Thanks "bro"
I'd rather write an essay that actually sends out a message than writing a LOL which makes me look retarded and lacking decent education.
@chainlock : You can use the pen-and-paper D&D style and find out how many seconds you need to kill Berderk with your set. Yeah, it can be done. It's all about math. Whatever calculations you do on paper, they will be valid in-game too, as the game uses the same math as you did. You don't need a video to prove stuff, it can be proven even outside the game.
Because you crit differently. If you would hit ONLY non crits or ONLY crits, you would kill faster with them at 9. But like this, having a mix of crit and non crit, the killing time is averaged and it appears to be the same. You can do the math with the average values and not the crit values of skills, and you can see the results then. Your colleague LastHour did a more than excellent job with his DMG Calculator, and it takes into consideration every aspect of the game (crit rate, crit damage, average dps, etc)
It can depend on your crit rate. If you have a lower critical rate especially, the total criticals for your bm2 can fluctuate a little leading to less total damage in the end. This can still happen whether your critical rate is 40% or 60% or w/e. Sometimes I've noticed finishing a boss within my last skill and other times with 5 seconds or so to spare using the exact same gear set. Its usually not a gigantic difference, since the more skills you use the closer you get to your true average [your critical rate], but its still likely going to go off way one or another a little. Ex: A full bm2 might be within the range of +/- 10 criticals from making your true average, those +/- 10 criticals could be masking the difference for your bm2 test.
When we use math and analyze like we did here, we would assume the average. Over a full bm2 and certainly over the course of all your bm2s, it is likely you will not be far off from your critical rate and thus safe to assume when we talk about long term usage. Your bm2 efficiency wouldn't change much from the theory ratza has backed up for going something like +/- 2% rate. The difference is large enough you can pretty much assume the conclusion [9-12 skills are most efficient for comp/gm in bm2] would hold true even with the standard deviation of your criticals.
It also sounds like your just eyeballing it, and especially if you have low damage, you won't notice as much difference and leave yourself up to human error. I'm sure someone might be able to record and put them side by side to show you the difference, but it'd be a waste of time proving something already proven.
Edit: Dang ratza beat me in typing up :P
Last edited by LastHour; 09-29-2010 at 11:41 PM.
Also, I'd like to point out that the %CR you see is not Mathematical Percentage of Crit Rate, it's actually Probability Percentage of Crit Rate.
Mathematical Percentage means that out of the total number of hits you critted half of them, while probability percentage means that EACH hit has a 50-50 chance of critting, which can result in 0 crits or all-the-time-crit.
What this means is that with 50% CR, you will not crit half of the time. The game doesn't know how many skills you will cast, whether you will cast 10 skills so as to make you crit 5 times or 200 skills, so as to make you crit 100 times.
Each hit you make has its own chance of critting or not, that's why you might crit twice in 10 hits with 50% CR or you might crit 8 times with 10 hits.
The misconception about this %CR is based on your number of samples. While 2/10 or 8/10 might not explain the 50% CR, once you increase the sample number of hits from 10 to 200, your fluctuation will drop (it won't fluctuate between 20% and 80% mathematical percentages) and you will get closer and closer to the real %CR shown in C stats.
I actually hit a friend of mine 500 times to prove this, with the same skill and 40% CR and 50% CR. On a smaller sample of hits, the 40% CR critted more often, but as I was reaching 500 hits with each set, the 40% CR yielded somewhere around 190 - 195 crits (which is almost 40%) while the 50% CR yielded 230 or so crits (which is close to 250, the 50%). Thus said, if you really want to see any difference in gear and crits and so on, you need a much bigger sample of hits than the ones from a single BM2.
However, this does not mean that there isn't any increase in DPS, it's just not that extreme so as to finish Derk with 1 bar instead of 2 with same gear. In dungeons though, when you go BM2 multiples times and you are running against the clock, you will notice the difference immediately.
I'm judging it by looking at how much time is left in bm when I finish. No matter how I arrange the skill levels of the 3 big skills it never differs by more than a few seconds, which can be explained by crits and non crits. Having the skills at 9 or 12 is supposed to be a HUGE increase, but myself and the people that have tested with me don't see those results in game.
I thought about this too. My friend and I 2 man IC daily and have tested in there too, the dungeon takes us pretty much the same amount of time no matter how our skills are set every run. He is also a wiz.
Last edited by chainlock; 09-29-2010 at 11:59 PM.
Bookmarks