2009-03-22, 12:37 AM
*does the necro post rain dance*
It's a sticky, so I think it's okay in this case.
Anyway, I was revising my damage calculator a bit and I became interested in the nature of the damage reduction formula. I don't think the one you have adequately describes the process that the game goes through. It may work for lower test values, but it doesn't work for all values.
Using *ahem* slightly dubious methods, I've determined that 1%/level worth of difference is absolutely correct - at 0 wdef. However, at extremely high levels of wdef, this is not the case.
I attacked a level 200 monster with a level 145 Night Lord. I set its evasion to 0 so accuracy would not be allowed to play a part of this.
The raw damage range I'm using for testing (I can give more specific stats if need be, but I don't see them as relevant to this issue at the moment) is 12622~25244 with critical Triple Throw hits.
At 0 wdef, the maximum observed critical over a few minutes of sampling constantly was roughly 11400, which is almost exactly 45% of 25244. At level 100, I did nothing but 1s. At level 101, I saw quite a few crits of 251, but no more. That's .99% or essentially 1% of 25244. This is why I believe the formula is correct in this respect.
However, when I upped the mob to 9000 wdef, the maximum critical I observed was 8678. The original formula projected 1~1, which is obviously wrong. The formula that Stereo/Technolink created projected 279~6859.
Max crits of 1 occur at level 110. Even at 111, it's very rare and the damage is very low, but damage can occur in excess of 1. The Stereo/Technolink formula projects this to occur 100% at 117.
The formula projects 2679~8859 for 5000 wdef. I witnessed 2735~8817. I spent almost 30 minutes throwing for this test. So it's pretty close. However, it projected 1~1 for normal hits and I observed a hit of 114 at level 111. I'd say it's still a bit off at this level.
It projects 1~1 regular damage and 1~514 for crit damage for a level 106 against 2000 wdef. I did not see anything other than 1 in 15 minutes of throwing. Crit or normal. Tester's note: Most boring 15 minutes of my life, hands down.
It looks like there's a more complicated relationship between wdef and level difference than we currently account for. I don't know if the formula changes at certain points, such as 2000 wdef, 5000 wdef, and 7000 wdef (arbitrary examples, but you get the idea), or if the formula is incomplete.
Given that there's still a big question mark on Anego, neither would particularly surprise me at the moment. If you have any numbers that would make the relationship clearer to you, let me know. And before you ask, no, I didn't save this post for months while adding to it.
It's a sticky, so I think it's okay in this case.
Anyway, I was revising my damage calculator a bit and I became interested in the nature of the damage reduction formula. I don't think the one you have adequately describes the process that the game goes through. It may work for lower test values, but it doesn't work for all values.
Using *ahem* slightly dubious methods, I've determined that 1%/level worth of difference is absolutely correct - at 0 wdef. However, at extremely high levels of wdef, this is not the case.
I attacked a level 200 monster with a level 145 Night Lord. I set its evasion to 0 so accuracy would not be allowed to play a part of this.
The raw damage range I'm using for testing (I can give more specific stats if need be, but I don't see them as relevant to this issue at the moment) is 12622~25244 with critical Triple Throw hits.
At 0 wdef, the maximum observed critical over a few minutes of sampling constantly was roughly 11400, which is almost exactly 45% of 25244. At level 100, I did nothing but 1s. At level 101, I saw quite a few crits of 251, but no more. That's .99% or essentially 1% of 25244. This is why I believe the formula is correct in this respect.
However, when I upped the mob to 9000 wdef, the maximum critical I observed was 8678. The original formula projected 1~1, which is obviously wrong. The formula that Stereo/Technolink created projected 279~6859.
Max crits of 1 occur at level 110. Even at 111, it's very rare and the damage is very low, but damage can occur in excess of 1. The Stereo/Technolink formula projects this to occur 100% at 117.
The formula projects 2679~8859 for 5000 wdef. I witnessed 2735~8817. I spent almost 30 minutes throwing for this test. So it's pretty close. However, it projected 1~1 for normal hits and I observed a hit of 114 at level 111. I'd say it's still a bit off at this level.
It projects 1~1 regular damage and 1~514 for crit damage for a level 106 against 2000 wdef. I did not see anything other than 1 in 15 minutes of throwing. Crit or normal. Tester's note: Most boring 15 minutes of my life, hands down.
It looks like there's a more complicated relationship between wdef and level difference than we currently account for. I don't know if the formula changes at certain points, such as 2000 wdef, 5000 wdef, and 7000 wdef (arbitrary examples, but you get the idea), or if the formula is incomplete.
Given that there's still a big question mark on Anego, neither would particularly surprise me at the moment. If you have any numbers that would make the relationship clearer to you, let me know. And before you ask, no, I didn't save this post for months while adding to it.