sk starting points

[DeathsSilkyMist]

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

Three things:

1. Thank you for the link to the code, that will be helpful
2. You didn't answer any of my questions
3. If my starting stat allocations results in one extra kill per hour, I'd be overjoyed. How many extra kills per hour will 20 int provide?

My own answers:
I'm expecting a uniform distribution of 20 unique hit values on an even-con mob. On a significantly lower-leveled mob, I'd expect the distribution to skew higher. I expect a similar increase across the entire distribution with added strength, and I also expect the distribution to skew very slightly higher, probably well within the realm of statistical noise for a sample of this size.

1. You are welcome.
2. I provided the formulas so you can plug in the numbers. You can then come back with your findings. That will provide the answers to your questions.
3. You are overvaluing 1 extra kill per hour. On an Iksar without Blood Ember Clickies, a bit of extra mana could save your life if you are trying to maximize kills per hour while minimizing meditation times. You will be getting closer to emptying your mana pool in that case. You can also run into scenarios where you fail FD multiple times, and need to life tap to stay alive while waiting for FD to cooldown. This will drain your mana pool. Not dying from a situation like that can also save you time and give you more kills per hour. You cannot simply assume you are gaining no benefit from INT. Both STR and INT are providing small bonuses that may or may not help, and will not be noticeable most of the time.

[Lune]

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

3. You are overestimating 1 extra kill per hour.

lol that was your estimate

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

Mobs around level 24 have about 800 HP I believe, so its 3 kills per hour at best. It is more likely to be something like 1 extra kill per hour, due to the fact that you will be taking breaks, travelling between mobs, etc.

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

You can also run into scenarios where you fail FD multiple times, and need to life tap to stay alive while waiting for FD to cooldown. This will drain your mana pool. Not dying from a situation like that can also save you time and give you more kills per hour.

This is such a hilariously bad take. What are the chances you spike your mana from 100 to 0 lifetapping to save your life during FD fails? If sitting there channeling lifetaps from 100 mana to 0 saves your life during chain FD fails 1. you're not even splitting something worth splitting, or 2. You're so bad at FDing through spells you shouldn't be trying to split anyway, just fucking kill it lol. And he weighs this one in a million, totally absurd situation more highly than the 1 kill per hour that he himself estimated.

Seriously, imagine a situation where you burn 100% of your mana on lifetaps while chain FD failing trying to split a room, think about the absurdity and rarity of that.

[DeathsSilkyMist]

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

lol that was your estimate

I meant to say he is overvaluing it. I agree my wording was a bit confusing there, thank you for asking me to clarify.

I have edited the post to remove further confusion.

[bcbrown]

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

I provided the formulas so you can plug in the numbers. You can then come back with your findings. That will provide the answers to your questions.

Jeez dude, your attitude here is incredibly unhelpful. Proper data analysis requires starting with a hypothesis before running any calculations. And, like, I get it, you're being attacked and ganged up on, and it's human to feel defensive about that. But I had hoped I had demonstrated enough good faith that you would be willing to enage with me.

When I started preliminary analysis, I was expecting results consistent with the "damage interval" concept, which would involve 20 unique damage values, where the distribution across those values is a function of the relative attack and defense values. Out of ~240 hits, I found ~110 unique values, so my mental model is incorrect. I could type up a lot more about why it's important to do preliminary hypothesis checking like this, but I know you'd just disregard it.

Sorry bud, but I'm done with you.

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

The 20 di thing looks to get a bit more complicated as you get higher level / higher str.

I'd really appreciate it if you (or anyone else) can explain the 20 di thing. The EQEmu code seems to use it:

auto roll = RollD20(hit.offense, mitigation);
hit.damage_done = max(roll * hit.base_damage + 0.5), 1);

I looked around and didn't see anything else in the codebase that would account for what I found in the log I examined. Would you expect exactly 20 unique hit values?

[DeathsSilkyMist]

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

Jeez dude, your attitude here is incredibly unhelpful. Proper data analysis requires starting with a hypothesis before running any calculations. And, like, I get it, you're being attacked and ganged up on, and it's human to feel defensive about that. But I had hoped I had demonstrated enough good faith that you would be willing to enage with me.

When I started preliminary analysis, I was expecting results consistent with the "damage interval" concept, which would involve 20 unique damage values, where the distribution across those values is a function of the relative attack and defense values. Out of ~240 hits, I found ~110 unique values, so my mental model is incorrect. I could type up a lot more about why it's important to do preliminary hypothesis checking like this, but I know you'd just disregard it.

Sorry bud, but I'm done with you.

You are taking this wildly out of context. I am simply saying you can use the formulas and data to see what is going on. There is no need to add extra complexity and confusion at this point by answering your questions.

If you are doing this in good faith, you would not assume I am being mean to you in some manner. I honestly am not.

[Jimjam]

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

I looked around and didn't see anything else in the codebase that would account for what I found in the log I examined. Would you expect exactly 20 unique hit values?

Okay, I’ve done some thinking for when I was engaging with the steel warrior… on reflection I believe the 20Di system was introduced with a melee revamp way, way after velious. Sorry!

[Troxx]

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

Okay, I’ve done some thinking for when I was engaging with the steel warrior… on reflection I believe the 20Di system was introduced with a melee revamp way, way after velious. Sorry!

I always assumed the DI20 system was in from the start of eq but I could be wrong. The great “ac vs hp” debates didn’t really kick off in a large way until augments were introduced to the game.

For those that are interested, ac won by a landslide in those eras. It didn’t stop the debate from raging for YEARS on the warrior and sk forums. The debates were nasty. I think I logged hundreds of hours parsing fights defensively to help settle the debate. It was possible, with sufficient ac to completely eliminate DI20 incoming hits even on standard length boss fights. The hp centric tank (warriors especially) had a few thousand more hp but the ac centric tank would receive 45-50% of incoming hits for minimum damage (vs 32-37% for the hp tank and like 0.001% or less hits for max vs 2-5% max hits for the hp focused tank … against raid bosses.

On trivial expansions old raid targets and all group content I would regularly never see a DI over 16 (4 highest hits completely eliminated). So yeah a few thousand more hp to hopefully survive being one rounded vs making it statistically nearly impossible to receive one of those tank killing max rounds to begin with. Healers clearly preferred the latter.

So yeah I am assuming we still follow a DI 1-20 model here … but not certain. It may not have been introduced until PoP/LDoN/GoD era. It was definitely in by Omens of War and epic 1.5/2.0 - that’s when all those fancy 90hp + whatever augs started appearing on minor raid mobs. I always went for the 20-25-30 ac augments.

[Jimjam]

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

Three things:

1. Thank you for the link to the code, that will be helpful
2. You didn't answer any of my questions
3. If my starting stat allocations results in one extra kill per hour, I'd be overjoyed. How many extra kills per hour will 20 int provide?

My own answers:
I'm expecting a uniform distribution of 20 unique hit values on an even-con mob. On a significantly lower-leveled mob, I'd expect the distribution to skew higher. I expect a similar increase across the entire distribution with added strength, and I also expect the distribution to skew very slightly higher, probably well within the realm of statistical noise for a sample of this size.

The 20 di thing looks to get a bit more complicated as you get higher level / higher str.

[DeathsSilkyMist]

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

This is such a hilariously bad take. What are the chances you spike your mana from 100 to 0 lifetapping to save your life during FD fails? If sitting there channeling lifetaps from 100 mana to 0 saves your life during chain FD fails 1. you're not even splitting something worth splitting, or 2. You're so bad at FDing through spells you shouldn't be trying to split anyway, just fucking kill it lol. And he weighs this one in a million, totally absurd situation more highly than the 1 kill per hour that he himself estimated.

Seriously, imagine a situation where you burn 100% of your mana on lifetaps while chain FD failing trying to split a room, think about the absurdity of that.

When you are soloing, you are not always killing a single mob, meditating back to full, and then killing another mob. You can also kill multiple mobs before you meditate. This means you are not going to be a 100% mana all the time.

I have had FD fail multiple times due to bad RNG on quite a few occasions. FD has a random chance to fail, it is not guaranteed to succeed. You cannot control that via player skill. You are simply making silly assumptions because you want to be right. Remember that I have actually leveled an SK to 60.

Mana is more important to Iksars, because they are the only race that cannot use Blood Ember Clickies. They have less ways to save mana, and thus are more likely to have mana issues.

[Lune]

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

When you are soloing, you are not always killing a single mob, meditating back to full, and then killing another mob. You can also kill multiple mobs before you meditate. This means you are not going to be a 100% mana all the time

In which case INT is completely irrelevant.

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

Mana is more important to Iksars, because they are the only race that cannot use Blood Ember Clickies. They have less ways to save mana, and thus are more likely to have mana issues.

This has little to nothing to do with INT, as iksars have a problem with mana USE; max mana does nothing for that, except for those extremely rare 100-0 situations that everyone tries their best to avoid.