• Aldriana

In the current version, it's pretty hard; It's Sword Calcs row 233, or Dagger Calcs row 240. Update the formulas in that row such that the 5000 is something else instead. Make sure to change it in every column, or the stat-value computation will break.

Making this parameter (and several others such as the weighting for "weighted total") more accessible is being planned for a future version (probably 0.7). We're thinking of adding an "advanced settings" page that will allow adjustment of these parameters directly.

 

sp00n

In the spreadsheet's introduction it still says http://rogue.bleedo.net/ when pointing to the Rogue DPS Spreadsheet instead of the new http://rogue.raidcal.com/.

Item Ranking Rogue [horribly outdated]

 

• Aldriana

I've been working on the new calculation sheets for a bit, and while so doing it occurred to me that it would really be nice to get the correct values for some of the proc effects built into the sheet on one of these iterations. In particular, I'd like to get the values of haste effects entered correctly, because they're a pain to change later. So if any of you are feeling bored and would like to help out, it would be nice if people could take some data sets on the proc rates of various items, most notably Dragonspine Trophy, although the Blacksmithing maces, The Bladefist, and other weapons with haste procs would be helpful as well.

For a helpful tip on how to get good data on these, swing by the [Rogue] Mechanics Testing Thread and check out post 16. (Heck, while you're there, check out the rest of the thread and test some of the other open issues - the spreadsheet can only be as accurate as our knowledge of the mechanics of the game)

Thanks for any help you feel like providing.

 

Oie

If you can wait a couple of days, I'll have some good data on Dragonspine Trophy for you by Sunday. I'm looking to get 10,000+ auto-attack swings each with weapons of varying speeds. I've checked the other threads on getting accurate data, and I'm following their guidelines.

Thanks for your work on the spreadsheet, and I hope I can help.

Edit: I updated the rogue mechanics testing thread with my initial results if you want to take a look at them.

 

Macblade

If you could revise your original post to reflect all the work that has been done (which is great!) that would be really helpful. Along these lines, you still have the link for the very first version in the original post. You also have the newest versions at the top, but there's not really any reason to keep the link to the old version and if you read through the post and come across the link in the body of the text it seems like it is the most up to date spreadsheet, which it is not.

Thanks for all the work on this. I was using the rogue DPS sheet for gear choices as well and was confused as to why it wasn't suggesting +8 hit gems when I am not at the hit cap

I also have a couple of questions.

First, I'm human and using swords but when I switch back and forth between human and NE the current hit value doesn't change. Does the spreadsheet take into account the human racial bonus to sword skill?

Second, how is optimal hit calculated? In particular it's saying that my current hit is 235 and my optimal is 251. To the best of my knowledge neither 235 nor 215 is hit capped even with the human racial and using swords with Latro's as an offhand. So shouldn't my optimal hit be whatever the cap is-294?

 

Pstar

Great work, one minor detail; you don't seem to have the Pristine Fire Opal in available gems; I coulda sworn it was there last version... sorry if it's already been mentioned

NM old version is linked off of raidcal... my bad

 

• Aldriana

Okay. I'll clean that up at some point.

I refer you to post 324, and the last line of post 332.

Post 201.

 

roq

Ald, approximatly how far out are we from the next version of the spreadsheet? Are you waiting on some more stuff to be tested, or are you working on it and going to add in the tested information as it pops up? There is no hurry of course, i am just wondering.

 

• Aldriana

It's sort of a combination of things that are holding up the show at this point.

1) Mechanics testing. For the most part, this is in decent shape; I was waiting on this for a while, but at this point I'm mostly at the point where I'm either going to take my best guess or leave out the offending piece of gear for the moment... and mostly the former, probably, given that testing so far has led me to the following rule of thumb:

All weapon procs are probably 1PPM in absence of significant evidence to the contrary

So, for things that haven't been tested yet (Siphon of the Nathrezim and Blacksmithing Hammers, for instance) I'll probably just fill in 1 PPM for now and will change it later if necessary.

The only outstanding thing here that I'm at all concerned about is the T5 4/5 proc rate. This turns out to be sort of a difficult effect to model, and I really have no good data on what the answer might be. The limited testing I've seen so far makes it appear that the answer is most likely not 1 PPM, so I don't have a good feeling on what to plug in as an estimate. Still, this by itself wouldn't be holding me up - I'd just take my best guess and move on.

2) The new calc sheets are a slightly larger job that I originally estimated - there are a few subtleties to implementing them that I didn't initially appreciate, and the planned number of sheets has increased from 7 to 11 (of which 5 are mostly complete, all of which are significantly larger than either calculation sheet in 0.6.3). So they're taking a bit more time than I originally expected.

3) The other thing holding up the show at the moment is a sort of communications mixup between Darlal and myself. In order to support the underlying mechanic changes, there are some UI improvements that need to be made; additionally, there are some general UI improvements that would nice to have made at some point (for instance, the unique gem issue). He and I discussed these about 2 weeks ago, and he indicated that he thought he might have time to make at least some of them... and I haven't heard from him since. I guess he got busy with work or whatever. Now, normally, this wouldn't be a problem; I'd just take a day or two, figure out how to do the UI stuff myself, and plow forwards. What I'm concerned about, though, is the possibility that he might have started work on some of the more sophisticated changes; if I make the basic changes and then pound out 0.7 based on that + the new calculation sheets, and then he shows up with a version with all sorts of neat UI fixes... we have sort of a problem in terms of merging two fairly sophisticated sets of changes back into one sheet. So I sort of hate to move forwards until I know what his status is.

Anyway: I haven't forgotten about this sheet, and I am still working on it, but I don't really have a good time estimate as to when everything will be done.

 

• tsigo

If you give me an exact setup of the best way to test this, I can give it a shot again.

 

Cybelirrae

Aldriana, would WWS parses help at all in at least narrowing the range of possibilities for the proc-rate? Lurana and I have been running with this set bonus for some time now and I can link you a number of our WWS parses that include the Coups de Grace bonus. I am guessing this will not help, but thought I would throw it out there.

 

• tsigo

I think I discovered a possible bug on the Armor sheet. Some of the "total gems" formulas include the meta gem column, some don't. I'm not familiar enough with the sheet to say whether or not this causes a calculation issue.

 

• Aldriana

WWS parses will perhaps helps some, but the fact that you're using two weapons of different speeds makes it hard to draw conclusions. It's useful for generating ballpark figures but you can't really do statistics to know how good your answer is.

In general, the most useful data collection for testing proc rates is a reasonably long (1k+ attacks, and more is better) period of attacks performed either with a single weapon or with two weapons of the same speed. Using combo point generating moves is fine, but avoid finishing moves (as I have a sneaking suspicion that finishers might obey slightly unusual proc rules).

When testing proc buffs (such as Coup De Grace) it is advantageous to use a buff-canceling macro like the one listed in the mechanics testing thread. This isn't essential, but it's defenitely helpful.

I generally like to see a combatlog of such data, but if you use a buff-canceling macro the following are sufficient:

1) How many attacks, including specials, did you land on the target?
2) How many procs occurred?
3) What is the shortest period of time between consecutive procs?

If you don't use a canceling macro, a combatlog is sort of necessary to analyze for overlapping procs.

To accurately pin down the proc rate of an effect requires two such data sets performed with different weapon speeds, and the more different the better; for instance, take a data set with a 1.3 speed weapon, and then another with a 2.6 speed weapon.

 

roq

Just as a quick note, there are some buffs that can't be cancelled with that Macro. For instance, the proc on the Blackout can't be cancelled. I found out a few days ago, that mongoose can't be cancelled either. So just keep that in mind.

I would suggest trying to manually trying to cancel the buff at first to see if it can be done before you go cursing the macro for not working .

Edit: oh and thanks for the response Aldriana.

 

Kol32

With 306 hit rating, why is it that Recap over several days raiding and misc. farming etc. shows a 7% melee miss rate? I have 5 points in precision, no Weapon Skill on any of my items.

 

neg^

Total miss rate, including dodge/block/parry. Should be able to see a breakdown splitting them up if you select 'melee' in your own stats instead of the 'personal details' view.

 

• Aldriana

I worked on the sheet a bunch over the weekend, and finally have it up and running on the new calculations sheets. I haven't had a chance to make all the other changes I'd like to make, but never fear - I do plan to work on them over the next week or two. However, before I took the time to implement all the suggestions, ideas, and other features people have had, I wanted to make the new calculation sheet available, so as to allow you fine folks to start playing with it and finding any mistakes I might have made (as, while I tried to be careful, it is highly probably that I missed something at some point).

So, anyway: version 0.7 is now available at the links below, but be aware that it's perhaps a bit buggy in spots. If you find anything that doesn't make sense/doesn't match your expectations from the prior version/whatever, let me know.

Excel Version: Free file hosting by Savefile.com
OO Version: Free file hosting by Savefile.com

Known Issues:
* Defensive contribution of Mongoose is totally neglected.
* Sheet does not include the latest weapon-skill findings; hence, if you operate with less than 5 weapon skill, your damage values will be incorrect.

Stuff planned to be added soon:
* Maces and fist weapons.
* User-edittable interface for offense/defense weighting, mob armor, mob resist rate, and similar parameters.
* More descriptive gem names
* Possibly BT and Unique gems, assuming I can find a way to implement them.
* Any other comments in the last 5 pages or so - I'll look through them at some point and add what I feel is worth adding.

 

LiteSabre

Dunno if this was mentioned, but I thought I might post it again just in case: in the helm slot, the Cursed Vision of Sargeras, which I think is probably better (and more stylish, to boot) than the T6 helm, is missing.

 

• tsigo

Please at least include the epic gems. Any BT guild should basically have enough of these to use them for every piece Hyjal and BT gear they get. Maybe you could only show them if they have Black Temple gear selected in the zone selection? And possibly only include them as the "best option" gem if it's for a piece of Hyjal or BT gear, since re-socketing something like Edgewalker boots might not be worth it for most people.

 

• Aldriana

Yeah, I noticed that was missing a week or two back - I do plan to add it in a future revision. Again, this release is more to let people play with + debug the new calc sheets than anything; expect more of the actually features + fixes in a week or two.

Yeah, the main thing holding up the epic gems if figuring out the best way of including them such that they can be used by the people with access to them, but can be turned off for those without them.

Again: I do plan to add them; the hold up is that historically I've done more of the calc stuff and not so much of the UI changes, so the UI stuff tends to develop somewhat more slowly.

 

Oscarvil

Just downloaded the sheet and I want to say a big thankyou for making and keeping this sheet current Aldriana (-:

I found an error with Romulo's Poison Vial (at least in my version) where the calculation for estimating proc value for each hand is not implemented, and each hand procs 277 times per minute. Going from the old sheet (and of course common sense), to fix it you just need to multiply the '277' value by the main hand and offhand strikes per minute (respectively) and I have done that for my own sheet.

 

• Aldriana

That's actually an intentional change and not a bug. I introduced a new way of computing the damage done by PPM procs which will be more accurate than the old way. What's encoded on Trinket sheet is that it procs for 277 damage per minute (i.e. 1 PPM at 277 damage per proc). It then evaluates the value based on the columns Q and R of the DamageCalcs page, which compute the incremental value of a 1 damage per minute proc. So while the computation is done a bit differently than in prior versions, the overall value of the trinket should be about the same in 0.6.3 and 0.7 - which is true, at least, in my working copy of the sheet.

Edit:

My apologies, there is actually a bug there - I failed to totally switch things over to the new calculation method. I'll post a fix in a few minutes.

Before anyone brings them up, the following are also not bugs:

1) Dragonspine Trophy scores even higher than it did before. It does so since I updated the proc rate to match the newest data on it's proc rate.

2) Windfury is a smaller damage boost than it was before. Again, intentional. Done to match the hotfix to Windfury.

3) Daggers do somewhat more damage than previously. Again, intentional. It's much smarter about computing dagger cycles than it used to be.

 

• Aldriana

New version (0.7.1) with the RPV fix:

Excel Version: Free file hosting by Savefile.com
OO Version: Free file hosting by Savefile.com

 

◊ Vulajin

Ald, I wonder if it would be possible for you to easily add support for fist weapons. They are equivalent to swords in most respects, with the difference being 5% crit instead of extra attacks. Additionally, I wonder if you might be able to support the mixed sword/fist spec I -- and other rogues who may have had shitty main hand sword drop luck -- use.

It's a lot to ask, but you make quite an enticing tool here, and I'd like to be able to use it effectively without needing to worry about whether the values it gives are accurate for me. :-P

 

• Aldriana

That's actually one of the major reasons for the rewrite of late - it's going to make it possible to add fists and maces. Which is why it's in the "stuff to be added soon" list I made in post 367. The new system is much more flexible and will allow fists, maces, mixed specs, hemo - the works. I'll just need a little time to add the new items in (which is slightly more sophisticated than one might initially think).