|
Su5/Divergence weapons - B path Subtle Blow II
Server: Asura
Game: FFXI
Posts: 1,454
By Asura.Pergatory 2025-03-27 13:56:36
Carbuncle.Maletaru said: »That's true, but are you prepared to take this all the way to the logical conclusion? Removing all forms of MA because they give too much TP? In theory you should use nothing but STP (because it feeds no extra TP to the mob) and have as-close-to-0 multi-attack as possible. Having delay reduction is increasing TP feed, are we going to use gear with +slow on it to reduce TP feed? I've had several scenarios against V25 Mboze and V25 Arebati (1st KI which is TP denial) where DDs ended up taking MA out of their subtle blow sets to give a more consistent TP feed because they kept having to turn around after a triple attack or it would get too much TP. (DNC often finds this impossible in Arebati without removing their MA. I think Shadowmeld was the DNC on my most recent clear and I think he removed all MA, feel free to fact check me on that one if you see this Shadow.)
So yeah it's not just theory, this is definitely what is optimal for TP denial strats.
Now as for delay and removing haste / adding slow, that's going too far. Just getting rid of MA so that you're consistently hitting once and only once is adequate. That said... if you're still feeding too much TP after removing all MA then this might be worth considering.
[+]
Phoenix.Capuchin
Server: Phoenix
Game: FFXI
Posts: 3,745
By Phoenix.Capuchin 2025-03-27 14:33:43
SAM DRK and WAR all rely on omen body to get respectable SBII without a weapon, though. Obviously you can buy it from trove, but there is some difficulty finding something like that exactly when you want it. It's also costly.
You can grab a Su4/Su5 weapon, buy heroisms, and be capped in 5 minutes with no luck or hassle.
I don't think everyone needs these weapons (honestly, most people don't need most of the gear they pursue), but they have a place. That place probably isn't for folks who already have masamune and dagon.
Not false that you can just buy a Su5 and heroism, but cost for that is over 100mil (assuming you just buy heroism and don't upgrade in Dyna). At that cost, maybe paying someone for Dagon does start to sound a little more appealing, and you can make use of that on multiple jobs.
I tend to think that people who care enough to be trying to intentionally put SB+75 sets together for that reason are likely the same kind of people who already DO have weapons like Masa, and at least are the more likely ones to have put some effort into obtaining Dagon (though yeah, that's easier said than done). So for them, the existence of B path isn't really that much better (if any better) of an option for a low TP feed fight than just going with a set with SB+65~70, better DPS, and attempting to not use excessive amounts of multiattack.
By Dodik 2025-03-27 15:05:00
I have Sam's SU5 B path as well as Dagon breastplate.
I much prefer the SU5 with a standard body when SB is needed than A. Nother Weapon that has to use Dagon body to hit cap.
That said, SB is never really needed on Sam and I only break that weapon out for funsies and when I want to do less dmg for whatever reason.
Don't use Sam on Mboze, that's a drk fight.
Asura.Sechs
Server: Asura
Game: FFXI
Posts: 10,323
By Asura.Sechs 2025-03-27 15:49:21
Interested in some math from the math pros...
[cut] I suck at math and everybody knows that, but I think there's a potential fallacy in your reasoning Capu.
Please allow me to elaborate.
Let's take into consideration Gandring MH and Kustawi+1 MH. It's a stupid option but I picked it simply because they have the same delay.
Same gear, same buffs, same setup, k?
Let's say with your current STP it takes you 7 hits to reach 1000+ TP, i.e. being able to WS.
It's the same STP, it's the same delay, so it's exactly the same amount of hits to reach 1000+, right?
Now the real difference might be that Gandring does that in like, what, 6 seconds, whereas Kustawi takes 8 seconds, numbers are just for the sake of making an example. But the point is it's going to be the same exact amount of hits.
What's the real difference then? Well, the TP generated on the enemy by your hits, that number will be of course 25% lower on Gandring compared to Kustawi, and that's it.
So yeah, your whole idea of "more hits generated over the same amount of time means more TP generated, despite higher Subtle blow!" doesn't make much sense, unless you're implying that on Gandring you intend to shoot WS at a higher TP threshold, that's a different story though, isn't it?
[+]
Server: Asura
Game: FFXI
Posts: 1,454
By Asura.Pergatory 2025-03-27 17:37:17
I feel like some confusion could be cleared up by clarifying what we talk about when we're talking about increasing or decreasing TP feed.
When talking about TP feed, it's usually in the context of the speed at which TP is given when attacking (amount of TP over a given timespan), not the amount of TP given per weaponskill used. This is because TP denial strategies rely on regularly draining an enemy's TP to prevent them from being able to use a move. Can we agree?
I understand that you can take the approach of "well if you're attacking too fast, you can just turn around, so what really matters is TP per damage dealt which will determine if you can kill something within the time limit" but who the hell wants to turn around every few seconds? I think the goal of a nice subtle blow set isn't to make it so you can fit enough weaponskills within the time limit of the fight, but to make it so you aren't having to constantly evaluate "gee have I fed it too much TP since the last drain? I better turn around just in case."
In my mind, a perfect subtle blow set for a given fight will not necessitate turning around during general combat. Sure you may need to turn around if, say, the BST is swapping pets for Killer Instinct, or maybe the SMN is reapplying a buff, such that there will be a gap in the TP draining. But barring that, the goal should be that you can just keep swinging. That's my 2 cents.
[+]
Server: Cerberus
Game: FFXI
Posts: 1,882
By Cerberus.Shadowmeld 2025-03-27 19:10:51
Carbuncle.Maletaru said: »
That's true, but are you prepared to take this all the way to the logical conclusion? Removing all forms of MA because they give too much TP? In theory you should use nothing but STP (because it feeds no extra TP to the mob) and have as-close-to-0 multi-attack as possible. Having delay reduction is increasing TP feed, are we going to use gear with +slow on it to reduce TP feed?
I've had several scenarios against V25 Mboze and V25 Arebati (1st KI which is TP denial) where DDs ended up taking MA out of their subtle blow sets to give a more consistent TP feed because they kept having to turn around after a triple attack or it would get too much TP. (DNC often finds this impossible in Arebati without removing their MA. I think Shadowmeld was the DNC on my most recent clear and I think he removed all MA, feel free to fact check me on that one if you see this Shadow.)
That's exactly what I had to do. Once I removed all that from my sets, we had a much easier time keeping Arebati under control.
I view path B as kind of a quantum mechanics kind of thing. You either care about the FUA or you care about the SB II on the weapon, but never both at the same time.
[+]
Lakshmi.Buukki
Server: Lakshmi
Game: FFXI
By Lakshmi.Buukki 2025-03-27 19:32:45
Let's say with your current STP it takes you 7 hits to reach 1000+ TP, i.e. being able to WS.
Sechs, why does how much TP you generate to be able to WS matter in this context? The argument Capuchin is making is "do the additional hits from FUA equate to overall higher TP Generation on the monster I am fighting vs simply using the highest Subtle Blow I can achieve with a better weapon and killing it faster". We should be talking about how much TP each weapon feeds over time, not how fast you can get TP to WS.
Now if the argument is "but more WS means the mob dies faster, which is the ultimate goal", then you would have to calculate things based on a few factors.
What's the real difference then? Well, the TP generated on the enemy by your hits, that number will be of course 25% lower on Gandring compared to Kustawi, and that's it.
There's more to it than that. Take for example a monster that has 1,000,000 health. Assume we execute a low TP feed scenario using two separate weapons with identical delay, and the goal is to reduce the risk of a mob getting a move off. One weapon is safer, the other is stronger. You'll kill it in either case, but there's risk-reward factor that should be taken into consideration for both options. Using Su5B, let's say your average WS damage is 25k and 5k white damage per cycle (so 30k damage per WS cycle), and it takes you 6 seconds for one cycle. Using a better <weapon> let's say your average WS damage is 35k and 8k white damage per cycle (so 43k damage per WS cycle), and it takes you 8 seconds for one cycle. In the first example, you'd need roughly 33 WS + white damage cycles to kill the 1m health monster, 200 total seconds. While the second example it would take around 23 cycles for the 186 seconds. (I'm using random numbers but they don't matter to illustrate the point). It's only a 14 second kill speed difference, but the "safer" weapon uses 10 more WS to get the job done. How significant is that?
Do the extra WS that I have to execute with Su5B to kill this thing feed more TP over time to the monster and raise the risk of allowing a TP move vs the increased WS/white damage with <weapon> using a slightly lower Subtle Blow set?
How fast do I feed TP to the monster with Su5B vs <weapon> to where the monster has enough TP to use a move and kill us? Is it every 5 cycles with Su5B vs 3 cycles with <weapon>? The stress factor plays a part in either case, but one scenario will be less stressful.
You guys are good at math and always post spreadsheets, it shouldn't be hard to compare weapons and determine this stuff like Cap asked. It's an interesting argument (you could say "14 seconds is insignificant" but that's not the point).
[+]
Carbuncle.Maletaru
Server: Carbuncle
Game: FFXI
Posts: 3,079
By Carbuncle.Maletaru 2025-03-27 20:06:44
Interested in some math from the math pros...
[cut] I suck at math and everybody knows that, but I think there's a potential fallacy in your reasoning Capu.
Please allow me to elaborate.
Let's take into consideration Gandring MH and Kustawi+1 MH. It's a stupid option but I picked it simply because they have the same delay.
Same gear, same buffs, same setup, k?
Let's say with your current STP it takes you 7 hits to reach 1000+ TP, i.e. being able to WS.
It's the same STP, it's the same delay, so it's exactly the same amount of hits to reach 1000+, right?
Now the real difference might be that Gandring does that in like, what, 6 seconds, whereas Kustawi takes 8 seconds, numbers are just for the sake of making an example. But the point is it's going to be the same exact amount of hits.
What's the real difference then? Well, the TP generated on the enemy by your hits, that number will be of course 25% lower on Gandring compared to Kustawi, and that's it.
So yeah, your whole idea of "more hits generated over the same amount of time means more TP generated, despite higher Subtle blow!" doesn't make much sense, unless you're implying that on Gandring you intend to shoot WS at a higher TP threshold, that's a different story though, isn't it?
This is literally just a copy of my first post on this topic so...I agree, but it's not 25% less TP, it's 50% less TP.
If you give the enemy 100 TP per swing naturally, then have 50 SB you give 50 TP. If you had 25 SB II, you give 25 less TP. (50-25)/50 = 50%
I see what you guys are saying about removing MA in favor of STP and sure, when it's a fair trade between the two I think it makes sense, but you're still missing my point. If your only goal is to reduce the TP you give the enemy per second and you're willing to completely ignore the damage you're doing to the enemy, then you should actively slow down your attack speed. This is functionally the exact same thing as removing MA, it's just reducing the number of attacks you have every minute.
Now, you could say that with DA you are getting those hits in more concentrated "bursts" which is...fair I suppose? But in terms of "amount of TP given to the big scary monster per second" removing all the DW from your equipment, wearing high delay weapons, and not applying Haste or Haste Samba are all effective tools you should be using to reduce the TP you give to the enemy.
What's that, you say, you have a time limit and need to still do damage at the same time? Well then I guess the amount of damage you do per TP fed is...kinda relevant.
FWIW I've done dozens of Arebati V25 attempts and none of the DNC in my group has ever worn the Su5 dagger or removed any MA from their standard DNC TP sets.
Server: Cerberus
Game: FFXI
Posts: 1,882
By Cerberus.Shadowmeld 2025-03-27 21:03:15
Carbuncle.Maletaru said: »But in terms of "amount of TP given to the big scary monster per second" removing all the DW from your equipment, wearing high delay weapons, and not applying Haste or Haste Samba are all effective tools you should be using to reduce the TP you give to the enemy.
When I removed all the MA from my sets in favor of TP it helped us due to the fact that I was no longer randomly spiking TP up to 3k and we no longer out paced the TP drain from our BST and SMN.
I agree with you to a point. It's not really about minimizing TP feed, it's about managing TP feed/damage within the constraints of how fast the BST and SMN can steal it.
I don't have any SU5 path B weapons except on cor. I would never swap out of twashtar or Mpu Gandring for the Setan Kober. I agree with you there.
When I'm on cor, and I'm using SU5 path B, there has literally never been a situation where I have cared at all about the subtle blow on the dagger. I'm using it because the FUA gets me to faster leadens.
[+]
Phoenix.Capuchin
Server: Phoenix
Game: FFXI
Posts: 3,745
By Phoenix.Capuchin 2025-03-28 02:37:40
Cerberus.Shadowmeld said: »I agree with you to a point. It's not really about minimizing TP feed, it's about managing TP feed/damage within the constraints of how fast the BST and SMN can steal it.
Yes, that. You worded it better than I did, thanks! Buukki's last post is also excellent and gets at a lot of what I'm thinking about.
Many of you are better at math than I am, and despite trying to get my head around all the intricacies of the stuff like the formulas at https://www.bg-wiki.com/ffxi/Monster_TP_gain, I'm not confident I am properly accounting for things. I also know there are a lot of complicating factors that can make it challenging to get great numbers. Depends on how many times the mob hits you, dAGI comparison between you and the mob, how much Subtle Blow you have in your WS set, whatever.
But let's just simplify things a bit for purposes of this comparison and say "What is the total TP fed to the mob in a 60 second period?" for two different setups for a DRK with no subjob (a job with no native muiltiattack, no offhand weapon, etc.)
1) Father Time B (a 513 delay weapon). Subtle Blow +75 set.
2) Apocalypse (chosen solely for ease of comparison because it's also a 513 delay weapon). Subtle Blow +65 set (i.e., Dagon + Niqmaddu)
Assumptions:
- Amount of damage you're doing is totally irrelevant for this exercise. I only care about how quickly the mob gets TP.
- Don't worry about WS sets having different SB for this comparison. Just assume you're either WSing in the same TP set, or don't even WS at all and just keep auto-attacking.
- I'll say we're getting max WHM Auspice (SB+29), so only need another SB+21 to hit SB1 cap. Easy peasy, let's go with Bathy Choker +1 (SB+11) and one Chirich Ring +1 (SB+10).
- Capped delay reduction (let's say we have marches and capped equipment haste).
- Fill our gear with non-MA pieces wherever possible. For the Father Time set that's gonna be every slot except the weapon itself (FUA+50%). For the Apoc, our two SB+II pieces do have some multi-attack (TA+5 on Dagon, QA+3 on Niq Ring) but otherwise can avoid multiattack. Or hell, just pretend DRK can get SB+65 with no multiattack if you want; the point of the exercise is more to compare mainhand weapons.
- Don't worry about mob TP generated by it hitting players. Just ignoring that for this simplified comparison.
So, how much TP have you fed the mob after 60 seconds with each weapon set?
Carbuncle.Maletaru
Server: Carbuncle
Game: FFXI
Posts: 3,079
By Carbuncle.Maletaru 2025-03-28 03:02:40
Since this is all %-based, the base number shouldn't matter, rounding aside. Say you normally feed 100 TP per swing for simplicity.
Father time swings 100 times (made up number) plus 50 FUA. Gives 25 TP per swing, total 3750 tp.
Apocalypse swings 100 times, gives 35 tp per hit, total 3500 tp.
Bonus, compared to 50 sb build (capped SB1, no SB2), you'd have 100 swings for 5000 TP fed. 2/3 as many swings for 1.4x as much TP fed.
If you're dual wielding and/or WSing at all, these numbers go even more in favor of path b weapons. The more multi-attack you have in your equipment, traits, etc, the better for path B.
These weapons will essentially always way more than cover the TP feed added by FUA. Maybe there's some extremely small corner case where it's not true, but I can't imagine what that is.
[+]
Shiva.Thorny
Server: Shiva
Game: FFXI
Posts: 3,072
By Shiva.Thorny 2025-03-28 07:29:04
Carbuncle.Maletaru said: »These weapons will essentially always way more than cover the TP feed added by FUA. Maybe there's some extremely small corner case where it's not true, but I can't imagine what that is. This is the answer to the question initially asked. It's been said at least 3x, with math supporting it from several different angles. There should be no debate on it. You cannot twist the math such that the FUA is more of a detriment than the SBII is a gain in terms of TP fed.
In practical situations, there are other considerations. Father Time being a scythe makes it so much weaker than GS options that you probably won't ever want to use it. The higher TP fed/time is worth it for the immense gain in WS damage.
You can make a great argument that many of the weapons are not worth it based on the fact that most are a substantial loss in DPS and you'd be better off turning X% of the time and using your best weapon. You can support it with math and logic [the math will vary with job, I'm not doing a dozen evaluations here].
The argument that FUA makes them a TP feed risk is just silly and unfounded.
But let's just simplify things a bit for purposes of this comparison and say "What is the total TP fed to the mob in a 60 second period?" for two different setups for a DRK with no subjob (a job with no native muiltiattack, no offhand weapon, etc.)
1) Father Time B (a 513 delay weapon). Subtle Blow +75 set.
2) Apocalypse (chosen solely for ease of comparison because it's also a 513 delay weapon). Subtle Blow +65 set (i.e., Dagon + Niqmaddu) For the sake of answering your question, 513 delay * .2/60 = 1.71 seconds/round. In 60 seconds, you have 35(.08) rounds. Each hit will give the mob 172 base tp. The father time will average 1.5 hits/round, for a total of 52.5 hits at 43 tp each (2257.5 tp). Apocalypse will average 41.65 hits at 60 tp each (2499 tp). So, in this case, Apocalypse gives more. If you delete the TA/QA from dagon/niqmaddu, it changes to 35 hits at 60 (2100 tp) and Apocalypse gives less.
This is a bad comparison for a lot of reasons, the biggest being that you don't care about total TP. You are constantly resetting, a 10 minute fight has ~30-40 windows for the mob to get TP before a reset occurs. Your concern isn't your average TP speed, it's your max possible TP speed: Quadruple attack is the biggest factor here. More in next post on why this matters.
Shiva.Thorny
Server: Shiva
Game: FFXI
Posts: 3,072
By Shiva.Thorny 2025-03-28 07:40:20
In a 15 second window, you get 8.77 rounds. Given players aren't perfect, you'll likely have some situations with 10 attack rounds between drains. Your concern in these fights is how much TP is fed in one of these windows, most of the time an extra swing or 2 means nothing besides that your drain will take more.
To model this, I used a simulation: Code local function TPTestApoc()
local tpPerHit = 60;
local tpFed = 0;
for i=1,10 do
local hits = 1;
if math.random(100) < 4 then
hits = hits + 3;
elseif math.random(100) < 6 then
hits = hits + 2;
end
for j=1,hits do
if math.random(100) ~= 100 then
tpFed = tpFed + tpPerHit;
end
end
end
return tpFed;
end
local function TPTestFT()
local tpPerHit = 43;
local tpFed = 0;
for i=1,10 do
local hits = 1;
if math.random(100) < 51 then
hits = hits + 1;
end
for j=1,hits do
if math.random(100) ~= 100 then
tpFed = tpFed + tpPerHit;
end
end
end
return tpFed;
end Code local results = T{};
for test = 1,10000 do
results:append(TPTestApoc())
end
table.sort(results, function(a,b) return a>b end);
for index,result in ipairs(results) do
if index < 10 then
print(result);
end
if result < 800 then
print('Total of ' .. index-1 .. ' results above 800 TP fed.');
break;
end
end
With apoc, dagon, and niqmaddu.. 1739 results out of 10,000 fed over 800 TP in 10 attack rounds. The highest results were 1380(once), 1320(twice), and 1260(7+ times).
With father time and nothing else, 86 results out of 10,000 fed over 800 TP in 10 rounds. The highest result was 860 (full 10 FUA and no misses), which occurred 10+ times.
Editing the simulation to remove niqmaddu, both the SBII and QA, gives 3638/10,000 results over 800 TP in 10 rounds.
You can tweak the numbers depending on what you think will cause the mob to WS. The point remains the same though: even if the difference in average TP fed is not particularly huge for SU5 vs other options, the difference in maximum and likely TP fed favors them by immense amounts. This effect is actually stronger if you remove other MA gear; the SBII gives a greater margin of error for procs.
tldr; SU5 feed slightly less on average. But, they drastically reduce the chances of having a particularly high feed round, which is what causes your loss.
Edit: Ran it again with more detailed output.
Apocalypse + Dagon + Niq Code 1440 TP fed 2 times. (0.02%)
1380 TP fed 1 times. (0.01%)
1320 TP fed 4 times. (0.04%)
1260 TP fed 14 times. (0.14%)
1200 TP fed 18 times. (0.18%)
1140 TP fed 32 times. (0.32%)
1080 TP fed 90 times. (0.90%)
1020 TP fed 133 times. (1.33%)
960 TP fed 277 times. (2.77%)
900 TP fed 622 times. (6.22%)
840 TP fed 549 times. (5.49%)
780 TP fed 1315 times. (13.15%)
720 TP fed 2292 times. (22.92%)
660 TP fed 256 times. (2.56%)
600 TP fed 3992 times. (39.92%)
540 TP fed 390 times. (3.90%)
480 TP fed 13 times. (0.13%)
Father Time Code 860 TP fed 9 times. (0.09%)
817 TP fed 78 times. (0.78%)
774 TP fed 397 times. (3.97%)
731 TP fed 1021 times. (10.21%)
688 TP fed 1854 times. (18.54%)
645 TP fed 2401 times. (24.01%)
602 TP fed 2202 times. (22.02%)
559 TP fed 1298 times. (12.98%)
516 TP fed 567 times. (5.67%)
473 TP fed 145 times. (1.45%)
430 TP fed 28 times. (0.28%)
Father Time + 30% DA Code 1204 TP fed 1 times. (0.01%)
1118 TP fed 1 times. (0.01%)
1075 TP fed 6 times. (0.06%)
1032 TP fed 29 times. (0.29%)
989 TP fed 128 times. (1.28%)
946 TP fed 269 times. (2.69%)
903 TP fed 654 times. (6.54%)
860 TP fed 1084 times. (10.84%)
817 TP fed 1503 times. (15.03%)
774 TP fed 1803 times. (18.03%)
731 TP fed 1722 times. (17.22%)
688 TP fed 1349 times. (13.49%)
645 TP fed 859 times. (8.59%)
602 TP fed 398 times. (3.98%)
559 TP fed 154 times. (1.54%)
516 TP fed 32 times. (0.32%)
473 TP fed 7 times. (0.07%)
430 TP fed 1 times. (0.01%)
Code local repetitions = 10000;
local results = T{};
for test = 1,repetitions do
local result = TPTestFT();
if results[result] == nil then
results[result] = 1;
else
results[result] = results[result] + 1;
end
end
local sorted = T{};
for k,v in pairs(results) do
sorted:append({TP=k, Count=v});
end
table.sort(sorted, function(a,b) return a.TP > b.TP end);
for _,entry in ipairs(sorted) do
print(string.format('%u TP fed %u times. (%.02f%%)', entry.TP, entry.Count, (entry.Count/repetitions) * 100));
end
Server: Phoenix
Game: FFXI
Posts: 223
By Phoenix.Gavroches 2025-03-28 13:00:33
Keep talking…. Extremely interesting
Interested in some math from the math pros...
People frequently talk about B path Su5 weapons as the ultimate low TP feed options, and neglect to consider a really important factor: they also have Follow-up Attack+50% on the same weapon, which means a whole lot of additional TP-generating hits. I think that a large part of why the B path weapons got SBII+25 is solely down to NEEDING that extra SB to mitigate the otherwise excessive TP feed from all those extra attacks (similar thought process to why they gave BST Tandem Blow trait when the master and pet are both attacking the same mob).
Capped SB+75 is great and all, and generates the lowest possible TP per hit. But who cares about that if you get "lucky" with 5 straight FUA procs and feed a dangerous mob enough TP to use some party wiping move before the next Mewing Lullaby or TP Drainkiss goes off?
Like, if I can build a SB+65 or SB+70 set on the same job without a ton of multiattack, isn't that preferable to a SB+75 set with Su5 and all of those extra FUAs? Especially considering that you may be able to use a stronger non-Su5 weapon, so the mob also dies faster (i.e., the less time it's alive, the less chance of a mistake leading to a nasty TP move).
So, I'd like to understand how it really breaks down for total TP feed over time. Does that data exist somewhere that I'm not aware, or can someone skilled with the numbers provide some insight?
|
|