I just went and checked it to make sure I was not crazy. I had just shy of 12k mana. Unbound Seers below Dalaran have about 13k. I was getting returns of 390 mana per tick of Viper Sting.
390x4 ticks = 1560. 1560 / 3 = 520. 4% of 13k is... 520. Therefore you gain 300% of what is drained. And the maximum you can drain is 8% of your maximum mana, and it returns 300% of that, or up to 24% of your maximum mana. This is why Chimera shot owns Viper Sting.
What you are seeing in your log is how much mana she is losing. You are actually gaining triple that amount.
Yeah, double checked my log and it says I drained 130 / gained 390.
Ok, that makes sense. worldoflogs.com was showing my "power gain" to be exactly the same as what I drained from Freya. It looks like this site is just reading a line like "[21:21:46.865] Esoth's Viper Sting leeches 282 mana from Freya" and assuming that's my power gain.
Originally Posted by Florrie
What flavour of hipster racism am i missing today?
I was recently investigating a MM build and noticed something a bit odd about the rotations I was seeing: there is no serpent sting at the beginning. Are the DPS numbers that are being output with MM shot rotation data just assuming that one is there? Or might the shot rotation data be missing the damage from the Serpent Sting DOT because it isn't in the rotation at least once (at the beginning, of course)?
The rotation seems off, as reported before. Putting readiness last and having it never go off is a DPS gain. Shandara is assuredly looking into it. I hope.
I switched to a MM build for our raid last night, and have a question regarding the damage distribution.
I used the standard rotation (cast the most damaging shot available first.) Since I was new to the build, I didn't cast rapid fire as often as I should have. Other than that, I believe I stuck with the rotation reasonably well.
Spreadsheet damage distribution (89a) (from the shot rotation page):
Total SS% Raid%
---- ----- -----
Autoshot dps 1576 26.5% 29.1%
Chimera dps 1453 24.5% 19.8% (Chimera + Chimera-Serpent)
Aimed dps 549 9.2% 5.4%
Arcane dps 696 11.7% 10.1%
Steady dps 856 14.4% 18.9% (Steady + Piercing)
Other dps 809 13.6% 16.7%
World of Logs for the raid%: http://www.worldoflogs.com/reports/hTMfUqbgIPFyB3km/details/6/
The question I ask is: do other Marksman hunters also see less damage than expected with their Chimera shot? The spreadsheet appears to say that the total autoshot damage should be ~8% more than the total chimera shot damage, but I'm just not able to do that in a real raid, both integrating over the entire raid, or looking at individual bosses. (There are also other variations, but I am focusing on Chimera at the moment.)
I switched to a MM build for our raid last night, and have a question regarding the damage distribution.
I used the standard rotation (cast the most damaging shot available first.) Since I was new to the build, I didn't cast rapid fire as often as I should have. Other than that, I believe I stuck with the rotation reasonably well.
Spreadsheet damage distribution (89a) (from the shot rotation page):
Total SS% Raid%
---- ----- -----
Autoshot dps 1576 26.5% 29.1%
Chimera dps 1453 24.5% 19.8% (Chimera + Chimera-Serpent)
Aimed dps 549 9.2% 5.4%
Arcane dps 696 11.7% 10.1%
Steady dps 856 14.4% 18.9% (Steady + Piercing)
Other dps 809 13.6% 16.7%
World of Logs for the raid%: http://www.worldoflogs.com/reports/hTMfUqbgIPFyB3km/details/6/
The question I ask is: do other Marksman hunters also see less damage than expected with their Chimera shot? The spreadsheet appears to say that the total autoshot damage should be ~8% more than the total chimera shot damage, but I'm just not able to do that in a real raid, both integrating over the entire raid, or looking at individual bosses. (There are also other variations, but I am focusing on Chimera at the moment.)
How do you get these numbers? I don't see them directly on the shot rotation page, but I clearly see how you can get them - it just seems slightly cumbersome to do this. Also, I think it's a really really nice idea.
Shandara, would it be possible to get an overview similar to this so we can compare the spreadsheet to WWS logs for example? It would be really interesting to see how close we can get to the predictions of the spreadsheet, both to understand where the spreadsheet is perhaps lacking in precision and also as a measure of how well we can keep the rotations going and so forth. Just need a view so we can switch between WWS logs and the spreadsheet to directly compare.
Also according to your numbers there the spreadsheet is saying your CS damage should be close to your Auto damage, you claim it says otherwise - I don't get it. The balance between my CS and Auto damage is slightly off, spreadsheet seems to think CS does more damage than it does in actual raiding. Possibly happens when pushing back CS?
The SS% were obtained by dividing the shot dps as stated on the shot rotation page by the total hunter dps (not including the pet.)
Also according to your numbers there the spreadsheet is saying your CS damage should be close to your Auto damage, you claim it says otherwise - I don't get it.
No, you misunderstand. I am observing I was not able to get my CS damage in actual raids anywhere near my autoshot damage (as the SS claims I can), and I was asking what other people experienced. If you're actually able to get the two close, could you please post a log so I can analyze the differences between your rotation and mine. (Hopefully the problem is my rotation or my interpretation of the spreadsheet, and not the spreadsheet itself.)
The spreadsheet says my autoshot damage should be about 26.5/24.5 = 1.08 times more than my chimera shot damage (which is recorded in the world of raid log as chimera plus chimera-serpent.) The log shows a ratio of 29.1/19.8 = 1.47, which is significantly different. This difference can be argued as statistically real (rather than just due to random bad luck), since if you look at each (successful) boss you see the following:
Ratio of autoshot damage to chimera shot total damage
Archavon (28.7/(14.7+6.0)) = 1.39
Emalon (30.5/(13.5+4.9)) = 1.66 (this is most certainly due to my not getting serpent sting up on the adds)
Freya (21.9/(11.5+4.1)) = 1.40
Mimiron (29.4/(13.2+5.1)) = 1.61
Note: actually the full raid report shows my auto/CS ratio as 27.1/(13.2+5.3) = 1.46; in my original post I had removed the non-ranged attacks and got a slightly different value. The conclusion remains the same, though.
Last edited by Cranch : 05/29/09 at 8:09 PM.
Reason: clarification
I'll see if I can get some good parses this week, had some issues with getting complete ones and not all fights are equally well for analysis. From what I could see of limited parses, hence also limited correctness, my total CS damage seems to be 3-4% lower than what the spreadsheet says it should be. What surprises me most with your numbers is how much your autoshot damage seems to be, mine are only around 23% of total damage.
"The spreadsheet appears to say that the total autoshot damage should be ~8% more than the total chimera shot damage"
That's what I meant, the spreadsheet is not saying that. Numbers you posted say 2%.
Edit: Ah, I thought you were talking about percentage points and not actual percentages >.<
Your labels indicate that Piercing Shots damage is rolled into Steady shot damage. I believe Shandara just updated the spreadsheet to calculate the Piercing Shots contribution within each shot it affects. So the spreadsheet is probably increasing Chimera, Aimed, and Steady by a smaller amount rather than all of it in steady as your label indicates your WoL data does. That could contribute to some of the "missing" Chimera damage, though is certainly not enough to explain the entire discrepancy.
EDIT: I guess the other thing to do is compare the average damage per shot listed in your empirical data to the average damage per shot listed in the spreadsheet. The discrepancies have to either be because of frequency differences (i.e. "your rotation," as you've guessed), or damage differences (caused perhaps by debuffs/buffs the spreadsheet is assuming 100% uptime on falling off, etc).
"The spreadsheet appears to say that the total autoshot damage should be ~8% more than the total chimera shot damage"
That's what I meant, the spreadsheet is not saying that. Numbers you posted say 2%.
The numbers I posted say that the spreadsheet autoshot would be 26.5% of my total damage and spreadsheet CS 24.5% of my total damage. That means that the damage ratio of the spreadsheet autoshot to CS is 26.5/24.5 = 1.08+; i.e., autoshot does about 8% more damage than CS.
I'm sorry if this is a nooby question cause this is my first time using the spreadsheet, but i was wondering why i can't import my armory page for all of my gear and talents... nothing happens when i click the button. Is it because the document is a read only document? Also, when inserting in a custom piece of gear or inserting my talents manually, it doesn't let me (for this, i'm sure it's because it's a read only document)... Any help is appreciated. Thanks
Shandara, would it be possible to get an overview similar to this so we can compare the spreadsheet to WWS logs for example? It would be really interesting to see how close we can get to the predictions of the spreadsheet, both to understand where the spreadsheet is perhaps lacking in precision and also as a measure of how well we can keep the rotations going and so forth. Just need a view so we can switch between WWS logs and the spreadsheet to directly compare.
This is a good idea. Also are people using the rotation test for comparisons? I really feel that is more accurate for shot intensive specs (i.e. MM and SV).
This is a good idea. Also are people using the rotation test for comparisons? I really feel that is more accurate for shot intensive specs (i.e. MM and SV).
That's an excellent idea! In the Rotation Test tab "Shot" area, e.g.,
we could add another column containing the expected damage per shot (i.e., noncrit * noncritrate + crit * critrate)? Then we can use the test rotation to simulate dps. (Presumably you already do this via hidden calculations.)
Would actually be really nice, set the fight length to what the log is and also see frequencies in the same comparison table. List average shot damage as well.
So you have predicted damage %, frequency and average damage by shot columns.
Tools > Macro > Security > Security Level "Low" (bottom option) could be your solution.
When i go to the tools menu, i don't see macro, i just see "Macros..." and then it starts setting up for me making a macro. I'm using office version 2008, so that might be why... and more suggestions?
When i go to the tools menu, i don't see macro, i just see "Macros..." and then it starts setting up for me making a macro. I'm using office version 2008, so that might be why... and more suggestions?
Office 2008 is for Mac. That version of Office removed support for VBA, which is intrinsic to many of the operations in the spreadsheet. You'll need to run a Windows version of Office to use all the features.
Office 2008 is for Mac. That version of Office removed support for VBA, which is intrinsic to many of the operations in the spreadsheet. You'll need to run a Windows version of Office to use all the features.
oh ok... so does that mean i can't use the spreadsheet if i dont have a PC?
oh ok... so does that mean i can't use the spreadsheet if i dont have a PC?
In a word, yes.
In response to Cranch/Whitemane. I'll have a look at what I can do. I've been toying with the rotation test a bit more because it really feels that a simulation is more accurate when it comes to rotations, rather than bland frequency based calculations.
The downside of only having a simulation is that it is slow and non-deterministic. It'd be nearly impossible to determine a real gear upgrade unless we did many iterations (which is very slow).
oh ok... so does that mean i can't use the spreadsheet if i dont have a PC?
A lot of people use Boot Camp/Parallels to run Windows on their Mac. Supposedly Office 2004 (which is also unique to Mac) supported a few more of the spreadsheet features (but not all of them). In any case, Office 2008 won't work with most of the scripted features, and you'd be best served by getting over to a Windows version somehow.
In response to Cranch/Whitemane. I'll have a look at what I can do. I've been toying with the rotation test a bit more because it really feels that a simulation is more accurate when it comes to rotations, rather than bland frequency based calculations.
The downside of only having a simulation is that it is slow and non-deterministic. It'd be nearly impossible to determine a real gear upgrade unless we did many iterations (which is very slow).
Agreed. While it is possible in principle to replace the Monte Carlo analysis that simulation is doing for us with a more detailed analysis of each shot (e.g., its mean and variance under differing conditions) that would allow us to get a damage distribution in one iteration only, that's too prone to difficult-to-find errors.
We already have a quick way to analyze gear upgrades (the current SS), and a more crufty but more accurate way (simulationcraft.) It's more crufty because it doesn't have a nice table of items; you have to set stats by hand. What we don't have is a way to compare your rotation versus the "ideal" rotation (I wrote a combatlog parser to do this but a lot more work is needed.)
A first step in that direction would be to print a log similar to worldoflogs as an output from the rotation test. I don't have a problem with pushing a button and letting that run for several iterations (in order to get variance estimates), as I wouldn't need to do that very often. The existing mean-based calculations are still useful for estimating gear upgrades in a way that's good enough for almost everyone.
Of course, the ultimate goal would be to have a nice perhaps Excel-based front end to simulationcraft so that you could just import your gear, talents, glyphs, specify a rotation, buffs, and boss, and push a button to get a damage distribution.
Of course, the ultimate goal would be to have a nice perhaps Excel-based front end to simulationcraft so that you could just import your gear, talents, glyphs, specify a rotation, buffs, and boss, and push a button to get a damage distribution.
I think someone working on SimulationCraft said that it was using the spreadsheet calculations anyway, so SimulationCraft is just presenting the information differently but it's all derived from the same calculations?
05/29/09, 2:13 PM
Similar Threads
