On the alignment issue: I'm going to talk to my friend who did the checkboxes in the first place, and see if he knows how to fix this; if not, we might have to go more towards dropdowns.
On the whole, though, I generally prefer radio buttons and checkboxes to dropdowns, simply because they require fewer mouseclicks to operate. However, I don't feel *particularly* passionately about it either way.
I like the by-class format that you've started on the sample sheet provided a few posts back. What do you think of providing spec-independent buffs first as either checkboxes or radio buttons, and then supplying talent modifiers below that, grouped according to talent tree, as drop-down menus that indicate the number of ranks spent in that talent. A brief description could be nested to the right of the talent for clarification.
For example:
Warrior
Battle Shout [ ] Solarian's Sapphire [ ] Warrior trinket that increases AP of battle shout by X.
Sunder Armor [dropdown]
Arms
Blood Frenzy [dropdown] Increases physical damage to target by X.
Fury
Commanding Presence [dropdown] Increases the AP of battle shout by X.
Shaman
Grace of Air Totem [ ]
Strength of Earth Totem [ ]
Air Totem Twisting [ ]
Enhancement
Enhancing Totems [dropdown] Increases Grace of Air and Strength of Earth totems by X.
Improved Weapon Totems [dropdown] Increases AP bonus of Windfury Totem by X.
Unleashed Rage [dropdown] Increases AP by X.
etc.
This would help people when entering buffs if they aren't entirely sure which buffs a particular class and spec would provide. It would also help in turning on/off all buffs supplied by a specific talent spec if that is something that changes frequently (e.g. being given an full-fledged enhancement shaman or a resto shaman just for totems). The downside is that additional rows are used and the talent-buffs are separated from the abilities that they modify.
Last edited by nelalas : 03/20/08 at 3:26 AM.
Reason: added Solarian's Sapphire
That would be another way of doing it... I can't decide which I like better. It would make it easier to figure out what people should have, but the separation of buff + talent isn't ideal... I'd be interested to get some opinions from users of the sheet.
Also, where would something like Solarian's Sapphire fit in under that arrangement?
If you are going to consider buffs by class, remember that Sunder Armor and [Imp] Expose Armor do not stack. There may well be other buffs I can't think of off the top of my head which also do not stack and are from different classes.
Ideally there should be a section for each "group" of buffs, from which you can only pick 1. Dropdowns give a nice way of enforcing only 1 is selected and cut down on space consumption. Radio buttons are fewer clicks, but chew up screen real estate fast.
Each group of buffs (GoA, Imp Goa, Twisted Imp GoA) should them be placed roughly near other buffs from that class. So the GoA group would be near the SoE group, which would be near the Unleashed Rage group. Also perhaps a # of Bloodlusts per 10 min.
One thing to consider, WF.. while it is a weapon enchant, is also an Air Totem. It would be nice to block using WF and GoA, but still allow twisted GoA. Messy, but we might as well throw out all the ideas now rather than go back and tweak and patch stuff later.
You could just toss Solarian's to the right of the regular Battle Shout as is done in the current sheet and the sample you provided. The reason I like separating the talents from the "core" abilities is that it makes it easy for people do properly assign buffs when they may know "hey I do raid with an enhancement shaman and feral druid" but may not know exactly which buffs they provide. It also makes it easier to answer questions such as how cycles and gear choices may change depending on group and raid composition since spec-related buffs are blocked together.
As for windfury, I think that should remain with other weapon buffs since they are mutually exclusive. Also, since it is generally more favorable than only grace of air, I think it makes sense to leave the option of twisting with the GoA buff.
Seems to me that the correct way of managing WF vs GoA is to have the sheet detect if both are selected and automatically reduce GoA uptime to 80-90% under the assumption that this means that they're twisted. This shouldn't actually be too hard to do.
For other nonstacking buffs, the most reasonable thing to do is probably just automatically take the stronger one, although that could be confusing as well.
My only concern with the listing improved talents by tree is that it may be a bit misleading; like, imp BS is fury, but most arms warriors are going to have it as well. You really want to list it under "DPS warrior", but I'm not sure we want to get to that level of granularity.
I would say 85% uptime for GoA and 100% for WF would be accurate for a twisting shaman. Shaman will use WF every 10 seconds, and GoA will only be down for the 1.5 second global after dropping WF. In 2.4 we should see 90% GoA uptime with the totem GCD being lowered to 1 second.
I think that would be great in a situation where the trees could be expanded and collapsed -- for example, the possible modifiers to a buff expand when you enable the core buff. Descriptions could be placed in the text of the talent/modifier to tell users what class and talent tree (or spec) provides it. I don't know if that's easily, or cleanly, done in Excel so perhaps it is something to think about for the programmed version. I'd imagine though, in the programmed version, that there could be the freedom for users to sort buffs however they wanted (or at least two or three pre-set views).
It seems that the main decision here is between 1) Sorting based on buffs with associated modifiers or 2) Sorting based on class and talent tree. If people want to send a PM to me with their vote, I'll tally the numbers and post the results for Ald to consider. I think that would be best so that we don't clutter the thread but, of course, keep posting additional thoughts or comments.
I think that would be great in a situation where the trees could be expanded and collapsed -- for example, the possible modifiers to a buff expand when you enable the core buff. Descriptions could be placed in the text of the talent/modifier to tell users what class and talent tree (or spec) provides it. I don't know if that's easily, or cleanly, done in Excel so perhaps it is something to think about for the programmed version. I'd imagine though, in the programmed version, that there could be the freedom for users to sort buffs however they wanted (or at least two or three pre-set views).
Ultimately the programmed version would allow multiple views, but such things require a fair amount of work... it's not the sort of thing that would make it into any early release of such an implementation. The thing about doing it in a programming language is that it gives you enough flexibility to do pretty much anything you'd want in terms of modeling... and there are a lot of cool ideas and features that one can think of. I could easily spend a year working on a programming-language implementation full-time and still have features that I wanted to add. Alas, since I have to work for a living, I won't be putting that much time into it anytime soon, if ever. So, just to set expectations: assuming I ever get to the point where I'm ready to release something, don't expect fancy UI features for the first couple... dozen... releases - those will be added in time, much as they were for this sheet.
Has it ever been conclusively shown whether attacks from Sword Spec and/or Windfury are subject to the Dual-Wield penalty or not? That is, do they use the yellow miss rate, or the white miss rate? I've been sort of assuming that they use the white miss rate, but it occurred to me today that I don't recall ever having seen that tested. Does anyone know if this has been tested, and, if so, where I could find a summary of the findings?
An interesting question. It is consensus on the Enh. Shaman thread that Shaman weapon Windfury (the kind they put on their own weapon, not the totem) uses the special miss rate, not the white miss rate. (This is part of the reason that crit is so valuable for Shamans while hit really isn't so much.) However, I don't know that the equivalent has been tested for the totem. It could possibly be parsed out of combat logs, but the way it is currently recorded in combat logs makes that hard. Come 2.4 (or from the PTR) it should be easier to test, given the combat log changes.
Last edited by Left : 03/20/08 at 6:08 PM.
Reason: Clarification
It's pretty clear it uses the white miss rate. Hand parsing and the added attack power make it pretty clear which is the WF attack when using the trusty 2-4 damage sword. I have observed "misses" on the WF attacks when I've been clearly "special" capped.
Unfortunately, the new 2.4 combat log as implemented on the PTR (last I looked), doesn't seem to distinguish the actual source of extra attacks (no more than the current one does). The source seems to just be one's own character ID. It will help significantly in spreadsheet parsing of the logs (because the delimited format is easier to separate into columns), and will distinguish between 2 of the same named mobs (because of the target IDs), but beyond that, it doesn't do as much as I would have hoped.
On the alignment issue: I'm going to talk to my friend who did the checkboxes in the first place, and see if he knows how to fix this; if not, we might have to go more towards dropdowns.
On the whole, though, I generally prefer radio buttons and checkboxes to dropdowns, simply because they require fewer mouseclicks to operate. However, I don't feel *particularly* passionately about it either way.
From a UI point of view, for a single option list, drop-downs are better as they take up less real estate (screen space). As your list of options grows you will quickly discover just how ugly having a giant collection of radio buttons is.
Also, it's inefficient, as it always shows what's not selected in addition to what is selected.
While levelling up my Shaman Rtu I recall Windfury and Stormstrike being yellow text in my SCT indicating that it is considered a special attack. Being as the totem doesn not cause us to get yellow text I would think that it is treated as a white attack. However, SCT is not the best thing to base this on I still wanted to share my observations.
Remember that a Shaman WF attack is quite different. Its 2 extra attacks, and they are yellow in colour and uses the "Your Windfury hits Target for X." instead of "You hit Target for X".
I suppose the easiest way to test would just be to have someone attack a lvl 73 with as little hit rating as possible, but still over the special-cap. A single miss that was conclusively a SS/WF attack would give us an answer.
Since it sounds like the answer is "there has been no formal testing of this", I took a few minutes to dig through the WWS of some recent fights - in particular, my recent fights with Mother Shahraz, given that my SR gear doesn't have much in the way of hit on it.
Relevant log snippet #1:
A20:24'15.382 Aldriana's Melee hits Mother Shahraz for 348
B20:24'15.419 Aldriana's Sinister Strike hits Mother Shahraz for 1169
C 20:24'16.883 Aldriana gains 1 extra attack through Windfury Attack
D 20:24'16.883 Aldriana's Melee crits Mother Shahraz for 1554
E 20:24'17.103 Aldriana's Melee dodged by Mother Shahraz
F 20:24'17.413 Aldriana gains 1 extra attack through Sword Specialization
G 20:24'17.413 Aldriana's Melee hits Mother Shahraz for 321
H 20:24'17.785 Aldriana's Melee misses Mother Shahraz
I 20:24'18.451 Aldriana's Melee hits Mother Shahraz for 221 (glancing)
J 20:24'18.807 Aldriana's Melee hits Mother Shahraz for 527 (glancing)
(I have labeled the lines for clarity while analyzing)
Lets do this in order.
Line A is an OH hit, and line B a SS; there are no other melee hits for almost 1.5 seconds. Thus, we can assume that anything triggered after this gap must be triggered by something in lines C or after.
Line C is the triggering of a WF attack; this message frequently precedes the attack that triggered it, so that's fine. Line D is the only MH hit until line J; thus, this has to be the attack that triggers WF (as a missed/dodged attack cannot trigger WF). Thus line D is the triggering attack, and line E would appear to be the WF attack, which was dodged. Now, line F is a sword spec proc. Line G and I are OH hits, so cannot be the sword spec proc; line J is 1.5 seconds later, so most likely not the proc. Hence, we can only conclude that line H is the Sword Spec proc... which misses. Thus, Sword Spec procs can miss even with Precision and ~150 hit rating, meaning that they must use the white hit rules (i.e., they're subject to the DW penalty).
Relevant log snippet #2:
A 20:25'05.753 Aldriana's Melee hits Mother Shahraz for 798
B 20:25'06.416 Aldriana's Melee dodged by Mother Shahraz
C 20:25'07.392 Aldriana's Melee crits Mother Shahraz for 739
D 20:25'07.392 Aldriana gains 1 extra attack through Windfury Attack
E 20:25'07.392 Aldriana's Melee hits Mother Shahraz for 499 (glancing)
F 20:25'07.567 Aldriana's Melee misses Mother Shahraz
G 20:25'08.314 Aldriana's Melee crits Mother Shahraz for 680
H 20:25'09.170 Aldriana's Melee hits Mother Shahraz for 673
The last melee hit before line C is 1.5 seconds before; thus, it's not triggering the WF proc in line D. Lines C and G are OH hits so can't proc WF. Line H is almost 2 seconds later so can't proc anything. Thus lines E and F must be the MH attack that procs WF, and the WF proc. Since there would be no proc if the initial attack missed, line E is the triggering attack, and line F is the WF proc. Thus WF procs can miss as well.
Hence: both SS and WF appear to use white hit rules, and are subject to the DW miss penalty.
I noticed that when equiping [Shoulderpads of the Stranger], the 0.9.4 version of the spreadsheet shows my Expertise (with Weapon Expertise talent) as 13. In-game my Expertise is shown as 12. This comment from the Roguecraft 101 thread sortof explains the in-game value:
...expertise rating comes in increments of 1, but expertise itself only increases in increments of 0.25% (~3.94 expertise rating). For example, if you have 0 expertise rating, [Shoulderpads of the Stranger] by themselves will only increase your expertise to 2, and [Grips of Deftness] by themselves will only increase your expertise to 3. However, equipping both items together will increase your expertise to 6.
If I change the Expertise Rating value for [Shoulderpads of the Stranger] on the Armor page of the Spreadsheet to simulate an Expertise gain of exactly 2 (i.e. 7.88 rating), overall dps decreases.
Is this a bug in the Expertise handling of the spreadsheet causing Expertise Rating to be slightly overvalued by allowing fractional values or is the Roguecraft 101 thread and in-game display not quite accurate? Thanks in adavnce for shedding some light on this for me.
I thought I'd addressed this previously, but I guess it must've been in a PM to someone. This spreadsheet models fractional Expertise, even though the game quantizes to full points. The reason for this is to get greater smoothness in the modeling of various effects. The problem with the approach is that, as you know, it does lead to certain items being slightly overvalued if they're not roughly at a multiple of 4 points of Expertise Rating; unfortunately, there's not a whole lot that can be done about this, as artificially lowering those items to the next smaller integer value would mean that multiple expertise items would not stack correctly. So yes, it's a concern, and it's good to be aware of it, but it's not likely to be fixed anytime soon if ever.
Regarding the 12 vs 13 display bug; this is because the sheet is internally using 12.6..., which in-game truncates to 12 but in the spreadsheet rounds to 13. This display bug is something I plan to fix sooner or later, but I wouldn't place any large wagers on whether it's going to be sooner, or later.
SS attacks have been white attacks since patch 2.1
Rogues
* Sword Specialization: The change to Sword Specialization making its extra attacks appear in yellow has been reverted. Extra attacks will appear in white and act like any auto-attack. They will no longer reset the swing time of your weapon.
Well, yes, we know they're white attacks. The question is whether they're considered 1H attacks, or DW attacks. It seemed *likely* that they'd be subject to the DW penalty, but we've found more and more that not everything in this game works the way you'd expect, so it's always worth testing these things.
I seem to be having some problems adding gear to the list.
I'm trying to get an idea of what my stats are with current gear, and some of it isn't on your list.
I added a couple of daggers and they seem to be fine, I just copied a dagger and changed all the stats appropriately.
Worked great.
I am trying to add Shadowprowler's Chestguard, but column H just says #VALUE instead of showing a calculated value and doesn't show up in the selection box regardless of which other chest piece I copy from.
Am I doing something wrong or can it just not calculate this value because the piece is weird or something?
For a quick reference:
Shadowprowler's Chestguard (Crafted)
Arm 333
Str 32
Agi 40
Sta 12
Hit 8
1R, 1Y, 1B, +4 Agi socket bonus
03/20/08, 2:43 AM
Similar Threads
