Iksar Shaman - Solo Challenges?

[DeathsSilkyMist]

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

What about rear strikes?

Minimal. If I needed to end a fight, I would FD instead of turn and run. Also not relevant. The chance to stun is the same regardless, and a properly played Ogre Shaman would not willingly put their backs to a mob while slowing.

[bcbrown]

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

With regards to Troxx's specific point on stun rate. I'll use my Monk's logs from levels 12-52 to get some hard data to back up the rates. He almost exclusively fought unslowed warrior mobs via face tanking in twinked gear.

It's super helpful when you provide concrete numbers backed up by logs like this. Can you provide a youtube video of your shaman where FSI is applicable? Ideally one where at least one slow is resisted. If it's one where you could also post the log in a text file, that would be best, but just the video would be helpful.

I don't want to do further analysis on a random video on your channel, as you may have reasons to call that fight atypical in some way, which is why I'm asking you to nominate a fight to analyze.

[DeathsSilkyMist]

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

It's super helpful when you provide concrete numbers backed up by logs like this. Can you provide a youtube video of your shaman where FSI is applicable? Ideally one where at least one slow is resisted. If it's one where you could also post the log in a text file, that would be best, but just the video would be helpful.

I don't want to do further analysis on a random video on your channel, as you may have reasons to call that fight atypical in some way, which is why I'm asking you to nominate a fight to analyze.

I usually do provide concrete numbers backed up by videos/logs, or I am referencing concrete numbers I have posted many times already, or I am referencing proven game formulas that can be backed up by videos/logs. Often times it gets drowned out in troll posts unfortunately. Many posters (especially trolls) generally do not post concrete numbers to back up their points by comparison.

As for your question about when FSI is applicable:

https://www.youtube.com/watch?v=uEgFcImQ9XU - A fight where I marked all of the places where a bash occurred in the description. There were 12 occasions. The number of bashes is lower because the mob was slowed. FSI would have prevented at least one stun statistically using the data from my previous post: https://www.project1999.com/forums/s...&postcount=267

https://youtu.be/igpbLuNe0ls?feature=shared&t=438 - At this mark I was bashed by Bravatar like half a second before Slow finished casting. If that bash had a stun component, I would have lost a full 6 seconds of cast time, and not have him slowed. I would need to attempt to slow again and take more unslowed damage.

https://www.youtube.com/watch?v=JLhe3KQXke0 - Here is a video where my first attempt at slowing was resisted, even after malo. I landed the second slow, and the dragon bashed me right after. If I was a bit slower to react on casting the second slow or had gotten a fizzle, I could have been down two slows if that bash had a stun component.

In all three of these videos, Iksar/Troll Regeneration wouldn't have helped at all during the preslow phase. It would have only saved me 30 seconds of recovery time after the fight, as you get 1200 HP from Iksar/Troll Regeneration over 15 minutes, which is the equivalent of 1 Torpor.

[Troxx]

Thanks for the numbers. Though not strictly controlled and though there is a chance erroneous occurrences (ie spell stuns) - they are likely negligible. Not worth splitting hairs over. Did you ever hunt mobs like sonic bats with a stun proc? Probably not but worth asking.

1). Good to know that common knowledge is correct- the majority of landed bashes do not stun.

2) Also good to know that you got stunned 41.525 times on average per level. Or another way of thinking about it you spent 41-85 seconds being melee stunned per level on unslowed mobs.

I’m curious enough now that I will go pull up some logs of one of my melee to look at the average duration of a single stun.

If a shaman were leveling and with a similar stun rate, reduce the number of stuns in a manner that correlates for the mob being slowed … oh and all that time playing a shaman where you weren’t face tanking (groups/duos, root rot, pet tanking). A shaman leveling will not be face tanking unslowed mobs from 12-52 the same 100% of the time.

We’re making good progress. Now show us some situations where your circumstances were tenuous enough that a brief stun would have a meaningful impact on fight outcome.

Bear in mind that we already know that the botb shaman soloist (barbarian) had all that success without FSI.

Keep it up (not sarcastic here). I’m interested to see what you come up with.

[Naethyn]

I'd be upset too if a barbarian was the best shaman and I was really into shaman.

[DeathsSilkyMist]

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

Keep it up (not sarcastic here). I’m interested to see what you come up with.

Thank you for a serious post. I also appreciate it, I am not being sarcastic.

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

Thanks for the numbers. Though not strictly controlled and though there is a chance erroneous occurrences (ie spell stuns) - they are likely negligible. Not worth splitting hairs over. Did you ever hunt mobs like sonic bats with a stun proc? Probably not but worth asking.

No problem! No, I did not hunt sonic bats. My leveling route was FoB/Kurns -> Highkeep Guards -> Crystal Caverns -> Tower of Frozen Shadow (Floor 3 and then Floor 6) -> Bloodgills -> Kelethin Guards (current location).

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

We’re making good progress. Now show us some situations where your circumstances were tenuous enough that a brief stun would have a meaningful impact on fight outcome.

I already did with my previous post https://www.project1999.com/forums/s...&postcount=273. Look at the Bravatar video: https://www.youtube.com/watch?v=igpbLuNe0ls&t=438s . I was bashed right before my Slow landed. I had taken ~20% of my total life in damage from Bravatar when he wasn't slowed. If I had to cast another slow while he was unslowed, that could have taken another ~20% of my life total, which means I would be at 60% life when the fight started. 160 HP from two minutes of Troll/Iksar Regeneration while pulling is ~6% of my life total by comparison. So in this instance FSI saved 20% life vs 6% life via Troll/Iksar Regeneration. This is further supported by this video https://www.youtube.com/watch?v=JLhe3KQXke0 - where casting malo + 2x slows caused me to take roughly 40% damage. I was bashed right after my second slow. If that bash had a stun component and I was a bit slower at casting slow (or had a fizzle), there is a good chance I would have needed to try casting slow again.

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

Bear in mind that we already know that the botb shaman soloist (barbarian) had all that success without FSI.

Everybody agrees racials are not required on a Shaman. Nor have I ever said they were required. Everybody agrees Barbarians can do the same content. We are simply looking at which racial is the Min/Max option. You can objectively determine which racial is best, even if it isn't needed to do content. This information is useful for people who like to Min/Max for fun. I am not trying to get people to regret their race choices or anything sinister like that. I just want the correct information to be out there for people interested in it.

[bcbrown]

https://www.youtube.com/watch?v=uEgFcImQ9XU

Two malo casts, first interrupted (by a spell?)
Two Turgur's casts - first resisted.

https://youtu.be/igpbLuNe0ls?feature=shared&t=438 - first Turgur's lands

https://www.youtube.com/watch?v=JLhe3KQXke0

1 malo cast
Two turgur's casts - first resisted

So assumptions are that each fight has one Malo and two Turgur's, and the first Turgur's getting resisted.

Turgur's has a 3s cast and a 6s recast. Malo has a 5s cast, recast not applicable since it's unresistable.
That means the pre-slow portion of the fight is 5 + 3 + 6 + 3 or 17 seconds. Using an 8 second interval between bash/kick attempts, that's 2.125 bash/kicks per fight pre-slow. Using your 70/30 split, that's 1.4875 bashes.

Your numbers were 21044 total kick/bashes and 6718 kicks, so 14326 bash attempts. 6157 bashes hit, so the success rate was .4298 (6157 / 14326). Your data shows a stun rate of .2698 (1661 / 6157) on successful bashes.

1.4875 expected bashes pre-slow results in .639 successful bashes, or .1725 stuns, or about one in six fights.

You with me so far? If so we can proceed to damage calculations. Two numbers that would help greatly are average stun duration and Turgur's resist rate. Here I'm using a 50% resist rate (data shows 3/5 or 40%), but it's gotta be closer to 10-20%, right? Until Troxx or someone else digs up some stun duration from logs, my wild-ass guess is that it'll be about 3 seconds?

Providing a full log of one of these fights will also help with damage estimates.

[DeathsSilkyMist]

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

https://www.youtube.com/watch?v=uEgFcImQ9XU

Two malo casts, first interrupted (by a spell?)
Two Turgur's casts - first resisted.

https://youtu.be/igpbLuNe0ls?feature=shared&t=438 - first Turgur's lands

https://www.youtube.com/watch?v=JLhe3KQXke0

1 malo cast
Two turgur's casts - first resisted

So assumptions are that each fight has one Malo and two Turgur's, and the first Turgur's getting resisted.

Turgur's has a 3s cast and a 6s recast. Malo has a 5s cast, recast not applicable since it's unresistable.
That means the pre-slow portion of the fight is 5 + 3 + 6 + 3 or 17 seconds. Using an 8 second interval between bash/kick attempts, that's 2.125 bash/kicks per fight pre-slow. Using your 70/30 split, that's 1.4875 bashes.

Your numbers were 21044 total kick/bashes and 6718 kicks, so 14326 bash attempts. 6157 bashes hit, so the success rate was .4298 (6157 / 14326). Your data shows a stun rate of .2698 (1661 / 6157) on successful bashes.

1.4875 expected bashes pre-slow results in .639 successful bashes, or .1725 stuns, or about one in six fights.

You with me so far? If so we can proceed to damage calculations. Two numbers that would help greatly are average stun duration and Turgur's resist rate. Here I'm using a 50% resist rate (data shows 3/5 or 40%), but it's gotta be closer to 10-20%, right? Until Troxx or someone else digs up some stun duration from logs, my wild-ass guess is that it'll be about 3 seconds?

Providing a full log of one of these fights will also help with damage estimates.

I have full logs of a fight posted in the description of my Ionat video: https://www.youtube.com/watch?v=oPxeOVuX0G8 . I'll need to dig through my logs to find the 4+ Dragons. Average DPS was ~19 over 20 minutes, but the mob was slowed by 75% basically the entire time. That means Ionat is probably doing around 80 DPS unslowed. Ionat is a 6+ Dragon, so let's drop the DPS by 20% as a rough estimate. At 64 DPS unslowed, you would lose 512 HP over 8 seconds (3 second cast + ~2 second stun + 3 second cast) if you got interrupted at the end of your slow, as I could have been in the Bravatar video. So with the Bravatar video you are looking at saving 160 HP from Iksar/Troll Regeneration (2 minutes of pulling under 100% HP) vs. saving 512 HP with FSI (not having to cast another slow). This rough estimate is supported by the two videos I posted, where I took roughly 20% of my life per slow attempt in both videos. I also take roughly 20% of my life per slow attempt in the Ionat video when he is unslowed. My life total is 2620.

[bcbrown]

Great, 2 second stun duration and 64 DPS.

Since we have an expected stun rate of 17%, I'll discard the possibility of two stuns in a fight, and focus on the expected damage added given one stun occurs. Since I'm not going to build out a full Markov Chain for you, I'll asume the stun timing is fully stochastic, meaning it's equally likely to occur at any point of the 17 seconds pre-slow.

If it occurs during the malo: the malo phase goes from 5s to <duration of malo cast before stun> + 2 + 5. The stochastic assumption lets us take the midpoint of the cast as the expected time, so it goes from 5s to 2.5 + 2 + 5, or 9.5, adding 4.5 seconds.
If it occurs during a slow, using the same logic, we have it going from 3s to 1.5 + 2 + 3, or 6.5, adding 3.5 seconds.
If it occurs during a recast interval, it's slightly different. We add the 2s stun duration, but subtract time from the recast time; if you get stunned immediately after a resisted slow, the stun will end before the recast duration, adding zero time.
So, if it occurs in the first 4s of a recast, there's zero impact.
If it occurs in the last 2s of a recast duration, the expected duration impact will be 1 second saved from recast duration, plus 2 seconds stunned, or 1 second added.
Over 17 seconds, we have:
5 seconds casting malo, with 4.5 seconds expected added
6 seconds casting slow, with 3.5 seconds expected added
4 seconds in recast duration for the slow, with 0 seconds expected added
2 seconds in recast duration for the slow, with 1 seconds expected added

So the total expected time added will be:
5/17 * 4.5 + 6/17 * 3.5 + 4/17 * 0 + 2/17 * 1

This is an expected time added of 2.68 seconds, which at 64 DPS is 171 expected damage added.

Worst case will be a stun in the last second of a malo, which would be adding a 2 second stun and a 5 second cast, or 448 damage. This will occur 1/17th of the time. Next worst case is in the last second of a slow, which would add a 2 second stun and a 3 second cast, or 320 damage. Since there's three casts in this scenario, there's a total of 3/17th chance of this worst case happening.

So:
There's a 1 in 6 chance of a stun happening pre-slow.
If there's a slow, the expected damage is 171.
There's a 3/17 chance the expected damage is either 320 or 448.

This means the worst case will happen around .1725 * 3/17 or 3% of the time.

Edit: hold on, double-checking math. Ok, satisfied now.

[DeathsSilkyMist]

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

Great, 2 second stun duration and 64 DPS.

Since we have an expected stun rate of 17%, I'll discard the possibility of two stuns in a fight, and focus on the expected damage added given one stun occurs. Since I'm not going to build out a full Markov Chain for you, I'll asume the stun timing is fully stochastic, meaning it's equally likely to occur at any point of the 17 seconds pre-slow.

If it occurs during the malo: the malo phase goes from 5s to <duration of malo cast before stun> + 2 + 5. The stochastic assumption lets us take the midpoint of the cast as the expected time, so it goes from 5s to 2.5 + 2 + 5, or 9.5, adding 4.5 seconds.
If it occurs during a slow, using the same logic, we have it going from 3s to 1.5 + 2 + 3, or 6.5, adding 3.5 seconds.
If it occurs during a recast interval, it's slightly different. We add the 2s stun duration, but subtract time from the recast time; if you get stunned immediately after a resisted slow, the stun will end before the recast duration, adding zero time.
So, if it occurs in the first 4s of a recast, there's zero impact.
If it occurs in the last 2s of a recast duration, the expected duration impact will be 1 second saved from recast duration, plus 2 seconds stunned, or 1 second added.
Over 17 seconds, we have:
5 seconds casting malo, with 4.5 seconds expected added
6 seconds casting slow, with 3.5 seconds expected added
4 seconds in recast duration for the slow, with 0 seconds expected added
2 seconds in recast duration for the slow, with 1 seconds expected added

So the total expected time added will be:
5/17 * 4.5 + 4/17 * 0 + 2/17 * 3.5 + 6/17 * 1

This is an expected time added of 2.09 seconds, which at 64 DPS is 133 expected damage added.

Worst case will be a stun in the last second of a malo, which would be adding a 2 second stun and a 5 second cast, or 448 damage. This will occur 1/17th of the time. Next worst case is in the last second of a slow, which would add a 2 second stun and a 3 second cast, or 320 damage. Since there's three casts in this scenario, there's a total of 3/17th chance of this worst case happening.

So:
There's a 1 in 6 chance of a stun happening pre-slow.
If there's a slow, the expected damage is 133.
There's a 3/17 chance the expected damage is either 320 or 448.

This means the worst case will happen around .1725 * 3/17 or 3% of the time.

Using your own calculations, you can see that FSI gives you hundreds of HP in the first 17 seconds of the pre-slow phase of a fight when it triggers. Iksar/Troll Regeneration would consistently give you 24 HP in the same amount of time.

Realistically speaking 24 HP is basically never going to help during the pre-slow phase of a fight. The chances of 24 HP saving you from death is much lower than 3%. This is why FSI is better from a Min/Max perspective, and why people pick Vindi BP over Fungi Tunic. The chance to reduce damage spikes via Vindi BP or FSI is higher than the chance that 24 HP is going to save you during the pre-slow phase of a fight, which is the most dangerous phase of a fight. Once the mob is slowed, Iksar/Troll Regeneration is just going to help you with recovery time at the end of the fight, which ends up being 30 seconds in a 15 minute fight. Less if the fight is shorter.