Frost Talents, Stat Weights, Racials, Trinkets, FAQ

[oomy]

Is it possible that SimC is overvaluing the Bleed enchant? My sims show the bleed enchant to be worth about 615 dps but my logs are showing me getting anywhere from 450-550. I realise that rng is rng for the amount of damage you get from any proc. But I would expect to see numbers well above 615 as well as below instead of all below.

Does anyone have any logs where you're getting over 650 or 700 dps from this enchant?

rng is rng but simcraft shows 2.4% dmg for the bleed and in my logs it ranges from 1.8% to 2.5% . Mostly around 2.3 - 2.4% . So , fairly close?

[BasicBetch]

Is there a bug with Simcraft over-valuing Comet Storm? I ask this because it is common for crucial bugs to exist in the "sims" that even blizzard doesn't catch causing them to actually never be good...or rarely - too good. In this case, Comet Storm is utter garbage. Don't give me the excuse, "Well it's an instant cast on the move."

I've played around with Comet Storm after seeing it's supposed 2 target Sim. According to the Sim, it isn't drastically behind making player error or perfection come more into play. I can not...get Comet Storm to come even close to a TV build on 2 target. Either something is up with the Sims or something is wrong with the spell itself in the game.

On that note:

I find TV to be overall more reliable as you step into Mythic Raiding. Even fights like Kargath have the boss moving a lot more than the encounter journal would let you assume. The time I actually ran PC for his fight - I dropped the 2nd and 3rd PCs on the Tigers when they were in the flame because that was more reliable than the boss staying in place. Once you can ignore most mechanics when all fights are on farm, PC will be our parsing gem - but until then I'm staying with the more reliable TV.

Someone asked earlier about issues with Splitting Ice Cleave off of PC - Brackenspore is a good example where there's potential for PC but the crystal has to be at the dead center of his hit box for Ice Lance to cleave off the crystal. So...good luck lol.

[Komma]

Is there a bug with Simcraft over-valuing Comet Storm?

I'm not aware of any bugs with Comet Storm. I've just checked and made sure that it is splitting damage correctly.

CmS is behind TV. It's not a huge margin with low gear levels, because you don't get a whole lot of FoFs with low gear levels. When it comes to multitarget, TV honestly isn't that great. The reason why it's "best" is because CmS and PC are even worse.

[Interimo]

I have been having issues with SimCraft with a lot. It will sim me doing 22.4k dps on butcher yet when we killed him on heroic I did 24k. Simcraft is also showing me having dps losses when I equip very obvious upgrades like a 640 to 670 chest piece or 640 to 670 legs. Overall I am having a really hard time taking anything SimCraft has to say as hard fact. So I am taking it with a grain of salt.

Maybe it is just me, I have no idea.

[Komma]

I have been having issues with SimCraft with a lot. It will sim me doing 22.4k dps on butcher yet when we killed him on heroic I did 24k.

Did you set the fight time correctly? Even so, simulations are averages. You're supposed to do better than it sometimes - otherwise how would it be an average?

Simcraft is also showing me having dps losses when I equip very obvious upgrades like a 640 to 670 chest piece or 640 to 670 legs.

This just shouldn't be happening. Did you keep talents and all other configurations the same?

[Frosted]

You may also have used the wrong item ID, or the wrong bonus ID (or used improper syntax) when changing out gear.

You guys really need to include a pastebin of the exact profile you are running if you want help. There are a lot of variables, it's impossible to help if you don't give us what you had problems with.

No, you cannot just link your armory. You must pastebin the exact profile you ran. Armory is not helpful.

I'm not aware of any CmS bugs. When I implemented it during the beta, it was working fine. I suppose I can re-check it, but none of the code it's built off has changed.

[Interimo]

Did you set the fight time correctly? Even so, simulations are averages. You're supposed to do better than it sometimes - otherwise how would it be an average?

Fight time is set for 300 seconds. His 5 minute enrage timer, we downed him at 4 minutes and 55 seconds.

This just shouldn't be happening. Did you keep talents and all other configurations the same?

Nothing was changed in options. I logged out with gear swaps, and imported from armory. I checked that armory had the gear I wished to sim on, on. I then ran the next sim.

You may also have used the wrong item ID, or the wrong bonus ID (or used improper syntax) when changing out gear.

You guys really need to include a pastebin of the exact profile you are running if you want help. There are a lot of variables, it's impossible to help if you don't give us what you had problems with.

No, you cannot just link your armory. You must pastebin the exact profile you ran. Armory is not helpful.

I'm not aware of any CmS bugs. When I implemented it during the beta, it was working fine. I suppose I can re-check it, but none of the code it's built off has changed.

Is the pastebin the import data? If so I have linked that before when asking about rings and how they were not upgrades for me. If it is not then can you please tell me how to go about acquiring the pastebin for the profile?

[Frosted]

I saw it, the googledocs is what we need to see. I'll try to give it some time later this weekend (hopefully will have all the sims I've wanted to run finished by then and can start looking into issues more).

[Interimo]

I saw it, the googledocs is what we need to see. I'll try to give it some time later this weekend (hopefully will have all the sims I've wanted to run finished by then and can start looking into issues more).

Okay, thank you. I'll run a few more sims trying to show different gear swaps like the 640 to 670. I'll upload that import data as well. Another mage in my guild was running into the same issue so I am trying to figure this out. I appreciate the help.

[Frosted]

In regards to you doing different DPS than SimC on a pull, that's absolutely going to happen. There is a section in the SimC guide I wrote that may help explain why : http://altered-time.com/forum/viewtopic ... t-anchor-5" target="_blank

Scroll down to where you see the DPS distribution graph, and begin reading at the paragraph above it.

[Fireholder]

In regards to trinket ranks. Does the sim take into account the 20 seconds delay that you have for the on use coplands and shards together.

[Frosted]

In regards to trinket ranks. Does the sim take into account the 20 seconds delay that you have for the on use coplands and shards together.

Yeah.

[Pappus]

Can someone explain to me in words, why HC shards is better than copelands for frost mages?

For me it feels something is odd about this. A good portion of our damage comes out of the cd phases, which are way stronger with a copeland active, especially pull and second IV.

Haste only help if we have more procs than we can dish out, which ST rarely happens.

[Xenost]

They sim pretty close and if we don't take burst into account Shards Of nothing(HC) is better on paper (hence the higher Ilvl) giving 263 INt + 256 Haste (mean), when Copeland Clarity gives 290 Crit + 244 SP (mean) .

So if we take the mean Shards is way better, but that's not how it works because frost is all about burst and the SP on use is definetly better than the haste.

So I think you will see higher peaks (only the peaks on which you use the trinket of course) with Copeland but lower valley and vice versa, furthermore with the buff to FB valley are a little more meaningful that it used to be, so the 263 perma Int that Shards are giving are more worth than before patch.

It's also worth noting that haste is a really good stat with MI, better than Crit rating in most sims, the fact that you can use the trinket during MI (2 min on both CD) is great, not as good as copeland "on use" but definetly good.

There is probably a way to make the calculation easy, like giving a coeficient to the peaks and using something like

Stats (passive+on use) from the trinket * stat weight * coef on peak
+
Stats (passive) from the trinket * stat weight * coef on valley

For both trinkets and compare the score.
That may not be perfectly accurate but it gives an idea (if we can find good enough coefficients)

Edit: Can someone explain how AMR compare trinkets, it seems to compare them on a mean while accounting the peak only to know if it goes higher than a stat cap, which would explain why my DMFTrinket is rated so low?
I stopped using AMR to compare trinkets after that.

[Pappus]

MI also scales with SP if I remember correctly and they should also benefit more from copelands than from haste.

[TheGreyPaladin]

Awesome work Frosted you rock! The racial image on the front page shows fire race information - can we get a rerun of your sims for race updated for todays WOW? I'm assuming Draenei is still best frost DPS race? Plus self heal! I think I read on another forum its best for fire too so 2 in 1!

[Frosted]

Nah it's the frost stuff (you can tell cause the base profile is Mage_Frost_Highmaul_mythic, which is what all the other profiles are copied off of). The other names saying "Fire" in them just means I forgot to copy/replace their names to Frost before re-running!

[Rinoa]

The OP says that holding FO for PC is a dps increase in almost every scenario. Does this include 2-target cleave such as Brackenspore and Twin Ogron? What about 3-4-5 target cleave/AoE like Tectus?

[Frosted]

2 target


The logic behind it being a DPS gain is not dependent upon a single target assumption. You would consider FO not PC sync'd if there is a portion of the fight with a bunch of adds up for a short period of time. (Imperator transitions).

I suppose I could sim that actually....>.>

[Rinoa]

Fairs.

I'm just thinking that there must be a magic number of targets where casting FO on cooldown is better than saving it for PC. Perhaps I am missing a vitally important point here, that is very possible.

Okay, I'm not good with coding or maths at all, so bear with me here. I just want to convey an idea, hopefully my sleepy self can make sense of it.

From what I understand, on single target, a lot of value can be added to both Frozen Orb and Prismatic Crystal by delaying Frozen Orb to sync up with PC. This is because FO will do twice as much damage, and - not counting the one free FoF proc - also give twice as many FoF procs. On the other side, you get a lot more value out of PC as you will be able to fire more FoF-ILs into it. Okay, simple enough.

Now, if we increase the amount of targets outside of PC from one to two, it seems to me that the increase both FO and PC gain by lining them up with eachother is dimished by a bit. As you showed above, it's still worth it, but there should in theory be less of a difference on 2 target cleave than single target.

Here's the question for me, at what number of targets does the optimal FO/PC usage change from being based around PC (syncing FO up with PC by delaying FO) to being based around FO (syncing PC up with FO, by delaying PC)?

Perhaps it's a silly question, but I am quite curious about this, as I don't entirely know the optimal way to use these cooldowns when there are intermittent add phases such as on Imperator, and long-lasting multi target cleave such as on Tectus.