| Author | Hamster soon |
|---|
| Because there is one unit doing 5-10 damage even if you times that by 5 it can still only be 25 damage if it does min |
| Slightly irrelevant, but interesting. A tribal goblin with 1000 attack against someone with 20 def would do only 50 dmg. |
| Different story is when attacker has 30 attack and target 200 def - it won't help you much when there is huge number of units in attacker's stack ;) Still lot of dmg can pass and not big difference when difference is 100 parameters or 200 parameters when att<def |
Different story is when attacker has 30 attack and target 200 def - it won't help you much when there is huge number of units in attacker's stack ;) Still lot of dmg can pass and not big difference when difference is 100 parameters or 200 parameters when att<def
How so? |
not big difference when difference is 100 parameters or 200 parameters when att<def
Umm that is somewhat incorrect. In terms of difference in dmg it's not as much as when att>def but in terms of ratio of difference it's the same.
It just means that when you have +20 att you do 2x (which is should not be seen as 1+1 but rather 1*2) dmg, whereas when you have -20 att you do 0.5x dmg, not 0 dmg obviously (so not 1-1 but 1/(1+1)) which makes more sense.
Now obviously difference in magnitude of dmg when you have +20 att is +1x, whereas when you have -20 att is -0.5x, but it's not correct to say +20 att made more of a difference. Obviously because what matters is not change in magnitude of damage but change in proportion of damage. |
for randomr1:
I understand what you want to say, however my view is from different view. Imagine theoretical extreme case when you can distribute 100 parameters while you can use only 1 unit with base 0 attack, 0 defense and 100 damage. Imagine you would have to fight only against 1 enemy unit with 25 attack, 25 defense and 100 damage.
Before some maths we should look at this graph: http://prntscr.com/kik8s1
And now we ca distribute our 100 parameters:
(Advantage in the calculation stands for (damage dealt)/(damage received) ratio)
Case 1: All stats in defense
You have 0 att, 100 def; you deal 44 damage and receive 21 damage. Your advantage is 2.1
Case 2: All stats in attack
You have 100 att, 0 def; you deal 475 damage and receive 225 damage. Your advantage is 2.1
Case 3: balanced stats
You have 50 att, 50 def; you deal 225 damage and receive 44 damage. Your advantage is 5.1
So in this example balanced stats is the best choice (when your enemy has low balanced stats). Now imagine your enemy has 100 stats like you but he has 50/50 stats.
Case 1: All stats in defense
You have 0 att, 100 def; you deal 29 damage and receive 29 damage. Your advantage is 1.0 (equal)
Case 2: All stats in attack
You have 100 att, 0 def; you deal 350 damage and receive 350 damage. Your advantage is 1.0 (equal)
Case 3: balanced stats
You have 50 att, 50 def; you deal 100 damage and receive 100 damage. Your advantage is 1.0 (equal)
In this example it doesn't matter what you use. Thus only fight duration is affected when 2 units have the same amount of stats.
The last example is when enemy would have 50 attack and 50 defense, but you have only 50 stats:
Case 1: All stats in defense
You have 0 att, 50 def; you deal 29 damage and receive 100 damage. Your advantage is 0.29
Case 2: All stats in attack
You have 50 att, 0 def; you deal 100 damage and receive 350 damage. Your advantage is 0.29
Case 3: balanced stats
You have 25 att, 25 def; you deal 44 damage and receive 225 damage. Your advantage is 0.20 (the worst)
All these examples can be taken for adjusting your stats based on what tiers are your core units. For example (might) UN have huge T6 and T7 stacks who have higher stats than the most of enemies. From 1st example you can see that a bit more balanced stats can be more suitable for cases where you have more stats while it is not good idea to have balanced stats when your enemy has more stats.
For good overview of (damage dealt)/(damage received) ratio you can look at this link (when you can have 100 stats and enemy only 50):
http://prntscr.com/kilmjl
Another example when you can have 50 stats and enemy 100:
http://prntscr.com/kilnes
From these intensity graphs you can see how attack you should optimaly have from your 100 stats according to enemy's distribution of parameters. Yellow area is the best. |
for Nowar:
Ahh yes I am aware of stat optimization but thanks for the clear graphs and detailed analysis!
First of all, perhaps I was unclear but I never meant to say that no matter how you distribute stats it's the same. Obviously the closer your att or def is to enemy def or att respectively, the larger the rate of change in dmg for each att or def difference, which is why the calculation you showed give those results.
Damage wise what you said is correct, however, obviously there is more which goes to deciding stats other than simply comparing the ratio of dmg dealt to dmg taken. For example, in hunts (especially when you have mainly shooters), defense is almost meaningless. When your attack is say 40 more than the target def and defense is 10 more than target att, +1 attack only increases your damage by 1.67% whereas +1 def reduces dmg taken by 2.47%, but still it makes much more sense to increase attack.
Similarly, tribals usually have max def stats in PvP since tribal spirit overcompensates for the lack of stat optimization.
What I was talking about is the simple comparison between attack and defence. In other words, you don't have say 100 stats and are allowed to go up and down, you either have 100 att or 100 def. I have seen many people think attack is better than def just because the percent growth shown for att stats is higher than the percent reduction for def stats. A trivial misconception since the dmg calculations for att>def and def>att are different. I was just trying to point out that while the look different they mean effectively the same thing. That is NOT to say that no matter how you distribute them it makes no difference, but just that an attack stat is not necessarily superior to a def stat. |
All these examples can be taken for adjusting your stats based on what tiers are your core units. For example (might) UN have huge T6 and T7 stacks who have higher stats than the most of enemies. From 1st example you can see that a bit more balanced stats can be more suitable for cases where you have more stats while it is not good idea to have balanced stats when your enemy has more stats.
It's negating the fact that unholy recovers health back, tends to attack twice but be retaliated once (hit, raise, hit) and that max health shrinks quickly, so defense is not really interesting, unless playing with no KN but then the necro starts by shooting himself in the leg :)
You do have a point though in that most factions have tier 1-4 with overall low stats while the increase per tier skyrokets for last 3 tiers. That said it simulates the extreme of a T7 hitting a really low stat creature. You are taking a theoretical case where the ennemy has balanced stats which is an hedge case. So if I got it right, if I have more stats, I'm likely to face full ATK or full DEF. If I face full ATK, I'm supposed to have the optimum by having as much DEF as the ennemy has ATK and puting all the rest evenly beetween ATK and DEF. If ennemy is DEF then I want as much ATK and then spread the rest evenly too.
Although the ennemy with DEF does tend to have nasty side effects like spells, spirit or summons that make it desirable to have full ATK to take it down ASAP.
Eh, I tried to work something out like that years ago, now I remember why I kind of gave up on having a clear cut conclusion, there's too much for my little brain to sort out :( Thanks for the analysis nevertheless. |
| Lets not close this thread , instead of making new thread for every hamster , lets just keep this continue :p |
| Can someone please explain what the hamster is all about. |
| Hamster shows up every millionth battle and Awards Something like 100k gold if you beat it. |
| and 1 million gold for every 10 millionth battle. |
| Anyone here who encountered one ever? |
| Apurv anand its not allowed to post from alt :D |
for Chiki:
https://www.lordswm.com/forum_messages.php?tid=2658668 |
| wouldnt mind 100k |
for Apurv Anand:
uh i had twice, though i am primarily from ru server, but i like yours :)
there was a topic like this today on ru server and awkward moderator deleted it(
Actually i was thinking about making a clan where to invite people who had caught it and also those who are only trying to catch:)
So we can send a mail when hamster is coming soon and just help people and just chat a bit)
It's just i'm a bit on tight budget now and now i just plan to see how many people are interested in this) so wait for your ideas
Around 33k battles left for new one, anyway |
| about 6k away gl everyone |
| now 5. remeber the counter stops between 3-500 away so try to plan it well |
| 3k |
