Leviathan.Comeatmebro said:
»Yea, I don't understand what saevel is trying to say honestly.
going to try lynx baghnaks, lowest delay i can get, if a cap is applied post-weapon delay you're looking at 510 * .2 = 102delay = 1700ms
if it's applied pre-weapon delay, you're looking at 480 * .2 = 96 delay = 1600ms
that alone makes the margin of error pretty small, while trying to measure for something like spharai is much bigger.. but i will do a larger test with lynx baghnakhs and 44 martial arts, capped magic haste since that seems to be what was requested
....
Pretty easy to see from the first log how it's not really random and it will quickly normalize, for those who couldn't understand the explanation. Since the server is calculating your attacks constantly, the time the packets may come in will vary, but you're guaranteed that they even out significantly faster than a true random distribution because they're representing a fixed speed. If any specific requests, will run them tomorrow. The numbers are consistant enough for me to be confident they're correct, but they don't seem to match our delay models.
So in all tests, you saw different numbers than what the formula predicted. Assuming that latency or plugin errors weren't the issue, and that the plugin was distinguishing single hits from multi-attacks, counters and kicks, it implies that there is an error in the formula.
To summarize, and keeping in mind 96 delay and 1.6 sec between attack rounds is the absolute H2H floor:
Tests 1 & 8) spharai, capped magic/equip haste with 210 MA:
formula predicts 120.5 delay (2.0107 seconds) but you see 2.09 seconds. That's still pretty close (off by 0.08 sec). The nice thing is that Comeatmebro got the same number on two separate occasions.
2) " " with 236 MA:
formula predicts 112.5 delay (1.875 seconds) but you see 2.016 seconds. Now we're off by a bit (0.141 sec). Could the MA part of the calculation be wrong, or is there a different floor?
3) " " " with 10% job haste:
formula predicts 77.34 delay (1.29 seconds) which is below the floor! You would expect the floor to be 1.6 seconds, but you see 2.013 seconds. That's the shortest interval yet, and implies the floor for Spharai might be 2.013, but since this is nowhere near 1.6 seconds, it suggests that the weapon you use might affect your floor regardless of the formula.
4) lynx baghnaks, capped magic/equip haste with 254 MA:
formula predicts 80 delay (1.33 sec), which is below the floor. Again, you expect the floor to be 1.6 seconds, but you see 1.759 sec. This isn't too far off, and is closer than you managed to get with Spharai. Since the weapon delay on the lynx is only 30, and is 116 on the Spharai, maybe there's some additional way that the extra delay is factored in, and is more noticeable the further you deviate from the base of 0.
5) " " with 210 MA:
formula predicts 93.75 delay (1.56 sec), which is also below the floor. Again, you expect the floor to be 1.6 seconds, but you see 1.76 sec, which is basically the same number you saw in the test above. That helps support that with the lynx, your floor is 1.76.
6)Spharai, capped magic/equip haste with 254 MA:
formula predicts 106.88 delay (1.78 seconds) which is close to the expected floor, but you see 2.016 sec, just like you did in test #2 with only 236 MA. In test #3 we saw that 2.013 might be the true floor for a Spharai when you should otherwise have reached the 1.6 floor. We weren't expecting to be below the floor in this test though, so its not surprising we're not at 2.013.
7) " " with 220 MA:
formula predicts 117.5 delay (1.959 seconds), but you see 2.016 sec, just like you did in test #2 with only 236 MA, and test #6 with 254 MA. It seems that adding more MA really doesn't help you past a certain amount, but since we know (assume?) MA doesn't have a cap, it means we are reaching an arbitrary floor despite the formula, and that we are reaching that floor sooner than expected (which puts into question whether the floor is linear, or happens in tiers).
Again to summarize:
-Lynx Baghnakhs (30 delay), when expected to be below the 1.6 sec floor, was actually 1.76 sec. That's the best delay we've seen yet (Comeatme hasn't tested barehanded yet, which might be nice just to prove 1.6 sec is still the absolute floor)
-Spaharai (116 delay),
---when expected to be above the floor by a lot (2.0107 sec, so 0.41 sec more than floored), it was actually pretty close to that calculated value (2.0899). This is the closest we've seen the formula hold true.
---When it was close the expected floor (1.78 sec, so only 0.18 sec more than floored), the delay suddenly jumped higher than expected (2.016). This value was still seen at slightly higher delay, when you were expecting to be at 1.875 and 1.959 sec.
--When it was below the expected floor (77.34 delay, or 1.29 sec), you didn't reach the expected 1.6 sec value, or even the 1.76 seen with the lynx. You saw 2.013 seconds.
If delay is actually floored separately for each weapon based on some inherent property (probably the delay itself, which is further multiplied into the formula somewhere), that's truly depressing for high-delay weapons. The fact that the same value was seen whether the delay was meant to be 1.78, 1.875, or 1.959 per the formula, suggests that it's working more like a tier system, and that the tiers are applied differently depending on the weapon (probably based on their delay, much like how the store tp tiers work).