When I set a 34 agi gem as the Preferred Dragon Eye, and a 34 hit gem as the Full +hit gem, the regemmer puts four Dragon's Eyes in my gear (some hit, some agi). If the Full +hit gem is set to a 20 hit gem instead, the regemmer works fine (which in my case means: it does not regem at all).

So obviously I'm not supposed to use a Dragon's Eye as a full +hit gem, but at times it may make more sense to use one (instead of say two Rigid King's Amber; which is what the regemmer does, even if only 34 hit from the cap).

The gear/gem combination this occurred with was somewhat similar to my current gear (as per armory). However I replaced Grim Toll with a non-hit trinket, and put another point in FA and used five Glinting Ametrines (10 agi/10 hit) instead of Deadly Ametrines to get back to 2 over the hitcap.

Please remove.

I've made a small re-upload of V92b fixing the shot rotation test issue. You'll have to input the wait time in milliseconds now (default 800 ms). Any imported profiles will have this value zeroed on load.

Issues in the latest version:

Regemmer switches out my Nightmare Tear in a Belt of Dragons to a Favor of the Dragon Queen, replacing a 12 AP bonus with a 8 AP one.
I have shot priority set to KS, RF, SrS, CS, AiS, AS, SilS, Readiness, and SS. The rotation test does not attempt to use Readiness after the first 'shot set', waiting 180 seconds until Rapid Fire gets popped again to use it.

Odd error... when loading item "Mask of Distant Memory" I received a run-time error across several attempts. When I selected "Custom" from the drop down list (before going to the custom tab and modifying anything), it loaded a Relentless Earthsiege Diamond into the Meta slot (which, also, didn't have a "None" option). Then when I went to manually enter the statistics, everything was already there except for the 2nd gem slot, which was red.

This is what shows up in the debugger: socketBonusValue = Left(socketBonusNode.Text, InStr(socketBonusNode.Text, " ") - 1)

Also, when messing around with the Shot Rotation tab, the analyzer claims the highest DPS for me would be Kill Shot > Serpent Sting > Steady Shot. Thoughts on this?

It does this so it can use Rapid Fire twice in a row the second time too. It should be doing this at the start as well.

Use Rapid Fire, Readiness to reset CD, use Rapid Fire again once the first runs out.

Repeat ~180 seconds after.

I think I found 2 bugs in the spreadsheet:

1st: The values of the serpent sting effect on the Shot Rotation tab are too low. The problem is in the Calculations tab cell H163. It doesn’t take into account any talents. That’s only a display bug however the calculations for the damage done with chimera shot are fine.

2nd: The calculations for the pet talent rabid are most definitely wrong. The spreadsheet says that the pet should have 5 stacks of rabid 87% of the time. I looked through the logs of our last week raiding and a lot of times it doesn’t even reach 5 stacks before it runs out. And it’s not only my pet. It’s the same for all the hunters I looked up on WoL

EDIT
-----

So, I did a quick rabid test today. Here is the combat log: World of Logs - Real Time Raid Analysis

I went to the hero dummy with my pet, disabled all attacks but rabid, and let it attack for some time. Here are the numbers:

Pet attack speed: 1.54
# of rabid cast: 45
# of rabid procs: 166

So, my napkin math is that my pet was 900 sec under the rabid buff. In 900 sec it cast 585 melee attacks. And 166 procs out off 585 casts is a chance of 28.38%. I knew that’s to low because WoL don’t count procs that only refresh a 5 stacks buff but out of 45 rabid cast my pet only gets to 5 stacks 16 times.

EDIT 2
-------

I did another very short test and this time with bite active: World of Logs - Real Time Raid Analysis

I used the lvl 60 dummy so my pet gets enough focus from GftT. The results are very similar to the test above. Even so my pet attacked more than 2 times as fast as before it only get to 5 stacks of rabid 5 out of 12 times. So it seems like bite does not affect the proc chance of rabid.

Pet attack speed: 1.54
# of rabid cast: 12
# of rabid procs: 43

Napkin math tells me that the rabid buff last 240 sec. That’s time for 156 melee attacks. 43 procs out of 156 melee attacks is 27.56%.

If I had to guess I would say rabid only procs from melee attacks with a proc chance of 30%

If I start out with a "blank" Nightelf level 80. I.e. No gear, no talents, no buffs.

I end up with a negative Base Critical Hit Chance (-3,86% in cell Calculations!B42).

Shouldn't the lowest crit chance I can have be 0%. Negative chance for something to happen seems a bit odd.

...as a followup.

If I equip only the BRK-1000 gun... and chose MM talents so that I reach the Piercing shot talent (Talents!H column from 2 and downwards with spaces to mark tiers: 032 230 2300 32, 113, 23, 100 00 000 0 0)

I get -0.57 "added" dps due to having the Piercing Talents at Calculations!N147

Which I think is due to the negative Base Crit chance

Did you change the level of your target to 80?

It's because of the crit depression applied.

Note that hunters _do_ have a negative base critical strike chance, which is made slightly positive by the contribution from our base agility.

I was formulating my response while you answered.

And yes, I have no problem with Hunters having a negative base crit level... since everything you list and calculate before the bottom line is "modifers" and a modifier might have a negative as well as a positive effect.

But the bottom line crit chance cant be negative it can only go down to zero.

The effects otherwize becomes a bit weird as in the Piercing Shot case with negative dps effect.

Ok, I agree it should be capped at 0%. I'll take this along for the next version. On a practical note that isn't a situation that's anywhere near the intended use of the sheet though.

EDIT:
In fact I'll cap at 0% lower bound and 100% upper bound (since the upper bound check is missing too). Also had to cap Viper/Hawk uptime because of it.

Currently both the Sample SV and Sample MM spec are the MM profile, btw. I was confused why armor piercing and hearty rhino were being used as survival til I checked the talents, heh.

When looking at the code for the rotation test. Even though "The Beast Within" is not selectable as a "shot" in the shot rotation table, there is code on a couple of places that checks the table if the shot type is "The Beast Within".
e.g.
... If ShotTable(selectedShot, 1) = "The Beast Within" Or ShotTable(selectedShot, 1) = "Bestial Wrath" Then ...

Being "or" statements it has no effect on the execution but it looked a bit odd.

In the calculation sheet cell K21 There is the following to handle the Marked for Death/Glyph of SS bug.

... should it be >= 1 or >0 instead of the current >1?

Feature request:

On the Overview tab, would it be possible to allow "Haste to Cap" to show negative numbers
like "Needed to cap hit"? (I'd like to see how much over the haste cap someone is)


Edit: spelling

The Heroic version of Armor of Shifting Shadows have "Colliseum-10" listed as its location. Should be "Colliseum-10 Heroic"

The Sentinel Scouting Greaves (non-heroic version) have haste 32 should be 39.

I was wondering if there is way to enable macros for a Mac-based Excel. It seems to disable them for me and I can't update/load/change anything. The .xlms and other recommended "Save-As" templates are only causing errors when trying to open them and not really solving any problems.

Hi guys

Sorry if this has been mentioned before but I can't seem to find anything about it. In a previous version of the spreadsheet there was an option to 'assume hit capped' or something along those lines. I personally think that's a brilliant idea especially the way spreadsheet seems to treat hit for me.

I am 72 over the hit cap and yet hit is, according to the overview tab, more valuable to me than agility or any other stat for that matter. I have tried everything I can think of to fix it. The only thing that seems to work is giving myself custom gear with ridiculous amounts of hit (actually it has to be > 10^19). Then the attribute value jumps from 1.6 down to zero and the Needed to Hit Cap value can't be dislayed (you just get a cell full of #####). Then ofc, the recalculate dps button needs debugging as the line from the regemmer macro can't be understood. Using the debugger to change hitToCap = 0 just broke the gear planner completely but using a negative number seemed to work. Using a large number in the order of 10^4 again broke the line but after testing with various different powers of 10 it doesn't seem to make a difference to the gear planner. Any negative number gives the same results. But not in the way I'd expect. I've ended up using -1 and -9999. Items with hit didn't become less of an upgrade in fact mostly the opposite happened. My neck, for example, [Sparkling Onyxia Tooth Pendant] was given a better delta value than before. Another interesting thing happened to the trinkets. Suddenly [Bandit's Insignia] and [Gnomish Lightning Generator] are the two BiS with undisplayably large delta values. Another thing to note is that although changing the line in the regemmer code does then allow the recalculate dps button to work, it does seem to have broken parts of what I assume are the regemmer. Basically the gear planner no longer shows 'Current Item with Regem'.

As far as I understand it hit vs. a lvl 83 boss isn't a soft cap. Once you have 0% chance to miss surely that's it, having more hit is useless. Ofc, I know very little about the inner workings of this spreadsheet, but I assume that these attribute values help calculate dps upgrades in the gear planner? So I thought you might like to know. Sorry if its working as intended and I'm actually an idiot but if anyone has the time I'd love someone to explain why this is happening... just for interest's sake.

Thanks

Well assuming you had the boss still at level 83 you need a lot of +hit to compensate the level difference. And it's not precisely a situation the sheet can cope with (or was designed to cope with).

Using it out of the bounds defined for use (i.e. being level 80 attacking a level 83 boss) is not something I've really kept up-to-date or bugfixed, sorry. It is very likely a lot of the calculations will bug out or show undesired results.