Just adding to Neo's post
I will say however, that with the 20% SI Atmacyte, a tank can luckily just gear for 82% SI and still obtain non interuptable casting.
10% merits.
10% Impatiens.
8% Kaiser body augment.
8 % Resolute Belt
15% Kara feet.
9 % Knightly Earring
8 % Magnetic Ear
10% Valor legs.
5 % Willpower Torque
8% from rings (Augmented Diamond or some lucky Dark rings).
That's 91 % = just 11 SID short ;; Or way over with SI Atmacyte.
Also dont forget you can get 19 % more per merit (95 % cap) under Rampart / Ironwill but can only use for 45 seconds every 4 mins 25.
Note :
Could investigate Cors - Choral Roll to make up the 11 SID shortfall (or more) ... but i dont know what % it boosts SID by - does anyone else ?? or needs testing ??
... i like the fact it lasts 5 mins but means your Cor needs to be on the ball and not leave you without it for ages - or cry about giving you the roll xD
Paladin FAQ, Info, And Trade Studies.
Paladin FAQ, Info, and Trade Studies.
It just seems like to much to give up too much defensively, offensively, and your inventory to increase your interruption rate from 65%~ to 100%.
The 65% was just a number I threw out there considering aegis block rate, shadow mantle, and not getting interrupted anyway.
The 65% was just a number I threw out there considering aegis block rate, shadow mantle, and not getting interrupted anyway.
Those are two completely different variables.
Considering the recent findings in shield block rates/skill effect. A proper SI set will outweigh all of those "losses"
This is simply for macro casts. Not something that is full timed by any means.
Considering the recent findings in shield block rates/skill effect. A proper SI set will outweigh all of those "losses"
This is simply for macro casts. Not something that is full timed by any means.
What is two completely different variables? You're using a shield right? Facing mob?
It just seems like to much to give up too much defensively, offensively, and your inventory to increase your successful cast rate from 65%~ to 100%.
When you throw in stuff like cureskin, fanatics, sleepga, silence and knock back it will make even less of a difference.
It just seems like to much to give up too much defensively, offensively, and your inventory to increase your successful cast rate from 65%~ to 100%.
When you throw in stuff like cureskin, fanatics, sleepga, silence and knock back it will make even less of a difference.
Inventory space is one factor that I'd consider. SIR is an all or nothing kind of thing so you're going to need most all of this to make it worth your time and much of it is very difficult to get or very time consuming. Some won't have the option of a magnetic earring either way.
I don't see how you'd be giving anything up defensively other than the lack of pdt or shield skill gear for 1 or 2 hits during casting time. Offensively, you loose nothing because you don't swing during casting. EDIT: If you consider that you're guaranteed to get a cast off, this actually improves your DPS since you won't have to recast for a failed cast.
Now if you're speaking of opportunity cost because you had to give something up to carry this gear, I suppose I can see that but if you're fighting something that you'd bother using this set on, Aegis's block rate isn't going to be too hot.
I don't see how you'd be giving anything up defensively other than the lack of pdt or shield skill gear for 1 or 2 hits during casting time. Offensively, you loose nothing because you don't swing during casting. EDIT: If you consider that you're guaranteed to get a cast off, this actually improves your DPS since you won't have to recast for a failed cast.
Now if you're speaking of opportunity cost because you had to give something up to carry this gear, I suppose I can see that but if you're fighting something that you'd bother using this set on, Aegis's block rate isn't going to be too hot.
I would assume that the offensive loss here is because you need to use an Atmacyte for SI to hit 102% that could have been used to improve DMG output instead.
So compare an atmacyte's DD enhancement to never having to recast post interrupt, for the offensive side.
Inv space might not be too big an issue here. Assuming this is for use with /nin non-Ochain.
You wouldn't need any shield skill gear for casting/block purposes, and typical physical defenses are what shadows are for. So that frees up some space.
Then you'd just use a full PDT/terror set for shadows/recast down moments and such. Not to mention reprisal and palisade.
You might or might not be a bit more vulnerable when shadows aren't up. But it's kinda hard to directly compare shield skill vs PDT considering the lack of solid precise testing. You'll certainly take more dmg on hits while casting due to lower PDT/shield skill. But there should be less hits taken overall since you won't fail a cast.
Still, I think it's a viable option, depending on your inv. Being uninterruptable is pretty powerful.
So compare an atmacyte's DD enhancement to never having to recast post interrupt, for the offensive side.
Inv space might not be too big an issue here. Assuming this is for use with /nin non-Ochain.
You wouldn't need any shield skill gear for casting/block purposes, and typical physical defenses are what shadows are for. So that frees up some space.
Then you'd just use a full PDT/terror set for shadows/recast down moments and such. Not to mention reprisal and palisade.
You might or might not be a bit more vulnerable when shadows aren't up. But it's kinda hard to directly compare shield skill vs PDT considering the lack of solid precise testing. You'll certainly take more dmg on hits while casting due to lower PDT/shield skill. But there should be less hits taken overall since you won't fail a cast.
Still, I think it's a viable option, depending on your inv. Being uninterruptable is pretty powerful.
Yea Martel I was referring to the atmacite and using a cor for a sir roll (lol).
What spells are we needing to get off 100% of the time? Ichi? Cure4?
The only time I can remember being annoyed with not getting a cast off is because of knock back, silence/sleepga, petrified/terror, or stun/enstun.
What spells are we needing to get off 100% of the time? Ichi? Cure4?
The only time I can remember being annoyed with not getting a cast off is because of knock back, silence/sleepga, petrified/terror, or stun/enstun.
Odin.Aable said: »
You have over 400 shield skill. You do not have 100% block rate. It is clearly more complex then that.
As it stands though, showmo's findings are about the only evidence pointing this out (that I have seen at least) and it could very well be possible that only your base blockrate decreases, the higher-lvl the mob is. And that every 4 skill still gets you 1% more blockrate.
Those are all assumptions for the time being, but they certainly give us an idea about how shield skill might be valueable or not, compared to other options.
On a side-note, 100% blockrate has probably never been intended by SE (until ochain), regardless of shield/skill/mob lvl, so there were/are caps for that with normal shields.
Which still doesn't mean that 4 skill =/= 1% blockrate (if not capped).
Personally, I'm just thankful for showmo sharing this with us, helping us to understand it and maybe inspire ppl to do more testings into that direction.
Odin.Aable said: »
You have over 400 shield skill. You do not have 100% block rate. It is clearly more complex then that.
The rule of thumb for hit rate is that every 2 accuracy is equal to +1% hit rate. However, having a total of 200 accuracy does not provide a 100% hit rate; you must first calculate the differences between your level & accuracy with the mob's level & evasion to get the base hit rate. Every 2 accuracy on top of that will increase your hit rate by 1% (with a cap of 95% and a floor of 20%).
Using the above example, there is likely a formula which calculates the difference between your level & Shield skill with the mob's level & possibly other factors (such as mob's accuracy for example) to get the base block rate. After that, adding +x amount of Shield skill will grant +x% amount of block rate on top of the existing block rate (until you hit the cap).
From the parse done on Fear Deargs, it appears that every +4 skill is equal to +1% block rate. More tests may need to be done though on different mobs and levels to verify if this remains true.
Phoenix.Gustavve said: »
Yea Martel I was referring to the atmacite and using a cor for a sir roll (lol).
What spells are we needing to get off 100% of the time? Ichi? Cure4?
The only time I can remember being annoyed with not getting a cast off is because of knock back, silence/sleepga, petrified/terror, or stun/enstun.
What spells are we needing to get off 100% of the time? Ichi? Cure4?
The only time I can remember being annoyed with not getting a cast off is because of knock back, silence/sleepga, petrified/terror, or stun/enstun.
For me:
Ichi (Ni even sometimes with some of the mob hit rates), Cure4, Enlight, Phalanx.
Ichi being the big "reducing total damage taken over time significantly", while the others are "increasing DoT via not having to recast after interrupts over and over".
Since you're essentially immune to magic damage with Aegis, 100% interrupt drastically improves your time with shadows up (and thus reducing Emnity lost), as well as increases DD a bit via not having to attempt to recast shadows multiple times.
I agree however that being able to ride fanatics for about half of each VW fight when you have good proccers weighs much heavier in the no SIR argument.
Looking at showmo's test and speculation again, I decided to look through some of the test data I had lying around. I did these test back at lvl 90 but never analyzed the data closely enough to figure block rate vs skill.
I'm only posting results atm, not raw parse data. If there are doubts about any of this, I can dig the parses out. But I pulled this from the spreadsheet I recorded the results to. Besides, that'd be a hell of a lot of parse data. 16k~ hits worth. It'd make things messy.
The size 3 data fits pretty nicely with the 4 skill=1% idea. 'Cept the +10 sample, but we could explain that as variance or lower sample size.
But the rest of the data points match pretty well. Skill+ / 4 = Block rate gain.
Now we need to know how mob lvl fits into that so we can predict block rate.
I still want more data. And more testing on various mob lvls to see if things change at different skill/mob levels. But this looks fairly good.
Now the Size 4 test doesn't quite match. It doesn't really have enough data points. And while 2% isn't a big difference, it's enough to make me wonder if skill works differently for different shield sizes. Although, since Showmo's test was with Aegis(size 5) and mine are size 3 there may be some evidence against that idea. Unless Aegis has the same skill mechanic a size 3, but with a higher base block rate.. ugh.. Too many possibilities.
Still, this is getting pretty interesting. Might do some more tests later. Getting more data on the lvl 95 skill range could be useful.
I'm only posting results atm, not raw parse data. If there are doubts about any of this, I can dig the parses out. But I pulled this from the spreadsheet I recorded the results to. Besides, that'd be a hell of a lot of parse data. 16k~ hits worth. It'd make things messy.
Code
LVL 90 shield tests Even Match Monitors, Abyssea Grauberg Koenig(size 3) Skill 369(+00) 379(+10) 389(+20) 399(+30) 417(+48) 420(+51) Block% 45.0% 46.5% 50.5% 52.4% 57.1% 57.6% gain 00.0% 01.5% 05.1% 07.4% 12.0 12.6 hits 2860 2126 3086 3034 2049 3295 Tower Shield Skill 369(+00) 420(+51) Block% 29.4% 43.9% gain 00.0% 14.5% hits 3093 3137
The size 3 data fits pretty nicely with the 4 skill=1% idea. 'Cept the +10 sample, but we could explain that as variance or lower sample size.
But the rest of the data points match pretty well. Skill+ / 4 = Block rate gain.
Now we need to know how mob lvl fits into that so we can predict block rate.
I still want more data. And more testing on various mob lvls to see if things change at different skill/mob levels. But this looks fairly good.
Now the Size 4 test doesn't quite match. It doesn't really have enough data points. And while 2% isn't a big difference, it's enough to make me wonder if skill works differently for different shield sizes. Although, since Showmo's test was with Aegis(size 5) and mine are size 3 there may be some evidence against that idea. Unless Aegis has the same skill mechanic a size 3, but with a higher base block rate.. ugh.. Too many possibilities.
Still, this is getting pretty interesting. Might do some more tests later. Getting more data on the lvl 95 skill range could be useful.
Phoenix.Neosutra said: »
Phoenix.Gustavve said: »
Yea Martel I was referring to the atmacite and using a cor for a sir roll (lol).
What spells are we needing to get off 100% of the time? Ichi? Cure4?
The only time I can remember being annoyed with not getting a cast off is because of knock back, silence/sleepga, petrified/terror, or stun/enstun.
What spells are we needing to get off 100% of the time? Ichi? Cure4?
The only time I can remember being annoyed with not getting a cast off is because of knock back, silence/sleepga, petrified/terror, or stun/enstun.
For me:
Ichi (Ni even sometimes with some of the mob hit rates), Cure4, Enlight, Phalanx.
Ichi being the big "reducing total damage taken over time significantly", while the others are "increasing DoT via not having to recast after interrupts over and over".
Since you're essentially immune to magic damage with Aegis, 100% interrupt drastically improves your time with shadows up (and thus reducing Emnity lost), as well as increases DD a bit via not having to attempt to recast shadows multiple times.
I agree however that being able to ride fanatics for about half of each VW fight when you have good proccers weighs much heavier in the no SIR argument.
At the very least, an Aegis/SIR/NIN SJ setup has merit while playing catch-up during irregular event hours for anyone who's missed a clear. The better the proc spread, the less advantageous it becomes, but I'm sure your setup isn't perfect every, single time, so its probably still handy to keep around.
Shield data is excellent, thanks for the diligent testing. I'll have Ochain(yay Tarudin) finished relatively soon, and I'll try to get some numbers of whether or not Ochain's block rate is dropping as drastically as being suggested on certain targets.
I need to find some mobs of an extremely high lvl to do some Ochain tests on. But it's hard to control tests on NMs/Vw etc. And you can't really stand around for hours with a VWNM letting it beat your face in.
I'd also really want to know what lvl any test mob is. As that could be helpful in figuring out exactly how mob lvl affect block rate. Kinda hard to find a good mob that can uncap Ochain block rate.
I'd also really want to know what lvl any test mob is. As that could be helpful in figuring out exactly how mob lvl affect block rate. Kinda hard to find a good mob that can uncap Ochain block rate.
Ya. I'd probably advocate the capped SIR build in pickup groups and when you're fighting tougher stuff with less support. More aggressive Atmacyte combos for LS and other event related stuff.
Hi all, I'm back with some more shield testing.
I thought this was an interesting theory, so I decided to do some tests with different shield sizes. The shields I used were: Aegis (95) (Size 5), Januwiyah +1 (Size 4), Koenig Shield (Size 3), & Strike Shield (Size 2). I wanted to test a size 1 shield as well, but my play time was cut right as I finished the size 2 test. All tests were done on the same exact mob (yes, I held the same mob for 12 hours). Fortunately, I had a few friends farm time extensions for me on & off while I was able to perform these tests. Anyway, here is the data info:
My Job/Level: PLD95/WAR47
Buffs: Grabbed all HP/MP/Stat cruor buffs to avoid VIT down in order to prevent skewing the results (when comparing which data was blocked and unblocked hits).
Atmas: Mounted Champion, Impenetrable, & Vicissitude (for regen to sustain HP)
Mob: Fear Dearg (only one), Very Tough
Base Shield Skill: 404 (capped & includes 4/4 merits)
No equipment other than the Shield itself were used for the initial tests (completely naked).
No special buffs (such as Phalanx or Protect) were used other than Cruor buffs and Atmas.
Any attacks that occurred while slept from Dream Flower were excluded from the parse.
Each individual Shield parse has at least 2,000 samples worth of successful landed hits.
For each parse, equipment remained completely static (no gear swaps were used at all)
Aegis (95)
Blocks: 969
Unblocks: 1,167
Block Rate: 45.37%
Aegis (95) (with Shield Skill +48)
Blocks: 1,252
Unblocks: 1,045
Block Rate: 54.51%
9.14% Increase
Januwiyah +1
Blocks: 475
Unblocks: 1,538
Block Rate: 23.60%
Januwiyah +1 (with Shield Skill +48)
Blocks: 733
Unblocks: 1,391
Block Rate: 34.51%
10.91% Increase
Koenig Shield
Blocks: 869
Unblocks: 1,286
Block Rate: 40.32%
Koenig Shield (with Shield Skill +48)
Blocks: 1,026
Unblocks: 993
Block Rate: 50.82%
10.5% Increase
Strike Shield
The difference between blocked hits & unblocked hits for the Strike Shield were hard to determine with the initial parse alone, so a second parse with my character's back turned away from the mob (but shield still equipped for the DEF) was done to show which hit ranges were unblocked hits.
Edit: Strike Shield data is invalid due to the low damage reduction overlapping with normal hits. I'll keep the parse here though in case anyone wants to examine it.
Strike Shield (with back turned towards mob)
Blocks: 691
Unblocks: 1,348
Block Rate: 33.89%
Strike Shield Data is invalid, see note above.
Strike Shield (with Shield Skill +48)
Strike Shield (with Shield Skill +48 and back turned towards mob)
Blocks: 923
Unblocks: 1,183
Block Rate: 43.83%
9.94% Increase
Strike Shield Data is invalid, see note above.
It's interesting to note that Aegis (95) appears to have roughly a 5% higher block rate than the Size 3 Koenig Shield.I also found it interesting that a Size 2 shield had a lower block rate than the Size 3 Koenig shield; I was under the impression that smaller sized shields had the highest block rate in return for lower damage reduction. (Edit: Strike Shield data is skewed due to normal hits overlapping with blocked hits).
While there is some variance with the block rate increase in each data set, I think that the increase amounts are close enough to each other to say that the size of the shield does not effect how much block rate increase you receive from shield skill. If it is effected by shield size however, the difference seems to be very small (Koenig Shield received a 10.5% increase while Januwiyah +1 received a 10.91% increase, which would only be a 0.41% difference from +48 shield skill).
If you average the 3 block rate increase samples together (excluding Strike Shield data), you get a +10.183% block rate increase from +48 Shield Skill. This would suggest that every 1 Shield Skill is equal to +0.21%-0.22% block rate on VT mobs, which would mean you would need +5 Shield Skill for every 1% block rate increase. A previous parse on a Decent Challenge Fear Dearg suggested that every +4 skill is equal to +1% block rate. This can be explained from 2 scenarios:
1) Level of mob has a big impact on how much block rate increase you receive from Shield Skill.
2) Variance skewed the data towards one direction
Also noting the previous and current tests:
On a Decent Challenge Fear Dearg, Aegis (95) had a 53.27% block rate with no equipment or Shield Skill + gear.
On a Very Tough Fear Dearg, Aegis (95) had a 45.37% block rate with no equipment or Shield Skill + gear.
The block rate difference is 7.9%.
According to Wiki, a Decent Challenge mob is 1-7 levels below you while a Very Tough mob is 4-7 levels above you. So the level difference was at least 5 levels apart, and at most 14 levels apart.
What do you guys think? Any ideas on how total Shield block rate may be calculated? The formula would likely only consist of your level, the mob's level, and your total shield skill (from base + merits + gear).
Ragnarok.Martel said: »
Now the Size 4 test doesn't quite match. It doesn't really have enough data points. And while 2% isn't a big difference, it's enough to make me wonder if skill works differently for different shield sizes. Although, since Showmo's test was with Aegis(size 5) and mine are size 3 there may be some evidence against that idea. Unless Aegis has the same skill mechanic a size 3, but with a higher base block rate.. ugh.. Too many possibilities.
I thought this was an interesting theory, so I decided to do some tests with different shield sizes. The shields I used were: Aegis (95) (Size 5), Januwiyah +1 (Size 4), Koenig Shield (Size 3), & Strike Shield (Size 2). I wanted to test a size 1 shield as well, but my play time was cut right as I finished the size 2 test. All tests were done on the same exact mob (yes, I held the same mob for 12 hours). Fortunately, I had a few friends farm time extensions for me on & off while I was able to perform these tests. Anyway, here is the data info:
My Job/Level: PLD95/WAR47
Buffs: Grabbed all HP/MP/Stat cruor buffs to avoid VIT down in order to prevent skewing the results (when comparing which data was blocked and unblocked hits).
Atmas: Mounted Champion, Impenetrable, & Vicissitude (for regen to sustain HP)
Mob: Fear Dearg (only one), Very Tough
Base Shield Skill: 404 (capped & includes 4/4 merits)
No equipment other than the Shield itself were used for the initial tests (completely naked).
No special buffs (such as Phalanx or Protect) were used other than Cruor buffs and Atmas.
Any attacks that occurred while slept from Dream Flower were excluded from the parse.
Each individual Shield parse has at least 2,000 samples worth of successful landed hits.
For each parse, equipment remained completely static (no gear swaps were used at all)
Aegis (95)
Code
Melee Damage Taken
Player Melee Dmg Melee % Hit/Miss M.Low/Hi M.Avg #Crit C.Low/Hi C.Avg Crit%
Showmo 73270 97.50 % 2136/118 1/86 32.68 108 12/124 61.35 5.03 %
Showmo
Melee
1: 45
2: 94
3: 94
4: 110
5: 106
+ 6: 125
7: 101
8: 107
9: 96
10: 29
11: 4
12: 5
13: 5
38: 16
39: 46
40: 29
^ 41: 19
42: 32
43: 32
44: 42
45: 41
46: 37
47: 20
48: 36
49: 34
50: 41
51: 30
52: 41
53: 24
54: 41
55: 31
56: 29
57: 26
58: 41
59: 36
60: 25
61: 23
62: 34
63: 25
64: 37
65: 36
66: 28
67: 39
68: 35
69: 24
70: 24
71: 20
72: 14
73: 11
74: 1
76: 4
77: 2
80: 5
82: 1
83: 2
84: 2
86: 2
Melee Crits
12: 1
13: 6
+ 14: 12
15: 6
16: 3
17: 6
18: 3
19: 5
20: 4
22: 1
24: 1
81: 1
83: 2
84: 2
^ 85: 2
86: 2
88: 5
89: 3
91: 1
93: 3
94: 5
95: 3
96: 1
98: 2
99: 4
100: 4
101: 1
102: 2
104: 1
105: 1
106: 1
107: 2
108: 2
109: 3
111: 3
112: 1
113: 1
117: 1
124: 1Blocks: 969
Unblocks: 1,167
Block Rate: 45.37%
Aegis (95) (with Shield Skill +48)
Code
Melee Damage Taken
Player Melee Dmg Melee % Hit/Miss M.Low/Hi M.Avg #Crit C.Low/Hi C.Avg Crit%
Showmo 57537 97.43 % 2297/108 1/75 23.43 130 10/119 50.65 5.64 %
Showmo
Melee
+ 1: 358
2: 115
3: 173
4: 133
5: 159
6: 137
^ 7: 88
8: 10
9: 8
10: 4
38: 165
39: 142
40: 27
41: 18
42: 32
43: 23
44: 25
45: 32
46: 43
47: 31
48: 40
49: 20
50: 22
51: 26
52: 41
53: 31
54: 34
55: 32
56: 32
57: 38
58: 32
59: 23
60: 16
61: 22
62: 7
63: 15
64: 7
65: 3
66: 2
67: 3
69: 1
70: 2
71: 1
75: 2
Melee Crits
10: 4
11: 14
12: 5
+ 13: 17
14: 8
15: 6
16: 7
17: 4
^ 18: 2
74: 1
75: 3
76: 1
78: 4
79: 2
80: 1
81: 1
82: 2
83: 4
84: 1
85: 1
87: 2
88: 4
89: 5
90: 1
91: 2
93: 4
95: 5
97: 4
98: 4
99: 2
100: 2
101: 1
102: 2
103: 1
110: 1
115: 1
119: 1Blocks: 1,252
Unblocks: 1,045
Block Rate: 54.51%
9.14% Increase
Januwiyah +1
Code
Melee Damage Taken
Player Melee Dmg Melee % Hit/Miss M.Low/Hi M.Avg #Crit C.Low/Hi C.Avg Crit%
Showmo 95458 97.67 % 2013/101 0/91 45.65 95 10/133 76.99 4.69 %
Showmo
Melee
0: 15
1: 57
2: 62
3: 66
4: 54
5: 64
+ 6: 70
7: 47
8: 7
9: 4
10: 2
40: 4
41: 15
42: 28
43: 36
44: 54
45: 59
46: 51
47: 52
48: 59
49: 43
50: 31
51: 36
52: 29
^ 53: 40
54: 41
55: 50
56: 46
57: 38
58: 48
59: 59
60: 41
61: 30
62: 40
63: 47
64: 53
65: 54
66: 42
67: 37
68: 32
69: 45
70: 41
71: 31
72: 38
73: 41
74: 35
75: 16
76: 5
78: 2
79: 5
80: 4
81: 3
82: 4
83: 2
84: 6
85: 2
87: 3
88: 2
89: 1
90: 1
91: 1
Melee Crits
10: 4
11: 4
12: 1
13: 3
+ 14: 9
15: 5
18: 1
85: 2
86: 1
87: 1
88: 2
89: 1
90: 1
91: 5
92: 2
93: 1
95: 2
^ 96: 3
97: 6
98: 1
99: 1
100: 1
101: 2
102: 3
104: 4
105: 4
106: 3
107: 2
108: 2
109: 3
110: 2
111: 1
112: 3
114: 2
116: 4
129: 1
133: 2Blocks: 475
Unblocks: 1,538
Block Rate: 23.60%
Januwiyah +1 (with Shield Skill +48)
Code
Melee Damage Taken
Player Melee Dmg Melee % Hit/Miss M.Low/Hi M.Avg #Crit C.Low/Hi C.Avg Crit%
Showmo 72657 98.07 % 2124/114 0/77 32.88 89 7/115 63.04 4.18 %
Showmo
Melee
+ 0: 211
1: 114
2: 116
3: 97
4: 84
5: 68
6: 6
7: 6
38: 165
^ 39: 166
40: 33
41: 24
42: 38
43: 37
44: 48
45: 46
46: 43
47: 41
48: 38
49: 46
50: 34
51: 34
52: 50
53: 40
54: 49
55: 34
56: 34
57: 47
58: 32
59: 46
60: 48
61: 26
62: 23
63: 29
64: 24
65: 28
66: 13
67: 1
68: 1
69: 3
70: 4
71: 2
72: 5
73: 1
74: 2
77: 2
Melee Crits
7: 2
8: 3
+ 9: 6
+ 10: 6
11: 5
12: 4
13: 3
14: 1
16: 1
75: 2
76: 3
77: 2
+ 79: 6
^ 80: 2
81: 1
82: 2
83: 1
85: 2
88: 2
89: 1
91: 3
93: 2
94: 3
95: 2
96: 2
97: 3
98: 3
99: 4
100: 2
101: 2
103: 1
104: 1
105: 3
106: 1
109: 1
115: 1Blocks: 733
Unblocks: 1,391
Block Rate: 34.51%
10.91% Increase
Koenig Shield
Code
Melee Damage Taken
Player Melee Dmg Melee % Hit/Miss M.Low/Hi M.Avg #Crit C.Low/Hi C.Avg Crit%
Showmo 91182 97.91 % 2155/103 7/88 40.57 105 25/113 72.49 4.85 %
Showmo
Melee
7: 22
8: 33
9: 70
10: 46
+ 11: 84
12: 46
13: 81
14: 40
15: 66
16: 45
17: 59
18: 77
19: 54
20: 70
21: 13
22: 8
23: 5
24: 4
25: 4
26: 1
40: 10
41: 24
42: 26
43: 34
44: 24
45: 36
46: 36
^ 47: 40
48: 35
49: 50
50: 30
51: 37
52: 40
53: 37
54: 20
55: 34
56: 41
57: 55
58: 30
59: 36
60: 36
61: 37
62: 26
63: 39
64: 48
65: 30
66: 25
67: 28
68: 47
69: 37
70: 45
71: 34
72: 38
73: 29
74: 19
75: 13
76: 2
77: 2
78: 3
79: 2
80: 2
81: 1
82: 3
84: 5
86: 1
87: 2
88: 3
Melee Crits
25: 3
26: 3
27: 6
28: 4
29: 1
30: 4
31: 3
+ 32: 7
33: 1
34: 6
35: 1
36: 1
37: 1
84: 1
85: 2
86: 1
87: 1
88: 1
89: 2
^ 90: 4
91: 5
93: 3
94: 2
95: 1
96: 1
97: 3
98: 2
99: 1
100: 4
+ 102: 7
103: 2
105: 1
106: 2
108: 3
109: 3
110: 5
111: 2
112: 4
113: 1Blocks: 869
Unblocks: 1,286
Block Rate: 40.32%
Koenig Shield (with Shield Skill +48)
Code
Melee Damage Taken
Player Melee Dmg Melee % Hit/Miss M.Low/Hi M.Avg #Crit C.Low/Hi C.Avg Crit%
Showmo 62623 97.05 % 2019/104 6/76 29.76 93 21/104 52.66 4.57 %
Showmo
Melee
6: 137
+ 7: 164
8: 47
9: 95
10: 67
11: 98
12: 55
13: 80
14: 60
15: 73
16: 59
17: 18
18: 3
19: 3
20: 7
21: 4
^ 38: 118
39: 116
40: 19
41: 21
42: 19
43: 22
44: 37
45: 37
46: 33
47: 36
48: 28
49: 34
50: 34
51: 27
52: 34
53: 23
54: 27
55: 31
56: 34
57: 34
58: 38
59: 41
60: 25
61: 25
62: 17
63: 18
64: 15
65: 15
66: 5
68: 2
69: 1
71: 1
74: 1
75: 1
76: 1
Melee Crits
21: 3
22: 3
23: 5
24: 4
25: 5
26: 5
27: 4
28: 4
29: 7
30: 4
+^ 31: 8
32: 3
39: 1
76: 2
77: 1
79: 1
80: 1
81: 1
84: 3
86: 1
87: 1
88: 2
90: 2
92: 1
93: 2
94: 5
95: 3
97: 4
98: 2
99: 1
101: 1
102: 1
103: 1
104: 1Blocks: 1,026
Unblocks: 993
Block Rate: 50.82%
10.5% Increase
Strike Shield
Edit: Strike Shield data is invalid due to the low damage reduction overlapping with normal hits. I'll keep the parse here though in case anyone wants to examine it.
Code
Melee Damage Taken
Player Melee Dmg Melee % Hit/Miss M.Low/Hi M.Avg #Crit C.Low/Hi C.Avg Crit%
Showmo 106244 98.11 % 2039/118 16/94 49.77 118 41/130 85.56 5.76 %
Showmo
Melee
16: 1
17: 23
18: 41
19: 36
20: 20
21: 36
22: 43
23: 35
24: 22
25: 37
26: 38
27: 27
28: 17
29: 45
30: 35
31: 48
32: 11
33: 41
34: 43
35: 32
36: 4
37: 6
38: 1
39: 2
41: 2
42: 1
43: 4
44: 13
45: 31
46: 34
47: 41
48: 36
49: 31
50: 32
51: 47
52: 28
^ 53: 39
54: 38
55: 47
56: 37
57: 48
58: 33
59: 31
60: 26
61: 42
62: 35
63: 42
64: 42
65: 43
66: 46
67: 36
68: 37
69: 31
70: 40
71: 37
72: 35
+ 73: 52
74: 41
75: 30
76: 39
77: 20
78: 13
79: 6
80: 1
81: 1
82: 2
83: 1
85: 6
87: 1
88: 4
89: 2
90: 1
91: 1
92: 1
94: 1
Melee Crits
41: 2
42: 1
43: 6
44: 2
46: 2
47: 6
49: 3
50: 5
51: 3
52: 1
54: 1
55: 2
56: 3
57: 1
58: 1
68: 1
88: 3
89: 4
92: 1
93: 5
94: 1
95: 1
96: 4
^ 97: 4
98: 1
99: 1
100: 3
101: 2
102: 4
+ 103: 7
104: 1
105: 3
106: 3
107: 4
109: 1
110: 3
111: 4
112: 2
113: 6
114: 2
116: 1
117: 1
121: 1
122: 1
128: 2
129: 1
130: 1Strike Shield (with back turned towards mob)
Code
Melee Damage Taken
Player Melee Dmg Melee % Hit/Miss M.Low/Hi M.Avg #Crit C.Low/Hi C.Avg Crit%
Showmo 11301 98.34 % 178/9 44/94 62.05 7 95/104 98.71 3.93 %
Showmo
Melee
44: 1
45: 2
46: 4
47: 6
48: 6
+ 49: 11
50: 3
51: 3
52: 4
53: 3
54: 10
55: 4
56: 5
57: 7
58: 4
59: 2
60: 3
61: 6
62: 1
^ 63: 2
64: 9
65: 5
66: 6
67: 3
68: 4
69: 4
70: 10
71: 5
72: 8
73: 5
74: 10
75: 2
76: 2
77: 4
78: 2
84: 1
85: 1
86: 1
93: 1
94: 1
Melee Crits
95: 1
+ 96: 2
^ 97: 1
100: 1
103: 1
104: 1Unblocks: 1,348
Block Rate: 33.89%
Strike Shield Data is invalid, see note above.
Strike Shield (with Shield Skill +48)
Code
Melee Damage Taken
Player Melee Dmg Melee % Hit/Miss M.Low/Hi M.Avg #Crit C.Low/Hi C.Avg Crit%
Showmo 83411 98.35 % 2106/97 13/80 37.51 118 35/107 72.47 5.58 %
Showmo
Melee
13: 79
14: 109
15: 44
16: 13
17: 56
18: 61
19: 49
20: 24
21: 54
22: 41
23: 43
24: 27
25: 62
26: 43
27: 71
28: 26
29: 37
30: 25
31: 2
32: 2
33: 2
35: 2
36: 1
38: 101
+^ 39: 124
40: 24
41: 20
42: 31
43: 38
44: 30
45: 33
46: 29
47: 42
48: 28
49: 41
50: 35
51: 30
52: 39
53: 37
54: 29
55: 28
56: 30
57: 38
58: 38
59: 45
60: 22
61: 28
62: 24
63: 38
64: 37
65: 28
66: 20
67: 15
68: 8
70: 2
71: 1
72: 3
75: 1
76: 1
77: 3
79: 1
80: 1
Melee Crits
35: 1
38: 2
39: 4
40: 3
41: 5
43: 4
44: 3
+ 45: 8
46: 1
47: 2
48: 3
49: 5
50: 3
51: 1
52: 4
53: 1
65: 1
77: 1
78: 1
80: 2
81: 4
^ 82: 2
83: 4
84: 3
85: 1
86: 2
87: 1
88: 3
90: 2
91: 3
92: 1
93: 1
94: 4
95: 2
96: 5
97: 3
98: 1
99: 1
100: 2
101: 2
102: 1
103: 6
104: 2
105: 2
106: 3
107: 2Strike Shield (with Shield Skill +48 and back turned towards mob)
Code
Melee Damage Taken
Player Melee Dmg Melee % Hit/Miss M.Low/Hi M.Avg #Crit C.Low/Hi C.Avg Crit%
Showmo 11361 96.76 % 211/14 38/80 51.83 10 79/118 94.30 4.74 %
Showmo
Melee
+ 38: 14
39: 13
40: 4
41: 2
42: 7
43: 6
44: 9
45: 8
46: 9
47: 5
48: 7
49: 3
50: 8
^ 51: 7
52: 7
53: 6
54: 5
55: 4
56: 9
57: 8
58: 6
59: 4
60: 5
61: 7
62: 4
63: 7
64: 4
65: 9
66: 3
67: 5
69: 1
70: 1
71: 1
72: 2
80: 1
Melee Crits
+ 79: 1
+ 84: 1
+ 85: 1
+ 87: 1
+ 92: 1
+^ 93: 1
+ 98: 1
+ 102: 1
+ 105: 1
+ 118: 1Unblocks: 1,183
Block Rate: 43.83%
9.94% Increase
Strike Shield Data is invalid, see note above.
It's interesting to note that Aegis (95) appears to have roughly a 5% higher block rate than the Size 3 Koenig Shield.
While there is some variance with the block rate increase in each data set, I think that the increase amounts are close enough to each other to say that the size of the shield does not effect how much block rate increase you receive from shield skill. If it is effected by shield size however, the difference seems to be very small (Koenig Shield received a 10.5% increase while Januwiyah +1 received a 10.91% increase, which would only be a 0.41% difference from +48 shield skill).
If you average the 3 block rate increase samples together (excluding Strike Shield data), you get a +10.183% block rate increase from +48 Shield Skill. This would suggest that every 1 Shield Skill is equal to +0.21%-0.22% block rate on VT mobs, which would mean you would need +5 Shield Skill for every 1% block rate increase. A previous parse on a Decent Challenge Fear Dearg suggested that every +4 skill is equal to +1% block rate. This can be explained from 2 scenarios:
1) Level of mob has a big impact on how much block rate increase you receive from Shield Skill.
2) Variance skewed the data towards one direction
Also noting the previous and current tests:
On a Decent Challenge Fear Dearg, Aegis (95) had a 53.27% block rate with no equipment or Shield Skill + gear.
On a Very Tough Fear Dearg, Aegis (95) had a 45.37% block rate with no equipment or Shield Skill + gear.
The block rate difference is 7.9%.
According to Wiki, a Decent Challenge mob is 1-7 levels below you while a Very Tough mob is 4-7 levels above you. So the level difference was at least 5 levels apart, and at most 14 levels apart.
What do you guys think? Any ideas on how total Shield block rate may be calculated? The formula would likely only consist of your level, the mob's level, and your total shield skill (from base + merits + gear).
[+]
Nice testing. Feels like we're getting a little closer to figuring this stuff out.
About the size 2 tests. The Issue with testing size 1 or 2, is the low dmg reduction on block. When the reduction goes below 50%'ish, your blocked and unblocked hit's dmg starts to overlap. I'm not sure how to determine whether hits for the same dmg were blocked or unblocked in parse data. <,<;
Using the unblock parse you can determine that hits less than 38 Must be blocks. But that doesn't mean that all hits 38 or above Are Not blocks.
If you look at the huge number of 38 and 39 dmg hits compared to those of 40 dmg and up, it's very likely many of those hits are blocks whose base dmg was on the higher end of the dmg range. And even some hits above that may be blocks. Since with such a low def shield the dmg reduction could be very low indeed(-30%~40% dmg perhaps?)
This is skewing your block rate badly. So I suspect the block rate for a size 2 is most likely much higher than you're seeing.
For more accurate testing all I can think to do, is use a much higher def shield of the same size, and creed hands +2, to increase block dmg reduction. Then gather much larger samples, so the hit/dmg distribution can be better studied. Although honestly, I'm lacking the math skills to really analyze that kinda data. Without a clear block/not block gap, I get lost.
With skill vs block rate gain there could be a formula change at certain lvls of skill. Akin to the acc from combat skills after 200. Or, like you said, it could be a change based on mob lvl.
Regarding the effects of mob/player lvl. We Know lvl is a factor, but is it lvl difference, or just mob lvl vs skill? Does the player's lvl actually factor in, or is it just the difference in skill that makes being higher lvl help?
Hopefully, I'll know the answer to that soon.
I've actually got a Massive amount of parse data I collected recently. 25+ parse samples worth, at varying skill. with each sample being at least 3k hits. Usually more though. Mostly Ochain data, with a few size 3.
I lvl synced to 90 for the second half of the tests, and parsed samples with the same skill points as at 95. This should get us useful data on a lvl difference increase of 5 lvls while maintaining the same skill.
Considering how much data there is, I've been putting off compiling it till I get Excel reinstalled so I can use my shield spreadsheet. ^^;
I hope to have something put together in the next couple of days.
About the size 2 tests. The Issue with testing size 1 or 2, is the low dmg reduction on block. When the reduction goes below 50%'ish, your blocked and unblocked hit's dmg starts to overlap. I'm not sure how to determine whether hits for the same dmg were blocked or unblocked in parse data. <,<;
Using the unblock parse you can determine that hits less than 38 Must be blocks. But that doesn't mean that all hits 38 or above Are Not blocks.
If you look at the huge number of 38 and 39 dmg hits compared to those of 40 dmg and up, it's very likely many of those hits are blocks whose base dmg was on the higher end of the dmg range. And even some hits above that may be blocks. Since with such a low def shield the dmg reduction could be very low indeed(-30%~40% dmg perhaps?)
This is skewing your block rate badly. So I suspect the block rate for a size 2 is most likely much higher than you're seeing.
For more accurate testing all I can think to do, is use a much higher def shield of the same size, and creed hands +2, to increase block dmg reduction. Then gather much larger samples, so the hit/dmg distribution can be better studied. Although honestly, I'm lacking the math skills to really analyze that kinda data. Without a clear block/not block gap, I get lost.
With skill vs block rate gain there could be a formula change at certain lvls of skill. Akin to the acc from combat skills after 200. Or, like you said, it could be a change based on mob lvl.
Regarding the effects of mob/player lvl. We Know lvl is a factor, but is it lvl difference, or just mob lvl vs skill? Does the player's lvl actually factor in, or is it just the difference in skill that makes being higher lvl help?
Hopefully, I'll know the answer to that soon.
I've actually got a Massive amount of parse data I collected recently. 25+ parse samples worth, at varying skill. with each sample being at least 3k hits. Usually more though. Mostly Ochain data, with a few size 3.
I lvl synced to 90 for the second half of the tests, and parsed samples with the same skill points as at 95. This should get us useful data on a lvl difference increase of 5 lvls while maintaining the same skill.
Considering how much data there is, I've been putting off compiling it till I get Excel reinstalled so I can use my shield spreadsheet. ^^;
I hope to have something put together in the next couple of days.
Since there isn't a direct skill relation (like accuracy versus evasion), I'd wager that lvl difference sets the base block rate, and any skill over that lvl adds to that %.
(Mob lvl - Your lvl) = Base %. (Based on shield type).
(Shield Skill over Cap)/4.5ish = Added block rate.
Total - Base % + Added block rate.
I would assume however that the Added block rate could be negative if your skill is underleveled, and that the upper and lower bounds of the block rates are governed by set upper/lower block rates for each shield.
If someone could test block rate with like, 0 shield skill (abyssea burned pld), that would confirm whether the lower bounds are there regardless of skill lvl).
(Mob lvl - Your lvl) = Base %. (Based on shield type).
(Shield Skill over Cap)/4.5ish = Added block rate.
Total - Base % + Added block rate.
I would assume however that the Added block rate could be negative if your skill is underleveled, and that the upper and lower bounds of the block rates are governed by set upper/lower block rates for each shield.
If someone could test block rate with like, 0 shield skill (abyssea burned pld), that would confirm whether the lower bounds are there regardless of skill lvl).
Phoenix.Neosutra said: »
Since there isn't a direct skill relation (like accuracy versus evasion), I'd wager that lvl difference sets the base block rate, and any skill over that lvl adds to that %.
(Mob lvl - Your lvl) = Base %. (Based on shield type).
(Shield Skill over Cap)/4.5ish = Added block rate.
Total - Base % + Added block rate.
I would assume however that the Added block rate could be negative if your skill is underleveled, and that the upper and lower bounds of the block rates are governed by set upper/lower block rates for each shield.
If someone could test block rate with like, 0 shield skill (abyssea burned pld), that would confirm whether the lower bounds are there regardless of skill lvl).
(Mob lvl - Your lvl) = Base %. (Based on shield type).
(Shield Skill over Cap)/4.5ish = Added block rate.
Total - Base % + Added block rate.
I would assume however that the Added block rate could be negative if your skill is underleveled, and that the upper and lower bounds of the block rates are governed by set upper/lower block rates for each shield.
If someone could test block rate with like, 0 shield skill (abyssea burned pld), that would confirm whether the lower bounds are there regardless of skill lvl).
I'm unsure on how this works as I don't have access yet.
It used to indirectly let you do so but they changed it to automatically cap your skills when you change jobs iirc.
Ragnarok.Martel said: »
Using the unblock parse you can determine that hits less than 38 Must be blocks. But that doesn't mean that all hits 38 or above Are Not blocks.
If you look at the huge number of 38 and 39 dmg hits compared to those of 40 dmg and up, it's very likely many of those hits are blocks whose base dmg was on the higher end of the dmg range. And even some hits above that may be blocks. Since with such a low def shield the dmg reduction could be very low indeed(-30%~40% dmg perhaps?)
This is skewing your block rate badly. So I suspect the block rate for a size 2 is most likely much higher than you're seeing.
If you look at the huge number of 38 and 39 dmg hits compared to those of 40 dmg and up, it's very likely many of those hits are blocks whose base dmg was on the higher end of the dmg range. And even some hits above that may be blocks. Since with such a low def shield the dmg reduction could be very low indeed(-30%~40% dmg perhaps?)
This is skewing your block rate badly. So I suspect the block rate for a size 2 is most likely much higher than you're seeing.
Also look forward to your parses; hopefully someone here who is great with mathing skills can examine them all & come up with a working formula to calculate block rate. Pretty much everything, including evasion skill, has a formula change after 200 skill, so I wouldn't be surprised if Shield Skill mimic'd that as well.
If you study the way hit rate works (link), you lose 2% hit rate for every level the mob has above you. Perhaps block rate follows a similar level correction formula, where you lose 1% or 2% block rate per level difference?
I suppose, if we just wanted a 0 skill test then you could just test on a job that can wear shields, but has no skill. BLU, DRK, etc.
It'd limit what shields you could test(no Ochain/Aegis) but it would show if there's a block rate floor.
It'd limit what shields you could test(no Ochain/Aegis) but it would show if there's a block rate floor.
I'm trying to think of a single slot that would provide superior damage reduction via shield skill over just plain -DT%, if we assume 4.5 skill = 1% block rate increase.
6-7 skill on head versus 5% DT from Valhalla.
10 skill on back versus 5% DT on Molluscus Mantle.
10 skill on feet versus either 2% DT or 3% PDT..
7 skill neck versus 5% DT twilight.
I can't think of a single spot where shield skill would win for total damage reduction. Obviously it's a different story if you're wanting to up your block rate for other things (like shield mastery or spell interruption).
6-7 skill on head versus 5% DT from Valhalla.
10 skill on back versus 5% DT on Molluscus Mantle.
10 skill on feet versus either 2% DT or 3% PDT..
7 skill neck versus 5% DT twilight.
I can't think of a single spot where shield skill would win for total damage reduction. Obviously it's a different story if you're wanting to up your block rate for other things (like shield mastery or spell interruption).
In the case of Ochain, there may be something to be said for adding skill to cap block rate at 100%. That way you're getting your -65% dmg reduction 100% of the time. Then you stack PDT afterwards. Of course this may depend on just how much skill capping takes, and how much PDT it's costing.
But yeah, the increase per skill is just so damn small.. Need some gear with like +20 to a slot.
Edit. It'd be 12 on the feet vs 2%DT or 3%PDT though. not that that's going to change the comparison I think.
DMG reduction may turn out to be a cap PDT then skill kinda thing... <,<
But yeah, the increase per skill is just so damn small.. Need some gear with like +20 to a slot.
Edit. It'd be 12 on the feet vs 2%DT or 3%PDT though. not that that's going to change the comparison I think.
DMG reduction may turn out to be a cap PDT then skill kinda thing... <,<
Phoenix.Neosutra said: »
Since there isn't a direct skill relation (like accuracy versus evasion), I'd wager that lvl difference sets the base block rate, and any skill over that lvl adds to that %.
(Mob lvl - Your lvl) = Base %. (Based on shield type).
(Shield Skill over Cap)/4.5ish = Added block rate.
Total - Base % + Added block rate.
I would assume however that the Added block rate could be negative if your skill is underleveled, and that the upper and lower bounds of the block rates are governed by set upper/lower block rates for each shield.
If someone could test block rate with like, 0 shield skill (abyssea burned pld), that would confirm whether the lower bounds are there regardless of skill lvl).
(Mob lvl - Your lvl) = Base %. (Based on shield type).
(Shield Skill over Cap)/4.5ish = Added block rate.
Total - Base % + Added block rate.
I would assume however that the Added block rate could be negative if your skill is underleveled, and that the upper and lower bounds of the block rates are governed by set upper/lower block rates for each shield.
If someone could test block rate with like, 0 shield skill (abyssea burned pld), that would confirm whether the lower bounds are there regardless of skill lvl).
Reporting for duty
[+]
A thought occurs to me. I not sure you can really test with a burned 0 skill PLD. After all how do you keep from skilling up mid test?
Campaign mobs maybe? but I've no idea what lvl those are. It'd be nice to have that info.
But well, 0 skill vs whatever lvl C mobs are would still almost certainly floor block rate, to what ever the floor may be.
And this would be the only way to test base Ochain block rate. Assuming there is a base, and shield type doesn't just adjust the formula somehow.
Well, for the sake of my curiosity, test it anyway! Ochain info does want.
Hmmm, if you were feeling reaaaallly masochistic, you could skill up in intervals and test again. lol.
0 skill, 50, 100, 150, 200, etc. God.
It'd be horrible, but this could help us see if, and where the skill/block rate formula changes. Or if any changes are based on mob lvl not skill.
Campaign mobs maybe? but I've no idea what lvl those are. It'd be nice to have that info.
But well, 0 skill vs whatever lvl C mobs are would still almost certainly floor block rate, to what ever the floor may be.
And this would be the only way to test base Ochain block rate. Assuming there is a base, and shield type doesn't just adjust the formula somehow.
Well, for the sake of my curiosity, test it anyway! Ochain info does want.
Hmmm, if you were feeling reaaaallly masochistic, you could skill up in intervals and test again. lol.
0 skill, 50, 100, 150, 200, etc. God.
It'd be horrible, but this could help us see if, and where the skill/block rate formula changes. Or if any changes are based on mob lvl not skill.
My initial post exceeded the per-post character limit so I'll be posting summaries and conclusions in this post, and reserving the next two posts for supporting parse data.
Ok, first some details on the tests.
Determining mob lvl
Tests were done on Mourning crawlers in the tree. I used various lvls of acc gear, and /checked and determined that the crawlers had the highest evasion. With 461 total acc(had to wear a club) there was the occasional crawler that didn't check low eva, while the rest did. Adding even slightly more acc would tip those to low eva as well.
Based on this, and the GoV page that gives a max lvl of 105 for the 3 mob types in the area, I judged that these specific crawlers were the lvl 105 mob.
Yeah, it's an assumption. But I think it's a pretty fair one. And how else am I gonna tell. They cap exp at 600 even with 2 ppl(no chain). And I dunno if wiki exp calcs are accurate for telling mob lvl anymore(probly not).
PDT
Due to the need to survive such a high lvl mob kicking me in the face, for days on end, I used as much PDT as possible for each test. The low skill tests having the most, and high skill the least. This shouldn't affect the block rate info, but it may make things look weird if you compare the dmg taken between different tests.
Player lvl and shield skill
I did two main sets of tests. A range of skill lvls while PLD 95, and a set while PLD 90. Skill was mostly tested in increments of 5 skill.
Absorbtion
During some early test I made the mistake of wearing two pieces of creed+2. As absorbs can be capped like cures by hitting full hp, they are not reliable. And they ignore PDT, so the dmg range is different from normal hits. The absorbs have thus been ignored in the block rate calcs. The number of discarded hits is noted in each parse sample. The value is small enough that removing it should not significantly affect the acc of the tests. Note: I didn't realize this, but while Kparser lists absorbs in the def details it apparently does not add them to the hits taken total in the def tab.
Refer to the parse data itself in the following posts for any further details.
Summaries and conclusions
Firstly, Ochain.
Player level and block rate
The thing that grabbed my immediate attention here was the overlapping skill samples between the 90 and 95 tests. The 404~419 sets had nearly identical block rates, with less than 1% variance going either way. This pretty much confirms to me, that actual player level has no direct effect on block rate. It's only the higher skill caps, thus higher skill that makes gaining lvls help block rate.
This also means that we can get accurate lower skill tests via syncing, since lowering player lvl won't directly change block rate. Although syncing a shield like Ochain too far could be bad. A good chunk of its dmg- is from def. syncing too far could lower it into the ambiguous range where block/not start to overlap.
Skill and Block Rate, again
This leaves our shield formula with only two factors. Skill and mob level.
First, another look at skill using this data.
The 95 sample fits well with Skill Increase/5 = block gain. taking 29 skill to rise 5.95%
But the 90 sample is a bit on the high end for that model. 52 skill /5 should have been 10.4, but we have a 11.27 increase.
What I think is happening here, is that the skill modifier changes at 400 skill.
Below 400 skillgain/4=blockGain.
Above 400 skillgain/5=blockGain.
After checking some skill ranges, it seems to fit pretty well.
If anyone more mathy than I see's an issue with this, or has a better theory, by all means, let me know.
Speculation
This makes me wonder if a change doesn't occur every 100 skill. Or perhaps every 200. Starting off like the rest of the combat skills. Then getting hit again at 400 skill. Although I've no idea what the pre-200 mod would be.
Next
Now we're missing 2 pieces of this.
How to apply mob level.
And Shield size base block/block floor.
I'm not at all certain how to figure the mob lvl bit. So my next step will probly be 0 or low skill tests.
Size 3 tests
I did a very few Size 3 tests while I was living in the tree. <,< Mostly these just give us a clearer idea of just how Bad non-Ochain Block rate gets on very high-lvl mobs.
Gleaming Shield(Size 3)
*The lvl90 404 skill test is incomplete. Being a mere 900~ hits, I'd hesitate to draw any conclusions from it.
Hmmmm..
Edit:The previous 12.61 Dblock for the 404~456 was due to an error in my spreadsheet. this has been corrected. Thanks to byrthnoth for bringing this to my attention
Well, the first range fits the 5 mod like a glove, but the second..A lot of variance? Or conflicting data already.. Post edit: 0.72 difference is a bit better.
well anyway.
High level mobs suck
As you can see, High level mobs trash your block rate. In one of my previous tests, 369 skill vs a EM yielded a 45.06% block rate(size 3 shield). Going up 15 mob lvls ripped 22.45% block rate off me. Pretty much halving it.
Unfortunately, I don't have a EM vs 404 sample to compare. I do have 379, 389, and 399. So I may have to go do parses for those. could be useful.
But still..
22.61% block rate on a mob 15 lvls above my skill cap. That's... Baaad. I feel like I'm wearing a tower shield.
30.49% for one 10 lvls above my skill cap. Still baaaad.
Add 52 skill and you can get back up to 43.1% Basically how you blocked a EM with base skill. And wear ***gear to do it.
If VWNM's are anywhere near this lvl.. Well, it makes that spell interrupt build look pretty good. Cause you aren't blocking ***.
Oh, and to the ppl running around in Castellan's shields. You realize that thing's block rate is gonna be even lower than this, right? So stop it. The only time that things usable is with reprisal or palisade... maybe only with both. /rant off.
Aegis
As this is the first set of tests I know of with such high lvl mobs, we may be able to infer some useful info even for shields I didn't test.
For example, what Aegis' block rate might be like on current high lvl content.
I don't have an Aegis or I'd have done a full range of tests on it too. But thanks to Showmo's tests we know that Aegis has about 5% higher block rate than a size 3.
So using our earlier Size 3 numbers, and applying that 5%.
Projected Aegis block #'s
27.61% block rate on a mob 15 lvls above my skill cap.
35.49% for one 10 lvls above my skill cap.
Add 52 skill for 48.1%
That's.. still pretty bad. Well, hurray for fanatics drinks, right?
Now it is just a projection, but that's all I've got to work with atm. Someone with an Aegis go spend a few days getting face raped by crawlers, k?
Rambling on
Wow, I think that's pretty much all I can do with the data I've got. I feel like I just wrote a book. Definitely gonna get some TL;DR ppl from this. I tried to keep it to the point, but well, I meander. Like I'm doing right now. 'Sides, that was a freakin' ton of data.
Hopefully This post will be useful to people. And we can make more progress toward completely understanding shield mechanics.
I think next up will be the 0 skill and low skill tests. But I dunno when I'll get around to it.
Oh, and thanks to Keitaro, for letting me borrow his alt as a sync for like 3 days.
Ok, first some details on the tests.
Determining mob lvl
Tests were done on Mourning crawlers in the tree. I used various lvls of acc gear, and /checked and determined that the crawlers had the highest evasion. With 461 total acc(had to wear a club) there was the occasional crawler that didn't check low eva, while the rest did. Adding even slightly more acc would tip those to low eva as well.
Based on this, and the GoV page that gives a max lvl of 105 for the 3 mob types in the area, I judged that these specific crawlers were the lvl 105 mob.
Yeah, it's an assumption. But I think it's a pretty fair one. And how else am I gonna tell. They cap exp at 600 even with 2 ppl(no chain). And I dunno if wiki exp calcs are accurate for telling mob lvl anymore(probly not).
PDT
Due to the need to survive such a high lvl mob kicking me in the face, for days on end, I used as much PDT as possible for each test. The low skill tests having the most, and high skill the least. This shouldn't affect the block rate info, but it may make things look weird if you compare the dmg taken between different tests.
Player lvl and shield skill
I did two main sets of tests. A range of skill lvls while PLD 95, and a set while PLD 90. Skill was mostly tested in increments of 5 skill.
Absorbtion
During some early test I made the mistake of wearing two pieces of creed+2. As absorbs can be capped like cures by hitting full hp, they are not reliable. And they ignore PDT, so the dmg range is different from normal hits. The absorbs have thus been ignored in the block rate calcs. The number of discarded hits is noted in each parse sample. The value is small enough that removing it should not significantly affect the acc of the tests. Note: I didn't realize this, but while Kparser lists absorbs in the def details it apparently does not add them to the hits taken total in the def tab.
Refer to the parse data itself in the following posts for any further details.
Summaries and conclusions
Firstly, Ochain.
Code
LVL90 gear+ +0 +5 +10 +15 +20 +25 +30 +35 +40 +45 +50 +52 SKILL 369 374 379 384 389 394 399 404 409 414 419 421 Block% 86.2% 86.92% 88.29% 90.12% 91.16% 91.14% 93.59% 94.18% 95.12% 95.66% 97.47% 97.51% LVL95 gear+ +0 +5 +10 +15 +20 +25 +26 +27 +28 +29 SKILL 404 409 414 419 424 429 430 431 432 433 Block% 94.05% 94.73% 96.5% 97.1% 98.11% 99.51% 99.55% 99.77% 99.92% 100% Comparison Skill 369 374 379 384 389 394 399 404 409 414 419 421 424 429 430 431 432 433 90block% 86.2% 86.92% 88.29% 90.12% 91.16% 91.14% 93.59% 94.18% 95.12% 95.66% 97.47% 97.51% 95block% 94.05% 94.73% 96.5% 97.1% 98.11% 99.51% 99.55% 99.77% 99.92% 100% Difference 0.13 0.39 0.84 0.37
Player level and block rate
The thing that grabbed my immediate attention here was the overlapping skill samples between the 90 and 95 tests. The 404~419 sets had nearly identical block rates, with less than 1% variance going either way. This pretty much confirms to me, that actual player level has no direct effect on block rate. It's only the higher skill caps, thus higher skill that makes gaining lvls help block rate.
This also means that we can get accurate lower skill tests via syncing, since lowering player lvl won't directly change block rate. Although syncing a shield like Ochain too far could be bad. A good chunk of its dmg- is from def. syncing too far could lower it into the ambiguous range where block/not start to overlap.
Skill and Block Rate, again
This leaves our shield formula with only two factors. Skill and mob level.
First, another look at skill using this data.
The 95 sample fits well with Skill Increase/5 = block gain. taking 29 skill to rise 5.95%
But the 90 sample is a bit on the high end for that model. 52 skill /5 should have been 10.4, but we have a 11.27 increase.
What I think is happening here, is that the skill modifier changes at 400 skill.
Below 400 skillgain/4=blockGain.
Above 400 skillgain/5=blockGain.
After checking some skill ranges, it seems to fit pretty well.
Code
SkillRange Dblock% Dskill mod result 369~389 4.96 20 4 5 369~399 7.39 30 4 7.5 404~421 3.33 17 5 3.4 404~433 5.95 29 5 5.8
If anyone more mathy than I see's an issue with this, or has a better theory, by all means, let me know.
Speculation
This makes me wonder if a change doesn't occur every 100 skill. Or perhaps every 200. Starting off like the rest of the combat skills. Then getting hit again at 400 skill. Although I've no idea what the pre-200 mod would be.
Next
Now we're missing 2 pieces of this.
How to apply mob level.
And Shield size base block/block floor.
I'm not at all certain how to figure the mob lvl bit. So my next step will probly be 0 or low skill tests.
Size 3 tests
I did a very few Size 3 tests while I was living in the tree. <,< Mostly these just give us a clearer idea of just how Bad non-Ochain Block rate gets on very high-lvl mobs.
Gleaming Shield(Size 3)
Code
LVL90 Gear+ +0 +35 Skill 369 404 block% 22.61% 28.48%* LVL95 Gear+ +0 +25 +52 Skill 404 429 456 block% 30.49% 35.56% 43.1%
*The lvl90 404 skill test is incomplete. Being a mere 900~ hits, I'd hesitate to draw any conclusions from it.
Hmmmm..
Code
SkillRange Dblock% Dskill mod result 404~429 5.07 25 5 5 404~456 11.12 52 5 10.4
Edit:The previous 12.61 Dblock for the 404~456 was due to an error in my spreadsheet. this has been corrected. Thanks to byrthnoth for bringing this to my attention
Well, the first range fits the 5 mod like a glove, but the second..
well anyway.
High level mobs suck
As you can see, High level mobs trash your block rate. In one of my previous tests, 369 skill vs a EM yielded a 45.06% block rate(size 3 shield). Going up 15 mob lvls ripped 22.45% block rate off me. Pretty much halving it.
Unfortunately, I don't have a EM vs 404 sample to compare. I do have 379, 389, and 399. So I may have to go do parses for those. could be useful.
But still..
22.61% block rate on a mob 15 lvls above my skill cap. That's... Baaad. I feel like I'm wearing a tower shield.
30.49% for one 10 lvls above my skill cap. Still baaaad.
Add 52 skill and you can get back up to 43.1% Basically how you blocked a EM with base skill. And wear ***gear to do it.
If VWNM's are anywhere near this lvl.. Well, it makes that spell interrupt build look pretty good. Cause you aren't blocking ***.
Oh, and to the ppl running around in Castellan's shields. You realize that thing's block rate is gonna be even lower than this, right? So stop it. The only time that things usable is with reprisal or palisade... maybe only with both. /rant off.
Aegis
As this is the first set of tests I know of with such high lvl mobs, we may be able to infer some useful info even for shields I didn't test.
For example, what Aegis' block rate might be like on current high lvl content.
I don't have an Aegis or I'd have done a full range of tests on it too. But thanks to Showmo's tests we know that Aegis has about 5% higher block rate than a size 3.
So using our earlier Size 3 numbers, and applying that 5%.
Projected Aegis block #'s
27.61% block rate on a mob 15 lvls above my skill cap.
35.49% for one 10 lvls above my skill cap.
Add 52 skill for 48.1%
That's.. still pretty bad. Well, hurray for fanatics drinks, right?
Now it is just a projection, but that's all I've got to work with atm. Someone with an Aegis go spend a few days getting face raped by crawlers, k?
Rambling on
Wow, I think that's pretty much all I can do with the data I've got. I feel like I just wrote a book. Definitely gonna get some TL;DR ppl from this. I tried to keep it to the point, but well, I meander. Like I'm doing right now. 'Sides, that was a freakin' ton of data.
Hopefully This post will be useful to people. And we can make more progress toward completely understanding shield mechanics.
I think next up will be the 0 skill and low skill tests. But I dunno when I'll get around to it.
Oh, and thanks to Keitaro, for letting me borrow his alt as a sync for like 3 days.
[+]
Test Data
Ochain(90) PLD95Ochain(90) PLD90
Ochain(90) PLD95Ochain(90) PLD90
Test Data part 2
Gleaming Shield(size 3) PLD lvl 90&95
Edit:I found an error in my spreadsheet regarding the lvl 95 456 skill test. Data has been corrected.
Gleaming Shield(size 3) PLD lvl 90&95
Edit:I found an error in my spreadsheet regarding the lvl 95 456 skill test. Data has been corrected.
All FFXI content and images © 2002-2024 SQUARE ENIX CO., LTD. FINAL
FANTASY is a registered trademark of Square Enix Co., Ltd.