 |
02/08/11, 2:16 PM
|
#271
|
|
Piston Honda
|
Originally Posted by Maje
I have no clue how is that related at all, only thing I did was tried to explain why both devalue crit, which is because of ignite munching 'handling' (modeled in simcraft and a parameter in Rawr).I don't remember saying that at all, so I'm not sure how that is related either.
|
I'm sorry, I should've made myself more clear I suppose. Simcraft has an ignite model which proposes that ignite munching increases beyond a certain haste level, especially with 4PT11 speed bonus to fireballs, which ultimately devalues haste. I just wanted to say that there is nothing like this in Rawr, which means in cases when it values crit over haste it doesn't consider haste devaluation by increased ignite munching.
|
|
|
|
|
02/08/11, 5:37 PM
|
#272
|
|
Bald Bull
|
I've tried to carefully reword several references in the OP pertaining to combat rating relative values. We all know many people come to EJ looking for a 'magic number' to take away. But any numbers given, as people continue to stress, are often only applicable to characters with specific stats/levels of gear. Rather than EJ trying to provide copious numbers for multiple levels of gear, the emphasis is on encouraging people to educate themselves by using Simulationcraft and/or Rawr.
A more accurate way to ascertain the Relative Values applicable to you - is to load your character into Rawr, import your profile into Simulationcraft - or choose/make a profile that more accurately reflects your current stats, gear and encounters you're doing. Not by noting the values listed in the OP and blindly assuming they are equally applicable to everyone.
It is a little disheartening to read comments elsewhere blindly quoting any numbers given, and implying they are applicable to anyone and everyone. There should be some sort of added caveat stating something such as, "But to find out whether these values actually apply to you - load your character into Rawr / SimC and find out yourself."
Last edited by Tyrian : 02/09/11 at 4:02 PM.
|
|
|
|
|
02/08/11, 6:20 PM
|
#273
|
|
Don Flamenco
Gnome Mage
Naxxramas (EU)
|
Originally Posted by Sinless
I'm sorry, I should've made myself more clear I suppose. Simcraft has an ignite model which proposes that ignite munching increases beyond a certain haste level, especially with 4PT11 speed bonus to fireballs, which ultimately devalues haste. I just wanted to say that there is nothing like this in Rawr, which means in cases when it values crit over haste it doesn't consider haste devaluation by increased ignite munching.
|
Firstly both Simcraft and Rawr value haste higher at the gear level we're talking about.
Secondly the Ignite munching model in Simcraft isn't directly related to haste, what I mean is that there might be some indirect relationship between haste and ignite munching but I would guess that the relatioship is tangential at best; the vast majority of the Ignite munching is due to Living Bomb and Pyroblast! ticking (and some fireball + p! sim.crits) I very much doubt there would be any decrease in valuation of haste simply due to Ignite munching.
|
|
|
|
|
02/09/11, 2:34 PM
|
#274
|
|
Von Kaiser
|
I feel as though this may be a bit specific to mages that are either troll, herbalists, or possess an on use haste trinket. Through Wizeowel's guidance :-), I feel a bit enlightened after spending some time with haste values. Just to state the basics of haste...
Spell Duration / ( Base Tick Time / ( 1 + Haste % in decimal form)) = Total Number of Ticks Rounding Up
My newfound enlightenment is in regards to on use haste abilities and Combustion. It is my understanding that haste will not have any diminishing returns with dots and will provide more ticks as haste increases. Using the above formula, each 10% haste will give Combustion another tick. I advise updating the OP and reflecting how powerful these on use haste abilities can be when used in conjunction with Combustion, and even more importantly Impact.
I have not yet been able to find a way to sufficently model the value of trinkets such as [Heart of Ignacious] and their relative value for both single target and aoe situations.
Last edited by Brandox : 02/09/11 at 2:44 PM.
|
Midwinter 4/8 HM
US 6th
|
|
|
02/09/11, 7:28 PM
|
#275
|
|
Glass Joe
|
Was looking over trinket choices, and i noticed something interesting.
As of right now this thread is saying the 2 BiS trinkets are Volcanic Deck (which is hard to argue against) and the Atramedes Trinket. I had a hard time thinking that this would be the case because this thread, and all caster threads to be honest preach "int is king", so why pass up an int trinket?!
Did the math comparing the normal versions, but the difference between the two would even greater if you added more intellect to the Mirror.
Following numbers are based of the "relative values" section of this thread.
359 Mirror - 321 int / 1926 Mastery proc (20 seconds / 2 min internal)
321 int's relative value = 1036,7016
321 Mastery's relative value = 394.4127
So the trinket's "relative value" dps wise is 1431.1143.
359 Bell - 321 crit / 1926 spell power proc (20 seconds / 2 min internal). And lets reforge the crit to hit, to give it the maximum value you can give an item (assuming you need hit, it is our best secondary stat).
128 hit's relative value = 314.8416
193 crit's relative value = 268.1928
321 Spell Power's Relative value = 761.0589
So the trinket's "relative value" dps wise is 1344.0933
Basically, Mirror > Bell, unless obviously you have 372 bell and 359 mirror blah blah. Just making a note because I personally was curious, and I thought I would share my findings. Also, DMC:V is our best trinket, considering it's 12 second's of 1600 int, on a 45 sec internal (which comes out to be over 400 int) and it still has mastery which can be reforged to hit. So unless you are in need of hit, I would use Mirror / DMC:V.
|
|
|
|
|
02/10/11, 7:27 AM
|
#276
|
|
Glass Joe
Goblin Mage
Kel'Thuzad (EU)
|
Hey,
I'm totally new but I did some research on the haste plateau(s). I've used my own gear, so this research especially applies for... well me ;-)
In the OP the value of haste is 1.46, I’m not sure if this is an average but I was curious if it applies before the first plateau (LB, pyro, FFB dots) or after etc.
My research:
Haste percentage------------>DPS Worth per haste rating
base-12,50% haste---------->1,092
12,51% haste----------------->1,82779
12,52-15.00% haste---------->1,092
15,01%haste------------------>1,2511
15,01-25,00%haste---------->1,092
The first plateau is pretty effective because I need just 3xx haste to get it (goblin), so each point haste is very effective. The effectiveness of haste is strongly effected by your buffs, your talents, race and things like the 3% haste from a WL.
Example:
If I need 300 haste to get to 12,51% each point is pretty high rated because the significant dmg buff at 12,51% is divided through 300 haste points.
If I need 600 300 haste to get to 12,51% each point is not so high rated because the significant dmg buff at 12,51% is divided through 600 haste points.
Important note!
I've used Rawr! and 12,50% haste was NOT ENOUGH to get the extra tick, I needed 12,51%!!! So for me it’s worth getting the 15,01% and nort worth to get more than 15,01%. The 1,2511 dps/haste is worth to get (for me) because crit and mastery is rated 1,1xx (for me, again). So if you gonna evaluate the haste plateaus for your gear, remember to aim for XX,01%
It would be awesone if the OP could calculate those values for the average mage, because I don't know how the original values are calculated.
edit: I was warned for this post b/c of hitting enter at the end of each sentence - so I edited it. pretty hard rules here..
Last edited by Drimacus : 02/10/11 at 10:27 AM.
|
|
|
|
|
02/10/11, 12:55 PM
|
#277
|
|
Piston Honda
|
Hey Kavan, has anything changed specifically with Fire modeling between Rawr 4.0.19 and 4.0.16? With the exact same gear and setup, 4.0.16 recommends mostly gemming/reforging/etc. for crit whereas 4.0.19 recommends haste. Once again, these are with exact same gear/talent setup loaded directly from the armory.
|
|
|
|
|
02/10/11, 2:00 PM
|
#278
|
|
Bald Bull
|
Originally Posted by Sinless
Hey Kavan, has anything changed specifically with Fire modeling between Rawr 4.0.19 and 4.0.16? With the exact same gear and setup, 4.0.16 recommends mostly gemming/reforging/etc. for crit whereas 4.0.19 recommends haste. Once again, these are with exact same gear/talent setup loaded directly from the armory.
|
Well the two most obvious changes would be Ignite Munching support and PTR changes added since 4.0.17.
|
|
|
|
|
02/10/11, 5:00 PM
|
#279
|
|
Glass Joe
|
@OP and other Haste>Crit followers
I don't see how Haste is favored over Crit even though the simulators suggest otherwise.
Tyrian I looked at your World of Logs from your raids on 2/8 and 2/9 and it would suggest the exact opposite. Dolcy's overall damage output is significantly higher than yours.
The other question is why hasn't any of the major EU guilds switched to this concept (Paragon, Method)? I don't necessarily think this is wrong I just don't see it in a live raid environment and it might just be situational. You say ignite munching is a major issue and contributes to a loss of DPS but the World of Logs raids say otherwise.
BTW thanks Tyrian for creating and organizing this compendium.
Last edited by VelvetX : 02/10/11 at 5:05 PM.
|
|
|
|
|
02/10/11, 5:59 PM
|
#280
|
|
old and slow
Human Mage
Nordrassil (EU)
|
Originally Posted by VelvetX
You say ignite munching is a major issue and contributes to a loss of DPS but the World of Logs raids say otherwise.
|
This is horribly flawed logic. Now, more than any time in WoW raiding, skill is a larger part of the combat log results than an extra 50-100 theoretical dps. In the past, when mage rotations were based on pressing fireball repeatedly, you could mostly only make a difference by cutting down on lag and getting the perfect combination of stats. Now with the client allowing you to precast, and fixed rotations becoming more complex priority systems, not to mention that boss mechanics are much more twitchy than for example the run away once or twice in 6 minutes at Brutallus, there is so much more scope for skill to largely influence dps. Think about the effect that getting just the right combustion has, or the simple difference of switching to scorch when anticipating incoming damage rather than breaking a fireball. I could give plenty of examples, but the point is that Paragon doesn't need to follow up to the minute theorycrafting, they could beat your dps in while wearing blue gear. It's like when Jeremy Clarkson was pleased with himself for a 9 minutes 59 seconds lap of the Nürburgring, and Sabine Schmitz then did that time in a van, a Ford Transit Diesel. The point is that Dolcy might do more dps than Tyrian because he is just a better player, regardless of who is reforging to haste or crit.
What SimC and Rawr tell us, is what's optimal for the parameters those models are given. If you have something productive to add to the discussion to improve those models, then you are welcome to post it. But coming here and saying "I don't see how Haste is favored over Crit" without any actual non-circumstantial evidence is actually just wasting space.
|
|
|
|
|
02/10/11, 6:25 PM
|
#281
|
|
Bald Bull
|
|
Tyrian I looked at your World of Logs from your raids on 2/8 and 2/9 and it would suggest the exact opposite.
|
Be careful when viewing logs and drawing conclusions. It made me a little nervous when I realised a few people are doing this, because i'm not a good candidate to look at via WOL to try to make comparisons against. It might be ironic given that I wrote the OP, I know. But i'm also usually leading my raids, and my own performance often suffers noticeably when doing so. The difference in personal performance for many, between 'raid leader mode' and 'I can relax and focus on me mode' can be pretty high. I don't get to slip into the latter too often unfortunately, but that's one of the things many GM/RL's have to live with. Like many raid leaders, the former will manifest via making seemingly silly mistakes. (Such as forgetting to reapply LB, low uptime etc). And, like Wizeowl suggested - Dolcy indeed happens to be a multi-class Gladiator, and a strong player - not the sort you want to compare yourself with unless you're playing at your very best yourself.
This isn't really about "haste VS crit followers". Nobody has some sort of personal agenda they're trying to push. The OP is trying to convey the current findings, conclusions from tools many of us have been using for years now. A little skepticism or cynicism towards said findings can be a good thing, it starts debate and gets people talking.
I happened to PM dedmonwakeen earlier (responsible for Simulation Craft) and he commented there was a spike of Simulation Craft downloads in the last 24 hours, since the new Rawr/SimC section of the OP was made. If the current Haste/Crit state of affairs has encouraged curious/skeptical people to whip into shape and finally get around to learning how to use these tools, or trying new ones - that's a good thing. I am thrilled to see a thread like this pop up on the Official Forums, for example: What are you getting for 4.0.6 simcraft value - Forums - World of Warcraft . Also, I'm confident that "Ignite Munching" has been much more thoroughly exposed to the community - people who previously were unaware or didn't understand it. More players can be seen making comments like, "Haste might be better than Crit for some - but you have to check XXX as it varies depending on your gear". The latter part of that sentence is the best part.
At the end of the day, the OP will continue to be updated with whatever the latest research and conclusions suggest from our leading players and trusted tools. dedmonwakeen was initially anxious about the Ignite Munching/Crit changes in SimC, people knew they would be met with skepticism and cause a little controversy. But you know what? The "Haste<>Crit" state of affairs so far, strangely enough, I think has actually been a gift to the community.
People are talking. People are educating themselves and asking questions, because people want answers. People are seeking answers from highly respected tools. All of that is a good thing for the greater community as we move forward.
Last edited by Tyrian : 02/10/11 at 7:41 PM.
|
|
|
|
|
02/10/11, 10:08 PM
|
#282
|
|
Piston Honda
|
Wrong time for combustion
I have searched for this specific advice in this thread and couldn't find it so I thought I should bring this up.
The OP has a great summary of when you should use combustion (as in you should have LB and ignite at the very least, it's even better to delay combustion until you get an ignite, etc. etc.). I just wanted to add from my own experience that it's also better that you don't use combustion when flame orb is up on the target. The biggest reason for this is, you guessed it, ignite munching. When you see a big fireball crit going off, between the time you cast combustion, it is more than likely that the ignite could be munched by a close-simultaneous tick crit from the flame orb. I noticed that I was getting significantly higher overall combustion damage when I mutually excluded flame orb and combustion. You already have LB running constantly on the mob, you don't want to add another source of possible munching until you fire up your combustion.
|
|
|
|
|
02/11/11, 3:08 AM
|
#283
|
|
Von Kaiser
Blood Elf Mage
Frostmourne
|
Originally Posted by afflic
359 Mirror - 321 int / 1926 Mastery proc (20 seconds / 2 min internal)
359 Bell - 321 crit / 1926 spell power proc (20 seconds / 2 min internal).
|
The mirror has a 10% proc rate, the bell is 20%. Both have an ICD of 100 seconds.
Last edited by Wizeowel : 02/11/11 at 6:17 AM.
|
|
|
|
|
02/11/11, 3:34 AM
|
#284
|
|
Von Kaiser
|
Originally Posted by Kolenzo
The mirror has a 10% proc rate, the bell is 20%. Both have an ICD of 100 seconds.
|
Also, if you really enjoy RNG, look no further then Mastery procs. Imagine if you're going through a dry spell on crits for even just 2 or 3 casts. That's a HUGE chunk of the proc being wasted on only Living Bomb ticks and the ignites that may come from it. And with such a low proc chance, you really can't hold out on Combustion for it either.
|
|
|
|
|
02/11/11, 5:27 AM
|
#285
|
|
Glass Joe
Troll Mage
Stormscale (EU)
|
I got a warning last time I posted here for begging, I sincerely hope this doesn't land me int he same situation because it's not my intention.
With regards to Ignite munching, Blizzard said that they were going to try and fix this by rolling up the damage lost to the next ignite. So for example if ignite is running and is overwritten by another ignite, the lost damage that would have resulted had the original ignite ran its full duration, would then be calculated and added to the next ignite, and so on and so forth.
Im not the greatest with all these mathy calculations you guys do here, I read your advice and try to understand it as best I can, but is this change something that the developers of rawr/simC, or even the players who calculate all these numbers, have noticed?
Can't find the blue post now but it was flagged up on mmo forums:
Ignite Munch Fixed!!
|
|
|
|
|
|