Lukon

Hi all,

I have a few questions regarding haste modelled by the spreadsheet (for reference I am using version 2.2.4). My "Buffed DPS" sheet, from cells L31 to M35, is as follows:

Firstly, why is the blade flurry formula essentially 1.2 * SnD? It seems to imply BF is always active. I would have expected something like 1 + (0.2 * 15 / 120) = 1.025 * SnD. This cell is in turn referenced by the "Cycle Energy Damage" column of the "Buffed DPS Cycles" sheet (J23-J31). Here, it is present in the factor: bblfl-sndfact = 1.56 - 1.3, which again appears to assume BF is always active.

Secondly, why is average haste calculated to be 37%? With SnD and the Dragonspine Trophy alone I would expect at least 1.3 * 1.11 = 44% haste.

finally, concerning combat potency in cells J1-L2 of the "Buffed Cycles" sheet. The energy per second formula in cell L1 uses 'Buffed DPS'!$M$33 as the OH haste factor, which in turn points to the permanent haste value of 111%. Why doesn't combat potency take into account SnD and other haste effects?

 

Backpain

I could be wrong, but I think procs and buffs of this type are averaged out over the duration of the cycles and then converted to a permanant haste percentage.

For example, on-use attack power trinkets show up as adding a flat amount of AP, even though they are not active through the whole fight.





On another note, playing with the spreadsheet has me a little confused this morning. I have not capped hit, yet the spreadsheet tells me I would get more dps trading out certain items with hit, for others with greater stats/AP. (spymistress bracers vs bracers of maliciousness; veterans musket vs sunfury bow, etc)

I guess there are exceptions to the "+hit adds dps all the way up to the hit cap" rule? Perhaps I'm overlooking the ilvls of the items?

 

pf

Blade flurry is modeled assuming that it is always used in conjunction with SND. This might not be the case, but to model otherwise would be a pain.

Average Haste is calculated from taking the results of the cycle sheet. Look to that sheet to see if there might be some SND downtime.

How about searching for references to $L$1 throughout the rest of that sheet. You might find out why it is the way it is, and learn more about how the cycle sheet works. In short, it is done in this manner to prevent circular references as you cannot calculate snd uptime without knowing what cycle you are using.




Anyway, looking back through the last few pages, I find it odd that blizzard would adjust how sword spec is treated now. If they make it proc combat potency, then it is clearly superior to fist spec. Also, if it no longer has a swing timer reset for offhand/special procs, then its also clearly superior. Anyway, if someone decides to model the new version into the sheet, then special care would be needed to make sure its done correctly with the next-stat sheet, or you will end up with lots of bogus +1 values. Also, the +3% crit gem would not be hard to model, just make a variable and carefully replace all crit *2 values in the sheet. The hard part would be making sure to find them all as there are probably like 50 places where it is used, and no easy way to search for them as there was never a need to change it since our talents/gear previously had nothing of this sort.

 

Aiva

Hello,

Was playing with the spreadsheet a bit and got some questions :

1. Is it true that the +4 agi / +4 hit gem is best for all those red and yellow socket slots ? I was happy with my +8 hit ones :P

2. The spreadsheet shows the optimal snd cycle as 5s/5e for my build ( combat swords 19/42) , i am a huge fan of 3s/5s/5e cycle and wanna know what makes 5s/5e superior to 3s/5s/5e.


Thanks in advance.

 

Cloak

1) Regarding gems, it's what the spreadsheet tells you. I needed +hit all the way to the hit cap, and then started to focus on AP.

2) For one, you're using Eviscerate slightly more frequently with 5s/5e.

Secondly, there's no more than 10 wasted energy on finishers with 5s/5e while 3s/5s/5e will be wasting 25 energy per cycle on average.

Since they both keep 100% SnD uptime and basically similar frequency of Sinister Strikes, 5s/5e is clearly superior.

The more you know, the less you understand.

 

roq

The sheet by default has the hit cap set a little low. In my last post, i think it is on the last page, i have spelled out how to fix the problem.

 

Hagis

Which is better Raid DPS over all 41/20 or 15/41/5?

 

Dread

you could just download the sheet.. but 15/41/5 is superior.

 

Sokkou

nt,

 

Hagis

Where is this sheet?
Can I get a link?

 

• Aldriana

Page 1, first post, near the top.

 

Oklemus

Long time reader, first time poster.

The spreadsheet shows that I, in fact, do slightly more damage with Combat Fists than Combat Swords post-patch. (When comparing Gladiator's Fists to Gladiator's Swords.)

Given all previous information on this thread, I would assume the contrary to be true. Is there perhaps something I've overlooked?

I was actually in the process of saving up for the Gladiator's Slicer/Quickblade.

 

Gogge

I don't think the sheet has incorporated the latest sword specialization changes.

 

Tyranis

I have been pouring over the theorycraft this pass week and it really bugged me that my agi, stam and other stats wouldn't display as this did in game. So I started to pick apart some things and I think I have corrected a few known issues.

First off my MH the Runic Hammer was listed at 14 stam when it actually is 16 stam

Second my race NE has completly wrong stats, I wasn't able to test this for other classes since all of our 70 rogues are NE. Stats should be. Str - 92, Agi - 166, Stam - 91

Then I moved onto the talents. The known issue if they don't round properly, and not just not the enitre story. The fact is Vitatlity and Sinister Calling do not include socket bonuses or base stats in their calculation. So you need to take the sum of all your bonuses Agi or Stam (excluding socket bonus) at the 2% or 4%, then add in your base stats and any socket bonuses. Also it appears that Blizzard don't round the numbers up it just drops the decimal. Here are the forumla I used.

Agi
INT((SUM(E10:E70)*(1+Talents!H26/100)*(1+Talents!L17*3/100))+E72+E73)

Sta
INT((SUM(F10:F72))*(1+Talents!H26/50))+F73

One that was done I noticed my HP in game and on the sheet were different so I start to look into that. Taking off all my gear I found with 91 base stam I was only getting 730 increase health which negated 18 stam basically. So my true base HP was 3704. Putting on all my gear and calculating vitality I gained an extra 3730, then I added the 91 stam after subtracting 18 which appeared to nothing to get my total of 8164. Here's the math.

3704 Base HP + 373 Gear and Talent Stam * 10 + (91 Base Stam - 18) * 10

I fixed my version of the sheet to get the right number but more testing needs to be done to ensure all races have the same base HP and is the first 18 stam don't add to one's hp.

 

Cloak

Regarding Shiv viability, I realize this has been beaten to death but some new things have come up and there just seems to be a lot of factors that are in Shiv's favor.

A) The majority of raid bosses seem to have higher than 6k armor unbuffed.
B) Misery is pretty much a given in a 25 man due to the viability Spriests have.
C) Shiv cannot be parried or dodged.
D) SnD cycles have been revolutionized with the advent of Xs/5r. With Shiv, seemingly impossible cycles such as 2s/5r/Xe seem feasible. 31 energy Shivs offer a lot of flexibility.
E) Gladiator Ripper/Quickblade are now viable shivving alternatives with the 1.5 speed change.
F) Poison charges are infinite now.
G) There's virtually no preparation involved with SS => Shiv switch. DP-MH IP-OH is the staple setup for any raiding Rogue (once Anesthetic gets fixed).

If WF were not present, I'd gamble that Shiv is comparable, if not better, to use for raid cycles. Sadly, I have little excel experience so I can't tinker around with any possible Shiv cycles.

(Edit: Apparently D doesn't come after F.)

The more you know, the less you understand.

 

koaschten

You lost me on this point, please elaborate.

 

• Aldriana

Incorrect. Those are the base stats assuming Vitality. The correct base stats are Str 92 Agi 163 Sta 88.

It also might be noted that Vitality doesn't compute your total agi and then add 2% of it - it increases your base agi by 2% (dropping the decimal) and your bonuses from gear by 2% (dropping the decimal).

 

Cloak

Give it a shot. I've never tested it formally. But I can Shiv an evasioned Rogue unabated. I can also Shiv a deterranced Hunter with no parries. I know it doesn't translate like that, from pvp to pve, but I've never experienced a parried or dodged shiv on a mob either.

The more you know, the less you understand.

 

laforce

I tested my new trinket on the spreadsheet (Warp-Spring Coil) and it showed me LESS DPS than using Bladefist Breadth! So I thought: WTF?!

Then I went on the PTR to do some testing... my tests results were:

Test #1: http://www.lossendil.com/wws/?report=b4qr3w6uh1cns&m
Duration: 10:00 minutes
Trinkets: Bladefist's Breadth + Drakefang Talisman
DPS: 795dps

Test #2: http://www.lossendil.com/wws/?report=udu6fw4634brw&m
Duration: 10:04 minutes
Trinkets: Warp-Spring Coil + Drakefang Talisman
DPS: 813dps
Total procs of WSC: 9
WSC Uptime: 9 x 15sec = 135 secs..
% Time up: 135/604sec = 0.2235 x 100 = 22,35% of the time up.
Average armor reduction: 223.50 armor reduction.
Proc%: 9/(SS + Evis + Swordspec*) = 9/271 = 0.0332 x 100 = 3,32% proc-rate.

I performed my tests on a Servant of Razelikh, using a 5s/5e cycle, trinket (on test #1), BF and AR always when available, on both tests you can see that BF and AR were used equally (5 times and 2 times). Both tests were done using only poisons, Deadly on MH, Instant on OH.

*I also noticed that the trinket CAN proc off a Sword spec hit:

5/11 23:06:25.687 Your Sinister Strike hits Servant of Razelikh for 511.
5/11 23:06:25.687 You hit Servant of Razelikh for 172.
5/11 23:06:26.890 You hit Servant of Razelikh for 178.
5/11 23:06:27.312 You hit Servant of Razelikh for 360.
5/11 23:06:27.718 Your Sword Specialization hits Servant of Razelikh for 339.
5/11 23:06:28.125 You hit Servant of Razelikh for 192.
5/11 23:06:28.515 You gain Perceived Weakness.

Side note: There were NO missed hits on this mob WHATSOVER (well, he's like 15 levels behind me heh), so.. the +21hit did absolutely NO DIFFERENCE, and I still got +18dps using this rather than using Bladefist, can you imagine what difference would this make on a lvl 73 boss? So I think the spreadsheet is way off, since it gives me even less DPS with the WSC than with the Bladefist, but we do need more testing, this trinket is available for free on the PTR

And I say again, this results were against a low-level mob (lvl 57) and the +21hit did absolutely no difference, but against a lbl71-73 (bosses), would certainly do a lot of difference, that's around +1,3% hit... ;P

 

Elvenstein

Uhh.... the armor difference between a level 73 boss mob and a mob in blasted lands is quite drastic. Reducing 223.5 armor from a mob with 4500 armor after sunder and faerie fire is quite different from reducing 223.5 armor from a mob with like 1000-2000 armor in blasted lands. I don't think you took this into consideration.

 

laforce

Damn, I forgot that armor/dmg reduction is not linear.. that's fuckin retarded ;/

I guess the trinket sucks then, imagining that a lvl70 mob with 5000 armor (32,13% reduction) and with 1000 armor penetration, which brings it down to 4000 armor is 27.47% reduction... that's less than 5% more dmg..

On the mob Im testing, it's a lot more, since Im bringing his armor from, let's say 1500 to 500... 22,25% reduction to 8,70% reduction..

/cry

 

Lukon

The cycle sheet showed 100% SnD time. In this case I would expect (approximately) 11% from Dragonspine Trophy, 30% from SnD and around 4% from BF and mongoose, giving a total of 50% average haste. As a result the spreadsheet is underestimating procs including sword spec, poisons and mongoose by around 10%.

Ah, my bad. A little more investigation solved that one.

On that note, however, I did learn something interesting about the evaluation of hit rating. With my gear and talents and using standard raid buffs, the spreadsheet gives an AEP of 0.91 for hit rating. However as mentioned in the known issues section, the NextStat sheets don't fully incorporate mongoose and combat potency changes. By instead manually applying a +1 hit rating buff, the figure becomes 0.99 AEP.

But it gets better. The Dragonspine Trophy is modelled as a flat haste rating, so that the bonus effects of +hit rating are not registered. This means +hit is even more valuable, if you have this trinket. Additionally, as mentioned above, the value of procs seems to be underestimated. Taking all of these into account (the calculations are too cumbersome to repeat here) I came to a final figure of 1 hit = 1.10 AEP.

My point? A quick glance at the AEP section doesn't necessarily give the whole picture. For me, and perhaps others, +hit is significantly better than agility.

 

Sarlunas

Excellent thread, excellent spreadsheet, long time lurker... and so on.

Looking at the 100 dps Hyjal dagger: http://www.mmo-champion.com/images/n...ay/tracker.jpg

I just get this feeling that maybe Blizzard ought to have a look at the spreadsheet themselves, what are we supposed to do with a 1.5 speed dagger? Plugged it into the sheet and unless I've made some horrible mistake it appears to be a slight downgrade compared to sunblade, also a downgrade compared to malcha in mh (just to be sure). So is this supposed to be a Hunter weapon then for the stats? Couldn't they at least reduce the dps on weapons meant for Hunters like they do with caster weapons so that we don't have to commit suicide passing on a 100 dps dagger? *sigh*

Sorry about whining in this nice and usually whine-free thread. About shiv I've also noticed that it is never dodged or parried. One step further I've actually heard the parry sound while using shiv and it appeared to me that I got the energy refund (80%?) even though the shiv landed normally. At first I thought I was just seeing things but it repeated itself many times while I was grinding. Anyone else noticed this? This was a while ago so perhaps it's been fixed, looked interesting at the time though.

 

Telani

It comes out ahead by about 15 dps on my sheet versus sunblade. Maybe you were hit capped or (Something I've done) you forgot to input the average per hit damage of the weapon?

The dagger seemed pretty nice when I grabbed it on test (Ooh, check out the showoff), but I was also wearing some low hit gear (Yay for green gemmed T5), so the high hit on it helped more than it usually would.

 

Sarlunas

And there's the horrible mistake I spoke of, I indeed forgot that the hit cap on the sheet is quite a bit lower than in reality. Swapping out some hit the gain I see is only about 10 dps compared to sunblade but yes, it's definately a dps upgrade though some stam lost.

Nothing to see here, move along.