Ranger ac parsing (I'm serious)

[bcbrown]

I ran two experiments on my 54 cleric, one with 195 worn ac and one with 132 worn ac. Both had the full cleric buff line with ~64 ac. Here's the magelo. I'd fight Shiel Glimmerspindle, a level 40 rogue, until I had 4-500 hp, then root, CH, re-engage. I'd do that until oom, then root, atone, med up.

On the first experiment I took snapshots at 755 and 1399 hits before concluding the experiment with 1983 hits. Here's the hit distribution for those snapshots:

[You must be logged in to view images. Log in or Register.]

Some statistics on the full experiment:
Hit%: 55.6%
Average hit: 42.4
Average hit excluding min value: 51.0
Min hit %: 26.8%

The hit chart does not look normally distributed. Excluding the minimum hit, the rest of the DI intervals each had around 3-5% of the total number of hits, with an average of 29 hits per interval. Here's a chart showing each interval as a ratio of the 29 average number per interval:

[You must be logged in to view images. Log in or Register.]

The ratios for the full experiment ranged from 80% to 125%. So still pretty noisy but it does look like a uniform distribution is a better fit than a normal distribution.

For the second experiment I took 1150 total hits.
Hit%: 58.4%
Average hit: 48.5
Average hit excluding min/max value: 50.6%
Min hit %: 19.3%
Max hit %: 12.7%

And here's the chart:
[You must be logged in to view images. Log in or Register.]

[Jimjam]

I'm super interested in how the 'low' ac parse has a peak at min and max hits, but the 'high' ac parse has only the peak at min hit.

I wonder if intermediatery AC value parses were done we could see the second peak from the low AC parse move to the left as AC increases, eventually merging with the min hits peak as perhaps it has done in the high AC parse?

[bcbrown]

The 132-AC has a max-hit peak that's about 2/3rds the height of the min-hit parse, so I think this is in that intermediate range. I'll do a smaller 160ish parse and then I'm hoping a 100ish parse will either have same-height peaks or a higher peak on the max hit. Then I want to see whether there's a difference with spell-ac vs worn-ac and whether there's any difference with or without shield ac.

The other thing that jumped out to me was that the avg hit excluding min/max was almost unchanged in the two experiments at 51 and 50.6.

[Jimjam]

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

The 132-AC has a max-hit peak that's about 2/3rds the height of the min-hit parse, so I think this is in that intermediate range. I'll do a smaller 160ish parse and then I'm hoping a 100ish parse will either have same-height peaks or a higher peak on the max hit. Then I want to see whether there's a difference with spell-ac vs worn-ac and whether there's any difference with or without shield ac.

The other thing that jumped out to me was that the avg hit excluding min/max was almost unchanged in the two experiments at 51 and 50.6.

Do you think maybe the 132 is below the threshold for seeing mitigation improvements, and the 195 exceeds the threshold of maximum mitigation?

[bcbrown]

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

Do you think maybe the 132 is below the threshold for seeing mitigation improvements, and the 195 exceeds the threshold of maximum mitigation?

I think 195 ac exceeds the threshold, and adding more ac wouldn't do much damage mitigation for this specific mob. I bet when I test against a level 45 mob there will be room for further mitigation with more worn ac.

132 worn AC is squarely in the realm where adding more AC adds more damage mitigation. I ran a small test with 163 AC and although it's noisy (only 218 hits), it suggests that 163 is right around where maximum mitigation is reached for this mob. There's a small spike at the max hit, but barely more than the other hit values.

I ran a longer test (506 hits) at 106 AC, and that's right around the value where max and min hits occur equally frequently. Note that all of these also occur with ~63 spell AC as well.

I've run a small parse so far on my level 49 ranger with 137 worn ac and 35 spell ac, or 173 total. Only 139 hits; I'm actually killing Shiel on the ranger and there's a 30 minute respawn time, so harder to get a lot of hits in. I think I'll be able to add about 30 worn ac in upgrades and will run another parse when that happens. Compared to the cleric with 106 worn and 63 spell, or 169 total, ranger looks like it mitigates better. Looks most similar to the cleric parse with 132 worn acc, or 194 total. Cleric is using a shield, ranger is not. This makes me wonder if spell ac matters less than worn ac, but there's still far too little data to be confident in that.

[Goregasmic]

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

I think 195 ac exceeds the threshold, and adding more ac wouldn't do much damage mitigation for this specific mob. I bet when I test against a level 45 mob there will be room for further mitigation with more worn ac.

132 worn AC is squarely in the realm where adding more AC adds more damage mitigation. I ran a small test with 163 AC and although it's noisy (only 218 hits), it suggests that 163 is right around where maximum mitigation is reached for this mob. There's a small spike at the max hit, but barely more than the other hit values.

Very interesting. So bucket 2-19 are basically always flat and the more AC you get, the hits from the max hit bucket get transfered to the min hit bucket. I would have expected a smoother increase in the second half of the buckets.

So that means if your parse flattened the max hits you're at or above the squelch number for that mob.

I ran a longer test (506 hits) at 106 AC, and that's right around the value where max and min hits occur equally frequently. Note that all of these also occur with ~63 spell AC as well.

So 63 spell AC did nothing even though you're far below squelch point? That's odd.

I guess it is probable that for most group content, the softcap is mostly inconsequential (unless you're a heavy twink) since you'll reach the squelch point way before you reach the soft cap and charting the squelch points at various mob levels would be more useful. High end raid mobs probably cannot be squelshed even with the soft cap reached.

[bcbrown]

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

Very interesting. So bucket 2-19 are basically always flat and the more AC you get, the hits from the max hit bucket get transfered to the min hit bucket. I would have expected a smoother increase in the second half of the buckets.

So that means if your parse flattened the max hits you're at or above the squelch number for that mob.

That's what it looks like to me too.

So 63 spell AC did nothing even though you're far below squelch point? That's odd.

No, I don't think that's the right way to look at it. I had the spell AC on for every graph I've posted here. So it's better to think that, for example, the graph showing equal numbers of max-hit and min-hit is best thought of as being at (106 worn AC plus 63 spell AC). Maybe that's equivalent to 169 worn AC and 0 spell AC, I'm not sure.

I guess it is probable that for most group content, the softcap is mostly inconsequential (unless you're a heavy twink) since you'll reach the squelch point way before you reach the soft cap and charting the squelch points at various mob levels would be more useful. High end raid mobs probably cannot be squelshed even with the soft cap reached.

That's my conclusion too.

[Salaryman]

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

The 132-AC has a max-hit peak that's about 2/3rds the height of the min-hit parse, so I think this is in that intermediate range. I'll do a smaller 160ish parse and then I'm hoping a 100ish parse will either have same-height peaks or a higher peak on the max hit. Then I want to see whether there's a difference with spell-ac vs worn-ac and whether there's any difference with or without shield ac.

The other thing that jumped out to me was that the avg hit excluding min/max was almost unchanged in the two experiments at 51 and 50.6.

RED99

Well I feel bad for you because you have no shields

On my LEvel 60 EPIC RANGER

I have a:

Shield of the Red Dragon

Shield of the Green Dragon (420 every day)

Shield of the Water Dargon


and I think you should:

Learn 2 Play


[You must be logged in to view images. Log in or Register.]


RED99

[Ripqozko]

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

RED99

Well I feel bad for you because you have no shields

On my LEvel 60 EPIC RANGER

I have a:

Shield of the Red Dragon

Shield of the Green Dragon (420 every day)

Shield of the Water Dargon


and I think you should:

Learn 2 Play


[You must be logged in to view images. Log in or Register.]


RED99

i got an aary shield on mine, im sorry you never progressed past kunark

[Salaryman]

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

i got an aary shield on mine, im sorry you never progressed past kunark

RED99

Well then why dont you do some tests with the shield you supposedly have on your ranger, which you have no screenshots of or other proof of having?

[You must be logged in to view images. Log in or Register.]

Read that in your goblin tongue

RED99