I enabled the ability for people to browse tps reports by boss.
With this, you can get an idea of what kind of tps range other players of your class are producing against a certain boss.
Not all reports will be included in this index. For now, reports submitted using the new wws rendering engine are excluded, although that will change soon. I also plan to add some loose filtering to help strip out some of the bogus reports. Bogus report examples would be ones with obviously wrong/unrealistic dps time figures(generally a fault of wws), and people submitting reports of them dpsing/healing, not tanking.
Khordam produced 517 tps
Special characters are no longer being displayed correctly in WWS in either renderer. This affects quite a few of the german, french and spanish ability names with all their umlauts and accented characters.
Personally, when fixing this, I would allow for both the original name as well as the one you're seeing in the report linked above, as Lossendil may yet revert that change (or fix whatever bug it is that causes it).
Still love the tool, thanks for the constant improvements!
Edit: Just uploaded another WWS report and noticed special characters appearing correctly this time over, for whatever reason. If you plan on fixing this issue on your end, definitely do allow for both versions.
Last edited by Khordam; 06-09-2008 at 03:20 AM.
Would it be possible to provide percentage breakdowns going across rows (how much of the threat was from damage vs innate for each ability) and down columns (how much % threat did each ability produce). Perhaps even break down the column threat into damage vs innate (of your damage threat, what % came from each ability; of your innate threat, what % came from each ability).
Quick question: I did a tps report for a druid in my guild where she tanked azgalor and under the power gained section it listed mana spring even though she was in bear before the boss got there and never left form. Is that accurate, or just a bug with the WWS?
Anonymous Player produced 953 tps
I've been considering this for a while, but I don't want to clutter it up too much. I'm probably going provide this info as a title/tooltip attribute for certain cells.
Originally Posted by Cicero
I'm thinking hover your mouse over the ability name to see what % of total threat this ability contributed, and hover over the total threat cell of the ability to see which % of that abilities threat came from dmg vs innate.
I couldn't tell you specifically how wws determines when a fight "starts". My guess would be the first hostile action, but maybe it backtracks a few seconds or so before that point for good measure. It could also be overlap from the end of the fight after the boss died.
Originally Posted by OnourisofRavencrest
Another possibility, although remote, is that maybe the druid shifted out real quick to do something w/e(I don't see any healing pot buff though) and the moons aligned and the mana spring ticked at that split second before she shifted back? I do see furor (10 rage) and leader of the pack in her buffs, so there was definately a shapeshift into bear at some point, although maybe it was before the fight started and wws recorded it.
Last edited by bagelbite; 06-13-2008 at 10:16 AM.
Best to put this here. Gehn = bad searcher in Kansan.
I am trying to use this tool (WowWebStats Tps Calculator) to compare the threat output of my guild's tanks.
What I am unsure of is what exactly is the TPS total?
This battle states that my battle time tps is 555, but my dps time tps is 585.
Which one should I use to base my tps off of? Or, can I cheat and add them both together?
You should generally use dps time.
WWS sais they calculate dps time as time where you made combatlog entries within the last 5 seconds. Battle time is the encounter duration.
Both of these are calculated by wws using heuristics, which means every once in a while wws may get the numbers wrong.
Added to guides directory.
Any new thoughts/status on Prayer of Mending? Someone pointed out on another board that you don't have it being modified by Righteous Fury (1.9 multiplier). I know you were trying to figure out what to do with it and was wondering if you came to a conclusion/decision on it.
Is there a way to have it filter out threat from windfury attacks? A checkbox for that would be awesome.
I don't want to make the change unless I can see someone provide some reasonably solid proof that this is the case. I could be wrong, but last time I checked omen's source code, this was not the case. It was quite a while ago and I checked quickly, so I could definately be wrong. It's also very possible omen is just wrong here.
Originally Posted by jere
Did someone test it themselves?
I'm having some difficulty with the calculator. I was checking my TPS from gorefiend on your calc, then I checked it in Coolyo, and then I went back and checked it in yours again. The first report I did was using the 2.4 renderer, the second one I did using your calc was without the 2.4 renderer and I get a difference of almost 60 tps.
With 2.4 Renderer - Anonymous Player produced 1537 tps
With out - Anonymous Player produced 1599 tps
I was going to link the coolyo report as well just for reference but I don't know how =/
Since wws's 2.4 renderer is in beta, it's going to be undergoing changes often. Certain changes will cause bugs in my calculations. I'm working on fixing them right now, as I've found a few problems.
Don't expect dps time, battle time, and effective heals to match up, although they should be reasonably close between the different renderers.
Thank you very much for the bug report OnourisofRavencrest.
Ok, I beleive I fixed all the problems.
Note- If a spell/ability is listed twice in wws due to different spell id's, it will also be listed twice in my calc now. For example, bloodrage may now shows as 2 seperate entries, since the initial payload of 10+ rage is a different spell id from the ticks of rage which follow.
I don't think I will do this.
Originally Posted by veyd
wws doesn't list the damage statistics for windfury seperately. While they do list how many procs you got, you don't know how many times they landed, or how much dmg they did. I would have to make a total guess based off average swing damage and miss rate to implement this feature.
I tested it with a priest and the threat difference proves this without doubt.
Originally Posted by bagelbite
You can see the combatlog of 2 tests recorded with TankadinTps.
Schugy (priest) (with salvation in both tests)
Both in melee-range and without any other threat modificators (enchants).
Gebet der Besserung = Prayer of Mending
Einklang des Geistes = Spiritual Attunement
While the RF test seems a little off(should have pulled aggro at 2079, and aggro was actually pulled between 2124 and 2195), it still seems blatantly in favor of POM receiving the RF 1.9 modifier.
I applied the changes.
Recently viewed reports may take up to a day to show the new values.
Some buffs produce threat too. You can do aggro ping-pong only by buffing each other with might (but not with wisdom, funny ..)
So some buffs generate threat, others not.
I had 2 times the procc from Band of the Eternal Defender in my test with RF and no procc in the second test. This could be a solution (but it's speculation. Threat testing is not entertaining, so I probably won't test this out *g*).
Last edited by Zappi; 07-26-2008 at 04:00 AM.
I want to add a new filter method to help reduce bogus reports from being added to the boss index. I'm pretty happy with the results from the current filters(WowWebStats Tps Boss Index Filtering, although druid filter is lacking), but some reports have very borked numbers for battle time and dps time, which makes resulting tps just plain silly. Merged reports are also problematic, often resulting in 2 or even 3 combatlog entries for the same ability at the same time.
I want to validate them by comparing abilities with cooldowns to time. For example: If a warrior used shield slam more than time ÷ 6 or revenge more than time ÷ 5, or even devastate more than time ÷ 1.5, then either the time figures or number of times an ability was used is innacurate, and this report shouldn't be indexed. I'll need to play with this a bit, and will probably give a little leeway(like maybe 10-20% or so), I just want to identify the ones which are blatantly wrong.
For druids, it looks like I could use
Mangle (Bear) time ÷ 6
But what could I use for paladins? Looking at a wws report for a paladin, how would you find out something isnt quite right?
Maybe count judgements? Is there anyway a judgement could occur more often than every 10 seconds?
What about holy shield? Does haste affect this?
Last edited by bagelbite; 07-31-2008 at 02:55 PM.