#1




OHNOES S&F theorycrafting
So, after finally putting together what I can best describe as a ton of data points, here is a fairly reasonably mathematically supported breakdown.
First off, each class is different, and I'm going to start with warriors, as they, are right in the middle of mages (most predictable) and scouts (least predictable). PVP: The basic goals in pvp seem to be fundamentally driven by ratios. By this, I mean ratio of stats in specific categories, which do not drive initial damage, but rather how much the damage will increase in later rounds. Take, for example, a mage with 4000 INT and 450 str vs a warrior with 3000 str and 1100 int. The warrior has a strength ratio of over 6 to 1 against the mage, while the mage is LESS than 4 to 1 with int. As a result of this, the following will occur. By round 3 of the fight, whereas the mage will experience between a 110120% jump from initial damage (for example, round one 300k, round 2 440k, round 3 650k), the warrior will experience between a 200260% jump in damage (100k, 190k, 350k). This is particularly important for warriors against mages, as most warriors will kill a mage in roughly 34 hits of equal or near level. They will want to maintain more than 25% of whoever they are attacking's intelligence, or by round 3, they will be sustaining triple damage or more. By round 4, the damage ratio is close to 550%. The same applies to scouts as well, however, the dodge and block percentages override this and turn almost every battle into a more drawn out encounter. Against warriors, what you will notice is that if you have less strength, even though your initial hit might be stronger, as the fight draws out, they will be gaining damage faster than you do. For instance, against boss 92, when I attempted the fight with 3700 str, in round one, we both averaged about 40k damage done. By round 13, the boss was doing almost twice as much damage. At 3800 str, this bonus for the boss disappeared, which leads me to believe that the game has a built in margin of roughly 10% (meaning if you and your opponent have the same primary stat, if you are within 10% of them, your damage and their damage will increase over rounds at similar rates. Anything more than 10%, and a big dropoff can be expected). Also, after averaging almost 37,000 total data points out, I have determined that luck is broken as it appears on the character sheet. Over the course of 51 total attempts on boss 92 (myself having 1860 luck, the boss 1050), I had 148 critical hits (a ratio of roughly 19% of my attacks, or a little more than half of what the 32% on my sheet showed), and the boss had 113 critical hits against me (an average of 15%). In PvP style encounters, I have not hit a single target enough to determine anything mathematically significant, however, the eyeball test shows that the results are similar. Therefore, as it appears, luck is relatively useless. Taking this model and applying it in an fmodel test, the return spits kvalues like this: Str: 1.0 Int: .18 Dex .21 Con .59 Luck .29 What does this mean? It means that, if the next point of strength costs you 1000 gold, you should upgrade your intelligence until the next point costs 180, your dex until it costs 210, your con until it costs 590 and your luck until it costs 290. These ratios are relatively static metrics. Also, from a gear budgeting ratio, ideally you should have 3 all stats items, 2 or 4 3 stat items, and the remainder should be pure main stat items. 2 3 stat items is more ideal for dungeon boss encounters, 4 for encounters against other players. Considering the gear budgeting has 3 stat items giving you roughly your level in all stats, 3 stat items giving you approximately 1.2 times your level in 3 stats, and main stat items giving you 2x your level in that stat. This budgets to PvP (from gear): Str: 13.8 Con: 7.8 Luck: 7.8 Int: 3 Dex: 3 PvE: Str: 15.4 Con: 5.4 Luck: 5.4 Int: 3 Dex: 3 Now granted, the PvE setup is meant for scouts and mages. Against a warrior dungeon boss, you will want 34 3 stat items and 67 pure main stat items, as dex and int both have a marginal value of 0 in those encounters. More to come: 
#2




xodi
for president 
#3




More stuffs:
Ok, so, I gathered some data points yesterday off of Shemale, Hrahrahra and Kinia, to test some of the theories at work. (so sorry guys, it wasn't personal, I was collecting some data off of you, and I figured I might as well get SOME gold for my troubles, since it cost me over 70000 for the pots required) Against hrahrahra, as a scout, even though the results, eyeball wise, were about what I expected, the evades and blocks make it mathematically irrelevant, so we'll focus on the mages. I started with a base int of 1080. This was > 25% of Shemale's Int, and less than 25% of Kinia's int. Shemale had a base average damage of roughly 340k (270k410k). Kinias average base damage was around 355k. In 12 total relevant attempts on Shemale: I cut critical hit values in half to get the base damage from the hit: Round 1 average hit: 332k Round 2 average hit: 427k Round 3 average hit: 611k Round 4 average hit: 780k (only 3 datapoints) 12 relevant attempts on Kinia Round 1 average hit: 352k Round 2 average hit: 511k Round 3 average hit: 894k Round 4 average hit: 1.431M (4 data points) So then, I drank an int pot with 280 or so int, bringing my int up to 1360 or so, which was still less than than a third of Shemale's int, but brought me above 1/4 of Kinia. In 12 total relevant attempts on Shemale: I cut critical hit values in half to get the base damage from the hit: Round 1 average hit: 341k Round 2 average hit: 407k Round 3 average hit: 581k Round 4 average hit: 728k (5 datapoints) 12 relevant attempts on Kinia Round 1 average hit: 348k Round 2 average hit: 431k Round 3 average hit: 606k Round 4 average hit: 821k (4 data points) not that much of an improvement against Shemale, but the difference against Kinia was immense, cutting round 4 damage almost in half, and the third hit by almost 50%. SO then, I went back again, and I clicked off my int potion, and drank a crappy int potion (74 int) I was just above 25% of Kinia's int: In 12 total relevant attempts on Shemale: I cut critical hit values in half to get the base damage from the hit: Round 1 average hit: 329k Round 2 average hit: 437k Round 3 average hit: 619k Round 4 average hit: 778k (2 datapoints) 74 int made absolutely NO statistical difference against Shemale 12 relevant attempts on Kinia Round 1 average hit: 358k Round 2 average hit: 462k Round 3 average hit: 649k Round 4 average hit: 883k (7 data points) with just 74 more int, my numbers with Kinia were very similar to my numbers with a 280 int pot, but more importantly, going above 25% of my opponents int seems to make a HUGE difference in that combat. I'm going to try and do some testing later on what the difference is at 1/5th, 1/6th, and 1/3rd, but for the first two, I cannot really use a mage, since at those ratio differences, I will probably get blasted in 3 hits (the third hit would be over 1.8M at 6 to 1 ratio according to what I am figuring), so if anyone knows a scout volunteer at or near level 160 or so who's willing to help, lemme know 
Thread Tools  
Display Modes  

