Forum Settings
       
Reply To Thread

Viability of 5.0 tp/hitFollow

#1 Oct 01 2009 at 9:22 AM Rating: Good
Sage
***
1,281 posts
I am posting mainly because I am bored, and the forum hasn't been very active lately. So, I have been thinking about sTP and how I would love to have Enkidu's Subligar or Skadi's Chausses, but I have also been thinking about the viability of a build around sTP for thf.

What are your thoughts on this?

Thinking about it, 5.0 tp/hit gives 20 hits to 100 tp, meaning theoretically, that you can WS every 10 rounds (or WS + 9 rounds if you are being technical), however, requiring 20 hits to 100 tp, with a max hit rate of 95% means that on average you will miss one of your hits to 100, leaving you at 95% tp. This is not too much of an issue if you happen to get a 15% tp return on DE or Evis, but for normal situations it wouldn't actually lower your # of rounds to get 100 tp.

What about the viability of a build that nets 5.3 tp/hit. I can't think of one off hand for blau/sk, but maybe that is why X's/blau works so well against blau/sk in practice. 5.3 tp/hit nets 100 tp in 19 hits, not 20, so ideally, in 10 melee rounds, you get the requisite TP for your WS and essentially add the equivalent of ~10% haste to your build.

What do other thieves think of this?

Is it possible to boost blau/sk enough to net 5.3 th/hit?

Is it viable to do so?
#2 Oct 01 2009 at 9:35 AM Rating: Decent
***
1,067 posts
stp is good if you dont have to give up anything to get it. its way too random on nin or /nin for it to be really effective though. single handed double/triple attacks as well as single handed misses mean you cant guarantee it will help. one attack round you may only get 5tp, the next you may get 15, the next you may get 10, then you may get 20.

dual weilding staggers so much that you probably have like a 50/50 chance (pulled that out of my ass) that it will reduce one attack round.
#3 Oct 01 2009 at 10:43 AM Rating: Good
Sage
***
1,281 posts
That is actually my point iknoweverything. We make such a huge deal about getting blau/sk up to 5.0 tp/hit, but realistically there is no way for us to average out to less rounds of TP with that number. 2handers have it fairly easy because they are dealing with a number of hits less than their projected accuracy, and if they miss a hit, it doesn't throw their whole build off.

I think if we are talking averages though, if you were to get up to 5.3 tp/hit, it would reasonably increase damage output due to truly lowering the feasible number of rounds to 100 tp. That could account for why X's/blau loses on paper to blau/sk, but doesn't lose in practice when it is parsed. Was just thinking about it, and thought I would bring it up.
#4 Oct 01 2009 at 3:44 PM Rating: Excellent
***
2,270 posts
Randomness in TP gain =/= STP useless. It still is more TP/Time no matter how you swing it. If you are at 90 TP and do a 6 hit round, sure it didnt do anything THAT time, but what about those times you DO land at 100-101 etc. You do gain from those. Its not as clear cut a bonus as 2handers, but its still an improvement.

But, i gotta side with
Quote:
stp is good if you dont have to give up anything to get it.

Skadis is hawt (as you already know) because its on par with homam WITHOUT the STP. So its homam...+TP. I wouldnt TP in a non-haste leg like enkidu if I have an option like homam though. Id leave enkidu for WS/SA and Rngs going for an Xhit build.

Also, 19 hit isnt "better" than 20 hit because you have a 5% chance to miss 1/20. Its better because (assuming the same delay) you get more TP over time than a 20 hit. Sometimes it wont land RIGHT on 100 and overshoot by 5-10 and it feels wasted, but hey, at least you still get another couple points of acc on DE, % or 2 on Evis, 0.XX FTP on SB and such. Getting more TP is ALWAYS better than getting less even if it doesnt make your WS count skyrocket to obseen levels. WHen it lines up, you will be greatful. When it doesnt, its still adding to your damage in the background.

(I doubt seriously that a 5.3 build on B/S is even in the same ball park as a normal haste build with skadi 5.0)
____________________________
Aanalaty, Mithra of San d'Oria
99Thf/99Drg/99Rng/95Pld/Secret Job 92
Bonecraft-100+3 7/29/08
Lu Shangs-5/28/08
Twashtar-90
Ryunohige-75: 1/7/2012
Mandau-75: 3/9/2012

Maat-1/1 Thf Meleed (pansies steal)
Some of mah gearz: http://www.ffxiah.com/player/Lakshmi/Aanalaty#item-sets
Join 'People Against The Heart Snatcher' and help put Thfs on the right P.A.T.H.S. today!
#5 Oct 01 2009 at 8:15 PM Rating: Default
Scholar
**
485 posts
iknoweverything wrote:
Quote:
stp is good if you dont have to give up anything to get it.

I think we all agree here. So lets build on this.

Lets list some proper tp gear for thf with store-tp on it:
Rajas Ring +5
Brutal Earring +1
Skadi's Chausses +7
---------------------
13 store TP

I think most thf agree that the above is good gear to tp in.


Now lets add some less conventional store-tp gear/food.
Rajas Ring +5
Brutal Earring +1
Skadi's Chausses +7
Ecphoria Ring +1
Carbonara +6

---------------------
20 store TP

I put in Ecphoria Ring (4 acc) where you normally would use Toreador's Ring or Sniper +1 (7 acc). So giving up 3 accuracy for +1 store tp.
I listed Carbonara where you normally would use Yellow curry buns or Pizza +1.

Yellow Curry Bun: (Cap: 75@375 Base Attack) STR+5 AGI+2
hidden effect : attack caps at 85 when eaten in a party of 4 or more.
Lets assume 85 attack.
Carbonara: (Cap: 65 @ 361 Base Attack) STR+4
So giving up 20 attack STR+1 AGI+2 for +6 store tp.

Pizza+1 Attack (Cap: 55) Accuracy (Cap: 44+)
Giving up 44 acc but gaining 10 attack +6 store tp STR+4. Ouch! Giving up 44 acc alone is pretty big in itself. But lets continue for the sake of argument.

And since I am being silly, lets toss in Vajra also. A dagger that benefits more from extra tp/time then other daggers. Because at 300% you unlock the Aftermath effect: Occasionally attacks twice. I do this just to favour/justify the store TP build.

Vajra delay 200 - TP Per Hit: 5.5%
X's knife delay 201 - TP Per Hit: 5.5%

Going with X's. Since it is pointless to use Blau with Vajra. Hidden effect and all that jazz.

TP/hit from Store TP is calculated as (Base TP/hit) * (1 + (Total Store TP / 100))
11 * 1+ 20/100
11 * 1.2 = 13.2

I know it is silly to think in tp/hit. Or X-hit build for thf. You should think in tp/time. Other posters already mentioned this. I just list the tp/hit as a formality. Notice I did not even factor in Delay Reduction from suppa, mirke or /nin.

My point is: In situations where your need the acc from pizza it would be silly to go for carbonara. But giving up Toreador's Ring for Ecphoria Ring and Yellow Curry Buns for Carbonara does not seem that bad of a trade.

In any case it would be interesting to test this (20 store-tp) build just for Vajra. It might make Vajra a tad more interesting.


Edited, Oct 2nd 2009 7:06am by Breaze
#6 Oct 02 2009 at 1:13 AM Rating: Good
***
3,775 posts
First off before going into commentary on store tp

Quote:
In any case it would be interesting to test this (20 store-tp) build just for Vajra. It might make Vajra a tad more interesting.


Nothing makes Vajra appealing when you consider the amount of effort to get it as well as the fact that the cost of alexandrites rivals the Mandau in every way shape and form. Vajra will NEVER be worth upgrading instead of Mandau.
--------------------------------------

Now, with that said store tp is a difficult beast to gauge for a thief. I'm going to link to two recent posts made by banalaty that, while a tad long summarize thief tp building very accurately

Part 1
Part 2

I am in agreement with what Banalaty said in that thread. Generally speaking, more tp will mean more weaponskills. On many jobs with higher delay an X hit build is perfectly viable, but because we have such low delay and low tp/hit to start with, compounded by all the other factors Banalaty mentioned (triple attack and double with brutal being a big one) our tp building routine IS the most unstable of any job. Banalaty already said all this so all credit goes to him for his original statements, but it still warrents being mentioned again.

Now, my own thoughts on the matter are along the lines of adding store tp on pieces that are already good to begin with, similar to others. I wouldn't go out of my way to look for them elsewhere. The rajas ring, brutal earring, and skadi's chausses are easy because no other tp piece is as good as them for the slot so you would use it anyway. There is one IMPORTANT thing about store tp that you MUST remember. I'm going to bold this for emphasis.

While store tp increases tp/hr it does NOT increase non weaponskill DoT.

If you add accuracy you will increase tp gained by virtue of raising hitrate, but that comes with the added bonus of also HITTING MORE. Every additional hit on the target is more damage. If you raise store tp you do NOT get this effect. Store tp does just that... it raises your own tp on a per hit bases. Raising accuracy not only increases TP, it increases DoT DAMAGE as well. THAT is why I value accuracy and hitrate over store tp, and I don't advocate going out of the way to look for it.

The ecphoria ring is nice when it breaks a tier and I'd use it over a woodsman/snipers in that case, but the toreadors raises hitrate by 1.5% and that's better (This is why I was so happy with my augment on my own toreadors. Even with strength -1 I got dexterity + 2. The dex affects all targets with 1 more point of accuracy and the fSTR only goes down by 1 on 25% of mobs). Ecphoria won't be better than toreadors even with a broken store tp tier unless your accuracy is already capped. But this argument fails anyway because if your accuracy is capped you would be better off with a ring that increases damage directly. I only advocate swapping to a ring of this nature when you KNOW your accuracy is capped in melee, but if that IS the case that's the way to go. By this I mean a ruby/flame/triumph or blitz ring. I WAS going to advocate the blitz immediately, but I just noticed the price went through the %^$#@ ROOF in the past half a year. It was 10-30k earlier this year and now it's 500k. Given that, when you weigh the boost of fSTR +1 and 2 attack versus 1% haste worth of tp gain and DoT + weaponskills it's going to be fairly close. On a budget I'd say accuracy capped scenario a ruby is fine. For the spenders, I'd still grab blitz before a triumph/flame though.

ATM I'm too tired to think straight and comment on the matter of carbonara versus yellow curry buns/coeurl subs/marinara pizza (+1) so I'll wait till later to do that. I'd just regret saying something and edit later anyway >.>.

Edited, Oct 2nd 2009 5:17am by Melphina
____________________________
[ffxisig]56619[/ffxisig]

There are 10 kinds of people in the world. Those who understand binary, and those who don't.
#7 Oct 02 2009 at 4:09 AM Rating: Decent
Scholar
*
207 posts
With Blau / SK combo, all you need is sTP+12 to get 5TP/hit (if my calculation are correct, if not well do the math ^^)
Brutal Earring +1
Rajas Ring +5
Skadi's Chausses +7
Total = sTP+13

That should be the objectif no ?

Well now, it's true that not everyone have Skadi's (if you get interest in this build, you should have brutal & rajas still). I personally don't :(
Carbonara will help on this, but it's still 10 Attack less than Yellow Curry Bun / Coeurl Sub.
So it go down to : 20 Hit Build vs 21 Hit with 10 atk more.
Without doing the math, I'll go for the 2nd option.
____________________________
Prepare your barbecue! Tarutaru eater is back!
#8 Oct 02 2009 at 7:30 AM Rating: Good
Sage
***
1,281 posts
Wow, thanks guys.

Banalaty: That is more or less what I was thinking about x hit, but hadn't really fleshed it out much. You seem to have dedicated a lot of thought to it. Thanks for your input and thanks Melph for the links. Obviously, I still want either Skadi's or Enkidu's, but more because of the total package that they offer, and not specifically for their sTP.
#9 Oct 02 2009 at 12:38 PM Rating: Decent
****
7,094 posts
So where does Cobra Unit Harness (STP+6, ACC+11) fit into all of this?
____________________________
Terraxia - RNG/COR/THF - Midgardsormr/Quetzalcoatl (boxed)
Viper Beam - ARC/CRP - Fabul
#10 Oct 02 2009 at 3:38 PM Rating: Good
Scholar
****
6,580 posts
TP/hour is equally bad a model as rounds/100TP for Thief for the simple fact that all TP is lost when you WS.

"TP/hour" is so popular in this forum but I really don't see why. I've argued it befor and I'll argue it again: It is just not a good or accurate model to use.

You all argue about "outside factors that make stp useless", but you can not deny that having 10TP/round rather than 9.whatever will over time yield more weaponskills. I don't know why you're all so against it. It may not be properly quantifiable, but it is very obvious if you try it (don't need skadi's. Enkidu's gives enough STP with blau/SK)


I will say again exactly what I said last time this was brought up: TP/round/rounds to 100TP/STP and X-hit considerations may not play out exactly due to missed hits/dual wielding and triple attack, but over time it will be largely accurate. TP/hour will never be an accurate model.

You are favouring a system that is occasionally right, over one that is occasionally wrong.
____________________________
[HTID]
DRG BRD TH4 RNG RDM75
Dragoon Equipment Guide
ArsDraconis wrote:
Ultima spends 10 whole seconds shouting "I'MA FIRIN MAH LAZER" like a 2004 valkurm melee waiting for SC
#11 Oct 02 2009 at 4:51 PM Rating: Good
***
3,775 posts
I'm not going to try and debate too heavily on tp/hr versus rounds to 100% like the other thread because in all of honesty, both models are faulty. My earlier post was made in the morning and as I mentioned I was tired so I held off on commenting too much beyond that. I'll add a little more general input now though.

While I did say that I was in agreement with Banalaty on tp/hr, I'm going to say that Noodles is equally in the right. When you add in all the factors that Banalaty said most will go both ways and favor both models. As an example if you consider getting struck in battle as a tp source that can happen more than once or not at all in any given round, and random numbers can and will throw out 100% tp with both scenarios of tp/hr and rounds to 100 with this. However there is one thing that I disagree with in your comment noodles

Quote:
TP/hour is equally bad a model as rounds/100TP for Thief for the simple fact that all TP is lost when you WS.


Depending on how many hits land and which weaponskill you use, you will always have tp immediately after you weaponskill. THIS is the biggest setback to creating an accurate model for thief tp and the cause for the debate. Whereas a samurai with a 6 hit build has such huge tp gain per hit they have very stable tp building routines. If we use multi hit we can end up with a good chunk of tp, or a much smaller amount. If we use shark bite/mandalic stab that's going to be less, but even then it can vary. This "starting" tp is the cause for the current debate, but I WILL agree that while this works flawlessly for slow weapon/2 hand jobs, it is NOT a perfect representation for a thief. The one and only thing I can say is that building tp on thief is erratic. However

Quote:
You all argue about "outside factors that make stp useless", but you can not deny that having 10TP/round rather than 9.whatever will over time yield more weaponskills.


This I will agree with Noodles on. Store tp is a GOOD thing, make no mistake. There ARE critical points that store tp will become worth a lot more, and the 5.0 mark is definitely one of them. I have skadi's, brutal and rajas and I'm extremely happy with my tp gain. The OP however spoke on the matter of other store tp pieces, and that means choosing between other alternatives. There are only a few pieces that are non standard that that applies to then.

Coobra unit gear -- only the body is a contender against better alternatives, but homam/homam +1 mirke/enkidu and probably skadi's are still going to be better choices (I say skadi's because of the 5 attack and the consideration that 8 dex will probably raise crit rate by at least a percent or two). That leaves

Ecphoria ring --- I already touched upon this
Chiv Chain -- This is pretty well known. But I have love torque so that's better
Sphaghetti carbonara -- That's the only thing I haven't tried to break down, but with my recent revelation that yellow curry buns cost the same as coeurl subs, but the hidden effect raises the attack cap to 85 while used in a party of 4+ (and that it has 2 agility on it for whatever that's worth) while carbonara MAY be better than coeurl subs, it's definitely NOT better than yellow curry buns. There is no question on that.

Other than that, there aren't too many pieces to look at. Barring a store tp FoV augment (is this possible) despite an inaccurate model of tp gain, store tp is pretty easy to figure out what to look for and when and why.... simply because the choices are either no brainers or so few in number.

Edited, Oct 2nd 2009 9:01pm by Melphina
____________________________
[ffxisig]56619[/ffxisig]

There are 10 kinds of people in the world. Those who understand binary, and those who don't.
#12 Oct 02 2009 at 6:10 PM Rating: Decent
Scholar
****
6,580 posts
If over time you get a mode average of 102TP generated per WS, and for simplicity's sake we will say 102TP every 30 seconds, then you are getting 12240tp/hour, that's 122 WSs! No it's not. That 2TP over 100 doesn't add up. It will be 120WSs. Whatever variable that was changed to increase your TP/hr from 12000 to 12240 appears to be an increase of 240TP/hr. But as I mentioned: It's not. that 240TP gained does absolutely nothing.

That is what I was getting at when I said "all TP is lost when you WS".
____________________________
[HTID]
DRG BRD TH4 RNG RDM75
Dragoon Equipment Guide
ArsDraconis wrote:
Ultima spends 10 whole seconds shouting "I'MA FIRIN MAH LAZER" like a 2004 valkurm melee waiting for SC
#13 Oct 02 2009 at 6:54 PM Rating: Good
***
3,775 posts
Ahh, so by that you meant "tp over 100% is lost". I gotchya.

Honestly that's another factor that's hard to pin down. Even when I have 100% tp I don't weaponskill immediately because of human reflexes, or positioning if I wanna stack, or to recast cast shadows, or for whatever other reason tbh. In fact when I look at parse data I see an average of approximately 20-30% tp wasted in my daily use. This comes by multiplying tp/hit by total number of hits and then looking at my total number of weaponskills from that. When I consider that the extra tp from weaponskills varies and can't be perfectly calculated like the normal( tp/hit x total hits) I see more than just what's being argued here, I see actual performance beyond what works on paper.

This is just another reason I'm an advocate of gearing smart. Numbers can show a large portion of what's going on, but human error is something that cannot be ignored in a situation like this. I value store tp greatly, but I wouldn't lose my love torque or toreadors for chiv chain or rajas, similarly carbonara follows suit. The "big 3" have already been covered so they needn't be rehashed to death, leaving only cobra gear, and only the body and maybe legs are really nice to use. So no matter whose model you want to go by, the gear you would pick ends up the same anyway. And since the way it actually plays out in game is already set, that makes store tp for thief pretty simple in my eyes.
____________________________
[ffxisig]56619[/ffxisig]

There are 10 kinds of people in the world. Those who understand binary, and those who don't.
#14 Oct 02 2009 at 8:05 PM Rating: Decent
*
96 posts
Noodles wrote:
But as I mentioned: It's not. that 240TP gained does absolutely nothing.

That is what I was getting at when I said "all TP is lost when you WS".


Does that 240tp not have a direct impact on WS performance, however small it might be? Whether it's adding a scrap of accuracy on DE or direct DMG to SB, it should do something. Rangers don't often sit on a little spare tp before they sidewinder for no reason at all and I don't see how helping land an extra hit on DE is completely useless either.

I am unclear about a few things though. How much extra tp do you need to gain a usable boost? For example, will 102tp net you at least 1 point of accuracy for DE, or a boost to the modifier % on something like SB?

I will agree that it's certainly not efficient. WSing as close as possible to 100tp is optimal. I'm only 50 thf atm, but when I play rng or sam extra tp on my WS's does give a noticeable boost to performance. I expect it wouldn't be too different for dagger WS.
#15 Oct 03 2009 at 3:16 AM Rating: Decent
*
118 posts
Rajas and Skadi legs give you 10% TP return, using BLAU/SIR.

The bad thing about Store tp is when you start boost Store tp and you take acu off and haste, for Store tp and you do big sacrifices.

But Thief is not a 2 hander job for 6 hit build, or 5 hit build polearma. I think Monk, thief and ninja need get focus on haste. and acu before Store tp

#16 Oct 04 2009 at 6:22 AM Rating: Good
Sage
***
1,281 posts
Noodles: I know what you are saying, and when I did the tp body comparisons a while back, I used a tp/hr model to compare them. BUT, in order to account for not WSing @ 100 tp, I calculated # of WSs per 115 tp. Probably a little over board, but it does account some for wasted tp. Due to my lack of any sTP, I find that I WS usually between 104-120 tp.
#17 Oct 05 2009 at 9:06 AM Rating: Decent
Scholar
**
996 posts

Personally, while I agree that gearing specifically for sTP is idiotic, a good thief should know how much he gets per hit. So that you can know how many attacks to get to 87-88. (WS for me is 14, to allow for a miss 100-13 = 87.)

The key would be if your off by just a few sTP it might make sense to drop a "better" piece for a piece with sTP.

A clear cut win would be to drop a scorp ring for a ecphoria ring if your off by 1.

But I always wondered for example if the 6 sTP from Cobra Harness would make it win over Homan Body on a few dagger combos.

____________________________
Arcari wrote:
It's the animation. All Drakesbane does is poke a mob and make lights appear. Ukko's Fury smashes a hole through existence itself to damage an opponent.

#18 Oct 14 2009 at 11:51 AM Rating: Decent
***
1,274 posts
Mistress Melphina wrote:
Ahh, so by that you meant "tp over 100% is lost". I gotchya.

Honestly that's another factor that's hard to pin down. Even when I have 100% tp I don't weaponskill immediately because of human reflexes, or positioning if I wanna stack, or to recast cast shadows, or for whatever other reason tbh. In fact when I look at parse data I see an average of approximately 20-30% tp wasted in my daily use. This comes by multiplying tp/hit by total number of hits and then looking at my total number of weaponskills from that. When I consider that the extra tp from weaponskills varies and can't be perfectly calculated like the normal( tp/hit x total hits) I see more than just what's being argued here, I see actual performance beyond what works on paper.


For starters this is irrelevant when comparing gear. Human error is not a factor in comparing gear. That is to say in an ideal situation which one of these is better. After looking at the answer saying piece A is negligibly better than piece B its valid to throw a dose of reality X factor in and decide maybe its not or maybe it really is. Or maybe its just not worth the effort. I don't think its an unreasonable argument entirely, but its an aside to actually comparing gear and therefore irrelevant.

For 2nd, many things in this game are made up of non-linear functions (pdif, tp gain come to mind immediately, and there are more such as weapon rank and any place there's a cap on something not to mention dTP at different amounts). Trying to compare them using a linear function (such as % increase/decrease) is completely idiotic. It works for some things some times, such as Accuracy or DA/TA, but fails at many things by varying degrees. Its just sad to see it misused in these forums so much.
____________________________
FFIX Melee Damage Comparitor
Brimstone
Reply To Thread

Colors Smileys Quote OriginalQuote Checked Help

 

Recent Visitors: 25 All times are in CDT
Anonymous Guests (25)