sk starting points

[Jimjam]

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

Can you show the formula you are using to make this assumption? If not, you are just guessing at the mechanics.

If you know the mechanics so well, it should be trival to provide the formulas and plug in the data from the logs + videos I provided here: https://www.project1999.com/forums/s...&postcount=300

Yes, I am guessing at mechanics.

Neither recognising a bell shape on a graph nor understanding that ac / attack weight the damage roll should mean it should be trivial to provide formulas. That seems outlandish. It may be trivial for you, but not me.

I’m a bit confused what you are asking tbh. On one hand you are saying adding str would do nothing determinable on the other hand you’re disagreeing you’ve hit a soft cap for that mob. How are the two different? Also, bringing back to the conversation would the equivalent int made a more apparent difference? Lets not forget monks have many influences on their dps which make it far better than Sks, so using monks to demonstrate how little value starting str has over int is already a but of a stretch imo.

[Jimjam]

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

If and when you find those old thesteelwarrior threads - I’m Bedavir. Yoda and I did some pretty epic investigations into hp vs ac. I was fairly active on the SK forums alongside Brael as well.

To my knowledge, TSW website no longer exists. RIP

A saint posts amongst us.

[DeathsSilkyMist]

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

Yes, I am guessing at mechanics.

Neither recognising a bell shape on a graph nor understanding that ac / attack weight the damage roll should mean it should be trivial to provide formulas. That seems outlandish. It may be trivial for you, but not me.

I’m a bit confused what you are asking tbh. On one hand you are saying adding str would do nothing determinable on the other hand you’re disagreeing you’ve hit a soft cap for that mob. How are the two different? Also, bringing back to the conversation would the equivalent int made a more apparent difference? Lets not forget monks have many influences on their dps which make it far better than Sks, so using monks to demonstrate how little value starting str has over int is already a but of a stretch imo.

Thank you for admitting you are guessing. Please stop telling people STR will give a significant DPS boost during the leveling process. You have admitted that you do not know this.

I am willing to review any evidence you are willing to provide, if you choose to do so. If it shows I am wrong, I will be happy to admit I am wrong.

[Troxx]

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

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

[Jimjam]

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

Thank you for admitting you are guessing. Please stop telling people STR will give a significant DPS boost during the leveling process. You have admitted that you do not know this.

I am willing to review any evidence you are willing to provide, if you choose to do so. If it shows I am wrong, I will be happy to admit I am wrong.

My guess is based on extensive reading, past parsing, first hand experience and the evidence presented in this thread. Particularly mr turtle.

You’re throwing out the baby with the bath water.

Earlier you demanded i provided full code, arguing it would be trivial. Telling me (again) to shut up cos I didn’t (and won’t) is a bit extreme. I don’t have p99’s server side stuff and a big part of these discussions is to look at the limited evidence and work together to draw conclusions.

JSYK, when people say stats don’t matter they are being hyperbolic. Basically making your point that a dps increase from str will be small (just like the benefit of 20 int imo).

That said, plenty of evidence has been given for the benefit:

1) improvement to max hit formula
2) improvement to di multiplier formula
3) (flawed) parses showing 4.3% increase (I’m looking forward to doing my own investigations).

We’ve discussed the widely held belief that str impacts attack and in turn how likely the damage roll will do well (something you seem to refute?)

We’ve discussed how damage is normally distributed and the curve can be skewed left or right (take a look at your turtle logs)

1-2 alone mean the improvement is statistically significant. If you are using a different definition of significant please define it, and argue against that for how much more significant help 20 int would be.

[DeathsSilkyMist]

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

My guess is based on extensive reading, past parsing, first hand experience and the evidence presented in this thread. Particularly mr turtle.

I also have extensive first hand experience, past parsing, and evidence. I am not sure why you can just assume it is greater than my own. You haven't provided any evidence, and are at a disadvantage.

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

Earlier you demanded i provided full code, arguing it would be trivial. Telling me (again) to shut up cos I didn’t (and won’t) is a bit extreme. I don’t have p99’s server side stuff and a big part of these discussions is to look at the limited evidence and work together to draw conclusions.

Please stop assuming I am telling you to shut up. I didn't say that. I am asking you to stop posting things that you do not know, while trying to pass it off as if it as factual. You do not know that 20 STR will provide enough of a DPS boost to be meaningful to OP.

https://github.com/EQEmu/Server/blob...one/attack.cpp

The EQEMU source code is available. P99 is based off of this code. I can show you the P99 max damage formula is unchanged when compared to the EQEMU code:

Damage Table for a level 60: { 285 Max Extra, 23 Chance, 65 Minusfactor }

1. 225 Offense Skill + ((2 * 231 Strength - 150) / 3) + 21 ATK (1x AoB item + Firefist) = 350.
2a. 350 - 65 (Minusfactor from damage table) = 285
2b. (285 / 2) + 100 = 242.5
2c. ((46 Weapon Damage * 2) * 242.5) / 100 = 223.1
2d. Add 2h damage table bonus. Total is 37 according to https://lucy.allakhazam.com/dmgbonus.html?setcookie=1 (this includes the formula 3 + ((60 level - 28) / 3) + ((44 delay - 40) / 3)). 37 + 223.1 = 260.1

My highest damage hit on the turtle parses was 258, so this lines up so far when 260 max damage is what the formula came up with.

You can look through the code and show us where your assumptions are coming from.

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

JSYK, when people say stats don’t matter they are being hyperbolic. Basically making your point that a dps increase from str will be small (just like the benefit of 20 int imo).

If you are being hyperbolic, new players may not KNOW you are being hyperbolic. Please stop saying "starting stats do not matter", and then proceed to show why STR is going to matter. Either it does, or it doesn't. Pick a side and bring evidence.

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

3) (flawed) parses showing 4.3% increase (I’m looking forward to doing my own investigations).

You keep denouncing my parses, but refuse to give any of your own. I never said my parses were perfect, but right now they are much better than your admitted guessing. I would love to see your data when it comes out!

[Jimjam]

I'm not saying your experience is worse than mine, merely ours are different. We clearly play and gear differently, which produces different results.

By combining our experiences we have a bigger dataset. I in no way meant to demean your experiences, particularly because I've been actually looking at your experiences you've reported and seeing how they match or mismatch my own knowledge and experience.

I never said starting stats don't matter. I was explaining the context of something which you keep getting hung up on, and tbh you seem to be saying it about dumping str.

You keep telling me to stop participating in the discussion, which is the forum equivalent of telling someone to shut up.

Imo the entire point of discussion is to explore and argue ideas, it would not be ideal for people lacking knowledge to take part. In fact, such participance is a great way to have a fresh set of eyes, ask questions and improve knowledge. I mean you don't know the full code of the game, yet keep posting. Should you relent posting just because you don't know the full details of how str impacts the game? I suggest not. You clearly have a lot of knowledge to share and it would be a shame if you weren't to ever post again!


Perhaps I made a mistake to have mirrored your language in the use of the word 'guessing'. I didn't mean guess as 'completely fabricated from nothing'. TBH I feel a little bit like you may have set a trap there by using the word then latching on to it so hard after I repeated it.

For clarity, I meant guess as 'make an inference based on the limited information we all have'.

Edit: i don't mean to denounce your parses - I'm not saying they are worthless, merely acknowledging the limitations. That is normal to do when discussing a dataset. Don't take it as an insult as your parse is very good. As previously mentioned I can get a parse to you in maybe a month. Remind me.

[bcbrown]

DSM can you post your turtle logs? I'd like to do some parsing.

Also, open invitation: if anyone wants to do some strength testing I'm happy to provide a healer.

[DeathsSilkyMist]

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

I'm not saying your experience is worse than mine, merely ours are different. We clearly play and gear differently, which produces different results.

By combining our experiences we have a bigger dataset. I in no way meant to demean your experiences, particularly because I've been actually looking at your experiences you've reported and seeing how they match or mismatch my own knowledge and experience.

I never said starting stats don't matter. I was explaining the context of something which you keep getting hung up on, and tbh you seem to be saying it about dumping str.

You keep telling me to stop participating in the discussion, which is the forum equivalent of telling someone to shut up.

Imo the entire point of discussion is to explore and argue ideas, it would not be ideal for people lacking knowledge to take part. In fact, such participance is a great way to have a fresh set of eyes, ask questions and improve knowledge. I mean you don't know the full code of the game, yet keep posting. Should you relent posting just because you don't know the full details of how str impacts the game? I suggest not. You clearly have a lot of knowledge to share and it would be a shame if you weren't to ever post again!


Perhaps I made a mistake to have mirrored your language in the use of the word 'guessing'. I didn't mean guess as 'completely fabricated from nothing'. TBH I feel a little bit like you may have set a trap there by using the word then latching on to it so hard after I repeated it.

For clarity, I meant guess as 'make an inference based on the limited information we all have'.

Edit: i don't mean to denounce your parses - I'm not saying they are worthless, merely acknowledging the limitations. That is normal to do when discussing a dataset. Don't take it as an insult as your parse is very good.

I apologize if I sound mean. That isn't my intent, I simply have to defend myself from trolls. There are multiple people trolling and attacking this thread, so it can sometimes be difficult to determine if you are being sincere or not. Posters like Troxx are certainly not being sincere.

You are putting words in my mouth. I am not telling you to stop participating.

I have not set any traps for anybody. You have simply admitted that you do not have a factual basis upon which to base your current claims. I appreciate that you are willing to admit that. You are doing what most people in this thread cannot, which is awesome!

If you want to bring meaningful discussion to a factual conversation, you need to provide evidence. That is all I am asking. You are confusing people by trying to say "20 STR is going to give you a noticeable boost to DPS". You do not know this by your own admission, and therefore you should not say it in a manner that sounds like it is a fact.

You can say it is your guess and opinion that the DPS boost from 20 STR would be significant. That way people know you are not making a statement of fact. There is nothing wrong with that. Then people can look at the evidence I have provided, and the opinions you have provided, and make a decision. Or even better, you provide evidence proving your claim, so we can all benefit from the knowledge. You need to understand that I would love to be proven wrong. If we can get concrete evidence one way or the other, it will help everybody.

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

DSM can you post your turtle logs? I'd like to do some parsing.

Also, open invitation: if anyone wants to do some strength testing I'm happy to provide a healer.

They are already posted!

https://www.project1999.com/forums/s...&postcount=300

Links to the logs are in the descriptions of the youtube videos I have provided. The post above links to another earlier post I made which explained the DPS videos. That is where the video are linked.

https://www.youtube.com/watch?v=P5ouLQOBAoE
https://www.youtube.com/watch?v=Py96jk2NflU

Here are the video links directly.

[bcbrown]

Ah, I didn't know the actual logs were linked from the videos, thanks.

Before I start analysis, I'd like to know what you expect I'll find.

The 20 damage interval thing implies there should be 20 different unique hit values in each log, right? Do you expect a uniform distribution (each roughly equally frequent), a normal distribution (bell curve), or a beta distribution (in this case, a bell curve that's been skewed either higher or lower than usual)? Do you expect the same sort of distribution in both logs?

Do you expect there to be a difference across all 20 unique hit values? If so, do you expect it to be uniform (i.e. max_hit_more_strength - max_hit_less_strength == min_hit_more_strength - min_hit_less strength), or more of an impact at the top or bottom end?

I'll give my own answers before I start parsing.