Armor Class and Raid Tanking

[DeathsSilkyMist]

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

I'm referring to this two posts of yours:

https://www.project1999.com/forums/s...3&postcount=94

https://www.project1999.com/forums/s...&postcount=105

From the parses it seems there's a similar difference between 177 to 200, 200 to 217 and 217 to 250 but 250 to 300 there is barely anything. So it seems the softcap would be somewhere between 200 and 250 against that mob for chain classes, if armor type actually plays a role pre-softcap. In the ranger AC thread lvl 40 mobs softcapped at 180ac worn for both druid (leather) and ranger (chain) and lvl 45 at 200ac worn, regardless of if plate/chain. At lvl 50 it seems between 217 and 250. Wouldn't that disprove there's an actual armor type softcap? I was under the impression we pretty much proved that wrong unless there's a squelsh point AND a soft cap.

Not sure if I'm missing or misreading something.

The thread is a bit chaotic, as there were a few possibilities that I was experimenting with. So it isn't your fault, I wish I could tidy threads up afterwards to reduce confusion.

To clarify, I don't think the softcap is based on what armor you can wear in the strictest sense (i.e. Shamans are not a plate class simply because they can wear plate). The distiction is class specific.

If you want to know what class is assigned which armor type, look at what armor pieces are needed for their Velious Armor quests. Shamans use chain armor for their Velious Armor quests. So they are a chain class, even though they can use plate. There is a second distinction in terms of what softcap return you get:

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;

Since Shamans are a chain class, their Softcap is around 200. Furthermore, it lools like they are using the LowChainLeatherACSoftcapReturn softcap return. So any AC above the softcap is multiplied by 0.17.

This is why Rangers have an AC softcap around 200. They use chain for their Velious Armor as well.

Here are the last two datasets I posted:

https://www.project1999.com/forums/s...&postcount=198

https://www.project1999.com/forums/s...&postcount=199

Looking at it again, I may just need to get a larger data set for each value. In post 198, you can see that going from 100 AC to 200 AC is significant in terms of damage reduction. I took 53k damage with 100 AC, and 41.8k damage with 200 AC. Past 200 AC the damage reduction differences are significantly smaller, so a softcap is clearly in place around 200 AC.

1000 hits per test can still be noisy enough to make it look like the softcap is a bit above 200. The 235 AC test with lodi shield in the secondary slot took a bit less damage than the 250 AC test with lodi shield in the secondary slot, so we know that variance can still mess with the data a bit. Obviously 250 AC with a shield should reduce damage a tad bit more than 235 AC with a shield, but RnG is always a factor.

Either way, even if the softcap is 220 rather than 200, that is fairly close to the 200 value from EQEMU. I would be suprised if the Warrior AC softcap on P99 was like 400 or 500. It could be 320 instead of 300. Testing will need to be done to know for sure though.

[Vivitron]

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

At this point my guess is that p1999 adopts these eqemu values, except for the monk values -- monk ac seems to be significantly more effective here. I think your shaman parses are the best public attempt to check this, so there's significantly more uncertainty on the other classes.

[Goregasmic]

I can dig it up again but in the ranger thread bcbrown parsed a druid capping at 180, same as his ranger with chain, on a lvl 40 mob. A warrior, SK and ranger also capped at 200 on the level 45 mob. Either we're missing something or p99 isn't using those values. Druid should softcap at 100 according to that chart.

[bcbrown]

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

I can dig it up again but in the ranger thread bcbrown parsed a druid capping at 180, same as his ranger with chain, on a lvl 40 mob. A warrior, SK and ranger also capped at 200 on the level 45 mob. Either we're missing something or p99 isn't using those values. Druid should softcap at 100 according to that chart.

Yeah, that's what I recall the conclusions being as well. However, I think there were enough flaws and shortcomings in those parses to not treat them as definitive. I plan to reparse those experiments in a couple months when football starts again.

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

I have a 60 warrior with moderate raid gear on green 6650hp and 1350? Ac buffed last I checked. I can also use a different warrior with 1500ac. Did you just auto attack and parse logs? I guess ide have to find someone bored enough to heal me.

I'm always willing to heal someone for a couple hours of parsing. Send me a pm and we can coordinate something.

Originally Posted by Stonewallx39 [You must be logged in to view images. Log in or Register.]

I have a level 60 warrior with 6350 hp and ~1275ac (depending on buffs) wearing aggro weapons and fully raid buffed

I’m looking at future upgrades and trying to decide on priorities. I wanted to see if anyone could share testing that shows the impact of additional armor class on the damage dealt by raid targets (in particular NTOV dragons, AOW, Tuna).

I had heard previously that HP was king and raid targets had such high attack additional AC was mostly trivial. However I’ve been hearing more and more conversations on the effect of AC and it seems to be a significant consideration.

Danth gave a good summary of why HP is more important than AC: CH chain timings need to use worst-case scenarios. However, if you want to see what impact adding AC might have against raid targets, I'd suggest looking at some of your logs while tanking raid targets and looking at the proportion of hits that are for max value. Adding AC turns what would be max hits into moderate or min hits, and removing AC turns min hits into max hits. For a well-geared tank most blue cons will have something like 30% min hits and <5% max hits, but against a level 70 raid target it's probably mostly max hits.

The way I think about it is that if most hits are already for min value, adding AC isn't going to make a significant difference, while if most hits are for max value, it might take a lot of additional AC before you start seeing an effect. I suspect additional AC will have the most impact when there's roughly an equal number of max and min hits. I suspect part of the reason why for raid tanking HP is far more important than AC is that since the mobs are so much higher level, you need a much higher amount of AC before you start seeing any significant effect on mitigation.

[CrazyPro]

P99 is using an out of era AC system and raid mob ATK values are higher than they were in classic so until that changes:
Vyrinn's Earring of Insanity + Essence Pearl > double Hammered Golden Loops
Vyemm's Left Eye > War Bow of Rallos Zek
are the only real situations where you'd go for the higher AC item over the high HP item
and of course if you have 2 items with the same HP but one has more AC than the other, the one with more AC is better
high AC will just naturally come with high HP gear

at the end of the day just go for doze bracers and your HP and AC will skyrocket as those are the 2 biggest items for tanks because all of the other bracer options are way worse and it's not close

[Goregasmic]

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

Yeah, that's what I recall the conclusions being as well. However, I think there were enough flaws and shortcomings in those parses to not treat them as definitive. I plan to reparse those experiments in a couple months when football starts again.

Even if the parses weren't perfect leather shouldnt softcap anywhere near 200.

And if every armor type softcaps at 200 on shiel then I guess the main conclusion is shiel is too low level to know if softcap is a fixed value or dynamic depending on mob's atk value.

I have a cleric now so almlso willing fo help on parsing on green.

[bcbrown]

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

Even if the parses weren't perfect leather shouldnt softcap anywhere near 200.

And if every armor type softcaps at 200 on shiel then I guess the main conclusion is shiel is too low level to know if softcap is a fixed value or dynamic depending on mob's atk value.

I have a cleric now so almlso willing fo help on parsing on green.

I reread the thread you're referencing to refresh my memory. Here's the relevant posts:
Druid, 146 worn + 61 spell ac vs 99 worn + 61 spell ac: https://www.project1999.com/forums/s...10#post3724910
146 is much more min hits than max hits, 99 is slightly more max hits than min hits

Cleric, 195 worn ac +64 spell ac vs 132 worn ac + 64 spell ac: https://www.project1999.com/forums/s...34#post3723434
195 is mostly min hits, 132 is about a 2:1 ratio of min to max hits

Cleric, 163 worn ac + 64 spell ac vs 106 worn ac + 64 spell ac: https://www.project1999.com/forums/s...39#post3724039
163 worn ac is mostly min hits, 106 ac is equal numbers of max and min hits

One of the main flaws in those parses is that I'm not calculating specific metrics like avg dmg/hit or % of min/max hit. It's just graphs and qualitative observation. I guess my conclusion is that whether or not druid (as a leather class) has a 100 ac softcap, going from ~100 worn ac to ~150 has a big impact vs a lvl40 mob. And it seems like both druid (leather class) and cleric (plate class) have similar-ish benefits to adding AC in the 100-200 range vs a lvl40 mob.

This implies that either the leather AC softcap is much higher than 100, or that there's not much of a penalty for adding AC above the softcap, or that there isn't a softcap. In other words, I agree with you [You must be logged in to view images. Log in or Register.]

[DeathsSilkyMist]

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

This implies that either the leather AC softcap is much higher than 100, or that there's not much of a penalty for adding AC above the softcap, or that there isn't a softcap. In other words, I agree with you [You must be logged in to view images. Log in or Register.]

There is certainly a softcap based on all evidence and data thus far. There is no evidence of the "squelsh point" theory from what I have seen.

I have been doing some parsing today. I haven't finished yet on all the sets, but here is my 60 SK getting hit by Shiel Glimmerspindle with 300 AC. I realized I was able to just hit 300 AC without a shield using a few alternate pieces in the bank:

=========================
60 SK 1000 hits - 300 AC, No Shield
=========================
DV, Count
19, 453
22, 34
25, 36
28, 23
32, 42
35, 25
38, 26
42, 35
45, 32
48, 28
52, 24
55, 38
58, 27
62, 26
65, 33
68, 24
72, 29
75, 27
78, 32
82, 6

Total Damage = 35754

When we compare this to my Shaman's test:

=================================
60 Shaman 1000 hits - 363 AC Test, No Shield
=================================

DV, Count
19, 381
22, 22
25, 33
28, 28
32, 35
35, 39
38, 27
42, 37
45, 37
48, 38
52, 28
55, 32
58, 23
62, 41
65, 41
68, 41
72, 44
75, 31
78, 33
82, 9

Total Damage = 39236

We can clearly see that the SK takes significantly less damage, due to a higher softcap.

I don't have a level 60 leather class to determine the leather softcap, but as Bcbrown said, his data wasn't great. Ideally some better testing will be done.

Also, we can see that both an SK and a Shaman basically take the same damage at 200 AC. This means neither class has their Softcap below 200 AC, and the mitigation calculation is the same for both classes. To be clear, this is just the AC mitigation calculation. Obviously an SK will take less damage in a similar period of time due to parry, dodge, riposte, and a higher defense skill:

=========================
60 SK 1000 hits - 200 AC, No Shield
=========================
19, 291
22, 40
25, 33
28, 37
32, 37
35, 31
38, 39
42, 30
45, 41
48, 54
52, 53
55, 34
58, 54
62, 35
65, 31
68, 31
72, 41
75, 35
78, 37
82, 16

Total Damage = 41544

=============================
60 Shaman 1000 hits - 200 AC Test, No Shield
=============================

DV, Count
19, 289
22, 41
25, 41
28, 33
32, 32
35, 43
38, 43
42, 38
45, 40
48, 33
52, 38
55, 38
58, 37
62, 39
65, 33
68, 39
72, 41
75, 41
78, 34
82, 27

Total Damage = 41805

[DeathsSilkyMist]

In the AC softcap thread I added a third dataset, 351 AC with a 55 AC shield equipped:

https://www.project1999.com/forums/s...&postcount=209