Announcement

**Caspian** · 05-16-2006, 08:37 PM

Re: Strange Subjob Testing.

Isn't there a +5 atk earring for /drg too? That might help in between SATA. I have to say if anyone is going to take advantage of the 5% haste in one battle it could be thf. You still have to swing 20 times in one battle to get the extra hit, but if anyone can do it, it would be thf. For that matter, how fast were your mobs dieing?
I think /WAR definately has more merit now than before, I know berserk isn't permanent, but it is 60% of the time.
I still think when it comes to fast mobs (birds, raptors, and couerls) that /nin just wins. You'll just get raped trying to get back to camp after pulling.
Good luck with further testing though. Interested to see what else you find out.

**Spinnthrift** · 05-16-2006, 08:53 PM

Re: Strange Subjob Testing.

No +5 attack earring (that's for /Drk iirc), however - there is a +6 attack back piece, as well as a +1% haste shield. Not that we use shields much.

Using /Drg you can hit a maximum of +16% haste in total, which beats out Dual Wield II, with 0 tp loss, however, you'd lose some potential attack with using the Wyvern Shield, as I think the Viking Shield would be an informed choice of shields (leaving the job at +15% haste still).

For a good number of levels however Wyvern Mantle cannot be beaten as a back piece - with the exception of the Resentment Cape (for hate management) until level 60 by the Behemoth Mantle +1 (nice if you can get it

).

When I get to post 60, I'll also have Acc Up I - which will help with DE (I'm sure).

I definately agree with Bird/Raptor and /Nin - however, many parties that I've pulled in as Thf, Utsusemi wore after time more often than not, and my evasion has been left sorely underlevelled.

Only thing I'm left wondering about - is how would High Jump work with SATA? Would it lower my hate or the person I tricked onto? If it lowers the person being tricked onto - this could definately be used as a rescue move, if not - it's relatively pointless as Thf is rarely high in the hate queue.

**Armando** · 05-16-2006, 09:12 PM

Re: Strange Subjob Testing.

I'd like to point out that with a Hornetneedle, you still gain TP faster than usual as /NIN. Also, WAR provides Attack Bonus all the same, Double Attack is better than Jump with daggers, and Berserk is too good to pass up. I'll write more tomorrow, because it's late. But both /NIN and /WAR have the upper hand on /DRG.

Oh, and where are you getting 15% Haste from /DRG? The earring is 5%, so I don't see where the other 10% is coming from. If it's from equipment, you can't attribute that to /DRG because it's also available to /WAR and /NIN.

**Spinnthrift** · 05-17-2006, 09:37 AM

Re: Strange Subjob Testing.

Originally posted by Armando

I'd like to point out that with a Hornetneedle, you still gain TP faster than usual as /NIN. Also, WAR provides Attack Bonus all the same, Double Attack is better than Jump with daggers, and Berserk is too good to pass up. I'll write more tomorrow, because it's late. But both /NIN and /WAR have the upper hand on /DRG.

Firstly - without meaning to be condescending regarding the TP gain - you're wrong. Recently - the TP Floor was removed, meaning that the TP per swing no longer returned a minimum of 5 tp per swing. So - when using a dagger in your main hand at 150 delay, you now return 4.5 tp per swing, instead of 5. That in itself is a 10% reduction in tp gain over previous benchmarks.

With Dual Wield I, you knock off a further 10% tp gain when dualling and a further 5% off that wield Dual Wield II, so a Corsairs + Hornetneedle (post 50) will give you a combined delay of 206 + 150 = 356 * 0.85 = 302 delay/2 = 151 delay per swing, near as can be 4.5 tp per hit, or 9tp per swing. Further adding in Suppanommi (sp?) and/or Sarashi further reduces this down. There is no more exploiting the TP floor for quick WS gains.

A singular Bone Knife +1 will still return 5.23% tp a swing.
A lone Corsairs Knife will still return 5.61% tp a swing.

... Regardless of how much +haste% you add.

So - a singular Knife can gain more tp per swing and give greater DoT, instead of wielding a weaker offhand weapon and feeding the mob tp (which nets you less tp than it did previously)

Yes - in the last stages of the game - Thf/Nin using Suppa, a Sirroco or Mercurial Kris offhand with fully merited Triple Attack Up - is still the best way to gain TP, *however* - getting to this point takes levelling up - a lot more than just getting to 75. However - sadly, in the last stages of the game, Thief is also relegated to TH whore who sits on the sidelines till the last 3% of the mobs health and is swapped in.

For much of the game, Hornetneedle isn't the best TP gaining tool now.

So - a month ago, you were right in your thinking. However - the TP gain system has been tweaked and your facts are no longer up to date.

Originally posted by Armando

Oh, and where are you getting 15% Haste from /DRG? The earring is 5%, so I don't see where the other 10% is coming from. If it's from equipment, you can't attribute that to /DRG because it's also available to /WAR and /NIN.

Granted - I was displaying a theoretical 16%, 5% Earring, 1% Shield (already mentioned ftw), Then Reparee Harness 4% Haste, Bravo Subligar 2% Haste, and Swift Belt 4% Haste.

Now - assuming at lvl 50, you wish to match the same base attack with /Nin or /War, Sword Belt would be a *consideration*, to put in this piece instead of the Swift Belt (losing 4% haste and additional accuracy to match the base DoT), since neither sub offers an Attack Up bonus at this point. At 60, to replicate the Accuracy Up bonus and Attack up bonus from /Drg alone with /Nin you need to use at least two pieces of gear. War only needs one.

Secondly - with Ninja, to not hit 0's a lot, you need to invest a lot of gear into Attack/Str (where possible), this is less of a consideration with /War, but Accuracy now plays a part.

Pre 50, Drg potentially outweighs War by a considerable margin in sheer terms of damage over time (I'm yet to test it as a comparison - I'll try and get a friend to come heal me in Phominua Aquaducts this weekend), post 50 - I don't know. Pre 50 - the tp gain from /Nin isn't quite as noticable as it is post 50 - this is known, by many Thf players who either weren't close to abusing the TP floor *or* weren't paying attention.

I'll be setting up some tests when I get a parser in Promyvion/Phominua to compare 30 and 40 caps respectively, and plan to be as objective as possible, listing full gear setups for DoT tests, Race to 100 tp tests - and when I finally hit 50 Thf, Double Attack vs Haste comparisons. I do not deny the potential for /War to beat out /Drg on a double attack proc, however - since I've never *seen* anyone try this before (and I've searched for it) - I can only surmise that /Drg has been largely ignored as a potential subjob choice.

The above notion - that Dragoon has been ignored as a sub is supported by the 75 Mnk/Drg parsers which showed it clearly outparsing /War - that was met with ridicule and disbelief on Allak.

I will run some tests and give some hard data to be analysed, down to gear used and mobs fought - and I'll try and stick to neutral mobs to give a clear picture.

Expect some data after the weekend.

- Saeriel

**Armando** · 05-17-2006, 10:48 AM

Re: Strange Subjob Testing.

Oy...this is going to be an annoyingly long post (very same reason I made it so short last night, it was already past 12 AM and I wanted sleep.) There will be a LOT of calculating and mathematical analysis going on, so if you don't like it, just take my word for it. Also, I'm going to try to be as blunt as possible to save time; that bluntness can easily be misinterpretted as condescendention, arrogance, or just plain meanness. Try not to interpret it that way, it's nothing personal, I just want to make this brief and concise.

I'm right about the TP. Hornetneedle is still one of the best TP-gaining tools because of its low Delay. It's true that the TP floor was removed, and that TP now scales with Delay even below 180, but that does NOT mean it scales down at the same rate as the rest of the Delays. There are multiple TP formulas for the multiple Delay ranges, and S-E deliberately made TP scale down slowly for Delays under 180. The end result is, in short, that you get faster while remaining in the same TP "bracket," thus gaining TP faster. What I mean by "TP bracket" is that while TP is gradual, the number of hits required to reach 100 TP is not - therefore, there are certain ranges of TP in which it'll still take you the same number of hits to reach 100 TP. Back to the math...*sigh*

TP formula for Delays below 180:
5.0 + [(Delay - 180) * 1.5] / 180

Weapon setup
Delay calculations (for Dual Wield)
TP per hit
# of hits (# of attack rounds, when Dual Wielding) required to reach 100 TP
Ammount of time required to perform that many attack rounds / Prepatch time

180 Delay Dagger
5 TP
20 hits
60 secs/60 secs

180 Delay Dagger x2, DWI
(180 + 180)/2 * 0.90 = 171
4.9 TP
21 hits (11 attack rounds)
62.7 secs/57 secs

Bone Knife +1 x2, DWII
(190 + 190)/2 * 0.85 = 161 Delay
4.8 TP
21 hits (11 attack rounds)
59.03 secs/53.67 secs

Bone Knife +1 | Hornetneedle, DWII
(190 + 150)/2 * 0.85 = 144 Delay
4.7 TP
22 hits (11 attack rounds)
52.8 secs/48 secs

Thief's Knife | Sirocco Kukri, DWII + Suppanomimi
(194 + 150)/2 * 0.80 = 137
4.6 TP
22 hits (11 attack rounds)
50.23 secs/45.67 secs

Blau Dolch | Sirocco Kukri, DWII + Suppanomimi
(178 + 150)/2 * 0.80 = 131
4.5 TP
23 hits (12 attack rounds)
52.4 secs/43.66 secs

Sirocco Kukri | Hornetneedle, DWII + Suppanomimi
(150 + 150)/2 * 0.80 = 120
4.5 TP
23 hits (12 attack rounds)
48 secs/40 secs

That should be clear enough. In general, once you reach the 4.7 TP line, your TP gain hits a sudden jump, and keeps increasing as you go faster. It slows down a bit when you hit the 4.5 TP mark because the number of attack rounds increases by 1 simply because of the uneven number of hits; however, if you go faster within the 4.5 TP range, you still get TP faster than before. In general, if you go fast enough, you ARE gaining TP faster. However, you require a very low Delay weapon like Hornetneedle to make it possible. That being said, this makes Hornetneedle even more valuable, since it's now essential for faster TP gain, both when single-weilding and dual-weilding.

A very important factor is that Double Attack in itself is a (slightly different) form of Haste. In fact, it's better than Haste in the sense that it can improve WS damage. Double Attack's 10% proc rate translates roughly into 10% virtual haste.

Let's take a standard 180 Delay dagger and a three minute time range for the example.

Without Double Attack: 60 hits per three minutes.
With 10% haste: 66.66... hits per three minutes.
With Double Attack: 60 +6 hits per three minutes.

So, post-level 50, /WAR sub gains a virtual 10% haste on average. Now let's analyze Jump. Once we exclude the very first Jump of the party (which is free from any delays because of the timer already being at 0:00,) we're down to two Jumps a minute. Let's also throw in the Wyvern Earring, and a three minute time range. Again, we'll work with a 180 Delay dagger.

Normally: 60 hits per three minutes.
With 5% Haste: 63.16 hits per three minutes.
With 5% Haste and Jump: 63.16 + 2 hits per three minutes.

65.16/60 = 8.6% increase in attacking speed. However, if you add more Haste gear, the inherent increase in attack speed from Jump diminishes even further (because it's on a set timer, it'll be an even smaller fraction out of your total hits) whereas Double Attack's 10% "haste" never loses value. Of course, you can't melee for a straight 3 minutes, so the overall speed increase with Jump taken into account is somewhat higher. Let's say Double Attack and Wyvern Earring + Jump are close enough.

In terms of TP gain, it's fairly close. Now let's consider the other aspects. /NIN provides Utsusemi as a safety net, the Level 15 line of elemental Ninjutsu can be used to reduce the mob's resistance and thus help maximize skillchain damage, Dual Wield provides a 15% increase in speed, and the stats on the second dagger can help Accuracy/Attack a bit. /WAR provides Double Attack, which can be seen as 10% Haste, however, it can also apply to WS; /WAR also provides Attack Bonus, and Berserk, which is one of the best melee buffs in the game (25% Attack up/Defense Down 3 out of 5 minutes, as you know.) /DRG offers the same Attack Bonus as /WAR, as well as some Haste and Jump, which can bring its overall attacking speed on the same level as /WAR, but nothing more.

/NIN and /WAR can provide the same increase in attack speed and TP gain as /DRG while also providing meaningful ways to increase damage, whereas /DRG offers no meaningful additional bonuses.

/DRG becomes arguably more useful at 60 with Accuracy Bonus (Accuracy +10) and High Jump (more TP). Then I could see it as a viable alternative. However, in the 50-59 range, it is my opinion that /WAR and /NIN offer more.

**Spinnthrift** · 05-17-2006, 05:15 PM

Re: Strange Subjob Testing.

Having gone and tested this for myself - I am glad at least that my understanding has been corrected. So, my apologies - from what I've been told/read - the floor was gone completely, not set at 4.5 tp a swing.

However. I'm going to remind you - politely, that you're set in the "end-game" mentality, where what happens at lvl 75 merited - isn't always the same as what happens before then, never forget that.

I stated in the opening post - that it was a test... I also highlighted the things I was disappointed with (I've coloured the part in bright red where I said: "OMFGZ!!! THIS PWNZORZ NIN/WAR/MNK/BLM/AND ALL YOUR SUBS!!!1!ONE!).

My goal - if there is one, is to see whether it's possible to create a build where I can have some advantage over a similarly costed build of either /War or /Nin - the two standard subjobs for Thf. Not that you'd have known, but I've not exactly skimped on seeing if my whim worked or not.

Please - reread the first post again. I tested a sub that no one uses, found some pro's and some con's. Highlighted both. Stated my disappointment with some of what I'd found. I realise you've played Thf up to high levels, and have felt how disappointing it is to produce consistently worse and worse performance as other jobs close the gap on your previous benchmarks. So you'll forgive me if I spend my time and gil trying to eek a few more damage out of my job in a party. I'm also going to apologise that my trying to do a better job as a Thief offended your sensibilities of having the standard unwavering sub job that is always shown as the norm.

I can concede when I'm wrong, and I'm glad my knowledge has something new to work with, however - I'm still going to test things to destruction. That's my way. That's how I play now and always have. I don't follow what I'm told is the best thing to do (hence why I got my daggers out and went to work and went.." oh.. he's right, better go say sorry"), until I've worked it out for myself, and sometimes - it does agree with the "norm" and sometimes it doesn't *.

You can rebuke my playing around with this job all you want from now on. I'll do my tests and post them. Regardless of results. I did not state in the opening post that this is the sub to use from now on forever for all Thieves, nor did I deny the qualities of other estabilished subs, purely making notation on where /Drg would need to stand up to be considered a viable choice. I'm not answerable to you, and I don't want to get baited into defending something that I've not finished testing (after having already admitted some of the downsides found in my initial playing session).

- Saeriel.

* In Guild Wars - I was the creator of the Ele/Nec tank build, which was used repeatedly as a counter to the War/Mnk epidemic used by the major Korean Guilds at the time. Having a mage tank two melees to a standstill was not considered viable until it was proved repeatedly in the HoH. Actually it was considered the biggest joke I could play until I held two War/Mnk for a minute while my team finished up the other side. Then they realised that the combo was a serious one.

**seq** · 06-05-2006, 04:35 AM

Re: Strange Subjob Testing.

Armando, THF/DRG have always worked for me too from LV30-45
simple paper calculations can NEVER say if /NIN or /WAR or /DRG is better, unless you test by yourself.
who tested and said Double attack is 10%? it RARLY goes off on my DRG/WAR (Unless i have haste on etc) it's less than 10%.
/WAR gives:
20: Defense bonus1
30: Berserk
50:Double attack
60:Attack bonus1
70:War cry

/DRG gives:
20:Attack bonus1 & Jump (/WAR gives Attack bonus1 at 60, and it gives Double attack at 50 which /DRG gave at 20?)
30:Wyvern earring ,Haste+5% and wyvern targe Haste+1%... total Haste+6%
60:Accuracy bonus (helps in DE ofcourse more than Double attack or Berserk, and DRG gives more CHR which increases DE's DMG, right?)
70:High jump (Another way to get TP)

i'm not mentioning /NIN since Dual weild and hitting 0DMG never gives TP anymore. /NIN will cast Utsu,NIN debuffs as you said... all of this takes time to cast.

it's all about the player's playing style. you cannot just go out and say "THF/DRG is bad and you should /NIN or /WAR.".

180 Delay Dagger x2, DWI
(180 + 180)/2 * 0.90 = 171
4.9 TP
21 hits (11 attack rounds)
62.7 secs/57 secs

Bone Knife +1 x2, DWII
(190 + 190)/2 * 0.85 = 161 Delay
4.8 TP
21 hits (11 attack rounds)
59.03 secs/53.67 secs

Bone Knife +1 | Hornetneedle, DWII
(190 + 150)/2 * 0.85 = 144 Delay
4.7 TP
22 hits (11 attack rounds)
52.8 secs/48 secs

Thief's Knife | Sirocco Kukri, DWII + Suppanomimi
(194 + 150)/2 * 0.80 = 137
4.6 TP
22 hits (11 attack rounds)
50.23 secs/45.67 secs

Blau Dolch | Sirocco Kukri, DWII + Suppanomimi
(178 + 150)/2 * 0.80 = 131
4.5 TP
23 hits (12 attack rounds)
52.4 secs/43.66 secs

Sirocco Kukri | Hornetneedle, DWII + Suppanomimi
(150 + 150)/2 * 0.80 = 120
4.5 TP
23 hits (12 attack rounds)
48 secs/40 secs

most of this daggers will hit 0DMG, right?

**Spider-Dan** · 06-06-2006, 12:59 AM

Re: Strange Subjob Testing.

If THF/NIN is regularly hitting for 0 damage, THF/DRG (with 10 extra ATK) is going to be hitting for 0s too.

In any case, if you are a THF/anySJ and you are regularly hitting for 0 (excluding mob defense buffs that are waiting to be dispelled), your problems go far beyond your SJ.

**Spider-Dan** · 06-06-2006, 02:37 AM

Re: Strange Subjob Testing.

Here's the most interesting thing I've gotten from this thread:

Originally posted by Armando

Thief's Knife | Sirocco Kukri, DWII + Suppanomimi
(194 + 150)/2 * 0.80 = 137
4.6 TP
22 hits (11 attack rounds)
50.23 secs/45.67 secs

Blau Dolch | Sirocco Kukri, DWII + Suppanomimi
(178 + 150)/2 * 0.80 = 131
4.5 TP
23 hits (12 attack rounds)
52.4 secs/43.66 secs

When DWing with a Suppanomimi, what is the exact delay threshold for the jump from 22 hits needed to 23 hits needed (to reach 100%)? And where are you getting the new TP calculation formula from?

**Armando** · 06-06-2006, 09:31 AM

Re: Strange Subjob Testing.

Originally posted by Spider-Dan

When DWing with a Suppanomimi, what is the exact delay threshold for the jump from 22 hits needed to 23 hits needed (to reach 100%)? And where are you getting the new TP calculation formula from?

Assuming the formulas are 100% accurate (they can be off every now and then at the borderlines where the tenths place changes), 131 (average, reduced) Delay is where TP changes to 4.5 and you need 23 hits. 132 Delay and above should be 4.6 TP. I got the new formulas from Studio Gobli. You'll have to scroll down a bit.

Originally posted by Studio Gobli

â– æ–°TPè¨ˆç®—ï¼ˆæš«å®šï¼‰
___ï½ž180ã¾ã§ã€€5.0+(éš”-180)*1.5/180
180ï½ž450ã¾ã§ã€€5.0+(éš”-180)*6.5/270
450ï½ž480ã¾ã§ã€€11.5+(éš”-450)*1.5/30
480ï½ž530ã¾ã§ã€€13.0+(éš”-480)*1.5/50
530ï½ž999ã¾ã§ã€€14.5+(éš”-530)*3.5/470

Originally posted by seg

Armando, THF/DRG have always worked for me too from LV30-45
simple paper calculations can NEVER say if /NIN or /WAR or /DRG is better, unless you test by yourself.

I never said it doesn't work, or that it's gimp. I said other subs probably offer more. Calculations can't always reflect reality completely, but they can come pretty damned close.

Originally posted by seg

who tested and said Double attack is 10%? it RARLY goes off on my DRG/WAR (Unless i have haste on etc) it's less than 10%.

Pretty sure it's 10%. Lances are slow, it takes 8 hits to reach 100 TP with them, only 7 after a WS, and you'll Double Attack 1 out of 10 times. Thus, the illusion that you don't DA often.

Originally posted by seg

/DRG gives:
20:Attack bonus1 & Jump (/WAR gives Attack bonus1 at 60, and it gives Double attack at 50 which /DRG gave at 20?)

Jump alone (in this scenario) is nowhere near Double Attack. Using a 172 Delay dagger, you get 32 attacks instead of 31 every 1:30 minutes, which translates roughly into a 3.23% "Double Attack" proc rate. Double Attack is three times as good as Jump when using daggers. /DRG needs the Haste too to make it worth it.

Originally posted by seg

60:Accuracy bonus (helps in DE ofcourse more than Double attack or Berserk, and DRG gives more CHR which increases DE's DMG, right?)

Oh, come on. If there's a CHR difference, it's most likely 1 or 2 points. You're not going to see a difference in damage.

I know that despite my claims it seems I'm downplaying /DRG again. That's not it. I just like to put things in perspective.

**balvenie** · 06-06-2006, 09:37 AM

Re: Strange Subjob Testing.

This is a great post, thanks for all the work you guys have done. All of the data is well laid out and I don't see anyone claiming not to use one sub or another. I've always felt your sub should be based on the rest of the party and what you are fighting.

I am in a static of WAR, MNK, DRG, SMN, WHM, THF (me). Due to the lack of an MP refresh and any sort of debuffing, we've had to be creative in our subs. I subbed /RNG for the longest time simply to land more acid bolts on crabs when we had to level on those. /NIN (at that time) would have been better for TP, but at the expense of all of the melee hitting for much less.

The best subjob, is the one that gives the most EXP/hr.

Announcement

Strange Subjob Testing.

Strange Subjob Testing.

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment