for lack of better description (and certainly not a mathematical one) - it's in bad "taste" to do this. Because I'm intentionally avoiding such extremes, I'm not terribly worried about the 1.4 multiplier on ranged units (for example) because something non-trivial is always being sacrificed: you're going to end up with a garbage ranged unit with poor damage, or a ranged unit that is solid but falls quickly to lightning bolts. I mix it up a little - but I think the results on that front are reliable.
I think what you pointed out in the "converging to 0" schematic that's interesting is that the higher initiative values seem to get stronger as they go higher, and perhaps the value of 8 (I believe that's the one you used) and going lower renders more accurate initiative * damage results. I'll have to consider this more as I play-test. The way I'm thinking of it, experience and play-testing is obviously going to make alot of the results, but having a sheet and understanding what's going on doesn't hurt either. Regardless, I'm sure you'd plenty love my version of Zombies.
|
|
Elvin
Admirable
Omnipresent Hero
Endless Revival
|
|
posted April 23, 2020 08:02 AM |
|
Edited by Elvin at 08:02, 23 Apr 2020.
|
I'm calling it, dash and pao's scavenger 
____________
H5 is still alive and kicking, join us in the Duel Map discord server!
Map also hosted on Moddb
|
|
JoonasTo
Responsible
Undefeatable Hero
What if Elvin was female?
|
|
posted April 23, 2020 10:32 AM |
|
Edited by JoonasTo at 10:35, 23 Apr 2020.
|
We had a rough formula for quick estimation to land in the general ballpark for units in Elvin’s duel map balancing. In short, initiative is worth more than just the raw damage. And the more damage or initiative the unit has, the more extra initiative is worth.
So going from 10 init to 11 for 1 damage unit is worth very little but going from 15 to 16 for a 100 damage unit is massively broken.
Feel free to look around
The stats are all new so keep that in mind as you go. The numbers are not relevant for vanilla creatures but the formulas are quite good in our experience.
It’s abilities that are the real problem. A unit might be balanced on average very well but it might have massive potential if utilised correctly. So average player has some oooh, sweet moments while an expert will crush the map and/or opponent because of it.
____________
DON'T BE A NOOB, JOIN A.D.V.E.N.T.U.R.E.
|
|
Gidoza
Famous Hero
|
|
posted April 23, 2020 01:34 PM |
|
Edited by Gidoza at 13:36, 23 Apr 2020.
|
JoonasTo said:
We had a rough formula for quick estimation to land in the general ballpark for units in Elvin’s duel map balancing. In short, initiative is worth more than just the raw damage. And the more damage or initiative the unit has, the more extra initiative is worth.
So going from 10 init to 11 for 1 damage unit is worth very little but going from 15 to 16 for a 100 damage unit is massively broken.
Feel free to look around
The stats are all new so keep that in mind as you go. The numbers are not relevant for vanilla creatures but the formulas are quite good in our experience.
It’s abilities that are the real problem. A unit might be balanced on average very well but it might have massive potential if utilised correctly. So average player has some oooh, sweet moments while an expert will crush the map and/or opponent because of it.
Hey JoonasTo,
Thanks, this is interesting - I'd love to be able to download it and fiddle with it. Unfortunately I don't even understand most of the formulas I'm seeing.  But I'd love to compare notes and see what happens. Was this project of yours already finished? In any case thank you.
EDIT: Regarding abilities - I think I can understand you there in principle. For fun I decided to create a high-initiative Ghost Dragon with low damage and attack and heavy defence. It ends up running around the field casting Sorrow on everything and eating up enemy retaliations - that's the point of it. It doesn't exactly kill that much, but it serves its purpose. I like the design overall, but it needs some reigns pulled back on it a little.
|
|
frogo
Adventuring Hero
|
|
posted April 23, 2020 04:34 PM |
|
Edited by frogo at 18:19, 23 Apr 2020.
|
@Gidoza
Quote:
You're right that converging to 0 yields crazy results
Actually, in the example where i used the term converging to 0, which was where the ratio dmg/hp converges towards 0, the results dont go crazy, but they are exactly what you were using at the start.
Under the assumption that its single stack vs single stack, no hero, stacks only defined by HP, dmg, and initative, where dmg is proportional to HP (dmg = f(HP) where f is continous, rather than those "jumps" when a unit dies), then:
If dmg/hp -> 0 then: creature_strength -> hp*dmg*iniative.
Results actually go crazy as you get further away from 0 with dmg/hp.
Quote:
I'm not terribly worried about the 1.4 multiplier on ranged units (for example) because something non-trivial is always being sacrificed: you're going to end up with a garbage ranged unit with poor damage, or a ranged unit that is solid but falls quickly to lightning bolts.
Sure, when decreasing a units hp and therefore increasing its dmg, the unit both loses and wins something. I stated that for a ranged unit, having a highe dmg/hp ratio is more important than for a melee unit. As you are trying to come up with a formula that calculates a units worth, its not sufficient to state that the unit loses something in turn for its gain. What you care about is >how much< is lost compared to what is gained, when shifting dmg and hp.
An example:
Lets have the same assumptions as described above to simplify the situation.
As we dont care about initative in this scenario, lets simplify it even more by throwing that out of the equation aswell.
So we can easily calculate a units strength with hp*dmg.
Now lets have 2 units:
A: 1 dmg, 10 hp => strength = 1 * 10 = 10
B: 2 dmg, 5 hp => strength = 2 * 5 = 10
Suppose being ranged means that the unit gets in 1 hit before the other unit can engage. Lets say this is equal to a 1.4 multiplier to unit strength:
A: 1 dmg, 10 hp, ranged => strength = 1 * 10 *1.4 = 14
B: 2 dmg, 5 hp, ranged => strength = 2 * 5 *1.4 = 14
Lets define a unit C, that will be the enemy:
C: 3dmg, 5hp => 3 * 5 = 15
A vs C => C wins
B vs C => B wins
We can see B has gained more from becoming a ranged unit.
Sure B is more vulnerable to a lightning strike (or any dmg) than A,
but this was already the case before we made them ranged and has nothing to do with it.
____________
|
| |
|
|
