Originally Posted by Graecus
I would like to know why you chose the dagger build you did. I have always used the following:
http://www.wowhead.com/?talent=w0efooZ0eVbbEz0MMRtV
whereas you use something more inline with:
http://www.wowhead.com/?talent=i0gcoLZMIVbbEz0MMRtV
which I really don't expect to do more damage. Mostly based off my own experience, but also of course the original DPS spreadsheet. Is there numeric backing as to why yours will do more damage? You also seem to "hardcore" the original Assassination / Combat talents, whereas you leave the remaining to be changed. Could you provide an option to modify the talents a bit more?
|
Well, my initial comment would be: the two builds you describe are both described fairly well by the sheet. Yours only differs from what the sheet supports in that it has 3/3 Ruthlessness instead of 2/3 - everything else you can enter in just fine. And while that certainly is going to make *some* difference in terms of sustained dps output, I don't think it's crippling to the sheet - the answers you get will still be close.
In terms of the relative merit of the two builds: with the gear I will be wearing as soon as 2.1 comes out, the build you propose is only marginally superior against the targets that murder affects, and clearly inferior against everything else. In particular, pulling a point out of Lethality to get the last point of Ruthlessness drops me by ~3 dps both buffed and unbuffed. So while you can argue the merits of dropping Lethality in favor of Murder, I think it is correct to keep Ruthlessness at 2/3 - in which case, you're back in the realm of what the sheet can do.
As to allowing more talents to be changed: the fundamental issue is one of cycle determination. The rogue dps spreadsheet computes cycles by simply trying all possible cycles and figuring out which one happens to be best; the problem is, this creates all manner of issues with circular references once you deal with armor penetration procs and the like. This sheet takes a different approach; it takes the lessons learned in that sheet about which sorts of cycles tend to be optimal, and computes directly what the optimal cycle is going to be.
The problem with this is: cycles can reasonably be broken down into families (i.e. 5s/5r/Xe, Xs/5r, etc.) and while it is straightforward to compute the optimal member of a given family, it's extremely hard to compare across families. Hence, the talent selections are constrained so as to keep all possible answers in the same family.
In the case of swords, for instance: the calculations are built around using Xs/5r as the main cycle. However, if your energy regeneration is too low, this will not be sustainable. Hence, the energy-affecting talents (Relentless Strikes, Imp SS, AR, and Combat Potency) are all on the list of required talents; if you don't have these, your energy generation might not be sufficient to sustain 5s/5r, at which point you need to drop down to 5s/5r/3s, which the sheet just simply cannot handle. The other talents on the "required" list are there for similar reasons - it's because if they were allowed to vary, you'd risk moving out of the range of cycles the sheet is designed to handle. Hence, I just hardcoded them to what I believed to be the most common values, and let everything else vary.
Now, as for daggers: I haven't done much work with the cycle model, so I can't tell you for certain which talents are required - I just coded up 5/5/3 with the combat daggers build I use as a rough cut. Eventually, however, I hope to have a dynamically adjusted family of attack rotations for daggers as I do for swords, at which point what's required may change. Although, in all honesty, Ruthlessness has a fairly profound effect on dagger rotations, so that one is unlikely to be available for change.
Long story short: I made as many options as I could changable. However, the structure of the sheet coerces certain restrictions on talent builds, so I am forces to restrict some as well.