Auto-translated
Yes, that means he will win; I checked it myself!
Geksorin
Posts from - Units in Heroes 3
Geksorin
Zingel
Auto-translated
Yes, the table is interesting, but can Phoenixes be qualified if there are twice as many of them? This raises the question of how to judge then?
I'll post summary tables on weekly spawns, castles, etc. later.
Added 4 minutes later
And was it taken into account that creatures reduce defense? It seems not...
It doesn't even take into account who goes first - they go at the same time! The main thing is not who inflicts how much damage, but who can defeat whom in fewer turns!
speed, abilities, flight, double attacks... will be a little later.
Герои - это клёво.
Geksorin
Auto-translated
Let's do it! Overall, it's interesting statistics, but I don't think we should include additional turns, as they don't always happen!
Qwerty
Auto-translated
An unretaliating unit will attack twice as often.
X-plode
Auto-translated
Yes, but this table isn't 100% accurate for ranged units!!! Because some ranged units don't have a penalty in melee combat, while others do!!!
P.S. This is because I love ranged units (elves, titans, eye units, etc.)
Added 1 minute later
We can also create a similar table (a more complete one) for neutral creatures later!!! I want to know how strong snipers are.
P.S. This is because I love ranged units (elves, titans, eye units, etc.)
Added 1 minute later
We can also create a similar table (a more complete one) for neutral creatures later!!! I want to know how strong snipers are.
Zingel
Auto-translated
I didn't include neutral creatures because they have some incredibly powerful dragons that, when evaluated using my method, fall outside the acceptable range, resulting in completely biased information.
I suggest we have a discussion:
Regarding speed, I think it should be represented as a coefficient (I suggest 2% for each point of speed) applied to the overall score.
The "shot" and "unreturned strike" abilities: a coefficient of x2 to attack.
Double strike: something like a coefficient of (1.5-1.75), also applied to attack.
Flight: roughly 20% to the score.
Multi-target strike and biting multiple monsters: +15-30%.
Vampirism: (+ attack coefficient, converted into health and participating in the calculation of the defense coefficient) + (this coefficient, converted into attack and added to the initial attack coefficient) :)
Dragon strike (2 squares): +10-30%.
Remind me about the main abilities.
The rest: morale, magic resistance, double damage, mana drain and reduction, wall destruction :), etc. - somewhere around +5%.
In general, all these percentages should be calculated based on mathematical expectation (who is good at math here? :) ).
I'm waiting for your opinions! :)
Герои - это клёво.
AmberSoler
Auto-translated
I suggest we hold a council:...
1. Double Strike is no different from an unretaliated strike... A x2 coefficient would work for both...
2. Dragon Strike should be lower than multi-strike and "biting," because it hits a maximum of two units, while the others hit at least three. Archers who use area-of-effect spells (liches and magogs) can be equated to multi-strike.
3. Using mathematical expectation is the most accurate way to calculate coefficients, but the essence of the issue isn't so fundamental as to require bringing out the heavy artillery. The reason is that we still won't account for all influencing factors, and the margin of error will be much higher than the precision of the tool.
Qwerty
Auto-translated
No, AmberSoler, you can't equate a double strike with an unreturned attack. Because after the first strike, there will be a counterattack, and the second strike will only be delivered by those who survived the counterattack.
Zingel
Auto-translated
Here, this is closer to reality :)
1. Double Strike is no different from Retaliation... In both cases, a coefficient of x2 will work...
Double Strike is no different from Retaliation only in the case when... well, you know when.
In a normal case, after the first strike, the stack size decreases, and consequently, the strength of the next strike decreases (we are considering not a single unit, but a stack, otherwise it wouldn't make sense). As follows from the general trend, the average damage fluctuates within 1/2-1/5 of the unit's HP, i.e., after the retaliatory strike, the stack size will decrease by the average value :) 1/3.5, i.e., by about 30%, so the second strike will be performed with a strength of 0.6-0.7 of the initial value.
Considering that in the case when... this coefficient is slightly increased, we get approximately 0.75. :)
2. Dragon Strike should be lower than Multi-Strike and "Bite" because it hits a maximum of two units, while the others hit at least three. Multi-Strike can be compared to archers who attack with area-of-effect spells (liches and magogs).
3. Using expected value is the most accurate way to calculate coefficients, but the essence of the question is not so fundamental as to use heavy artillery. The reason is that we still won't take into account all the influencing factors, and the error will be much higher than the accuracy of the tool.
- Yes, colleague!
- Allow me to shake your hand!
- Please, colleague! :)
Mathematical expectation can easily replace the personal observation of a small group (even if it's 3-5 people). But "thoughtful" people. :)
Герои - это клёво.
AmberSoler
Auto-translated
...In a normal case, after the first hit, the stack decreases, and consequently, the strength of the next hit decreases (we are considering not a single unit, but a stack, otherwise it doesn't make sense).
I would like to draw your attention to the fact that it is not the attack that needs to be adjusted, but the damage... Because the developers have incorporated a completely different calculation mechanism for the situation when the attack is higher than the enemy's defense and for the situation when it is lower... By modifying the attack, we will unjustifiably change the calculation mechanism, which will not actually happen in reality...
Sorry for getting into the details, but since we've already started... :)
Added 7 minutes later
Flight - roughly 20% to the points
Zingel
Auto-translated
Then let's take the opposite point of view... How does a unit's double strike differ from a single strike of another unit, but with twice the damage?.. Essentially, we can make a simple substitution. Instead of a Crusader, let's take a Manticore: the average damage of both units within a single round is identical (all other things being equal). So, on what basis do we apply a coefficient of 1.75 to the first unit? Wouldn't it be simpler to include the doubled damage (not the attack) in the calculation scheme and forget about the boosting coefficient altogether? This would be more accurate...
Maybe I'm not expressing my thoughts clearly enough, but when considering a "one-on-one" fight, that's how it will be. But in this case, we lose an objective (real) in-game assessment of the unit's strength.
And when considering a "group" versus "group" fight, we find that a full-fledged double strike is only possible if the "attacked" unit doesn't retaliate. :)
In 70% of cases, after the first strike, there is a response, the number of warriors in the stack decreases, and the second strike turns out to be weaker.
The question is, what coefficient should be used for the second strike...
The average damage dealt by a unit is equal to 1/4 of the unit's HP, therefore the stack will decrease by a quarter, which is directly proportional to the damage. Taking into account the expected percentage of response to the first strike, we get a value for the strength of the second strike in the range of 70-80% of the initial value.
There. :)
I would like to draw your attention to the fact that it is not the attack that needs to be adjusted, but the damage... Because the developers have incorporated a completely different calculation mechanism for the situation when the attack is higher than the opponent's defense and for the situation when it is lower... By modifying the attack, we will unjustifiably change the calculation mechanism, which will not actually happen in reality...
Added 7 minutes ago
[/U]
Too much, IMHO...
And yes, it's a bit too much; I suggest reducing it to 10%. Well, here it's just a guess. :)
Герои - это клёво.
AmberSoler
Auto-translated
As follows from the general trend - average damage fluctuates within 1/2-1/5 of a unit's HP, meaning that after a retaliation strike, the stack strength will decrease by an averaged mean value :) 1/3.5, i.e., around 30%, and thus the second hit will be delivered with 0.6-0.7 of the original strength.
2. Regarding the 1.75 coefficient for the double strike, I take your side... We've sorted it out, so to speak... Everything is correct.
Zingel
Auto-translated
You've overreached here :) It turns out that with every retaliation, the number of units in the stack decreases by 30%... But that's not how it works... After all, damage dealt BY a unit and damage dealt TO a unit are not the same thing...
I haven't analyzed this situation yet... although, there is Excel. One moment.
Oops...
The total health of all (non-neutral) creatures is 7675, the sum of all average damages is 1625. Therefore, the average damage relative to average health is 0.21.
But after the upgrade, this coefficient increases slightly.
Then we take 23%
Consequently, the second hit will be (0.8-0.9) of the first.
Герои - это клёво.
AmberSoler
Auto-translated
The average damage dealt by a unit is equal to 1/4 of the unit's HP, therefore the stack will decrease by a quarter, which is directly proportional to the damage.
Your point will be absolutely correct only in the case of fixed damage, regardless of other circumstances. If you suggest considering the damage as fixed, I will accept your version, and in the future, we will simply keep in mind that we have included some error in our estimates... Or even better - I suggest introducing a progressive scale. For units with greater defense, introduce a certain compensating coefficient that reflects their higher resistance to damage received... I am sure you understand my idea... Otherwise, it will be too crude...
For reference: if the attacker's attack is lower than the victim's defense by 28 or more points, the attacker will deal exactly 0.3 of his damage... This limitation is set by the developers... For example, for a Lazurite Golem, this will be true in almost any battle where its opponent is below level 7... Only champions have enough attack to deal more than 0.3 of the base damage...
Zingel
Auto-translated
I still insist on a more thoughtful approach to the damage dealt... It is adjusted depending on the opponent's defense... That is, the stronger the opponent's defense, the lower the damage dealt to HIM, expressed in HP - down to 0.3 of the base damage...
And we see that the average attack skill is greater than the average defense skill (in total for all units, 1401 versus 1311). This means that the actual damage dealt is even slightly higher than stated in the unit characteristics.
Added 13 minutes ago
Your statement will certainly only be true in the case of fixed damage, regardless of other circumstances. If you suggest considering the damage as fixed, I will accept your version, and in the future, we will simply keep in mind that we have included some error in our estimates... Or even better - I suggest introducing a progressive scale. For units with greater defense, introduce a certain compensating coefficient that reflects their higher resistance to damage received... I am sure you understand my idea... Otherwise, it will be too crude...
Exactly! (regarding the progressive scale) This is very correct, but is it worth the effort... :(
Let me share my thoughts.
My "coolness" table for units was based on 2 points.
1 - how many hits it takes for this unit to "kill" an Archangel (power).
2 - how many hits it takes for a Peasant to "kill" this unit (durability).
It was not possible to create a linear scale of units in any other way, and if you take any 2 and compare them to each other (who will destroy the other in fewer hits), it didn't work. (If you check my table, you will notice only a few inconsistencies, but they are very insignificant and few) (honestly - I only saw 1 such glitch)
That is, the table is 99% accurate.
But this accuracy was achieved only with a slight adjustment of the skills, i.e., the weakest unit killed the unit being tested with 1 damage and 5 attack skill! And the most durable unit, which was killed by the units being tested, had a defense skill of 20!
That is, with values of 1 and 30, a less accurate table was obtained! Moreover, it also turned out to be even less accurate with the "most" average values of attack and defense skills.
This is where a differentiating progressive scale would be needed. (but I was too lazy) And again, we come to a contradiction: in the Elemental City, units of levels 2-5 are actually all of levels 3-4.
Okay, I'll continue tomorrow. The workday is ending. Good luck to everyone! :)
Added 3 minutes ago
For reference: if the attacker's attack is lower than the victim's defense by 28 or more points, the attacker will deal exactly 0.3 of his damage... This limitation is set by the developers... For example, for a Lazurite, this will be true in almost any battle where his opponent is below level 7... Only champions have enough attack to deal more than 0.3 of the base damage...
We know, we know.
But in this case, the "coolness" of a Centaur against a Demon will be lower than the "coolness" of a Centaur against a Lazurite (for example).
No, not quite. After comparing the Centaur with the Lazurite and the Demon with the Lazurite, we get a coefficient.
Then, these coefficients cannot be compared with each other, because the Centaur's coefficient will be slightly overestimated due to the fact that the Demon's coefficient will be underestimated. Something like that.
Therefore, this moment was not taken into account in the calculation. And I think this is correct! :)
Because we are not calculating damage, but calculating "coolness"!
:-)
Герои - это клёво.
Statistics
Welcome our newest member:
Emil