Thanks.

So after looking at the spreadsheet and doing some back of the envelope math, I'm not seeing a reason to switch from a MSD to the new CSD. And Hood of Hexing seems to be really good. Like better than Tier 6 good.

To be honest, I don't agree. I think CSD is more desirable than MSD. Let me elaborate.

MSD might be infinitely small dps increase over CSD (for most builds), but the problem that I have with it is that it doesn't increase DPM, whereas CSD does. Additionally, and this is the key point, since focus only lasts 6 seconds rather than 10, I could definately see procs getting lost due to movement or whichever reason. Or things like 'yay MSD proc while casting evocation...' then you get a free 45s cooldown.

On the other hand, after a MSD procs, the mage has 45 seconds worth of dead time. I honestly can't say which situation is better as I don't have any experience with trying to work around hidden cooldowns for procs.


Then again, I just reexamined it with a minor amount of haste. I realized that haste is probably better for CSD than MSD. It is true, but I really don't want to get in the math of evaluating how much my crit% + haste rating needs to be for a spreadsheet CSD to be better.

Bug report: the new boots Footpads of Madness, when picked from the Gear Selector, are missing the 25 spell haste rating listed on wowhead.

Regarding MSD/CSD, consider:

- Spirit is now even less valuable for mages due to the changes in how Evocate works.
- All of our tier helms still spend a significant amount of their item points on spirit.
- The only pve helms with meta sockets available to non-engineers before Illidan are said tier helms (ignoring the Consortium/Heroic Badge helms because, well, they suck).

Given all the above, I'd forgo a metagem and instead use of the many helms with 3 regular gem sockets such as [Spellstrike Hood]/[Hood of Hexing]/[Cowl of the Grand Engineer] etc.

I'll plug the values into the spreadsheet later to get definitive answers, but I'm guessing the dps loss is minimal, and you get the additional benefit of greater freedom with your gem choices (which is very satisfying if you are a terminal socket-bonus-matcher such as myself).

CSD is likely to be flat-out better because of the socketing requirements, opening up yellow gems for hit rating and socket bonuses easily outweighs the tiny potential damage loss.

Your results may vary.

I'm in agreement with Manly on this one, due to the requirements for the CSD, and the fact MSD scales worse the more haste you have I'm definitly going for the CSD.

Also with a 45 second CD I would hate it if my MSD went off when I desperatly needed to refresh that scorch stack.

This may be true for warlocks and for non-arcane mages at low gear levels.

You don't really socket for hit at T6 gear levels. The best gear is T6 hat/(shoulders)/robes/gloves, leggings of channeled elements, tempest/chronicle, skull of gul'dan, belt of blasting.
This puts you to 1.3% miss with fire spells. Capped if you have a draenei in your group.

You can use anetheron's noose or zhar'doom, but you lose huge amounts of hit for small damage gains. It's okay to use them if you're hit capped, but it's not worth it if you had to regem manually.

And with the setup above, you're at 1.3%. You need ~3.8 hit rating more to cap. So your 6dam/5hit gem becomes a 6dam/3.8hit gem, being the damage equivalent of a 12.4 damage gem.

So, socketing yellow/orange is nothing you do at high gear levels. And CSD needs 2 blues, instead of 1 for MSD.
From the sheet, MSD is better for fire by 5 DPS (8 if you count the additional blue gem), CSD is better for frost by 17 DPS (13 with the added blue gem), including a +2dmg set bonus from the blue gem.


If MSD procced right after the beginning of a cast, it would be better because you could use the proc on Evocation, saving you 4 seconds and making you less vulnerable to interrupts from damage (yay for 1.5k mana on Teron!)
It's neat if you spam AE and use a focus proc on Flamestrike or Blizzard.
It scales down with haste though and can mess up rotations/rhythm.

I think it depends on whether we can be fire again for the Sunwell, and how the fights will look like.

Edit: With 0.2s cast delay, I get Cowl of the Illidari Highlord - 1804, Leggings of Channeled Elements - 1807, Vestments of the Sea-Witch 1791, for a fire spec.
I'm at 1.3% miss with LoCE, CotIH and VotSW bring me too far over the cap and lose a lot of their benefit.

When I ran the gear in vontre's spreadsheet Illidary cowl with 4part t6 came out over Leggings of channeled elements with 4part t6 by around 10 DPS. Actually vestments of the seawitch with 4part t6 came out exactly the same or 1 dps above leggings of channeled elements + 4part t6. Maybe I did something wrong but that's what it showed to me.

The main reason for me to go with the CSD now is that I have to be "not-fire" for Illidan attempts. The problems it has with haste and the possibility of wasted procs are also fair arguments, even if it does spreadsheet better than the CSD. MSD would be an awesome gem though if I could put some thought into my next spell, instead of just blindly following the cast rotation and hoping it procs at the right time.

At this point in time though, I'd reccomend for mages to look at the same-ilvl three-socket helms, such as Grand Engineer or Hood of Hexing. Meta-gems are just weak compared to the raw stats that are available.

Just wondering, why do you use 1961 as the base mana value for all your base stats calculation in the excel sheet, but 933 for the buffed and talented values?

It thinks you are running out of mana. Might want to check your shadow priest and potion options.

At the SSC/TK gear level, you can use orange gems to work up to the hit rating cap because +4 hit/+5 dmg is slightly better than +9 damage in the first place.

At the BT/MH gear level, the +12 damage from Runed Crimson Spinels is better than the +6 dmg/+5 hit Veiled Pyrestones. It's easier to cap hit with the random gear bonuses than it is to work up to it with sub-quality gems.

Vontre, I have tried to add items to the ItemDB field in your spreadsheet. No matter what I do, I cannot get the items I create, to appear in the dropdown menu in the gear selector. If I for example add a chest armor piece in 'ItemDB', it can find it in the 'Chest' - section of the spreadsheet, but it still will not show in the dropdown menu in the gear selection section. I have tried redoing the dropdown menu reference, for the seet, by completely removing it, and remaking it with a new reference to the 'Chest' section, thus hoping to force excel to re-read the seet. But without luck. I use Excel 2007, but if the problem is caused by a compatability issue, I do not know.

I've spent 4 days reading through this entire thread looking for users with similar problems, but none seem to have gotten any specific answers. This is in no way meant as an offensive remark, it's simply an explanation of why I've created this post. I cannot completely outrule the fact that I have overseen a pos about it. =)

I wouldn't really be able to tell you, due to performance constraints the item dropdown was not really designed to be flexible as such. My advice would be to put new additions in the place of a useless item in the existing tables.

Totally agree with this. Testing it on ptr, it would always proc while building initial scorch stack. Some of this was due to being the only mage, but it is still wasted in this case. It then would have an 11% chance to hit a scorch in your standard 8/1 rotation. The proc was only lasting 4 sec on ptr, not sure if they fixed it but it was all too easy to lose procs all together at that duration.

Evocation is by far the best use of it if you need the mana. Saving 5 sec(with 2pc t6) is very nice. However, when it does proc on an evo cast, it is entirely wasted unless you interrupt the evo.

Faremma,

When you add a new item to the ItemDB sheet you then need to go to the corresponding sheet (chest/head etc) and add in another line in the same position as where you added the item on the ItemDB sheet. The drop down list ranges come from the individual sheets rather than the big-ass source sheet.

Hope that makes sense.

I know, and that's exactly what I tried, without success. Thanks for the pointer though.

Good things about gems is that they more easily let you convert 1 stat point to another, eg you lose 5 hit on an item somewhere you can easily regain that 5 directly through gems.

When I switch "Improved Scorch on target" to yes, the fire DPS stays the same, however the arcane damage goes up about 1%. I'm not smart enough to fix it myself, but I was wondering this:

Since the fire dps value is the same with Imp Scorch on or off, is the fire dps value I'm seeing with Imp Scorch or without?

otherwise, brilliant spreadsheet

I had a similar issue with the Imp. Scorch button not affecting fire damage. However, when I switched to a spec that didn't include Imp. Scorch, clicking the button on and off did have an effect. I wonder if the spreadsheet assumes that if your spec includes Imp. Scorch, fire vulnerability will always be up, but that if your spec doesn't, you have to specify that someone else is applying it for you?

In any event, I too could use some clarification on this issue.

Vontre could you do me a favor and add some new spells to your current sheet? I would like to find the optimal spell rotation for a 40/0/21 Spec that normally uses Arcane Blast x4 -> Frostbolt x3 rotations, while having Icy Veins active. I tried to add them manually to no avail. The new spells would just be Icy Veins versions of the ones already in the sheet. A switch that adds 20% haste to everything would work as well.

Thank You

This is exactly how it works.

Re: Icy Veins, I'll look into adding this, however currently the spreadsheet ignores minor timered abilities as their effect on overall dps is usually negligible. You're looking at maybe 1.5% overall from combustion, 3% or so from AP->PoM->Pyro, etc. Icy Veins by itself falls under the category of negligible (around 2.5-3%) by itself, though stacked with AP has some stronger benefit.