Randomer Question Thread

[SamanosukeShiva]

This is something that has started to catch my fascination recently. With Samurai's 240 skill unmerited, that is about 48 attack and accuracy short of my fully merited Great Katana, so does the +25% damage boost on normal hits and weaponskills exceed the kind of damage you can get with a Great Katana? Or would I need max out my merits before I saw Polearm pull ahead of my GK damage?

To Samurai Polearm users:
What is your Polearm(s) of choice?
What is your gear of choice (5-hit setups, anyone)?

1. I use Gondo-shizunori
2. Rajas, Hachiman kote, hachiman sune-ate

Polearm is very damaging but require at least 1 madrigal 1 minuet for it to work. The set up i usually go to bird camp with is SAM SAM BRD BRD COR RDM. Both polearm SAM and it's overkill as it is. Non-DA, Non-Berserk Penta with this set up can do up to 1600~, If you have 5-hit set up with additional SAM roll from COR it will become 4-hit which is awesome...

Songs+Rolls: 2x Minuet, March, MADRIGAL(very important for polearm), Chaos Roll, Samurai Roll

example of dmgs

http://i199.photobucket.com/albums/aa17 ... /penta.jpg

http://i199.photobucket.com/albums/aa17 ... ntarec.jpg

edit: I do have maxed polearm merit

[Amele]

ringthree: to the best of our knowledge now, it isn't actually variable (although we aren't 100% on this)

the approximation for FSTR2 is some formula - divided by a constant (2). the approximation for FSTR is some formula - also divided by a constant (another 2).

so if the equation is linear (at least the 'dstr' part actually be STR-VIT with no hidden multipliers) then the amount you need to increase FSTR by 1 is always 4.

of course, no one is 100% on the fstr formula (the approximations are a little messy) the 4~6 estimates were made before the formula was very well understood and have continued to be grandfathered into the information.

Ok, perfect, that is getting closer to what I want to know. Next question, how would one go about testing this to see if it is consistent?

Another related question: Would the full equation for a normal melee attack (assuming that the approximations are true and you are in a high STR situaton) look like this: ((dSTR+4)/2)/2)?

if discussing a real DD (and not say, a mage/nin) they will basically always be in the high strength situation for 90% of the activities in the game (I think you knew this already, but stating for posterity) the only stuff that comes to mind as being strong enough for you not to be is fafnir level or higher. (and even then the sorts of DD that actually DD on fafnir and higher will -still- be in that range)

the formula should be about that for a regular melee attack yeah. remember that it caps at wrank. (another thing that wasn't really understood when the 4~6 estimate was made)

the best way to test would probably be to manufacture a situation where you have as much control as possible and uncapped fstr (so can't do level 1 mobs) if you can get into a range where you see a significant mode at 1.0 pdif that will help.

identify the mode, then adjust your strength small increments. if you can go +/-8 in 1str steps that will be ideal. (preferably without moving your attack around much) the mode should shift 1 damage per fstr increment if you mode is around 1.0pdif.

if you can do str increments over an entire range from <VIT to much much more than vit (capped fstr) that'd probably be the best way for you to verify for yourself the full equation *shrugs* (personally, I did the above and it worked out to about 4str per point and I called it good. - remember to adjust your attack too to compensate the str if you can.)

What makes MNK a better tank for chariots opposed to NINs? H2H is better damage rather than katanas?

they do excellent damage (counter, asuran, etc) tend to have a very good TP/damage ratio and generally have higher HP (this is significant vs. some chariots, meaningless against others) and have an advantage over ninja on earlier floors because they have a better minimum damage threshold and significantly more hp nerfed. most groups that tank chariots with monks are doing so because they don't bring 'real' tanks at all. (because chariots aren't as hard as most worldspawns) so it's not really a case of monk being a better tank as it is that you don't need a better tank for chariots and monks are more useful generally in salvage.

if I had a ninja or paladin along for the ride for some reason, I would likely put them on the chariot to tank *shrugs*

[Tricen]

I always thought it was for the subtle blow lol >_> But NIN has more subtle blow traits than mnk. I could see NIN/DRK being fairly good for chariot tanking since stun can save lives and for the brainjack chariot, would be good to get hate back faster than a mnk :O

[ringthree]

What makes MNK a better tank for chariots opposed to NINs? H2H is better damage rather than katanas?

There are several things. MNK has high Vitality and high HP which is very helpful for the Chariots. NIN will take quite a bit of damage from regular hits which can be a large problem after Discharge and a paralyzed Utsusemi. A lot of NIN damage is from swinging very quickly to get to the next WS, but even with a good amount of Subtle Blow this can add up to more TP for the Chariot. That is why I generally prefer MNKs.

Really, there are actually a lot of jobs that can tank a Chariot. NIN, MNK, WAR, SAM, hell even DRK or DRG could follow the SAM pattern. It's really more about being conscious of what is going on and adapting. If you are giving the Chariot too much TP pull someone off of it. If you have the healing or are rushed, put an extra person on it. I have found that solo tanking is actually pretty effective if you have time on low-man runs, and at that point any job is fine for tanking a Chariot.

[ringthree]

ringthree: to the best of our knowledge now, it isn't actually variable (although we aren't 100% on this)

the approximation for FSTR2 is some formula - divided by a constant (2). the approximation for FSTR is some formula - also divided by a constant (another 2).

so if the equation is linear (at least the 'dstr' part actually be STR-VIT with no hidden multipliers) then the amount you need to increase FSTR by 1 is always 4.

of course, no one is 100% on the fstr formula (the approximations are a little messy) the 4~6 estimates were made before the formula was very well understood and have continued to be grandfathered into the information.

Ok, perfect, that is getting closer to what I want to know. Next question, how would one go about testing this to see if it is consistent?

Another related question: Would the full equation for a normal melee attack (assuming that the approximations are true and you are in a high STR situaton) look like this: ((dSTR+4)/2)/2)?

if discussing a real DD (and not say, a mage/nin) they will basically always be in the high strength situation for 90% of the activities in the game (I think you knew this already, but stating for posterity) the only stuff that comes to mind as being strong enough for you not to be is fafnir level or higher. (and even then the sorts of DD that actually DD on fafnir and higher will -still- be in that range)

the formula should be about that for a regular melee attack yeah. remember that it caps at wrank. (another thing that wasn't really understood when the 4~6 estimate was made)

the best way to test would probably be to manufacture a situation where you have as much control as possible and uncapped fstr (so can't do level 1 mobs) if you can get into a range where you see a significant mode at 1.0 pdif that will help.

identify the mode, then adjust your strength small increments. if you can go +/-8 in 1str steps that will be ideal. (preferably without moving your attack around much) the mode should shift 1 damage per fstr increment if you mode is around 1.0pdif.

if you can do str increments over an entire range from <VIT to much much more than vit (capped fstr) that'd probably be the best way for you to verify for yourself the full equation *shrugs* (personally, I did the above and it worked out to about 4str per point and I called it good. - remember to adjust your attack too to compensate the str if you can.)

So, at the high STR situation the problem is not so much that there is something hidden but more likely that it is relative hard to test, so people just took a stab at a range? I have been writing about the damage formula on my blog the last couple of days and I want to make sure I am getting this correct.

Another question, has anyone ever tried to plot their damage to see what the curve that the randomized algorithm makes looks like? Is there a parser that will just parse damage per hit and accumulate the results?

[Amele]

kirschy has been doing some work on that, I believe; it may have been suspended by the spate of damage formula patch changes recently.


I do not know if any of the parsers can provide just the numbers individually for you, but several parsers support raw chat log dumps that you could mine yourself.

[pyra_caitsith]

What is the approx drop rate on the Homan Body from Proto-Omega? I'm asking b/c my current Limbus ls has not seen one fall in probably 5 months and about 8 Omegas. Is this normal, or are we just unlucky?

Also, is there any difference in the Omega fight if you enter from Mea vs Dem?

thanks
-pyra @ caitsith

[ringthree]

kirschy has been doing some work on that, I believe; it may have been suspended by the spate of damage formula patch changes recently.

I do not know if any of the parsers can provide just the numbers individually for you, but several parsers support raw chat log dumps that you could mine yourself.

Well, since I have the programming skill of a 4th grader I don't know if mining will work for me, but I could just eyeball it if I set my filters right. It might not be that bad. How difficult would it be to adjust a parser to do that? Can you suggest a good parser?

[Tomiko]

What is the approx drop rate on the Homan Body from Proto-Omega? I'm asking b/c my current Limbus ls has not seen one fall in probably 5 months and about 8 Omegas. Is this normal, or are we just unlucky?

Also, is there any difference in the Omega fight if you enter from Mea vs Dem?

thanks
-pyra @ caitsith

It's all random (luck). When the group I was in was farming Omega, our rare drop was the Hat which made us very sad. >.>

No difference that I'm aware of.

[Amele]

kirschy has been doing some work on that, I believe; it may have been suspended by the spate of damage formula patch changes recently.

I do not know if any of the parsers can provide just the numbers individually for you, but several parsers support raw chat log dumps that you could mine yourself.

Well, since I have the programming skill of a 4th grader I don't know if mining will work for me, but I could just eyeball it if I set my filters right. It might not be that bad. How difficult would it be to adjust a parser to do that? Can you suggest a good parser?

raw chat log is all your logs just sitting there in front of you in text, instead of scrolling past *shrugs* I'd use direct parse, find zhais - he's got a link in his .sig

[Ryushii]

I guess I don't understand the damage formula well: for ranged attacks, how would 1 point of STR compare in increasing damage to 1 point of ranged attack? This question is for both normal shots and WSes. Could anyone point me in the direction of a formula to look at?

[Amele]

I guess I don't understand the damage formula well: for ranged attacks, how would 1 point of STR compare in increasing damage to 1 point of ranged attack? This question is for both normal shots and WSes. Could anyone point me in the direction of a formula to look at?

it's.. highly complicated.

for the section you're wondering about. until your str is greater than the mobs' vit by twice the value of your ranged weapon's damage divided by 9 (fstr2 cap) then you will always do better adding 1 str vs 1 ratt.


if you're really curious, vzx has great documentation at http://www.freewebs.com/vzx-01/

[Kaope]

1. I use Gondo-shizunori
2. Rajas, Hachiman kote, hachiman sune-ate

Songs+Rolls: 2x Minuet, March, MADRIGAL(very important for polearm), Chaos Roll, Samurai Roll

edit: I do have maxed polearm merit

1) Mind sharing us you WS gears? ^^;
2) Would this work with lvl 0 polearm?
3) Hachimans' ws accuracy only applies to the first hit of the ws? or it applies to all hits for multihits ws as well?
4) Love Halberd = Love it? Hate it?

[Arnwulf]

A lot of ppl talk about how a 75 BRD is the best Power Leveler for solo xp. What is the best way to pull this off?

[Suiram]

Has any BLU here ever tried Heat Breath on Temperance? Does it do 0, full damage, or partial damage?

[pyra_caitsith]

A lot of ppl talk about how a 75 BRD is the best Power Leveler for solo xp. What is the best way to pull this off?

It involves inviting the BRD to the pt and disbanding a lot. BRD songs don't suffer from level issues WHM spells do (protect 4 lasts ~10 mins on a lvl 20, haste wears very quickly, etc). So, Valor Minuet IV will give a lvl 10 and a lvl 75 +~54 attack (depending on the BRD and their gear, etc) and most importantly it will last the full time. 54 attack is huge at lvl 10 (I don't know all the math that well). BRD can also give highest ACC song too. Then, the BRD drops pt and can heal from outside. Another option is to use the highest Regen song. I'm not a BRD, and I only heard about this type of PL once, a long time ago.

[Max™]

Had a lvl 30 party of War, Sam, Mnk with a Bard giving us Madrigal and March, both top tier versions, then disbanding, we got to 33 getting chain 5 and 6 regularly, and were killing smithies and mandies like it was nothing. It rocked pretty hard.

Done this in the dunes too at 16-20.

[Therin]

If your songs wear off mid-EXP and you reinvite the BRD, you won't lose EXP as long as neither you nor the BRD uses any action on the mob while he's in party. Simple invite the BRD, turn away from the mob, have the BRD join the party, use the songs, then disband and continue killing.

[SamanosukeShiva]

1) Mind sharing us you WS gears? ^^;
2) Would this work with lvl 0 polearm?
3) Hachimans' ws accuracy only applies to the first hit of the ws? or it applies to all hits for multihits ws as well?
4) Love Halberd = Love it? Hate it?

1. Weapon: Gondo
Grip: Pole
Ammo: Tiphia Sting
Head: Ace's Helm
Neck: Chivalrous Chain (Love torque best)
Earring: Brutal/Hollow
Body: Haubergeon+1
Hand: O-Kote
Rings: Ulthalam/Rajas
Back: Amemet+1
Waist: Potent belt (AF Belt best IMO)
Legs: Byakko Haidate
Feet: Saotome sune-ate
2. Yes, but you wont get full hit penta constantly
3. I don't penta on hachiman, it's only for TP'ing
4. Hate it, you get TP too fast as it is without halberd.

[Arnwulf]

Thanks for the advice! Now to work on getting all jobs to 37 solo.