What Are The Appropriate Gears When Subbing /Ranger

[Spider-Dan]

How else would you eplain 95% Namas accuracy on things like Chariots, when I have 4 archery merits, almost no RACC gear (outside of agi) whatsoever in my WS build, using meat dishes.

Aftermath?

Just saying, if all ranged WS (sans Slugwinder) have RACC+100, I would have expected to land a lot more WS. And I'm not just eyeballing; on COR, from LV22-50 (i.e. pre-Slug):

ranged attacks: 3572/4717 (75.7%)
WS: 522/597 (87.4%)

There does seem to be some accuracy boost, but it certainly isn't +100.

[Drace]

I already covered the aftermath situation, not happening. With 324 total RACC on my SAM Namas gear I parse ~94+% Namas acc on Chariot bosses (50+ runs I have used Yoichi on bosses), 0 aftermath overlaps unless I am 2 houring. It is very obvious there is a significant accuracy boost to it, and I never said it was for all RACC ws, I said it was for Namas.

Vodou already showed some very convincing math that would put it in the 50+ range, and he was still at capped accuracy for that (suggesting that it could obviously be higher). I am telling you over 50+ chariot runs I have capped namas accuracy with 324 total RACC in my ws gear.

Not sure how you can definitively say it "certainly isn't +100" given the results we have shown. And again, you are basing this off of ranged WS's other than Namas, I don't think its a stretch to say Namas could be an exception to the standard. It is more than clear that there is an accuracy boost to Namas (which you at first doubted), it is very clear to me that it is similar to the boost that other single hit/ first hit of multi hit melee Ws's get.

[Spider-Dan]

Fair enough; I don't have Namas, so I can't speak to any accuracy bonuses that it may or may not have. If there is a ranged accuracy bonus, then it's specific to that WS.

[Vodou]

Aftermath?

Just saying, if all ranged WS (sans Slugwinder) have RACC+100, I would have expected to land a lot more WS. And I'm not just eyeballing; on COR, from LV22-50 (i.e. pre-Slug):

ranged attacks: 3572/4717 (75.7%)
WS: 522/597 (87.4%)

There does seem to be some accuracy boost, but it certainly isn't +100.

Wow. This is pretty interesting stuff.

Let me ask you this though, as I have no idea and I never thought anyone would have this data handy.

What do you think of the possibility that it is +100 but only at level 75? I've never thought about it before, but now that I do, it makes sense that we wouldn't get +100 on melee accuracy at level 30. So why would racc be that way either.

Example: its on a sliding scale from +25 to +60 from levels 20 to levels 50. That would fit nicely for the data you have. Crap, you may have the only proof it does work that way. Anyway, its a thought. I don't have data that says other ranged WS get the bonus, so your point is certainly valid.


Re: aftermath: I probably never used Namas during the aftermath of another Namas because I'd have to get the second one off in about 32-33 seconds and it would have been extremely rare for me to let my TP get that high. Aftermath is 20 seconds on 100tp scaling up to 40 seconds on 200tp, etc., so 32.5 seconds is around 165tp. My average WS was fired at 112.5tp and sidewinder was the one that I let tp get pushed to be higher on the tp scale. Namas was almost always fired at 100tp.

[Imladra]

It really depends what kind of event you want to do as sam/rng. If you plan to go on gods and high eva HNM, forget it unless you have Yoichi (maybe, i can't tell i don't have it).

But on events like limbus, salvage (but you need so many cells), sky farming, sea, and well, anything that isn't high eva or high def. It's pretty nice.

I've pushed the thing a bit further lately and here's what i'm using for sidewinder :

Soboro sukehiro
axe grip
shigeto bow +1
demon arrows
optical hat
aqua gorget
beater's earring
vision earring
nuevo corselete (racc+10 & ratk+10)
seiryu's kote
behemoth ring +1 x2
amemet mantle +1
buccaneer's belt
dusk trousers
askar feet

ontop of that i use Pot-Au-Feu, a food giving racc +ratk (equivalent of pizza imo)
i have 3 archery merits so far, so i'm not done with it. But i can tell that it doesn't work on everything.
With all this i stand at ~393 racc w/o food i think which is low and i'm wearing a full racc set. That also mean a low str on sidewinder, so ws damage is not really impressive compared to rangers (take the blank point in account though). You need bard or cor to make it shine too because 230 base skill means low base atk.

All that said, it's really fun and still very powerfull. But you will be disappointed if you don't reach 380+ racc imo. I'm looking forward at the next hat we can get in the coming mini expansion. str+4 wsacc+15 racc+10 & rat+5 is going to do wonder.

[Spider-Dan]

Wow. This is pretty interesting stuff.

Let me ask you this though, as I have no idea and I never thought anyone would have this data handy.

What do you think of the possibility that it is +100 but only at level 75? I've never thought about it before, but now that I do, it makes sense that we wouldn't get +100 on melee accuracy at level 30. So why would racc be that way either.

Example: its on a sliding scale from +25 to +60 from levels 20 to levels 50. That would fit nicely for the data you have. Crap, you may have the only proof it does work that way.

I have a good deal more data than that:

Just for the record, I'm a Mithra COR that's had maxed marksmanship merits during my entire COR career, as well as most of the +RACC gear in the game (I already had a great deal of it for my RNG). Although I have occasionally used other subjobs and eaten a coeurl sub or two, I would say that during the vast majority of my LV22-75 playtime (well over 90% of all kills), I was COR/RNG eating squid sushi. So I think it's pretty fair to say that my accuracy numbers are representative of the high end of COR accuracy.

With that being said:

LV22-40:
Ranged 1940/2512 (77.23%), WS 299/333 (89.79%)

LV40-50:
Ranged 1632/2205 (74.01%), WS 223/264 (84.47%)

LV50-60:
Ranged 2263/2917 (77.58%), WS 291/347 (83.86%)

LV60-67:
Ranged 2097/2485 (84.39%), WS 397/454 (87.44%)

LV67-72:
Ranged 1822/2132 (85.46%), WS 362/429 (84.38%)

LV72-75:
Ranged 946/1099 (86.08%), WS 232/269 (86.24%)

Total:
Ranged 10700/13350 (80.15%), WS 1804/2096 (86.07%)

--

Bullets- 5314
@current AH value of 14k/99 = 760,420G

Iron Bullets- 10132
@current AH value of 13k/99 = 1,364,579G

Total ammo cost to 75 = 2,124,999G

For the record, I got Slug Shot at LV53, and the ranged un-nerf (10/06) was implemented when I was LV72 (the 72-75 numbers represent post-update data).

[Vodou]

I have a good deal more data than that:


For the record, I got Slug Shot at LV53, and the ranged un-nerf (10/06) was implemented when I was LV72 (the 72-75 numbers represent post-update data).

Yes indeed that is a lot more than a lot of data.

Its interesting to see that the gap between ranged racc% and WS racc% that was evident before slugshot, seemed to disappear afterwards, but many things could cause that.

1) Do you know how many of each WS you were using and the racc% of each? That would be amazing if the non slugshot WS could be split from the slugs.

2) Do you know racc gear swaps you did on WS?

What I think you might have, if you do have that data available is some concrete evidence that could go a long way define some of the issues people are unsure of.

a) is there a racc bonus on all ranged WS and how much is it? That might be limited since your racc rates are so high, you were likely capped in many parties.
b) does slugshot (sidewinder) get that bonus at all or does the racc on those WS actually track ranged attack racc (i.e., no bonus or penalty)?
c) does the racc (and implicity the acc bonus) scale with job level

[Spider-Dan]

1) Do you know how many of each WS you were using and the racc% of each? That would be amazing if the non slugshot WS could be split from the slugs.

Prior to 53, I was using Split Shot virtually all of the time. Post-53, I was using Slug Shot literally all of the time (this was all before Detonator patch, IIRC).

I can try to break out the 50-60 data into splits and slugs, but I lost a lot of data in a hard drive crash a while back, and I'm not sure if I have that.

2) Do you know racc gear swaps you did on WS?

Didn't do any. I was a full /ra corsair, so there wouldn't have been any reason to swap out of max RACC at any time (particularly given my hitrate).

[Vodou]

Prior to 53, I was using Split Shot virtually all of the time. Post-53, I was using Slug Shot literally all of the time (this was all before Detonator patch, IIRC).

I can try to break out the 50-60 data into splits and slugs, but I lost a lot of data in a hard drive crash a while back, and I'm not sure if I have that.


Didn't do any. I was a full /ra corsair, so there wouldn't have been any reason to swap out of max RACC at any time (particularly given my hitrate).

Well that makes it much simpler than. The amount of data you have does a great job of narrowing the CI's and confirming a WS racc bonus plus might say something I had theorized from other nimbers I saw on slugwinder.

First, the data you have looks like this I think if I have my stats right:

(everything is at the 95% CI)

Level 22-40 racc is 77.2 +/- 2.5 WS racc was 89.8 +/- 3.2

Taking the midpoint 77.2 and adding 2.5 = 79.7 for ranged acc
Taking the midpoint 89.8 and subtracting 3.2 = 86.6 for WS acc. This is the smallest distance between these at the 95% CI. So the bonus is at least +14

Taking the midpoint 77.2 and subtracting 2.5 = 74.7 for ranged. Adding 3.2 to the WS% gives 93.0 for WS, so 93.0-74.7 is 18.3. ( a bonus of +36) That is the largest bonus we could expect. So the range is somewhere at +14 to +36.

The level 40-50 racc was 74.0 +/-1.7% and WS was 84.5 +/-4.4%

Doing the same things as above, the smallest and largest ranges are +9 to +33. So we know there is a bonus and we know its smaller than we can see at level 75. We don't know anymore than that really. We can speculate that it increases from level to level, but we'd need far more data to confirm that.


Second, the slugshot data looked like this from 60-67, 67-72, 72-75

Ranged was (84.4 +/- 1.3) (85.5 +/-1.4) (86.1 +/- 2.0)
WS was (87.4 +/- 3.0) (84.4 +/-3.4) (86.2 +/- 4.0)

What's 'bizarre' about this is that the ranges completely overlap in every set. That is, this data does not reject the hypothesis that the slugshots had exactly the same racc as normal ranged attacks.

I could only think of 3 ways this could happen:

1) slugshot has no penalty at any level of tp nor does it have any bonus.

2) the penalty and bonus exactly offset each other at all levels of tp

3) the penalty and bonus do not offset each other at all levels of tp, but do so at certain levels of tp (equilibrium point)

Any of these answers could give the same result, but for #3 to work, other than by chance, the slugs would have to be intentionally used at the particular tp that is the equilibrium point.


Btw, I was going to try to calculate the avg tp per slug, so here is the data on that:

If you divide the ranged hits + WS hits by the WS attempts you get the following pattern:

22-40 6.7 total hits per WS attempted
40-50 7.0
50-60 7.4
60-67 5.5
67-72 5.1
72-75 4.4

You were using /sam from 60-75 gaining faster tp and meditate? Perhaps meleeing a lot as well? I can't think of why else there is such a drastic shift at 60 other than /sam. (Quite dramatic and impressive.)

Were you firing slugs at any particular tp do you recall? (100, 150,?) If you had a consistent tp level, it would suggest all 3 answers above could be correct. If you didn't, it might suggest that only 1) or 2) are correct.

[Spider-Dan]

Btw, I was going to try to calculate the avg tp per slug, so here is the data on that:

If you divide the ranged hits + WS hits by the WS attempts you get the following pattern:

22-40 6.7 total hits per WS attempted
40-50 7.0
50-60 7.4
60-67 5.5
67-72 5.1
72-75 4.4

You were using /sam from 60-75 gaining faster tp and meditate? Perhaps meleeing a lot as well? I can't think of why else there is such a drastic shift at 60 other than /sam. (Quite dramatic and impressive.)

That's a good question.

As I recall, my gun progression was:

LV22- Platoon Gun (600 delay, 16.8 TP/shot)
LV27- Pirate's Gun (600)
LV30- Military Gun (548, 16.4 TP/shot)
LV46- Deluxe Carbine (600)
LV56- Matchlock Gun (600)
LV72- Coffinmaker (600)

...so that, in itself, doesn't really explain it; everything but the Military Gun is a 6-hit build.

The best explanation I can think of is something along the lines of the following:

1) In the pre-Slug levels, I'd be firing from the sweet spot (which had a HUGE impact on your ranged damage pre-unnerf-patch) virtually all of the time, and Split Shot isn't really strong enough to pull hate, so I wouldn't really have an opportunity to get free melee TP via being engaged while shooting. I would also have been more likely to hold TP for a SC at this stage. This explains the high shot counts pre-53.

2) In the 53-60 range, I would not have been shooting slugs with less than 150 TP (probably closer to 200), and I may have been more likely to get extra melee TP after a slug (because I would have pulled hate). This explains why my shot counts didn't drop in this range.

3) By LV60, COR gets several significant upgrades in RACC (AF, 2nd trait from /RNG, rings, knives) and I probably felt more comfortable shooting closer to 150 TP. Furthermore, I would have also gotten Barrage from /RNG. I think this should account for a good chunk of the shot count reduction.

4) Around 73-75, I do recall experimenting with some melee configurations, which obviously would have drastically reduced my /ra output. Since 72+ was under a completely different ranged system anyway (post-unnerf), I don't think that data is of any value to you (for purposes of this discussion).

[Spider-Dan]

Just as an addition:

3) the penalty and bonus do not offset each other at all levels of tp, but do so at certain levels of tp (equilibrium point)

Any of these answers could give the same result, but for #3 to work, other than by chance, the slugs would have to be intentionally used at the particular tp that is the equilibrium point.

In order for the above to be true, wouldn't I just have had to have fired slugs at the equilibrium point on average?

The working hypothesis among many RNGs is that slugwinder has RACC-40 at 100 TP, flat equal at 200 TP, and some-number-we-don't-care-about at 300 TP. So is it possible that my ~200 TP slugs in the fiftysomething range were at just about flat accuracy (no penalty or bonus), and that as I got into the sixties, my more-aggressive TP usage (as somewhat indicated by my reduced shot counts) and subsequent increased accuracy penalties on slugs were slightly-overcompensated-for by my big jumps in +RACC?

Seems to touch all the bases, but maybe it's a little too neat.

[Vodou]

Just as an addition:


In order for the above to be true, wouldn't I just have had to have fired slugs at the equilibrium point on average?

The working hypothesis among many RNGs is that slugwinder has RACC-40 at 100 TP, flat equal at 200 TP, and some-number-we-don't-care-about at 300 TP. So is it possible that my ~200 TP slugs in the fiftysomething range were at just about flat accuracy (no penalty or bonus), and that as I got into the sixties, my more-aggressive TP usage (as somewhat indicated by my reduced shot counts) and subsequent increased accuracy penalties on slugs were slightly-overcompensated-for by my big jumps in +RACC?

Seems to touch all the bases, but maybe it's a little too neat.

These are all very good points. No one can discount option #3. (My writing might have implied that I was discounting it, but I was only trying to describe how it might work. I certainly do not discount that it could be what is correct.) Without data, choosing any option over another would just be based upon opinion, since it certainly fits as well as the other options to the data.

What would 'prove' option #3 is data that shows slugwinder with a different racc% than ranged racc % (at any tp).

Technically, options 1 and 2 can be condensed into 1 option, since without access to the program, I don't think anyone could differentiate the data to 'prove' one from the other.

It boils down to: tp affects the WS accuracy or it does not. One thing we can say from data you have is that if tp affects the WS accuracy, then you must have on average had the correct amount of tp where the penalty was equal to the bonus.

[Spider-Dan]

It boils down to: tp affects the WS accuracy or it does not.

Without meaning to sound simplistic, it seems to me that this is a foregone conclusion; slugwinder is dramatically worse at 100 TP than at 200 TP, and this doesn't take very much testing at all to confirm (e.g. have SAM/RNG try it in dynamis).

One thing we can say from data you have is that if tp affects the WS accuracy, then you must have on average had the correct amount of tp where the penalty was equal to the bonus.

It's really too bad I no longer have the data, as (with a lot of time invested) I would actually be able to tell you what my average TP per slug was (it's part of my macro and that data was all pulled from logs).

Note to moderators: this thread will be necro'd (with good cause) if/when I decide to pay for some data recovery.

[Clave]

0 aftermath overlaps unless I am 2 houring.

Just wanted to remind you all that Relic aftermath is active during the weapon skill. This is why I get the Store TP effect of Kaiten on every Kaiten regardless of whether I Kaiten while the aftermath effect is still on from the previous Kaiten.

Namas does get more R.Acc than other WS due to Aftermath. Whether there is an additional R.acc bonus due to the single WS status is another matter and by all means continue to test that. But first, the aftermath r.acc needs to be tested.

Edit: For comparison's sake, Randgrith gives +20 Melee Acc to the user. I wonder if Namas is similar? Though it seems like it may even be more potent. Then again that may be to make up for the -40 eva on the mob.

[Drace]

Just wanted to remind you all that Relic aftermath is active during the weapon skill. This is why I get the Store TP effect of Kaiten on every Kaiten regardless of whether I Kaiten while the aftermath effect is still on from the previous Kaiten.

Namas does get more R.Acc than other WS due to Aftermath. Whether there is an additional R.acc bonus due to the single WS status is another matter and by all means continue to test that. But first, the aftermath r.acc needs to be tested.

Edit: For comparison's sake, Randgrith gives +20 Melee Acc to the user. I wonder if Namas is similar? Though it seems like it may even be more potent. Then again that may be to make up for the -40 eva on the mob.

That would definitely explain why Namas accuracy is an anomaly then - never really paid attention to it on Kaiten, but I'm sure if you have tested it then that would be the case.

That would make Namas and Randgirth the only 2 relic WS's that really have a self boost (I would estimate Namas accuracy boost is considerably more than 20), and I'm guessing this would hold true for all the mythic ws's as well.

Hopefully between the data on regular ranged WS accuracy and Namas accuracy we can identify any natural accuracy boosts they might have as well as the boost from Namas alone.

[Vodou]

Just wanted to remind you all that Relic aftermath is active during the weapon skill. This is why I get the Store TP effect of Kaiten on every Kaiten regardless of whether I Kaiten while the aftermath effect is still on from the previous Kaiten.

Namas does get more R.Acc than other WS due to Aftermath. Whether there is an additional R.acc bonus due to the single WS status is another matter and by all means continue to test that. But first, the aftermath r.acc needs to be tested.

Edit: For comparison's sake, Randgrith gives +20 Melee Acc to the user. I wonder if Namas is similar? Though it seems like it may even be more potent. Then again that may be to make up for the -40 eva on the mob.

I hear what you're saying. We know Kaiten does and the fact that you gain the aftemath whether it lands or not is strongly indicative (but not proof). That's why the split shot data was so interesting since it does show a clear +racc bonus. Afaik, no one has actually shown a test of the racc bonus on Namas that indicates the total bonus while uncapped. (My data was capped.) You raise a very interesting point which I did not perceive.

The total bonus on Namas is at least +50 (but could include both a relic bonus and a ranged WS bonus.) The bonus on split shot is simply a ranged WS bonus. It was at least somewhere between 9-33 at levels 22-40.

Does Namas get a bonus and no other ranged WS get one? Clearly no.

Does Namas get a bigger bonus or an added bonus over other ranged WS? Quite possibly yes. How much has not been quantified.

I think this is a more accurate way to say it then.

[Vodou]

Without meaning to sound simplistic, it seems to me that this is a foregone conclusion; slugwinder is dramatically worse at 100 TP than at 200 TP, and this doesn't take very much testing at all to confirm (e.g. have SAM/RNG try it in dynamis).

I know! I honestly edited this a few times to make it not sound simplistic, but failed. I think, my ignorance, is on what tests have been done to show what slugshot's penalty is at varying levels of tp.

Here are the 3 options I was mentioning:

1) slugshot has no penalty at any level of tp nor does it have any bonus.

2) the penalty and bonus exactly offset each other at all levels of tp

3) the penalty and bonus do not offset each other at all levels of tp, but do so at certain levels of tp (equilibrium point)


What I'm getting at maybe has been tested. If slug has a different racc penalty at different levels of tp, then options 1) and 2) are wrong. It very well could be that there is a range of TP (e.g., 150 or more, 140-200, etc) where data looks like options 1) and 2).

I think you're saying that at 100tp its noticeably worse, indicating that option 3) is likely and that there is a 'sweetspot' people aim for in tp where it gives the same results as all 3 options. Sounds good to me. Testing slug at 100tp as you suggested could clear that up.

BTW I'll also be adding to this thread when I get more test data completed.

[Spider-Dan]

Reading an old thread, I found an interesting post:

Both Gun and Bow relic WS have fixed hate. Namas Arrow is something like 160 CE / 480 VE while Coronach is 80 CE / 240 VE (the exact same as Hojo:Ni, which was used in the example above). Coronach as the addition enmity down stat aftermath which I don't have numbers for.

If, as Clave said, aftermath is active during WS, then wouldn't that seem to indicate that (presuming that Coronach and Namas have the same base VE/CE, which seems plausible) Coronach's Aftermath is capped -enmity effect (Enmity-50)?

[Drace]

Somewhat related (but still slightly off the recent topic) to that is that you will still get full hate from the skillchain damage. May sound pretty obvious, but I've never seen anything posted on it one way or the other. May not be as much of a problem from Coronach, but for Namas it is pretty noticeable on self chains/ closing chains from other openers.

[Vodou]

Somewhat related (but still slightly off the recent topic) to that is that you will still get full hate from the skillchain damage. May sound pretty obvious, but I've never seen anything posted on it one way or the other. May not be as much of a problem from Coronach, but for Namas it is pretty noticeable on self chains/ closing chains from other openers.

I would assume as you do that you get full hate on Yoichi on everything except Namas itself. I doubt that the skillchain damage has been tested with regards to Coronach. It's an interesting question.