yah i am sure, i have added it alot of times. because i have 13(helm)+23(shoulder)+13(cloak)+15(wrist)+17(gloves)+20(belt)+25(legs)+16(feet)+10(ri ng1)+10(ring2)+22(trinket)+9(MH)+9(OH)+10(Ranged)+30(gems) and it comes out to 242. i added in items and a few gems and they all seem to work, but i am missing something somewhere. and i have no socket bonus' for hit.

Addition of Items: yah, i understand for you not putting them in, thats why i added them myself.

Stealther's Helm: helm of the claw i understand, but i solo'd my kara key, and just don't have time lately for instances. i will be getting helm of the claw hopefully soon.

Band of Anguish: i wear it and the slayer's mark of redemption.

Cloak of the Craft: i use because its one of the top according to roguespot's pve armor chart only boe's above it are one from Doomwalker, and vengeance wrap. and then the 60 badge item.

Because of how easy the developers made this spreadsheet to add and remove items, theres no reason for you to do it, thats why i asked for a how-to, so i could tweak it for my own uses. I don't expect anyone to do the work for me, i'd rather do it myself. Its a great spreadsheet tho, my only problem is the 6 hit, so i was asking if anyone had any suggestions. and I am greatful for all the helpful people on this site.

Admittedly the relative values of these things change somewhat with gear, but off the top of my head, Blood Knight War Cloak, Auchenai Death Shroud, Vengeance Wrap, and Dory's Embrace usually all score higher than Cloak of the Craft. And next patch, Black-Iron Battlecloak will be BoE, and Cloak of the Coming Night will become available, both of which are also better.

Anyway, regarding the hit summation: my guess would be that the problem is something to do with gems, as adding gems has some subtleties to it which makes it easy to do wrong. That said, I have had the experience lately wherein I added up a set of gear by hand and got a different answer than the spreadsheet, but I haven't had time to figure out whether it was my addition or possibly some minor bug in the spreadsheet. Thus, I can't totally rule out an error in the sheet.

Best way to diagnose it is probably to look up the value of all the components of your hit rating, and see which one disagrees. If you could pull up the values for Ahit, Whit, Thit, EHit, GHit, and BHit, that would probably shed some light on the situation. If you're not clear on how to do that: just put =AHit in a cell somewhere, and record what the value is; do the same for each of the other values. These correspond to hit from Armor, Weapons, Trinkets, Enchants, Gems and Buffs, respectively, which will allow us to narrow the problem down.

well, now it is reading correctly, i deleted some of the ones i added and re-added using a diff item to copy since saw some 0 in the active slot, not sure where the missing 6 went or whatever, but its correct atm.

Cloaks: According to the MAEP chart, the blood knight war cloak is worse by 3 MAEP. Auchenai Death Shroud(sorry i will never do heroic crypts again), and not sure if cloak of coming night will be better, since it is -1Hit, +18 AP, -25Agil, +18Stam. i would assume the 25 agil would still make cloak of the craft better, but i am unsure. and Dory's Embrace is 60 badges, i have just hit 70 like, 3 days ago so i've been trying to get enough hit rating where i won't be dead last in a kara run.

Thank you all who answered, i appreciate your time and help in answering my questions.

Brood Knight War Cloak: it might be 3 MAEP behind, but in practice that's only about 2 DPS; and 21 stamina for 2 DPS is not an unreasonable tradeoff, as, despite what some might lead you to believe, stamina is not useless.

Auchenai Death Shround: Well, if you don't want to go where the good loot is, that's certainly your option. But I'd encourage you not to swear off the instance forever just because you had one bad group there; with a good group, it's not too bad an instance (and I don't mean "it's not bad in full T6" - I mean that back in Karazhan days, it was one of my guild's major badge-farm instances since it was quick and not too hard).

Cloak of the Coming Night: Don't forget the Coming Night also has 18 haste on it; in terms of ballpark figures, note that 2 AP ~= 1 agi ~= 1 haste ~= 1 hit, so trading 25 agi + 1 hit ( = 52 AP) for 18 haste and 18 AP (=54 AP) and picking up 18 stamina in the bargain isn't the dumbest thing you could do.

But, yeah. Cloak of the Craft is probably reasonable for now - I'm sure you have other slots with greater need of improvement. My point is not so much that you have an urgent need to replace it as simply noting that there's a reason it was omitted from the sheet in the first place. In fact, I might not have even bothered bringing this up, except that people's fascination with Stealther's Helm of Second Sight really bugs me.

Edit: It might also be noted that Agi is probably not as valuable for you as Shadowpanther's MAEP list makes it out to be; I believe that table assumes you have Blessing of Kings, which is not usually the case in Karazhan; thus, the value of Agi is 10% lower than that chart indicates, meaning any item with it will tend to be overestimated.

ok, thank you for your information, since i'm still learning my rogue again since i just brought him back since pre-bc. i am not sure if it assumes BoK. but i will have to check into that, and your right, the cloak is not the first thing i need to upgrade. i am right now working on my hauberk of karabor since its better then what i use atm.
Thank you for all your advice and help.

Is the lack of a Shard of Contempt choice in the spreadsheet awaiting real modeling from 2.4?

I have a two-part question regarding how we can interpret/use the theoretical DPS (tDPS) that the Gear (and DPS) Spreadsheet provides us after inputting our gear/talents. As stated oftentimes in this and other threads, the spreadsheets model an infinite, uninterrupted fight with a stationary boss.

Part 1 of my question is about assumptions: would it be accurate (or accurate enough) to assume a "theoretical damage done" (tDD) number for specific chunks of time within the undefined "infinite" of the spreadsheet's modeled fight?

In case that didn't make any sense, an example:

Let's say I end up with gear+talents that give me 1000 tDPS (for ease of math). Can we assume:

60,000 tDD in a 1-minute fight
600,000 tDD in 10-minute fight

Further, is it okay to assume that any 60-second segment of said 10-minute fight would equal 60,000 tDD (30 seconds equals 30,000, etc.)? That is, from 1m55s to 2m55s one did 60,000 tDD (at 1000 tDPS)?

The reason I ask is that in the discussion of modeling interrupted fights (and thinking about other things, like trash or solo work, and how to apply the spreadsheet data to those situations), by determining the total damage one would do over the course of an uninterrupted, stationary fight, you can extrapolate how much more dps that same rogue would have to do to do the same amount of damage in an interrupted fight. While that isn't the same thing, could it still be useful?

For example, with the same numbers: In a 5-minute spreadsheet-style fight, Rogue A does 300,000 tDD. In a second 5-minute fight, this time with a boss mechanic that forces a 12-second interrupt every minute, Rogue A would do 240,000 tDD (tDPS*300 - tDPS*60). To maintain the same amount of tDD, rogue A now needs to do 1250 tDPS. Of course, this all only applies to theoretical numbers, and we all know that any given minute's dps can be boosted or lowered in real-time with cooldowns, etc.

So, the second part of my question is: Can there be any applicability to looking not just at tDPS, but tDD, for maximizing performance? In the given 5-minute fight example, can one say that because of the interruptions, you need to do 125% of your tDPS for 80% of the time?

I imagine that rogues like Ald, Vulajin, DDM, Latito, Hanos, etc. could (and probably already) do that. Does this touch more on strategy and playstyle than theorycrafting? I don't think that this is revolutionary math, obviously, nor am I quite sure its about the math at all (as the only thing one needs to use is tDPS*(length of fight) - tDPS*(length of time lost to interruptions), but I have noticed, in reading the many threads dedicated to roguecraft, the focus is on DPS and its increase. Is there a place for thinking about or calculating how to maintain the same level of performance in more adverse situations? Can min/maxing for shorter time periods result in higher dps overall?

So, it's not clear to me how useful that number is. I mean, you do... however much damage you do... in a given amount of time. If your theoretical sustained DPS is 1000 as in your example, it's not like you can just "decide" to start doing 1250 DPS 80% of the time such that when you're unable to attack (during the other 20%) your damage still averages out to the same thing. It'd be great if you could, of course - but it just doesn't work that way. In practice, you do... however much... DPS during uptime, and that lowers your overall DPS average due to doing none during downtime.

Now, in terms of applying sustained theoretical DPS (tDPS) to interrupted real DPS (rDPS): the tDPS value applies, modulo 2 constraints:

1) Cooldowns. The uptime on AR, BF, Berserker's Call, etc. are all measured in terms of average uptime for endless combat. Blade Flurry is up 15 seconds out of every 2 minutes, or 12.5% of the time - period. Thus, any arbitrary minute of combat will have higher or lower DPS depending on which cooldowns happen to get blown during it.

2) Edge effects. So, the endless combat model assumes that your cycle is stable, you're spending energy (on average) as fast as you get it, and so on. Now, in the middle of any uninterrupted patch of combat, there's probably times when this is true; if you're attacking a boss for 4 minutes uninterrupted, and you look at the middle one minute of that 4, your average DPS is probably fairly close to the spreadsheet prediction. However, the average of the first minute will be different, as you have some extra amount of energy to spend (as you start with 100 energy).

Fundamentally, modeling arbitrary interrupted fights is going to be very hard, as every fight has a different interruption profile; you'd basically need a different spreadsheet for every fight (which I am *not* going to do - although there is an interesting question around whether this might be possible using modules in a programming-language implementation - but that's a story for another time).

Now, what *might* be possible/interesting to model is the notion of fights with duration. That is, we'll no longer look at a fight that's infinitely long and uninterrupted; instead, we'll look at an uninterrupted fight that's 2, or 5, or 10 minutes long. This is actually a somewhat solvable problem: one would simply optimize the first and last 30 seconds separately from the sustained portion, and allow one to come up with a real model for a fight of a given length. Note, however, that there are significant and important differences between this and the interrupted model; a fight that last 2 minutes is not the same thing as a 2 minute uptime period in a 10 minute fight (in a 2 min fight, you want to end with no combo points; and an interrupted fight, you want to end with at least a couple. In a 2 min fight, you want all your DoTs to end the instant before the fight does; in an interrupted fight, you can leave them ticking. In a 2 min fight, you start with no CP and need to ramp up your SnD; in an interrupted fight, you can use saved CPs to get kickstarted... and so on).

So, fundamentally: I think a finite-length fight model is probably not unreasonable to try to put together (not that I'm saying I'm going to do so - it'd be a *lot* of work) - and there would probably be value in such a thing. But modeling interrupted fights is fundamentally intractable with the framework we have in place.

Would it be possible to add the Might of the Scourge (14crit 26ap) shoulder enchant to this spreadsheet?

hi guys,

could you somehow add setup into the dps sheet? although it is not really a dps talent it generates cps on full resist/dodge/evade. this come in advance for deep subtley builds with enveloping shadows skilled or with active cos on encounters with aoe.

Re: Might of the Scourge. Yes, it's possible. I'll do it if I remember the next time I'm working on the sheet.

Re: Setup. The issue is that on a general-case fight, this provides basically no damage. The contribution of it ranges from 0 on many if not most fights up to "quite a bit" on some like Mother. So assigning a rate of combo point generation to it is... tricky. Thus while I'm not totally averse to the idea, it's the sort of thing that would be very hard to implement in any meaningful way, and risks breaking other things in the sheet. So for the moment I'm disinclined to worry about it.

In other news, I took another look at Rawr, and what it boils down to is: there's a lot of nice features there, that would in some ways simplify the calculation process; but, on the other hand, there's a lot of not-so-nice things about it - missing features, UI choices that I don't care for, and so on. Fundamentally, I think it's a great tool for doing sort of a simple high-level exploration of things, but is not-so-good at doing deep and subtle analysis with; and I think we're far enough along in the realm of theorycrafting that the deep and subtle is what we need more than anything right now.

This is not to say it couldn't be fixed, of course; but I'm not convinced that doing so would be any less work than starting from scratch, so at this point I'm somewhat disinclined to work within that framework. I may yet change my mind, of course, based on future development within that project; but that's the way I'm currently leaning.

Yeah, that sort of comes back to the "the whole buff sheet needs a revamp" point; regrettably, I don't think I'm going to have time to do much about that anytime soon. I have a pretty good idea what I want, I just don't really have time to do it. If some reasonably spreadsheet-savvy person would like to take a crack at it, let me know and I'll explain what I have in mind.

I'm certainly interested in helping with this -- if you would like to send me a list of changes you've collected that would be great; a sketched format would be even better.

The reason I am posting this instead of sending a PM is that I have an additional idea for the buffs sheet that I'd like feedback on. My idea is to group the buffs according to whether they are raid-wide (i.e. improved faerie fire) or party-only (battle shout, totems, etc.) or consumables (Drums, potions, elixirs). Then, under the party-only buffs, allow users to construct the classes in their desired raiding party from a drop-down menu, which would automatically select/deselect or even hide/unhide the appropriate buffs. The user would then just have to choose whether or not the buffs used are improved by talents. Thoughts?

Ald, if you want to PM me your ideas I can start taking a crack at it.

I'm using the 0.9.5 beta version of the sheet and haven't seen this mentioned since you posted that download. I currently have the Skulker's Greaves with three 8hit gems in them and the cobrahide enchant (non-epic) and the spreadsheet lists my dps as 1498.72. When I put on the Trouser's of the Scryer's Retainer my dps jumps to 1556.58. It now only displays the 2 sockets this item has and they remain 8hit gems and the leg enchant has not changed. What is odd is that when I put on the Legging's of the Immortal Night my dps then DROPS to 1547.4. This shows the three sockets for the item and they are all 8hit gems and again, the leg enchant remains the same. Is this a bug in the calculations for the new items? Or am I missing something obvious?

Summary
Skulkers - 1498.72
Scryer's Retainer - 1556.58
Immortal Night - 1547.4

It does show that the Immortal Night legs are an upgrade, as it should, but it seems to be missing part of the stats for the dps calc for these pants. Is the armor pen not being added into the dps calculation?

Note: I am never over the hit cap in any of these scenarios.

EDIT: Spelling and additional info

Ald, fyi, I'm starting to work on a new Mutilate model for the DPS spreadsheet, and if you ever get around to implementing Mutilate for your spreadsheet you may be interested in looking at it. Basically, it uses a state/path model with a probability for each path/outcome. I'm only in the beginning stages now, but here it is: http://elitistjerks.com/f31/t19926-r...23/#post680673.

I think I'll actually post my ideas here, instead, such that people can comment and propose alternate ideas.

So, the major problem in need of fixing is the fact that having separate entries for "improved" and "nonimproved" buffs is confusing and hard to use. The secondary problem is that yes/no dropdowns are a pain in the you-know-where. Thus, the proposed change is to replace the buff selection for a given buff with a checkbox (similar to what's used in zone selection right now) and one or more dropdowns to indicate any additional parameters of the buff. For instance, Blessing of Might would have a dropdown with the values 0 through 5, for number of points in Imp Blessing of Might; Battle Shout would have a similar dropdown for Commanding Presence, and a second one for Solarian's Sapphire. Drums would have a 0-4 dropdown for number of drums used per 2 minutes. And so on.

In a totally ideal world, these boxes would be linked, so as to automatically enforce allowed combinations. That is, only one of the 3 food boxes could be checked; if you had Talbuk selected and you clicked Warp Burger, it would select Warp Burger *and* deselect Talbuk. Not sure if this is possible or not, but it'd definitely be useful.

The issue I haven't yet resolved is how to label these boxes. Clearly each dropdown needs some help text associated with it; simply having a row that says "Battle Shout" followed by a checkbox, a 0-5 dropdown, and a yes-no dropdown wouldn't be very clear. As such, somewhere in there - either next to each dropdown, or above it, or below it, or *something*, there needs to be labels saying "Imp BS" and "Sapphire" - or such such thing. I'm not sure what the best way of arranging this is.

One thing that might help with this is to move the buff computation and the buff selection on to different pages. Just like the talents_equipment page performs lookups against, the armor, weapon, etc. pages, one could have a very clean, parameter-free selection page where you just select buffs, and then a modified version of the current page would read off which buffs were selected and add them up - basically, it would read off the checkboxes and dropdowns on the selection page rather than off the calc page itself. This would give a significant amount of horizontal room to arrange stuff, and might resolve the issue in question (see attachment for a possible arrangement under this theory). Depending on how big the buff selection turns out to be, it could either be put at the bottom/side/whatever of the talents_equipment page (which strikes me as better if it can be done in an elegant way), or given it's own; in either event, the buff calc page (which is probably just a modification of the existing page) is hidden.

Secondly, of course, there's a few new buffs - from MS warriors, Ret Pallies, twisted GoA, etc. - that need to be added. These have been scattered over the last 25 pages of this sheet, so someone will need to compile a list of those - I can do it over the weekend if no one gets to it first.

Finally, there's the issue of buff arrangement. I don't like the idea of simply selecting classes and allowing the sheet to figure out what buffs that means; I don't think it would be any simpler, and I think it would be less accurate. However, grouping the buffs by what class gives them would certainly make life simpler, so I think that that should probably be done. This leads to two obvious ways of organizing things:

1) Group buffs by "raid buffs", "class buffs" and "self buffs", and within each of those catagories break them down by class. The disadvantage of this is that if you have a class that gives both types of buffs, you have to go to two different places in the buff sheet to change it.

2) Group buffs by class, period. All ret pally buffs are together. The problem with this is that if you don't know whether buffs are group buffs or raid buffs, it could be hard to know which apply.

Note that in either case, weapon buffs needs to be it's own section.

I guess my inclination would be to do the second, and perhaps provide some indicator as to whether a buff is raid, group, or self in a box somewhere to convey that information to the user; again, see the attached spreadsheet for an example.

Clearly there's still some design work to be done here, so if people have comments/ideas/whatever, feel free to speak up.

Also, before you start working on the sheet, PM me your email address so I can send you the most up-to-date version of it.

That happens (to a lesser degree) to me on Excel as well.

I know this some rare case, but .. uhm .. i edited like the last 5 revisions to have the shadow resistance gear available to optimize around Mother Sharaz (Hit and Meta requirements). If it's asked too much, ok. Just thought i would ask for it.

For that I'd suggest something like a none slot since high end resist gears basically have no dps stats. Or, what I'd like comes from the DPS spreadsheet, an Unmodelled slot, and an easy line to punch in whatever stats I want for an item for a quick check of stuff not included.

I don't know what other gear you're wearing, but I had the same problem. In the gear sheet the Scryer's Retainer are marked as T6, so it might be activating the set bonus. Fixing that solved my problem.

That was absolutely it! Thanks. I just got the t6 gloves Sunday night. I hadn't had the problem prior to that and it was indeed now activating the set bonus. Removing them from t6 on the Armor sheet dropped the dps gain down to a far more reasonable 1514.05 that is in line with what I would have expected. Thanks!

Ah, yes, indeed, Scryer's Retainer is improperly marked as a piece of T6. There's actually a couple other bugs with those beta versions as well (that's why they're beta). I know I've fixed a glitch around glove socketing and maybe one or two others as well; but if you find other things continue to pass the information along, and I will incorporate fixes as I get ready for the next release version.