Option 1: Can you clarify how I would do this? Simply switch the expertise rating I input to a negative value? I.e, -145
Option 2: How do I 'normalize' the expertise value?
Appreciated
Nah, leave the ratings you input from your paper doll alone. The value that you will make negative is the one under the EP tab where it says Expertise. I use -4 for mine since I run capped. Then when you do your EP run you will get a value for Expertise like -2.57 just remove the minus sign and that is your EP value for expertise.
As for normalizing you are just working off the assumption that you will get as close to expertise capped as possible without going over and that you will use gems to get there as needed. Since you get 1 point of expertise rating for 2 points of AP on a gem you are just setting the EP value of Expertise to 2.
I am expertise capped too, however setting expertise to negative values under the EP tab seems to make no difference. I have tried from -4 till -20000 and the warning message still stays and expertise EP is not calculated. My config file below
Hey guys, just wondering if you could help me please, I've use Shock and Awe to get the stats for Enhsim, but I keep getting a message that it can't run it as there are errors (enchant not found, melee ans spell haste not found, set bonus not found). I've tried removing, adding and replacing a few of what I think are the right errors, but I'm not 100% sure on what I am doing. Could you take a look at my input and tell me what I'm doing wrong. Thanks
It seemed to work grand on v1.6.8 but i recently downloaded v1.7.0, so maybe i've missed something.
Also, when I was entering these stats into v1.6.8 it was telling me that I needed spell power, again not 100% sure what I'm doing, so it may be something really simple. Any help would be greatly appreciated.
Configlvl80 file that is with 1.7.0 enhsim is old and it has several mistakes in it, config file with explanations seems to be current. Compare to that to see what you must fix, it is same issue as to what Rouncer replied last page.
Two options
1. Run your EP values using a negative number. Remove the minus symbol from the EP value and use that value.
2. Normalize Expertise to 2 considering you will Gem to the cap and if an item has expertise you can gem for more AP instead.
Option number 1 is wrong and misleading.
Option number 2 is simple AND correct.
While optimizing with current raid itemization you end up in Case 1 90% of the time, 10% being strange luck of getting items with heavy expertise only (Case 3), case 2 is for new 80s:
Case 1: some of your gems are with expertise
This fundamentally means expertise EP = EP of other stats present on gems, otherwise you would not reach such situation while optimizing your dps, remember that gems are equal in terms of item budget which would lead to replacing other gems with exp gems. So, if you have expertise gems and hit gems and AP gems, exp EP =hit EP = 2.
Case 2: only expertise gems
This means that without all exp gems you cannot cover cap. If you end up in this case while optimizing -value of exp is indeed >2. This also means that optimal dps gear set returned from wowhead/lootrank with exp =2 is not covering cap. You have to manually increase its value until returned gear set will cover exp cap.
Case 3: no expertise gems
This means that you managed to optimize your dps while exp is inferior to stats on gems, otherwise you would have to replace gems. In this case you can start to lower EP value for exp beyond 2 and "true" value for exp will be reached just before cap is broken.
As for normalizing you are just working off the assumption that you will get as close to expertise capped as possible without going over and that you will use gems to get there as needed. Since you get 1 point of expertise rating for 2 points of AP on a gem you are just setting the EP value of Expertise to 2.
If I go to the EP tab, and change the expertise value there to 2, it puts me over the expertise cap, as expected. I must be misunderstanding what you mean here.
Option #1 shows the true value of expertise if someone wants to know how it compares to other stats. I find knowing the actual number has value to me, personally, since if there comes a time where it's value drops below another stat I want to know about it.
The person I was responding too had just been told that uncapped expertise had an EP value of 1.7 which is clearly wrong as shown by running the EP's using a negative value. Knowing the true value helps to ensure that people aren't dumbasses and stop maintaining capped Expertise.
I do like your simplified Cases though and, if it is ok with you, would like to incorporate that language into the TTT since it simplifies the concept and makes it possible for people to follow "rules" instead of requiring them to actually understand what the numbers are showing them.
Originally Posted by Saabu
If I go to the EP tab, and change the expertise value there to 2, it puts me over the expertise cap, as expected. I must be misunderstanding what you mean here.
no, you aren't setting the EP value of expertise to 2 in the simulator. You are setting the EP value of Expertise to 2 in whatever program/webpage/pawn you are using to evaluate the actual gear.
I think some people might be confused about terminology. The sim calculates EP values by using step values. The EP step value for expertise is set by default to 4 in the sim. This means the sim increases your expertise value (not expertise rating) by 4, and compares your dps with the new value to your dps with your current values. It then makes a calculation to tell you over that step size how much 1 point of expertise rating is worth in increased dps, presenting its findings for all stats in equivalancies to attack power. When Rouncer is saying input negative expertise into the sim, this means input a negative step size into the sim. The sim will then decrease your expertise, calculate dps with the new lower value, and tell you how much attack power you would have to lose to get a similar loss.
Think of it like calculating the slope of a function. Ideally you would use a derivative to find the exact value of the slope at the point on your graph. But if the function is an irrational plot of real world data, and graphs out pretty jagged, this won't provide a helpful prediction. So, you take 2 points on your graph that aren't right next to each other, and calculate the slope between them.
This works great for a steadily (if unevenly) increasing function, like the plot of armor penetration vs. dps. Hit rating has 3 points where the slope drastically changes (the yellow, spell, and white caps), and so the sim warns if the EP step size passes any of these 3 cusp points. Expertise rating has a point where the fuction plateaus, any value beyond this point providing no dps increase. The sim warns if your step size passes this point, and informs that increasing expertise won't add any dps, giving it an EP value of 0. But expertise is not useless, and if you're capped decreasing your expertise will hurt you badly. Thus, since EP is the slope of expertise over dps (as expressed as attack power), you simply find the slope of the function before the cusp point, which is done by inputting a negative step size.
Edit: I apologize for potentially adding confusion when my intention was to clarify. Some people were confused about what "negative expertise" actually means. Whether you use a positive or negative step size for expertise in the sim, whether your calculated expertise EP is 0 or -3, you should use 2 for your EP value for expertise unless you are far below the cap and can't reach the cap with gems. When you gain or lose expertise with a gear upgrade, you can put yourself back at the cap by changing gems at the conversion rate of 2 AP = 1 expertise.
Hey guys, just wondering if you could help me please, I've use Shock and Awe to get the stats for Enhsim, but I keep getting a message that it can't run it as there are errors (enchant not found, melee ans spell haste not found, set bonus not found). I've tried removing, adding and replacing a few of what I think are the right errors, but I'm not 100% sure on what I am doing. Could you take a look at my input and tell me what I'm doing wrong. Thanks
It seemed to work grand on v1.6.8 but i recently downloaded v1.7.0, so maybe i've missed something.
Also, when I was entering these stats into v1.6.8 it was telling me that I needed spell power, again not 100% sure what I'm doing, so it may be something really simple. Any help would be greatly appreciated.
You missed updating ShockAndAwe to the latest version. V4.33 will be released very very soon (ie: within hours) v4.32 is designed to work with EnhSim v1.7.0+
Originally Posted by Jessamy
This works great for a steadily (if unevenly) increasing function, like the plot of dps vs. armor penetration. Hit rating has 3 points where the slope drastically changes (the yellow, spell, and white caps), and so the sim warns if the EP step size passes any of these 3 cusp points. Expertise rating has a point where the fuction plateaus, any value beyond this point providing no dps increase. The sim warns if your step size passes this point, and informs that increasing expertise won't add any dps, giving it an EP value of 0. But expertise is not useless, and if you're capped decreasing your expertise will hurt you badly. Thus, since EP is the slope of expertise over dps (as expressed as attack power), you simply find the slope of the function before the cusp point, which is done by inputting a negative step size.
However given that Expertise rating is now shown to be a smooth and not stepped as previously imagined is there not significant merit in changing this in v1.7.1 to be expertise_rating ep values. The default can then just be changed to 32.
Author of ShockAndAweEnhancement Shaman max dps addon
Please use the EnhSim by Ziff & others to simulate what gear, priorities etc are the best dps. You can use ShockAndAwe to export your paperdoll stats to EnhSim.
This works great for a steadily (if unevenly) increasing function, like the plot of dps vs. armor penetration. Hit rating has 3 points where the slope drastically changes (the yellow, spell, and white caps), and so the sim warns if the EP step size passes any of these 3 cusp points. Expertise rating has a point where the fuction plateaus, any value beyond this point providing no dps increase. The sim warns if your step size passes this point, and informs that increasing expertise won't add any dps, giving it an EP value of 0. But expertise is not useless, and if you're capped decreasing your expertise will hurt you badly.
Correct.
Originally Posted by Jessamy
Thus, since EP is the slope of expertise over dps (as expressed as attack power), you simply find the slope of the function before the cusp point, which is done by inputting a negative step size.
Wrong.
The "true" value of expertise with cap reached is not the one that is just before reaching cap.
It reaches plateau with the help of gems - it becomes as valuable as other gems.
EP value of 0 = undervalued. Generated wowhead list will have low expertise, thus forcing exp value to rise in next iteration.
EP value of 2.7 or whatever is calculated with negative step = overvalued, Generated wowhead list will offer too much expertise, which will lead to replacing exp gems with ap gems which contradicts their pretended higher value.
EP value of 2 allows preservation of the cap and allows focusing on other stats.
In v1.7.0 108 expertise rating with unleashed rage 3/3 makes enhsim think that it is over cap. Still calculating EP gives positive values for expertise rating.
In v1.7.0 108 expertise rating with unleashed rage 3/3 makes enhsim think that it is over cap. Still calculating EP gives positive values for expertise rating.
Good grief not another one. Tukez I think you need to change the wording. How about...
"If you include the EP step value your expertise will go over the cap"
and of course just replace expertise for MH hit rating, and OH hit rating. If they are clueless enough to be over the cap even without the EP step values then this should be a
"Even before including the EP step values your expertise is over the cap, recommend setting the EP step to a negative value"
Author of ShockAndAweEnhancement Shaman max dps addon
Please use the EnhSim by Ziff & others to simulate what gear, priorities etc are the best dps. You can use ShockAndAwe to export your paperdoll stats to EnhSim.
I've been meaning to write a stub to the enhance TTT about exp, hit and EP but Yo! sums it up nicely.
The negative exp thing people do in the sim is only usable to prove that exp still is a superior stat to cap. Same with hit(spell softcap). For lootrank or wowhead you should always insert exp and hit as value=2 as Yo! explains. Using the actual EP value of exp is only usable when your gear is terrible and you are looking for gear to get closer to the cap. If using the actual EP value after reaching the cap, you inflate poor items with a lot of exp or poor gear with a lot of sockets. Same goes for hit.
Don't recommend setting exp to a negative value. Instead suggest to use the value 2 for reference when browsing gear sites. We're talking after reaching the cap here and thus a negative value is misleading.
Baby, you can hold my balls.
10:10 < buu_> Raut: You are a hero of the internet.
I'm having EnhSim crash whenever I try to calculate EP values. It gets as far as '6/14 Expertise skipped' and then crashes when trying to do the calculations for 7/14 (which i cannot for the life of me remember what is). All simulations however run without any problems at all. I have a fresh download of EnhSim 1.7.0 and the gui is linked to the correct enhsim.exe file. I have experimented with different ep steps on the various stats but I always get the same problem.
Are you running on Windows 7 by any chance? I have been having a similar issue recently on my Beta 7 install, where 1.7.0 crashes when it runs the first two EP values (sim works fine)
I'm having some issues with reconciling with the output I get from EnhSim and was basicly wondering if something is wrong with the input. I seem to be getting very high values for hitrating, past spell hit cap (ranging from 2.9-3.1) where I'd expect to get values a bit below 2 atleast.
13.46 + 3.00 (misery) = 16.46 < 17.00 (spell hit cap). As such it is normal that your hit will have a fairly high value (Simply changing my hit to 13.46 puts me at 2.49 EP for hit from 1.40ish)
As for numbers - those are fine, I think they represent wowhead IDs and the sim treats them correctly.
13.46 + 3.00 (misery) = 16.46 < 17.00 (spell hit cap). As such it is normal that your hit will have a fairly high value (Simply changing my hit to 13.46 puts me at 2.49 EP for hit from 1.40ish)
As for numbers - those are fine, I think they represent wowhead IDs and the sim treats them correctly.
He appears to be alliance which would provide him with a free 1% hit which should cover the difference.
Are you running on Windows 7 by any chance? I have been having a similar issue recently on my Beta 7 install, where 1.7.0 crashes when it runs the first two EP values (sim works fine)
I'm having some issues with reconciling with the output I get from EnhSim and was basicly wondering if something is wrong with the input. I seem to be getting very high values for hitrating, past spell hit cap (ranging from 2.9-3.1) where I'd expect to get values a bit below 2 atleast.
this is the config file related to the above outputfile:
mh_speed 2,6
oh_speed 2,6
The 2,6 will cause a problem it needs to be 2.6!!
Author of ShockAndAweEnhancement Shaman max dps addon
Please use the EnhSim by Ziff & others to simulate what gear, priorities etc are the best dps. You can use ShockAndAwe to export your paperdoll stats to EnhSim.
I think our EU bretheren aren't used to using decimal points.
Just a hint, when a new version comes out, try using the default config file first to do any calculations/simulations. If the default config file works fine, then something you're adding is screwing up the sim.
I generally will make a copy, and edit that with my current values, as much as it sucks to reenter data, sometimes it's safer.
I'll rerun it with . rather then , for the values, I assumed the problem wouldnt be there as those values got taken straight from Rawr's export function.