Diminishing returns on shield skill?

[kareface]

Question.

I play a friends pld/war account from time to time, he has other SJ's, but I'm more interested in making macros for /war build atm. He had a relic shield and can wear around 327 shield skill. I've been told not to equip all of the shield skill gear because of diminishing returns. So the question is, what is the best combination of gear for /war, assuming you have just about everything short of salvage gear? What is the ideal number for shield skill? Because if I'm using too much the shadow mantle or gigant would be a better choice.

[Arnwulf]

In my opinion, anything over about 300 sees diminishing returns.

[Neosutra]

I dont have pld lvled and dont have relic shield, and havent really paid attention to shield block rate, so In my opinion, anything over about 300 sees diminishing returns.

fixed.

And the amount of shield skill you need to reach the cap block rate varies on what your fighting.

300 techinically "will" get you to near the cap on anything up to VT to low IT (you/mob attributes also effect this), but if you are fighting anything of worth (higher lvl HNMs) then you should be stacking on the shield skill or you will definately struggle in reaching the cap.

Basicly /war

AF+1 feet all the time.
Boxers all the time if you dont have shadow mantle.
Shield Torque when you have plenty of mp, parade when you dont.

Those 3 and merits will push you to the block rate cap on most things, just add in kaiser head and shield earring if your having trouble hit block rate cap with just the above.

With some experience you will become comfortable with what amount of shield skill it takes to reach cap on every mob you fight.

[Therin]

Pieces that are too good NOT to have, for PLD/WAR:

Koenig/Kaiser Schaller: On anything that hits for over 100 or so damage non-blocked, you should leave this on full-time.

Gallant Leggings +1: +12 Shield Skill is the highest amount on any piece of armor in the game, and obviously you don't get shit from your feet slow otherwise (the second-best piece is Askar and it's not nearly as strong).

Shield Torque: Again, an excellent piece. You should never take this off if you're /WAR.

As for the rest, other stuff is typically better if you have Aegis. If you didn't have Aegis I'd say go with the full nine yards, Buckler Earring/Knight's (if you have it), and Boxer's, but you won't need the rest of that with Aegis. Go with Ethereal/Loquacious or Hades +1 and a Shadow Mantle.

[Lampoon]

While we're on topic of Shields; Sipar has a hidden block rate of 25% on top of being a Type 1 shield. With all the gear to reach cap, how does that hidden percentage affect over all dmg mitigation on anything HNM?

[Ferien]

While we're on topic of Shields; Sipar has a hidden block rate of 25% on top of being a Type 1 shield. With all the gear to reach cap, how does that hidden percentage affect over all dmg mitigation on anything HNM?

Prove it. I've heard so many people claim this to be true, but nobody has offered any kind of proof to back it up.

As for the OP's question, as the other posters said, it depends on what you're fighting. 300 will cap you up to around IT mobs, but for HNMs and such you're going to want to be in the 310~320 range.

[Lampoon]

Simple. Take Highlander's Targe vs. Sipar on same monsters. Both are same level, with only 3DEF difference. 3DEF should only give a 1.5% dmg mitigation discrepency. If Sipar indeed has additional 25%, it should block and mitigate more. I would gladly test it out if I had means of recording the data.

[Jeryhn]

While we're on topic of Shields; Sipar has a hidden block rate of 25% on top of being a Type 1 shield. With all the gear to reach cap, how does that hidden percentage affect over all dmg mitigation on anything HNM?

Problem is its still a Type 1 shield. You'll block a lot, preventing spell interruption, but you likely won't block much (if any) overall damage.

Probably a good shield for a PLD/NIN without a decent Haste set.

[Ferien]

Simple. Take Highlander's Targe vs. Sipar on same monsters. Both are same level, with only 3DEF difference. 3DEF should only give a 1.5% dmg mitigation discrepency. If Sipar indeed has additional 25%, it should block and mitigate more. I would gladly test it out if I had means of recording the data.

http://windower.faservers.net/wiki/plugins/logger

Get a WHM, find an EM Steelshell or something (don't think Korrigans would show enough variation with a size 1), aggro it and let it go to town. I've written a script that will parse your damage taken numbers from a text log, and some simple statistical analysis can give me proc rate and damage-%. Woozie can confirm that the script works if you'd like, I wrote it for him.

Just don't do anything that will change your damage taken over the course of the tests. No Protect, no changing of gear, no Phalanx. Just you and your shield.

[Neosutra]

While we're on topic of Shields; Sipar has a hidden block rate of 25% on top of being a Type 1 shield. With all the gear to reach cap, how does that hidden percentage affect over all dmg mitigation on anything HNM?

Problem is its still a Type 1 shield. You'll block a lot, preventing spell interruption, but you likely won't block much (if any) overall damage.

Probably a good shield for a PLD/NIN without a decent Haste set.

This is a myth.

Shield block rates do not increase for smaller types of shields, as the block rate cap % is around 65% for all types (except size 4 which cap out a bit lower, but have higher damage reduction). The only thing size does for a shield is factor into the damage mitigation equation (The 2 factors in how much damage % is reduced is Shield size (base % based on size 1-5 where as a size 1 has the lowest and size 5 has the highest, and the Def rating on the shield (i.e. Aegis has 40% extra damage reduction added to the size5 modifier via the 40 def rating of the shield). I dont remember the exact % for each shield size but its really that simple.

As for sipar, I have heard the rumor, but havent been able to test it. But the fact that it has 20 def and "apparently" 90% block rate, would put it way up there as one of the best DoT reducers, as well as keep shadows up.

As to the poster a few posts up saying "with aegis you wont need as much", that is false. Aegis doesnt have a different shield skill/ block rate ratio then standard keonig. The only difference between a keonig shield and Aegis is the % damage reduced on block and the -25% magic damage. If you fight a HNM with no shield gear or the bare minimum with Aegis, you will start to question if your shield is even equiped.

[Ferien]

Shield block rates do not increase for smaller types of shields, as the block rate cap % is around 65% for all types (except size 4 which cap out a bit lower, but have higher damage reduction). The only thing size does for a shield is factor into the damage mitigation equation (The 2 factors in how much damage % is reduced is Shield size (base % based on size 1-5 where as a size 1 has the lowest and size 5 has the highest, and the Def rating on the shield (i.e. Aegis has 40% extra damage reduction added to the size5 modifier via the 40 def rating of the shield). I dont remember the exact % for each shield size but its really that simple.

Small correction: 1 DEF = +.5% damage reduction, according to the 'zam tests (yes, they've actually produced something useful!) and partially confirmed by mine. Been wanting to get a whole slew of test data, but my playtime's been sporadic because of school and all. However, my semester's over and I'm done with finals soon, so I should have some free time to run some tests. Now let's see if I can con a mage into wasting a few hours of a few nights for the sake of science.

[Neosutra]

Shield block rates do not increase for smaller types of shields, as the block rate cap % is around 65% for all types (except size 4 which cap out a bit lower, but have higher damage reduction). The only thing size does for a shield is factor into the damage mitigation equation (The 2 factors in how much damage % is reduced is Shield size (base % based on size 1-5 where as a size 1 has the lowest and size 5 has the highest, and the Def rating on the shield (i.e. Aegis has 40% extra damage reduction added to the size5 modifier via the 40 def rating of the shield). I dont remember the exact % for each shield size but its really that simple.

Small correction: 1 DEF = +.5% damage reduction, according to the 'zam tests (yes, they've actually produced something useful!) and partially confirmed by mine. Been wanting to get a whole slew of test data, but my playtime's been sporadic because of school and all. However, my semester's over and I'm done with finals soon, so I should have some free time to run some tests. Now let's see if I can con a mage into wasting a few hours of a few nights for the sake of science.

Correct, my mistake. Do you have a link to that test, I forgot the actual size %s that were determined.

[Ferien]

Zam tests were here

Tested three size3 shields assuming a common base reduction percentage and found out the DEF/2 = 1% thing. From there, he was able to extrapolate the base and find the base for other sizes. Results were:

1: 22%
2: 40%
3: 50%
4: ??

Aegis was tested by myself to give 76% damage reduction (here), so this leads to either there being a 56% base for it, or it just having a defined 76%, not that it matters.

Also, my numbers for Koenig Shield nearly mirror zam's, so there's a point for them. Only problem with going out and testing this now will be that the damage functions are still unknown, so prior assumptions may be invalid now. I used the two most frequent values to determine proc rate and reduction, which I was able to confirm as valid with my Aegis tests, but depending on how random it is now, it may be impossible to find. We'll see how it goes.

[Seditedi]

I dont have pld lvled and dont have relic shield, and havent really paid attention to shield block rate, so In my opinion, anything over about 300 sees diminishing returns.

fixed.

See Arnwulf's sig, but yes you are right on him not having Aegis, lol.

[Lampoon]

Zam Post

Ferien if you do end up testing this, try Pucks in Den. Also I'd try /WAR with Defender and /RDM with Phalanx for curiosity's sake.

[Ferien]

Zam Post

Ferien if you do end up testing this, try Pucks in Den. Also I'd try /WAR with Defender and /RDM with Phalanx for curiosity's sake.

I want to do Pucks in Den (halves the time or doubles the data), but I have a feeling the variance may be too low for it to be worthwhile, since they don't hit as hard as other mobs. Also, Wild Oats would through another defensive variable in there, since it lowers your VIT.

As for Defender and Phalanx, I want to get a good-sized pool of unbuffed data before I play with those. Using either during the test opens it up to defense and -damage variations, since it's much easier to control when they're off than when they're on, and trying to keep them on leaves you open to periods when they're not.

[BeowulfFFXI]

This is a myth.

Shield block rates do not increase for smaller types of shields, as the block rate cap % is around 65% for all types (except size 4 which cap out a bit lower, but have higher damage reduction). The only thing size does for a shield is factor into the damage mitigation equation

Then why would anyone use a smaller type shield over a larger type? Not disagreeing, just making sure I didn't miss something.

[Neosutra]

This is a myth.

Shield block rates do not increase for smaller types of shields, as the block rate cap % is around 65% for all types (except size 4 which cap out a bit lower, but have higher damage reduction). The only thing size does for a shield is factor into the damage mitigation equation

Then why would anyone use a smaller type shield over a larger type? Not disagreeing, just making sure I didn't miss something.

A few possibilites:

1. They are using it for its other status effects (i.e spike shield).

2. Soloing pld/rdm (Get hit for 0 with size3+, hit for single digits with size1, which yields TP from shield mastery.).

3. They just dont know better.

4. They have sipar and the accusations are correct.

5. They are not a pld.

[Neosutra]

Zam tests were here

Tested three size3 shields assuming a common base reduction percentage and found out the DEF/2 = 1% thing. From there, he was able to extrapolate the base and find the base for other sizes. Results were:

1: 22%
2: 40%
3: 50%
4: ??

Aegis was tested by myself to give 76% damage reduction (here), so this leads to either there being a 56% base for it, or it just having a defined 76%, not that it matters.

Also, my numbers for Koenig Shield nearly mirror zam's, so there's a point for them. Only problem with going out and testing this now will be that the damage functions are still unknown, so prior assumptions may be invalid now. I used the two most frequent values to determine proc rate and reduction, which I was able to confirm as valid with my Aegis tests, but depending on how random it is now, it may be impossible to find. We'll see how it goes.

What cracks me up with this data is how close the Aegis damage reduction over time is to Keonig.

Aegis = 76% damage reduction on block.
Koenig = 50%(size3) + 11%(def of 22/2) = 61%.

So on shield block your taking 15% less damage with Aegis.

So for Melee DoT reduction with max block rate:

Koenig: .65 * .61 = 39.65% Physical DoT reduction.
Aegis: .65 * .76 = 49.4% Physical DoT reduction.

~10% difference in DoT.

So basically Aegis is a 200-300 mil gil defending ring.

And yes I know I am not factoring in the -25% magic damage that helps on Magic WSs, AoE, and just about everything in the game, as that does place Aegis >>>> Koenig. Its just a common misconception that Aegis -significantly- reduces the physical DoT you take, as the gap isnt really that big.

Just like any other relic weapon, Aegis is the best option for its class, but by no means game breaking. As always, knowledge of job and how you use it > all. Though I have been in quite a few solo/low man situations since I got Aegis where afterwards I was like "jesus, Aegis was the reason I am still alive", so no regrets here.

[Seraph]

While we're on topic of Shields; Sipar has a hidden block rate of 25% on top of being a Type 1 shield. With all the gear to reach cap, how does that hidden percentage affect over all dmg mitigation on anything HNM?

Prove it. I've heard so many people claim this to be true, but nobody has offered any kind of proof to back it up.

As for the OP's question, as the other posters said, it depends on what you're fighting. 300 will cap you up to around IT mobs, but for HNMs and such you're going to want to be in the 310~320 range.

Now, I have done some testing with my Sipar, and the results always seem to come back that the block rate is over 65% on everything from TW to VT. I have never had the chance to use this in a serious endgame situation (IT & ITG mobs) besides the Limbus Mandys because I am not going to lose 3 Enmity for the sake of doing a block rate test. I know people will demand a large database of proof, but i'm lazy and me eyeballing shield block rates over a sample of about 100 hits with the Sipar and 100 blocks with the Light Buckler always put the Blockrate (and DoT reduction) on every mob I tried higher with the Sipar. I usually just watched 100 hits and recorded how many of them were blocked and unblocked. My highest parse was a 92% blockrate on the lesser birds outside of Whitegate. That just seemed way too much of a variance to be luck over a 100 hit sample. (Well, it could be, but damn...) Other people I know have tried out the Sipar with similiar results, so it's not just my characters random number karma messing with me.

Now, I know it could all be luck with the rates, but it always seemed like luck sided with the Sipar, and more than enough for it to be just chance. I also can't understand why it would be R/EX considering its other stats and what it drops off of. Go go -20 ice resistance. The reason this hidden effect might be so challenged by so many as fake all the time is few people have or keep Sipar when it drops. Ask your tanks sometime about it, chances are most people don't even know what it is. It doesn't come from the auction house, it doesn't have any desirable visible stats on it besides a high defense rating for a small shield, and not that many people really religiously camp Adam/Aspi. In addition, the drop rate is complete crap on this too, it's the Defending Ring of the turtle world. How many people didn't believe there was a hidden bonus on the Company Sword for the longest time until someone religiously parsed out the data, and those things are super common and cheap on the AH.

Now, what I want to know, is assuming the Sipar block rate bonus is real (I am all but positive it is) is it just a flat % increase, or is it something more like raising the ceiling cap on shield blocking? Does it add ~+20% to whatever your shield block rate would normally be, or does it just increase the potential shield block rate cap from ~65% to something closer to ~85%? Depending on which it is, stacking shield skill with Sipar would be ideal, or completely unnecessary.

THAT is something someone with a good parser, a Sipar, some shield skill gear, and way too much free time should test. Ideally this would have to be on mobs where the standard blockrate with a small shield is less than 65% normally, and I am not sure if any easy to tank mobs like this exist. Otherwise, it would have to be determined if there was a noticable change in blockrate with the SIpar depending on shield skill. If anything, do it for the only shield that might be better than Aegis at something besides +3/4 Enmity.

And for the person asking why you would use a Sipar, assuming you were say a PLD/NIN, it would prevent spell interruption when casting anything assuming you blocked the hit. This is nice or recasting shadows and such. It could even be nice for PLD/RDM for recasting stoneskin, phalanx, and all those other fun /RDM spells. I know a bunch of people laughed when I presented this idea in the Newbie forums a few months back, but how many times has a tank gone down from an unlucky hit during a recast? How many times has that one hit screwed you on a long cast?

...I guess it could also help if you are subbing /PLD25+ and can equip it. Who does that though?

EDIT: Reference link for great justice:
http://bluegartr.com/forum/viewtopic...17&hilit=Sipar