Don't you remember that they added a secondary subtle blow effect to AGI? It has its own 50% cap, which is easily reached by pretty much every DD job with just natural AGI by 80~85, applied separately from regular subtle blow. The two together result in approximately 75% reduction in TP/hit. Add to that Penance or Yurin, and nin and mnk shouldn't have any problem with feeding TP. I won't deny that it's incredibly easy to cap, given that my TP set on mnk has 47 subtle blow with only one gear change (torero torque). And that's without auspice.
waitwhat? The monster gets the same TP you do +3 (before Store TP) each time you hit it. Assuming Martial Arts VII+ average delay of ~+50 on the better h2h weapons, and for nin ~400 combined delay and DW +50% (DW5, AF body, Suppa, AF3+2 head)....
h2h: 4.8 TP/hit, thus 7.8 base given TP/hit; -75% before penance/yurin is 1.9 given TP/hit, 1.4 with penance (can't remember what yurin's inhibit TP is if it's different than penance)
katana: 4.3 TP/hit, thus 7.3 base given TP/hit; -75% before penance/yurin is 1.8 given tp/hit, 1.3 with penance.
Ignoring abyssea, the mnk would have 5~15% double attack (depending on sub, assuming brutal earring), and 22~25% kick attack rate (because I can't find anyone who agrees on what the KA2 % rate is, and accounting for AF3+2 set bonus procs), while nin would have presumably, 15~20% DA depending on gear and accounting for AF3+2 set proc, (I'm just spitballing because I don't know how used the double/triple attack gear is due to being gone for a while),
Say a 5 minute fight where all individuals have significant haste, ignoring the possibility of the NM having regain. Additionally, due to the fact that 1) no one that I've seen has actually pinned down exactly how agi effects subtle blow aside from the 75% net cap, and 2) it would apply to everyone in this situation due to level, I'm going to remove it from the equation (and thus use only actual subtle blow+), so it should be noted that the actual given TP/hits should be a bit lower than they are shown below. Comparing three sets: (Mnk and War are paired due to Counter and Retaliation, Nin and Sam due to lack of counter/retaliation, since the counter procs from Seigan/Third Eye and AF3+2 Augmented Yonin should cancel each other out) All will also ignore penance/yurin since those would apply to all melees. For the war and mnk, counter rate outside of abyssea with merits should be effectively the same as retaliation rate, and will depend heavily on the delay of the mob being fought, accuracy, shadows, and the player's ability to keep or lose hate. Thus, I've paird mnk and war together to more or less negate that variable (This isn't in depth analysis, I'm just trying to point out a baseline difference)
1: Mnk/nin to War/nin
2: Mnk/war to War/sam
3: Nin/war to Sam/war
Group 1:
Length of the Recorded Fighting time: 300 seconds
Mnk/nin: Capped Subtle Blow, TP/hit = 4.9, Given TP/hit = 3.9, lightly less than 2 seconds between attack rounds (should be ~106 delay at delay floor for 530 combined delay before MA traits and haste, unless I borked the numbers somewhere), so 169 attack rounds; 354 punches, 42 kicks (using 5% DA 25% KA), 406 total attacks, -5% due to acc ceiling, 385 connected attacks, for a total granted TP of 1501
War/nin: +20 subtle blow from /nin and rajas, 504 delay great axe will give 13.7 TP/hit, 16.7 given TP/hit, 13.3 given TP/hit. 504 delay at 80% haste is ~100 delay, 180 attack rounds, at a ~30% double attack rate (base+merits+gear, knowing that wars get DA2 from testing on BG) is 234 attacks, 222 successful, for 2951 granted TP
Group 2:
Mnk/war: Capped subtle blow, TP/hit = 4.9 given TP/hit = 3.9, 169 attack rounds, 10% DA difference should give ~430 attacks, of which 408 would be successful, for a total given TP of 1591
War/sam: losing the +15 SB from /nin, 16.7 -> 15.8 gTP/hit -> 222 successful attacks for 3507 TP
Group 3:
Nin/war: ~200 delay after DW, 4.3 TP/hit, 3.6 gTP/hit, capped haste, ~80 delay, ~225 attack rounds, 540 attacks with ~20% double attack, after 95% hit rate; 513 successful; total given TP of ~1846
Sam/war: 450 delay gkatana, 5 SB from rajas, 11.5 TP/hit (before Store TP), 13.7 given TP/hit, at haste cap and 20% double attack, 90 delay, 200 attack rounds, 240 attacks, 228 successful. No idea how to account for zanshin here, so I'll do this:
- no Zanshin procs or Zanshin allows the expected 95% hit rate: 3123 given TP
- Zanshin allows for higher than expected hit rate by 2.5% (234 successful): 3205 given TP
- Zanshin allows for a perfect 100% hit rate of the expected amount of attacks (240 successful): 3288 given TP
If the war or sam manage to cap subtle blow via auspice, rajas, /nin, and +10 SB from some other source, it comes out like this:
Before AGI:
Mnk: 1591 given TP (3.9 gTP/hit)
War: 1842 given TP (3.6 gTP/hit)
Nin: 1846 given TP (8.3 gTP/hit)
Sam: 1504~1584 given TP (7.2 gTP/hit)
...ok I see your point. With how easy it is to cap subtle blow now with Auspice and /nin, it does kind of steal mnk and nin's thunder, I'm also kind of surprised to see sam come out with less TP given than monk, though given the staggering difference the number of attacks, it does make sense. If I were to account for it, the AGI mod will effectively cut the given TP by about half, so that lessens the sting a bit but wow... I never noticed that before.