Monday, March 03, 2008

The lady doth protest too much.

Ok, first things first. Yes, Omoi did beat me on a parse this weekend. Dual BRD, and the other melee was a SAM. Omoi DRG: 37%, Ring MNK: 32%, Other SAM: 28%, Omoi's freaking Wyvern Nanaja: 3%, Blaize BRD: .01% This was on BIRDS! Ok, so she gets the 15% damage bonus or whatever. She barely ever got below 1k on a Pentathrust. Give me a freaking break. And her DRG isn't even fully pimped out. It's fully merited, but she only has 1 piece of Homam and she is still using freaking Barone pants... FREAKING BIRDS... here is another bird for ya... Crow, it's whats for dinner. Lets try a neutral camp next time, eh?

I fucking hate this worthless blogger software sometimes. It can't fucking figure out the difference between a smilie and a god damn link. Anyway, this is where I lost a good section of my blog on what I learned about the strangness from last weeks discovery that the results I had did not match what I expected. The first problem was with one of my assumptions, that the graph of damage per hit would be a normal distribution. You might remember that discussion from Friday and you can peek at the results here. Instead of a normal distribution, it is in fact, a flat distribution. For those that don't understand, a normal distribution is how a group of events (in this case damage per hit) should be distributed, there should be more events in the average (the middle of the graph) than the outlying areas. That would mean that there is a greater chance of getting a damage in the middle between the two limits. Bell curves are the example of this kind of distribution. For FFXI though, it appears that SE has used a flat distribution, which just basically means that there is an equal chance that any result will appear anywhere between the limits. The simple reason that SE is probably using this over a normal distribution is because of something I mentioned on Friday, it's just easier. A normal distribution by itself requires a more complicated algorithm to get good results, and that is only compounded as more influences are placed on the algorithms results, i.e. Attack and STR. A flat distribution is actually much easier to apply influences like Attack and STR to because they just become the curve itself. Since a flat distribution isn't effected by the curve that a normal distribution has naturally it should be easier to test how changes in attack and STR affect your damage. I have a baseline on which to test changes, though I will do a larger sample before the next test. As long as the levels of the birds are the same and my equipment stays the same I should be able to just add new baseline tests onto my old ones.

The other problem from Friday still persists though, but that is apparently not my fault. The old pDIF formula does not apply to two-handers any longer. This was likely changed with the two-handed update, and my guess is that it is in the level correction part of the equation. I believe this for two reasons. The first is that level correction is the place that is most likely to effect the range of damage, especially considering that it the range is not just wrong, but also too wide. The second is that SE mentioned they were making changes so that two-handers would have less of a penalty dealing with high level monsters. So, that is where I will be testing next. I think that the multiplier for the Attack/Defense ratio might be .030 for two-handers now, as that matches perfectly with my results for fMax but is providing a much too high limit for fMin and that doesn't match with my findings. Anyway, I am going to figure this out one way or another. Tomorrow, I will be looking at my MNK in different situations, calculating fSTR and Attack.

But enough with numbers for today, lets have a nice old school post with pretty pictures and neat stuff. :) And since the thing I like to talk about most is Salvage, lets do that. First, I have to say that doing Salvage with another group is great, its really helping me to run Versus Salvage much better. Last night for example, we did an Arrapago run (no drops, of course) but we did manage to do almost all the NM's and also do a quick run at the boss (timed out). We are now farming way more on the first floor which is making things easier in general, and we won't do Arrapago without a THF ever again. LOL We did 2FL QQA, 3FL QQTH, 4FL QQTH, 5FL QQA and Archaic Chariot, and on the sixth floor we started off really well but ended up with some crowd control issues and didn't clear the floor in seven minutes. We had about 5 minutes left at this point so we just zoned up and got Armored Chariot to 87% when we timed out. I have ways to shave time, and if we really need to we can get there with lots of time left over, but right now we have no Zhayolm 35's for it to matter very much.

That was Sunday's run, Saturday's run was another Arrapago. We did the big Bhoot also which didn't drop anything. In fact, Arrapago has been sucking a lot for drops as of late. LOL Friday night was an excellent Bhaflau, we did 10 minutes on the second floor Rampart with no spawns, then just killed it so we could move on. Cleared the first five gears on the fourth floor, waited for five minutes for wind weather to end then cleared the other five gears, and went into the fourth floor rampart. Which I had a screenshot of but apparently didn't upload. Fuck you again blogger for being completely retarded and ruining this post! Anyway, we went into the Rampart and killed a couple of Tragopan, but after about 10 pops we just killed the Rampart so that we could kill KBC with plenty of time. But once LBC is nerfed you usually have too much time! LOL Which is what happened with us. But man, I just didn't want to take any risks, and just get the win. Next time I will be way more lax about it though, LOL. A nerfed LBC was just ridiculous, it's damage is cut at least a 1/3 to a 1/4 and it took us less that 10 minutes to kill. Leaving us with about 25 minutes of wasted Remnant farming. Rygar was so used to moving for Homing Missile that he didn't stand in front of it to get hit with it and kept moving to the side. LOL Ah well, hopefully many more nerfed LBC's will be in our future. The results were a pair of Tsukikazu Haidate for me (I always feel bad lotting alone, but Rygar already has Usukane legs and Akanea and Hiryo, though now very consistent just haven't been going as long as I have and Hellz is lotting on Ares gear and he smells too) and another pair of Ea's Dastanas which went to Ashhhhh. That leaves us with all of the drops except for the bodies and the Ares's head piece. With three or four people waiting on that 25 body piece, I would appreciate if it would drop a little more often. LOL In other Salvage news, I did a SSR run with the other Salvage group that I go with and it went pretty well. We did all of the NM's except for Dekka (because its a forced choice) and Gyroscopic Gear (because once you choose to do Hammerblow you can't do Gyroscopic Gear). Gyroscopic Gear is a pain in the ass. Huge Restorals and huge AoE damage moves. And here is where blogger tried to fuck me a third time! At least I fucked it back. Anyway, we were onto the turtle after the Gears, which is actually surprisingly easy to pull if you don't need to get to the boss. We were limited on numbers for that run, but we I think there is a way with competent people to do the turtle, kill the Chariot to reach the boss all while avoiding the gear train. It just involves sacrificing Blaize to the gods of Salvage two more times. LOL We pulled the four Ramparts back to the start room, killed them then Blaize pulled the turtle. It is probably much more difficult when you need to get to the boss but this run wasn't too full so we just focused on the turtle. We killed it but of course it did not drop anything worth while. Ok, enough Salvage, on with other stuff. Well, things have been rather plain. We did a couple Ix'MNK and we finally got our second Merciful Cape which went to Crlmsonking, the luckiest person in the universe. LOL Hehehe, I think we are going to do a couple of Fortitudes soon to try to get some kind of torque drop from something. As if Torques existed in Sea! Hah! Finally, I had to work on my BLM a little more to be able to level my WHM, so that got to 27. And my WHM got to 49 I think, maybe I am just close to 49. Hopefully after Assaults tonight I will be able to get to 49 or 50 in Campaign. 51 still remains elusive!

3 comments:

Hellzfury said...

I think I'm going to just ctrl+f for my name in here before I actually read to see if you've hidden an insult in the wall of text... btw.... windows 95 bootup.... locked room... I'll make it happen...

Omoikitte said...

Bring it on *throws down the (Ares') gauntlet.

Anonymous said...

You parsed higher than a SAM? That SAM must not have been good. I'm no BB MNK but good GA WAR, SAM, DRG always seem to outparse me on Birds. The bad ones always get outparsed.