Quote:
Originally Posted by Ryba
[You must be logged in to view images. Log in or Register.]
That thin margin between outputs in the EQdpm chart is a result of a precise but inaccurate equation. Even though p1999 isn't using the live model, it is a decent approximation. So even though the EQdpm chart is wrong (and it is, e.g. damage bonus is off by up to 10), you can still see the weapon choices are roughly equivalent.
I also remember the (apocryphal) delay floor of 9, but since we are speculating, the rumor used to be that monk fists were the one exclusion to that limit. If this is the case on 99, epic fists trump tstaff.
I cannot imagine how you got a tstaff to parse at lower dps than a 9/20 1h. Also, not sure why you wouldn't factor an unresistable 120DD proc into your dps calculations. Can't really call my numbers suspect if this is what you're working with. |
EQdpm is irrelevant to this conversation. EQdpm may approximate how it worked on live, but has nothing at all to do with how it works on P99.
Further, EQdpm, Copeland's, and the rest of them didn't ever have access to live equations anyway and were trying to approximate what was going on in that world. A guess mapped on observed data. So that calls into question the precision and accuracy of their observations as well as the math they used to represent the replication of that data.
Therefore, I'm not calling your numbers suspect. I'm calling them non-existent. Until it's established that damage is calculated here exactly as it was on live (and that Copelands/EQdpm/et al was an accurate model - which was hotly contested in the Rangers Glade, Steelwarriors, the Safehouse, Monklybusiness among other places), EQdpm is useless. It's worse than useless when trying to compare two weapons (or weapon sets).
Finally, precision and accuracy are terms that are used in data
measurement (collection/acquisition), not in data interpretation (and the subsequent model building). So if you're talking about precision and accuracy with respect to EQdpm, you can only be talking about how its makers collected data use to come up with their approximation equation(s). If you're talking about the model itself, you're misusing the terms.
I never claimed I tested a T-staff vs. (only) the Sarnak Battle Hammer. I tested the T-staff vs a SBH in the main with an ADH in the off. What, did you assume that I'd put a GLS in my offhand for the parses? Seriously, though, the question's mostly been fist & good offhand vs. T-staff. I threw in the SBH as an alternative to the fist into the mix.
As far as procs go, recall the 60dd on the ADC when you're considering them. Further, even at (a too generous) 2 procs per minute (on each), the difference only works out to be 2dps. The reason that I leave out procs in my calculations is that they're too streaky to be considered when talking about dps (soloing I've had as many as 8 T-staff procs on a single mob - and gone some number of mobs in a row without a single proc - all hail the rng). If you're talking about DPM, I suppose I'd think about counting them. I'm wondering about sustained damage which is a different question from unreliable burst damage. YMMV, but whatever the case, I don't think it's enough to put it over the top. That said, the 2x ranger 625DDs deserve seperate consideration. So there's a threshold somewhere, obviously.
Finally mob level (and type?) need at least a nod in this conversation as do the previously mentioned ripostes (from the mob), your ripostes (or is this absolutely equiv curve to dps calculations?), and mob damage shields. Those should all be considered, at least, when choosing a loadout.
So the genesis of this all for me is that while I was leveling up, I thought the monks who had abandoned the T-staffs post-epic were nuts. So I decided to see for myself. My data showed that they were right.
I didn't mean to make this a pissy rebuttal to your post. I'm trying to understand the data that I've seen/am seeing and offering guesses about it.
This is what I guess so far:
1)- Speed doesn't kill the same way here as it does on live. (To wit, on live, an RFS owned an IFS by a wide margin - here not true.)
2)- H2H skill vs. 1hb (and 2hb) skill plays a significant role.
3)- There may be haste-percentage at which the in-game combat damage engine treats a 9/20 and a 9/16 as identical (save for the weapon skill check).