Damage type discussion

[zapbib]

Don't boost too much blunt weapon, they already have a damage bonus versus armored unit that make a lower damage blunt weapon to perform quite better then a sword.

 

[Leifdin]

I agree - I tend to give blunt weapons to either low tier units (vaegir peasants, for example), bandits or very specific units. There now actually is a unit called Watcher, who has a blunt polearm and Watchful Eye.

 

[kirkusmaximus]

I agree - I tend to give blunt weapons to either low tier units (vaegir peasants, for example), bandits or very specific units. There now actually is a unit called Watcher, who has a blunt polearm and Watchful Eye.

It is not that much better than piercing weapons against heavily armed targets. it only pulls a little ahead. Overall piercing has a slight advantage. Cutting weapons are laughable. only weapon master +savant keeps it level with the other 2 types and that is with a lot more investment. I have some charts I have been working on to illustrate the differences between the three types but i need some data to make it meaningful. Does anyone have a spreadsheet of all the different armors and their values?

I also found an elegant solution to the problem but it would require reworking cutting weapons and reworking weapon master + savant. essentially it would make cutting weapons work well against light armor, piercing already is best against medium but , and blunt already works best against heavy. As it stands currently cutting does not deal more against light without weapon master + savant. I don't want to release the charts and solution until i work with real values instead of hypothetical.

 

[theman007]

As it stands currently cutting does not deal more against light without weapon master + savant.

Piercing should be, in general, the most damaging, as it was and still is in real combat, in general. I'm not suggesting that the current balance is fine or realistic however.

Regardless, I think your focus may be wrong. There is the question of effectiveness rather than pure damage per hit. Cutting attacks tend to be much easier to use. If not, then they tend to have more range and speed (e.g. arming sword swing vs. military pick swing). Thrusts are more difficult to hit with and are slower because the player often needs to spend more time aiming. But the player can use any weapon type with ease and effectiveness, and therefore player use of weapons should not drive balancing. Damage types must be balanced for the AI, not the player, and the AI are simply inept at thrusting with anything. Pierce damage is perhaps the AI's least effective damage type, with cutting certainly being the most effective - think mounted sword swings, elite scimitar, huscarl axes, sergeant cleavers, etc. Yes, there are exceptions, and yes effectiveness is contextual - I'm talking generally here.

 

[Leifdin]

The solution was quite easy, I just changed damage values - cutting deals a lot of damage, but is not too effective against armour without a lot of power strike. Piercing deals less damage, but is much more effective against armour. Blunt weapons damage is a joke - hammers that are slow and short have like 45 for the highest tier. Warhammer is quite short, bit faster, but only has 32 (it is also a tier 5 weapon, not currently used by any troops). Peasant spear, used by low tier units has 25 cut and 25 pierce, cutting spear, which is used by t3 to t5 units has 36 cut and 27 pierce. Top tier spear has 25 blunt and 39 pierce. Awlpike has 27 blunt and 49 pierce, but is slower. Awlpike may need a nerf.

 

[kirkusmaximus]

Piercing should be, in general, the most damaging, as it was and still is in real combat, in general. I'm not suggesting that the current balance is fine or realistic however.

Regardless, I think your focus may be wrong. There is the question of effectiveness rather than pure damage per hit. Cutting attacks tend to be much easier to use. If not, then they tend to have more range and speed (e.g. arming sword swing vs. military pick swing). Thrusts are more difficult to hit with and are slower because the player often needs to spend more time aiming. But the player can use any weapon type with ease and effectiveness, and therefore player use of weapons should not drive balancing. Damage types must be balanced for the AI, not the player, and the AI are simply inept at thrusting with anything. Pierce damage is perhaps the AI's least effective damage type, with cutting certainly being the most effective - think mounted sword swings, elite scimitar, huscarl axes, sergeant cleavers, etc. Yes, there are exceptions, and yes effectiveness is contextual - I'm talking generally here.

Piercing is better overall. On light,medium,and heavy armor combined it has the best average damage output. It is at it's best against medium armor was all i was getting at there.

Point taken. But most of the things you are talking about are about the swing type and not the damage type. Give them a 2 hander with overhead swing with bashing and it becomes rather great. The spiked mace that deals piercing damage is superbly deadly due to being piercing and crushing and the fact that the AI knows how to swing much better than thrust. It was even better when it was a 1/2 hander.

The solution was quite easy, I just changed damage values - cutting deals a lot of damage, but is not too effective against armour without a lot of power strike. Piercing deals less damage, but is much more effective against armour. Blunt weapons damage is a joke - hammers that are slow and short have like 45 for the highest tier. Warhammer is quite short, bit faster, but only has 32 (it is also a tier 5 weapon, not currently used by any troops). Peasant spear, used by low tier units has 25 cut and 25 pierce, cutting spear, which is used by t3 to t5 units has 36 cut and 27 pierce. Top tier spear has 25 blunt and 39 pierce. Awlpike has 27 blunt and 49 pierce, but is slower. Awlpike may need a nerf.

This is the quickest fix but certainly not the most elegant. This causes issues in the original too.

Take for example 3 weapons, each of cutting, piercing and blunt. Each deals 25 damage. To someone new to the game they would assume that all will deal approximately the same damage. They would learn over time the rules that govern the damage types and realize that all 3 of these weapons deal vastly different types of damage. What comes next is that player trying to figure out which is better, a high damage cutting sword, a low damage blunt weapon, or a medium amount of piercing damage.

The solution I am putting forth would let a new player come in and pick from either of those 3 and realize that cutting deals best damage against low armor, piercing is best against medium, and blunt is best against heavy armor. Really Blunt damage in the main game is balanced around the fact that you can sell prisoners and that's fine. I just don't want people to need math degrees to compare two items of different damage types. Yes speed factors into this and so does proficiency but it should still not be as complicated as it is currently.

To be honest I don't expect this to be fixed, the rework of the weapons flanked me before I could finish a real analysis. I also assume some of my assumptions will be wrong when i figure out what light medium and heavy mean from the game's term, hence why i am trying to export the data from the game into a spreadsheet before i displayed.

Really just some food for thought at this point. Speaking of the warhammer, it can thrust for some reason but has zero thrust damage.

 

[theman007]

Piercing is better overall. On light,medium,and heavy armor combined it has the best average damage output. It is at it's best against medium armor was all i was getting at there.

Point taken. But most of the things you are talking about are about the swing type and not the damage type. Give them a 2 hander with overhead swing with bashing and it becomes rather great. The spiked mace that deals piercing damage is superbly deadly due to being piercing and crushing and the fact that the AI knows how to swing much better than thrust. It was even better when it was a 1/2 hander.

Swing type, range, speed, etc. are very important to factor in. We have to account for the fact that there is no blunt equivalent of the elite scimitar, great sword or decent length arming sword. This matters. If you went into the files and changed the damage type of a two handed to blunt or pierce, then damage output would increase significantly, but that is not what we're working with. We have to take into account that an extra 33 reach and 3 speed  makes the elite scimitar superior to the military pick in most cases, despite its weaker damage type. You couldn't have, for instance, a mace or warhammer that was as fast and as long as an elite scimitar - form absolutely matters here. We need to consider in-game effectiveness rather than what the values in the files tell us. This is demonstrably provable by creating identical multiplayer AI troops with the exception of weapon, and running team deathmatches whilst spectating. Give a unit a scimitar and they trump a unit with a mace, repeatedly, as long as average-level armor is used. Speed is one of the most important factors when considering how effective a weapon will be used by AI. The troop with the faster weapon will often get the first hit in, and likely win the fight if they are elsewise similar. This is because the AI is stupid and won't block, allowing faster weapons to win by spam. Again, there are exceptions and contextual variations. Of course, sometimes, a mace wins every time.

As it stands, I take the position that cutting is too powerful, and piercing is too weak.

 

[kirkusmaximus]

Swing type, range, speed, etc. are very important to factor in. We have to account for the fact that there is no blunt equivalent of the elite scimitar, great sword or decent length arming sword. This matters.

I was not dismissing the other variables merely pointing out that the differences in damage are unintelligible to players at a single glance.

If you went into the files and changed the damage type of a two handed to blunt or pierce, then damage output would increase significantly, but that is not what we're working with. We have to take into account that an extra 33 reach and 3 speed  makes the elite scimitar superior to the military pick in most cases, despite its weaker damage type. You couldn't have, for instance, a mace or warhammer that was as fast and as long as an elite scimitar - form absolutely matters here.

I don't see a reason we could not make an equivalent item actually. If you say there is no top tier 1-hander comparable to the best top tier cutting weapon then one should and could be added.

We need to consider in-game effectiveness rather than what the values in the files tell us. This is demonstrably provable by creating identical multiplayer AI troops with the exception of weapon, and running team deathmatches whilst spectating. Give a unit a scimitar and they trump a unit with a mace, repeatedly, as long as average-level armor is used. Speed is one of the most important factors when considering how effective a weapon will be used by AI. The troop with the faster weapon will often get the first hit in, and likely win the fight if they are elsewise similar. This is because the AI is stupid and won't block, allowing faster weapons to win by spam. Again, there are exceptions and contextual variations. Of course, sometimes, a mace wins every time.

As it stands, I take the position that cutting is too powerful, and piercing is too weak.

Agreed. Some more item balance in is order. Although i find the effectiveness of range varies depending on the situation and your opponents weapons as well. Also the situation changes heavily in er heavy armor  . Piercing weapons excel at siege defense and offense I find due to their short reach.

 

[theman007]

I don't see a reason we could not make an equivalent item actually. If you say there is no top tier 1-hander comparable to the best top tier cutting weapon then one should and could be added.

I mean in the sense can be no such weapon in terms of equivalent speed, reach, and capability, not in terms of power or damage output per second, etc. There could not be a nimble blunt weapon with decent reach because the point of balance of a blunt weapon must be close to the head. Boosting the attack of the two handed mace, for example, would not create an equivalent weapon to the great sword because it could still never be as fast or as long - of course it could be more powerful overall. 

 

[Leifdin]

I also found an elegant solution to the problem but it would require reworking cutting weapons and reworking weapon master + savant. essentially it would make cutting weapons work well against light armor, piercing already is best against medium but , and blunt already works best against heavy. As it stands currently cutting does not deal more against light without weapon master + savant. I don't want to release the charts and solution until i work with real values instead of hypothetical.

Sorry, I didn't notice your post. If you are willing to help with this, I am willing to put time and effort to write the code - now is the time to make items balanced, as it's the purpose of this release. There may not be time or space in future. I want to deal with this before I release 0.31.

 

[kirkusmaximus]

Sorry, I didn't notice your post. If you are willing to help with this, I am willing to put time and effort to write the code - now is the time to make items balanced, as it's the purpose of this release. There may not be time or space in future. I want to deal with this before I release 0.31.

Please see the link here https://drive.google.com/folderview?id=0B3nTCPcQZaOPUUxKSERweVlJd00&usp=sharing. I pulled the current values of the armor from the game and calculated the average light, medium, and heavy armor values. Likely will shift as more balance is done to the armor values. I also didn't remove tournament items or different colored variations so the value is likely off by a bit. Another thing to note is this is when hitting the body. Head and feet values were not taken into consideration. I will have to make compile that data later. I felt body was a good place to start since that is mostly where the AI hits.

Check the damage calc sheet. The first tab, original, is how M&B does it's damage. The second tab has my changes. Really I only changed the cutting values. Two ini changes makes a world of difference and you can see right away that numbers don't have to be quite so far away from each other now.

I'll add some fancy graphs and do a ramping damage sheet when i find more free time but those values there are the essence of the change I envisioned.

If you want access to the original sheet PM me and i'll give you the raw sheets.

 

[Leifdin]

Thank you, I'll read though this. BTW you don't need to make graphs for me, I can't read them very well. I prefer just plain text. Could you also please post some stats of dmg/armour before changes for reference?

 

[kirkusmaximus]

I posted 2 spreadsheets to the share. Armors and weapons. Armors just shows what i used to calculate the 3 levels of armor. Weapons has a few examples of how cutting weapons will change around line ~330 with a bunch of well known cutting weapons. It's not entirely filled in since I cannot decipher the capabilities string.

Armors will have no change in values, neither will piercing of blunt weapons. the only value that will change is on cutting attacks and those should go down 40% and still deal approximately the same damage, not counting weapon speed of course.

Take the western noblemen's sword. It is 22p 37c. with the new rules it is 22p 22c which is exactly what is meant to happen in a perfect world. This will not at all be an exact science as the amount of damage relies heavily on the raw damage dealt. As numbers get smaller the weapons tend to favor piercing and blunt and as they get higher they even out.

I'm honestly thinking this may be too much work, i think the numbers may have to change again. What would help immensely is if i could find the upper and lower levels of damage for each damage type and if it's 1h/2h/polearm but as such i present my findings incomplete as they are.