dinesh

FYI, when importing from 2.4.0.7a to 2.4.2.1beta, I get issues if either Talents or Options are selected to be imported.

 

Dontmindme

First off, given that the Talent and Option sheet was completely revamped, importing from versions prior to the revamp just won't work.

I'll look into the division by 0 errors under those circumstances not caused by an import.
Edit: I discovered the problem. The proc uptime modeling didn't like 0 finishers. I've applied a fix on my version that seems to work.

As to negative values in "Other DPS", that is certainly possible. One of the things being applied in "Other DPS" is a correction factor for proc uptime. Uptime starts with an estimate so the cycle model has something to work with in determining the cycle. Because the cycle affects one's proc uptime, you cannot get a totally accurate value until after the cycle has been determined. Thus, in "other DPS", I added a post cycle correction. Using the AEP Macro (or doing the copy/paste special in OO), makes this correction factor 0, but until you do this, you may have both positive or negative DPS corrections.

 

dinesh

Understood, I was just pointing it out for other users.

I didn't import options. But I find that if I switch from the default "melee boss" options to "caster boss", I also get a bunch of #VALUE errors (this was also true in 2.4.0.7a, but I didn't get around to reporting it then). Switching back doesn't fix anything either. I have to close the spreadsheet and re-open reverted to a version before I tried switching. Haven't had a chance to try to debug where it is going wrong yet too much, but it looks like BuffedDPS!Q4 (among others) may be a reference to a named cell which no longer exists?

[edit: the "Delete Option Build" button also errors out. I'm guessing this whole section is probably in need of some re-writes.]

 

beany6316

hello all.

New guy here. Been a rogue for abit now.
I use to use the information at shadowpanther.net for along time. I had never considered d/l'ing the speadsheet before.

May i say, damn good job!

The problem i am having is importing one of my rogues. I am able to upload all the others except this one:

Melana, uldum alliance.

Young rogues need to go out and learn about their class. This is a great place to do it.

thanks again,

/bow

 

Sneakerpimp

Just wondering about the accuracy of the estimated Rod of the Sun King proc rate, since according to 2.4.0.9 and 2.4.2.1 it is best (untalented) mainhand in game short of the warglaive.

 

Dontmindme

As to Armory import, the problem is likely on Blizzard's end. Every once in awhile, one's Armory data in inaccessible and the Armory page on Blizzard's site comes up with "An Error has been Encountered". My guess is if you try to check that character in the armory, you do not see it.

As to Rod of the Sun King, I'm not seeing that on my sheet. Changing to Rod of the Sun King seems to be ~58 DPS loss vs. Blade of Infamy in the default gear set while the MH Warglaive is a 61 DPS increase, both of which seems correct to me.

As to Delete Option build, found a typo in the Macro. Looks like I missed a cut and paste and it was trying to call ResetTalentUI instead of ResetOptionUI.

Edit: Addressing the Options. It's possible that through the copying of builds from previous sheets, that an error has been introduced in that Macro. Look for the following going down the Options on the Talent Sheet:

73
Any
Yes
Yes
6200
0
Overall
30.00%
Yes
No
Finishers
No
Finishers
No
No

If anything is blank fill in the appropriate value and see if that fixes it, then save the fix.

 

dinesh

Just to make sure we're on the same page, it's not just the macro which has problems. Even if I manually change the boss armor value by typing in 6200, I generate a boatload of #VALUE errors on Talents and Buffed DPS. As far as I can tell, many of the named cells referenced in column Q of the latter page appear not to exist any more. [edit: or maybe it's just on a hidden sheet... either way, it's causing me a #VALUE error]

 

dinesh

Man, I dunno, maybe my verysion of the SS has been corrupted from earlier imports or something, but here is what I find.

I do not have any blanks in the options list, all values are as you have them above, except I have 7685 for the Boss armor.

1- I can now type in 6200 instead of 7685 for boss armor, and it calculates correctly (it used to cause #VALUE errors).
2- If I try to select "Caster boss" from the options dropdown, though, I still get the #VALUE errors I received before.

3- I'm a little confused about the AEP values on the Talents page. I have T5 pants with 4 piece bonus, and the Trousers of Scryer's Retainer. If I switch from T5 to Scryer's, I get a DPS increase, but if I then recalc the AEP values the DPS turns into a decrease. But if I switch back to T5, the decrease becomes larger. Basically, the sheet is telling me that Scryer's is always better than T5 (even breaking the bonus), but if I optimize AEP after making the switch the absolute DPS reported using Scryer's is lower than the absolute DPS reported using T5 (optiized with T5). Perhaps this is expected behavior, but it has my brain in a twist.

 

sp00n

In regards of importing from an older sheet, maybe you could add a version string to the sheet, as well as a 'minimum version string', which identifies a compatible version of the sheet where the importing will work.

As to the #VALUE bugs, for me it seems to work just fine with 2.4.2.1beta. I can change to 6200 or Caster Boss without problem.

Maybe you're using some armor penetrating equipment that is somewhere broken, try to identify that piece.

 

Moona

Looks like there is a little bug in the racedropdown.When selecting "Blood Elf" dps differs from selecting the german expression "Blutelf".

 

Makdoh

Hi there.
I've had some sort of bug with the spreadsheet for some time now.
The tab for the buffed cycle which also includes dps, hit, miss and such seems to be blank.
The bug already persisted through several new versions (i always kinda hoped it would be fixed in the next one).
Here's a screen: http://www.abload.de/img/empty5ax.png
It looks like that without me doing anything specific to the spreadsheet (except opening it).
Help! :/

 

Left

It's not a bug; the sheet is currently in transition mode. DMM is moving away from having two analyses (buffed and unbuffed) to having only one (buffed). The Gear/Buffs tab is currently in a state of transition between those two display modes. Use the display on the "talents" tab for now, as it represents what the finished display version will look like.

Speaking of which, DMM, when were you thinking of just taking out Unbuffed Cycles altogether?

 

Ashere

Unbuffed cycles might be an issue again when WotLK is released, since a lot of your time leveling up may be done solo. With the previous spreadsheets it was easy to find how 16 AP would actually give slightly more DPS than 8 Agi while unbuffed, some knowledge that shouldn't be totally lost then.

 

• Vulajin

Just turn off all the buffs?

 

Moona

Use the last Option on the Talent page [O29]

Disable Buffs Yes

I bet 10G it is written on the first Page of this post

 

Makdoh

Err, thanks, didn't know that.

 

Dontmindme

A new beta candidate is up...

Changes from previous candidate:
1) Revamped the Deathmantle (T5) 4-piece bonus to fix a small error and to recode it so that it will show a DPS increase without hitting the AEP/Optimize Macro button.
2) Fixed the division by zero errors for cycles without finisher damage.
3) Fixed the error in the Delete Options Build Macro.

And a few other comments...
The stats on the Gear _ Buffs page are the buffed stats, I just moved them over to where the unbuffed were. I hope to, at some point, revamp that sheet and add the new design.

I found the problem with "Blutelf", not fixed in this candidate (just noticed the post) but will be when this becomes a release version.

Next version after this one, I'll probably be moving the Importing to the Options page where I will have more room to explain which options work in which versions.

I'd still be interested in some T5 4-piece testing. Right now, the 0.5 PPM estimate is based on the preliminary testing done plus looking at a few WWS logs and picking a PPM value that seems in the ballpark. It would still be helpful to have some testing (10000 hits or so, more the better) against ghosts in DM with a linked combat log (as opposed to WWS).
The ideal conditions:
1) Single weapon or both weapons the same speed.
2) No proc haste effects (DST, Mongoose) as this will affect the proc rate. Static haste rating is fine, I just need to know what it is.
If someone does this, I need:
1) Linked Combat Log
2) Weapon Speed used
3) Static haste rating

 

dinesh

It turns out it must have been an import error with the options builds from earlier sheets. They were coming in as Mob Level of "73 Boss", not just "73". There was also an issue of the "Disable Buffs" importing as a blank, though that didn't cause any errors, it just defaulted to not using buffs.

I am still having an issue where the Scryer's pants are always a 10+ dps upgrade over the T5 pants, but re-optimizing once I'm wearing the Scryer's pants causes my total dps to drop by 15+ dps, so my absolute, overall DPS decreases when I make the switch. Unless anyone has some suggestions, though, I'm just going to assume that's a weird case where their stats are massively different, and re-optimizing just causes the nextstat calcs to rebalance in an odd way. Even if it is a minor downgrade, it's probably within the overall sheet error, and I would definitely replace them once I get another upgrade because the set bonus will already be broken.

 

dinesh

Man, my timing stinks! The latest version now shows the T5 pants (with 4-piece set bonus) as an upgrade, and the DPS change is minor before and after re-optimizing with either piece equipped.

One thing I will mention, just in case: with the Scryer's pants equipped I am 1 point underneath the hit cap. Reoptimizing at that point doesn't change dps substantially, but it does cause the value of hit to drop substantially, from 1.1 AEP to .73 AEP, with corresponding increases in other stat values.

My best guess is that this is either and edge case problem or a rounding issue, since I am possibly only fractionally under the hit cap, and I think the nextstat calc only works on whole value increments, but I thought I would mention it to be sure. I may just switch to Agi food at that point, to be sure.

 

Dontmindme

Yes, that is what is happening. The Macro AEP actually just adds one to hit and compares how much the DPS changes. So if you are a fraction of a point under, the AEP is only showing a benefit for the useful fraction. I'm not sure how much this can be improved. I could subtract 1 instead (and calculate accordingly) which would help in this case, but still would have the same issue for those past the cap (where it will still show 0 AEP or a fraction of it at the cap).

I may readdress the calculations in the future (when I try to rebuild the AEP estimates for Open Office), but it's not a high priority as the AEP numbers aren't used in the calculations.
---------
Edit:
In other news, I just discovered a somewhat significant bug. It appears the spreadsheet has not accounted for crit % reduction for level difference. I went back to very old versions, and the 0.6% theoretical reduction vs. a level 73 boss is not present, although prior to the Expertise change, it applied it to weapon skill, just not level difference for the boss. It amounts to about a 10 dps reduction with the default gear set.

 

patten3054

Does the spreadsheet work with Mircosoft Works Spreadsheet? When i open the SS all I have is the info and faq pages.

 

Ferrat

I am having trouble with the new speadsheets, one of the macros is not working for me. The section on buffed DPS is blank for me. Maybe this it covered in another part of the forums. If so can you please direct me too it or maybe someone that has had this problem can give me a quick fix. As i think Buffed dps is the thing i look at the most.

 

Left

Which macro? "One of the macros" isn't very descriptive.

As for the buffed DPS section, if you had bothered reading the curent page of posts you would have seen the following two posts which are less than 10 posts before yours:

For more info, read the rest of the discussion starting at post 1061.

 

Ferrat

sorry for wasting the time and not reading through the posts, but thanks for clearing it up.

 

sp00n

According to your report, probably not.
However, try changing the tab you are looking at (there should be more than just 'Info' there).

If there are no, or they don't work as expected, download Open Office, it's free.