1. #1

    Fire Mage SimC stat weights

    ok I ran 7 simulations earlier today using simc and 10,000 iterations following a standard patchwerk format & it is telling me that @ 491 ilvl with my gear my best secondary stat is haste. (fire pve mage)

    I can link the results of all 7 tests but suffice it to say

    hit=3.59 crit=2.45 haste=2.63 mastery=1.90

    so I read that simc has some issue with mages. Is this what is happening here?

    I'd appreciate any suggestions. I've tried reforge lite wowreforge robot and now i am trying to input my stat weights using simc & it throws me this curve ball.

    Thanks

  2. #2
    Do not build fire mage it isnt worth it. Go for arcane.

  3. #3
    Without your armory, it is hard to do our own work to compare and see a reason why. Maybe it is trying to push you to the first haste plateau?

  4. #4
    i'm easily at 3056 haste, but thanks for the replies

    unless it's trying to push me to 4052 for extra combustion tick?
    Last edited by bobcageon; 2013-01-19 at 01:37 PM.

  5. #5
    The Patient
    Join Date
    Nov 2010
    Location
    Berlin, Germany
    Posts
    326
    Quote Originally Posted by bobcageon View Post
    i'm easily at 3056 haste, but thanks for the replies

    unless it's trying to push me to 4052 for extra combustion tick?
    You could be right...do you have an armory link for us?
    Fire Mage 14/14 25H <Integrity> @ EU-Eredar

  6. #6
    4052 combustion tick is for un-glyphed combustion, witch you should never use. Glyphed combustion is 3056.

    I havent run any simc's on fire since nerf as i went arcane/frost 1 week after. What i can tell you is that before the cm nerf haste was really close to crit when i was ~490 ilvl.

    scaled factor weights were about, int = 1, crit = ~0.55, Haste = ~0.46-51.

    it seemed back then that more movement(heavy movement, helter skelter and to some extend light movement) was causing simC to raise haste to as high as 0.51

    Now considering the Cm nerf, witch in essence is the same as a crit nerf, haste could be better for fire in some situations as they where pretty close pre nerf in ~490ilvl(30%crit 6.5k+haste)

    But as smaikii said its hard to judge when you dont post any armory =)

  7. #7
    Thanks again for the reply. I'm don't really like to post much, I prefer reading what others are doing, but i'm just kind of beside myself atm trying to get my toon optimized. I really do not want to go arc so pls bare with me ... i do read logs and stay current but the reforge issue (assuming crit > intel spec) is the route i am going for now until i can build a decent arc spec (cringe hate arc)

    My Armory Link

    THe 4052 combustion rating is from Windry's haste chart raid haste non goblin combustion .. I'm glyphed and its in my xml so simc is taking that into consideration. The type of simulation is patchwerk. I tried earlier in a 5 man to boost my haste to +4100 to see if it helped and from what i saw, mastery was giving me more dps with hit/crit maxed out.

    Windry's Haste Spreadsheet
    Last edited by bobcageon; 2013-01-19 at 05:23 PM.

  8. #8
    First of all, please re-read windry's spreadsheet Line 121. I am using the same =)... first it says combustion(line 81), then if you scroll down further it say combustion (glyphed).. So trust me the glyphed combust cap is 3056 raid haste buffed non goblin.
    There is also 1 Cap you can go for during Heroism 3457.

    But tbh you should not worry much about haste, before nerf it was said that it was not worth sacrificing crit to gain even the 3056 one.


    And tbh if you dont like Arcane, then go frost its way closer to arcane, if not on par in most fights. fire is way behind. and this is coming from a player that really only likes fire, but plays arc/frost ><

    Edit: oh about mastery (without having much knowledge), it was simmed(again pre nerf) to be better than haste at tier14 BiS. so the higher ilvl you get, and as your secondary stat pool(crit/haste) increases, it will become crit>mastery>haste. Though im not entirely sure of when this exchange happens.
    Last edited by stX3; 2013-01-20 at 12:57 AM.

  9. #9
    ok i see that on the spreadsheet & thanks. Ya, everything you are saying is what ppl are telling me and what I have gathered from reading some of the other sites. Now you are getting the dilemna b/c simc is supposed to be the tool to use , right? I mean, if simc isnt doing it correctly, what would be the definitive guide to stat weights to correctly reforge?

    According to the info, it seems that reforge lite is the turn-key app that is getting it much closer to the correct weights.
    Again, as i stated i'm not worried about haste as i know this is wrong. I am trying to understand where the mistake is, how to compensate for it, and what ppl are doing to correct it.

    Talking about frost or arc is not relevant to the fire issue I am asking about. Patch 5.2 arcane will not be where it is now with the exploit and I will play fire and enjoy my mage while everyone else spams blast and scorch. Ppl might do better dps, but they have to be hating every minute of it.
    Last edited by bobcageon; 2013-01-20 at 04:29 AM.

  10. #10
    I get the feeling you read through things too fast. Read my first post again.

    What i was trying to say in as few words as possible:
    Before crit nerf, haste was really close to crit already. It would therefore not surprise me that after the crit nerf Haste is even closer and/or in some cases better, Depending on current stats etc.

    Now with that said i just ran my own SimC on your char @ 3120 haste, 8138 crit, 2733 mastery.

    and i get these weights using SimC v. 510-9 standard patchwerk fight Length 450s (vary length = 20%) 25k iterations

    Scale Factors Int=4.39, Crit=2.51, Haste=2.25, Mastery=2.06

    Normalized Int=1.00, Crit=0.57, Haste=0.51 Mastery=0.47

    Witch is pretty close to what i saw pre nerf were i had roughly the same ilvl as you have now. only that haste is now a bit(from ~0.46-0.51) better, and mastery is way higher than what i ever saw, it was normally around 0.30ish for me back then.

    So i really dont know what you are doing wrong, as I seem to get results that is in line with what i would expect.
    Just noticed you only use 10k iterations that might be why your results is a bit off. what version of SimC are you using?

    Also only use SimC as a guide/help not the absolute answer. Fx i noticed that even in the 510-9 version its still using combustion any time ignite is higher than 12k witch you would never do ingame(you would go higher). Still also requires pyro dot... so it clearly has its flaws
    Use SimC as you use science; SimC is the Theory of WoW, now in science when you have a theory it is time for experimenting to ensure theory also apply to the (real)world

    actions+=/combustion,if=!set_bonus.tier14_4pc_caster&dot.ignite.tick_dmg>=12000&dot.pyroblast.tickin g

    ^Find that line in simC and try and change the ignite number to something similar to your threshold of when to pop combustion, use the lowest number of cause as if you set it too high, it might not use combustion all together. I am not saying this has anything to do with your current issue, but this is something you can play around with if you get it to work.

    About ReforgeLite, Dont import stat weights as you see them in SimC into this addon. The numbers you assign in ReforgeLite is more of a priority list. ie. if you want to reforge crit>haste>mastery. Give crit value 160 haste 140 and mastery 120, as a baseline. after that you can play a little with the numbers to get the reforge results you want, if 160/140/120 does not yield what you want. but again consider the values in the addon as priority not stat weights.

    And about choice of spec, of cause it is up to you, to decide what is fun and what is not, i was simply trying to suggest going frost, as fire gear translates into that easily, i also find arcane boring+getting fixed.
    Although i personally dont find it funny being in bottom 5 dps in a 25m raid when trying to push hc progress and knowing i could be top5. So i went frost witch is okay in fun factor, that stupid pet just drives me crazy =), and played around with arcane.
    Last edited by stX3; 2013-01-20 at 08:08 AM.

  11. #11
    Thanks kindly for the long and in depth reply. I think we're on the same page. I'm just trying to focus on this atm, and the other specs are off my plate, but I understand, and again, thanks.

    Version is 510-9 win 32

    Ok, the 12k threshold is an issue, from talking to other mages, and now your comment. Are we talking raw ignite tic or dps b/c there are 2 values and dps would be 24k @ 12k tick based on what mybigignite and combustion are telling me. The problem is, I seldom see ignite ticks that are above 15k (ignite not dps), so I set my tick at 12k but off course I try to stack procs and increase that value. Sadly sometimes it doesn't work. What would u suggest as a value for this as a baseline to combust (with proc tic)?

    Thanks for the 25k iterations suggestion, that's the sort of thing I am looking for. 10k is what the app suggests.
    And yes, with reforge lite I was speaking of base "turn-key" values, but I understand you suggestion. I will try that as well.

    Lastly, the number's from your simc results make more sense than mine, and seem appropriate from what I'm seeing in game with various reforging.

    So, with crit being relatively close to haste, how can 2 crit = 1 intel? It just seems the weight of the stat isnt there according to simc, yet Windry says he gathered his information from Simc. Some of the information starts to become convoluted when I try to make complete sense out of it. The practical is based on the theoretical prediction and falsified from that basis–it's just not lining up as such. What's funny is that the end result seems to work for these guys, and yourself. lol (kinda would like that part of the process documented sometimes, lol)

    I know what I see, I'm trying to find out of there is a better way, is all. I'm trying to exhaust all avenues.
    Last edited by bobcageon; 2013-01-20 at 05:04 PM.

  12. #12
    First of all, the ignite is the actual ignite tick. I only played fire 1-2 weeks after nerf, and cant really remember how high i could stack ignite there as the randomness of HS procs, made me cry. Any way i still think you should be able to stack up ignite to 20-25k+ on pulls with all procs etc, but again its heavily dependant on getting multiply crits within the 6s ignite window.
    Before nerf i reached ignite ticks for up to 100k on (lucky)pulls, and prolly average of 40-70k. Though you should not aim that high now, i still think 10-15k is a bit low. But as i said i havent played fire for a while, my suggestion is to use what ever you feel is best, as you properly know its a fine balance in hoping for more crits and higher ignite or to loose a decent ignite and having to rebuild it, loosing cd uptime on combust. You know better than I what is a decent and what is a good ignite.

    Any way playing around with the action list in SimC is only something you should do if you are able to get somewhat similar results as what i got(i am using standard action list). Are you still getting the same weights as in your first post or have you been able to get more reasonable results?

    The reason 2crit>1 int, should be obvious with the weights i posted, but ill explain.

    Normalized Int=1.00, Crit=0.57, Haste=0.51 Mastery=0.47

    this means 1 point of int = 1.00, 1 point of crit = 0.57

    So 2 points of crit: 0.57x2 = 1.14

    1.14>1.00 (:P)

    The reason we discuss 1 int vs 2crit is purely because of gemming, as we get double the amount of secondary stats compared to int(160int vs 320 crit)
    It has nothing to do with reforging as we cant reforge to int.

    Im sorry if you already know all this, and i over explained stuff, please dont take offence if thats the case

  13. #13
    Your results were different from mine which is why I said that I cant operationalize the results stated by the guide.

    Your suggestion of 25k iterations gives differing values from my 10k, so it would makes sense that your #'s will change @ 50k iterations, as well, regarding the weights. Either way the source is ALWAYS simc. I'm suggesting that if simc is wrong, so is all subsequent logic.

    Hopefully I haven't confused you further. Thanks again. Your input is appreciated.
    Last edited by bobcageon; 2013-01-21 at 05:44 AM.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •