AC softcap and low level hardcap discussion

[DeathsSilkyMist]

After taking a look at Haynar's post on AC softcaps, I decided to test if the formula given was a hardcap formula or a softcap formula:

I added a low level raw ac cap of level * 6 + 25.

To test this I put a Fungi and some AC gear on my level 5 Bard mule. I had two level 5 skeletons attack me so I could get some hit data. I did one test with 55 AC, and one test with 178 AC. I sampled 400 hits from each test.

===========================================
55 worn AC ("raw ac cap" of level * 6 + 25)
===========================================

DV = Damage Value

DV, Count
01, 167
02, 19
03, 25
04, 17
05, 27
06, 23
07, 28
08, 11
09, 34
10, 25
11, 18
12, 6

Total Damage = 1731

=================================
178 worn AC (123 AC above 55 cap)
=================================

DV = Damage Value

DV, Count
01, 160
02, 28
03, 27
04, 4
05, 32
06, 30
07, 31
08, 12
09, 24
10, 33
11, 12
12, 7

Total Damage = 1728

As you can see, the amount of damage is basically the same, as is the number of hits for 1 damage.

Therefore, the following formula is a low level hardcap (I don't know what level range this is):

I added a low level raw ac cap of level * 6 + 25.

Going off of this quote:

And we use a softcap system, not a hardcap based system.

This means there is a separate softcap system that is used as well.

The only things I saw that Haynar confirmed are the following:

Warriors get a 45% return above softcap.

Adding a shield increases ur softcap.

Using the information above, I will show what some of the EQEMU values are:

ClothACSoftcap = 75;
LeatherACSoftcap = 100;
MonkACSoftcap = 120;
ChainACSoftcap = 200;
PlateACSoftcap = 300;
WarriorACSoftcapReturn = 0.45;
KnightACSoftcapReturn = 0.33;
LowPlateChainACSoftcapReturn = 0.23;
LowChainLeatherACSoftcapReturn = 0.17;
CasterACSoftcapReturn = 0.06;

These are supposedly "old" values, but this is probably in the context of the EQEMU itself going past Velious. As you can see here, the Warrior AC softcap return is also 45%, so this is probably a good starting point for further testing.

[DeathsSilkyMist]

For completion's sake I did a test at 23 AC with another level 5 skeleton, so we could see what under the hardcap looks like:

DV = Damage Value

DV, Count
01, 61
02, 32
03, 30
04, 12
05, 35
06, 35
07, 24
08, 9
09, 38
10, 30
11, 16
12, 78

Total Damage = 2642

At low levels AC does indeed seem to help quite a bit. Going from 23 AC to 55 AC reduced my damage by ~35%. This also means 123 AC should have done something noticeable if it was softcapped instead of hardcapped.

[Jimjam]

Very interesting. We had an earlier thread which seemed to show that ac mainly moved hits from the highest possible value to the lowest value (suggesting counts min + max should be about the same, but these hit distributions don’t replicate that finding.

Did you ensure melee skills like defence were capped before capturing your data?

How do these results compare to your calculator? Are there any other explanations for the similarities between the two ‘high’ ac values results, eg by 55 raw ac the mob’s attack is already squelched?

[Snaggles]

Scaling is often the issue, or seems the debate on the forums. Do you feel this model can scale accurately?

Anecdotally, it seems a heavily geared alt in the early levels greatly benefits from AC but this quickly fades. While cleric and shaman AC-only buffs are questionably helpful, they do increase in effect per rank.

[DeathsSilkyMist]

Originally Posted by Jimjam [You must be logged in to view images. Log in or Register.]

Very interesting. We had an earlier thread which seemed to show that ac mainly moved hits from the highest possible value to the lowest value (suggesting counts min + max should be about the same, but these hit distributions don’t replicate that finding.

Did you ensure melee skills like defence were capped before capturing your data?

How do these results compare to your calculator? Are there any other explanations for the similarities between the two ‘high’ ac values results, eg by 55 raw ac the mob’s attack is already squelched?

My defense was capped.

To answer your question about min and max values:

1. The game has a Roll D20 function with two inputs. I will call these inputs "Wrath" (attacker) and "Mitigation" (defender). The code confusingly calls them "offense" (attacker) and "defense" (defender), but these inputs are more than simply the attacker's offense skill and the defender's defense skill.

2. The D20 roll is weighted based on the ratio of the attacker's Wrath to the defender's Mitigation. An unweighted D20 has an average roll of 10.5.

3. If the attacker's Wrath is 50 and the defender's Mitigation is 100, the D20 has a weighted average roll of ~6.5. This is when you see the cluster of rolls at the minimum damage value.

4. If the attacker's Wrath is 100 and the defender's Mitigation is 50, the D20 has a weighted average of ~14.5. This is when you see the cluster of rolls at the maximum damage value.

5. If the attacker's Wrath is 50 and the defender's Mitigation is 50, the D20 has the unweighted average of ~10.5. This is when you see a roughly equal amount of rolls at the minimum and maximum damage values.

The 23 AC test is a scenario where my Mitigation is roughly equal to the Skeleton's Wrath. This is why the number of minimum hits and maximum hits are about the same. The Skeleton's Wrath is slightly higher, which is why there are a few more maximum hits compared to minimum hits.

The 55 AC test and 178 AC test are scenarios where my Mitigation is significantly higher than the Skeleton's Wrath, which is why you see the cluster of rolls at the minimum damage value.

If my AC wasn't hardcapped, you would see an increase in how many damage values were at the minimum damage value.

My damage calculator shows the same pattern. It doesn't have the AC hardcap or softcap built in, so more AC will increase how many damage values were at the minimum damage value.

[kjs86z2]

surely testing on a level 5 bard on level 5 skeletons with an armor difference of 120 is the 100% accurate way to prove something nobody gives a flying fuck about

[DeathsSilkyMist]

Originally Posted by kjs86z2 [You must be logged in to view images. Log in or Register.]

surely testing on a level 5 bard on level 5 skeletons with an armor difference of 120 is the 100% accurate way to prove something nobody gives a flying fuck about

The reason why the level 5 test matters is it shows that the formula provided by Haynar is not the softcap formula. It is a separate hardcap formula for low levels.

So we need to look elsewhere if we want to find the AC softcap for a level 60 melee class.

This also probably explains why Rangers feel like AC doesn't do as much for them. If they are hitting the softcap, they will see significant diminishing returns. This is especially true if a Ranger's softcap is lower than a Warriors or a Knights, which is suggested by the EQEMU numbers.

[Snaggles]

IMHO, I would probably use a level 5 warrior, monk, paladin or SK.

I still am not certain if there is some kind of scaling issue to account for, but nobody cares how AC affects bards and rangers.

[Jimjam]

I care >:-(

But only about rangers though.

[DeathsSilkyMist]

Originally Posted by Snaggles [You must be logged in to view images. Log in or Register.]

IMHO, I would probably use a level 5 warrior, monk, paladin or SK.

I still am not certain if there is some kind of scaling issue to account for, but nobody cares how AC affects bards and rangers.

You are correct that a Warrior would have less diminishing returns on the softcap, but my Bard hit the hardcap. Haynar did not say that the hardcap formula was different between classes.