i have a question about spellhaste, doing experiments with the spreadsheet.
on the official blizz forums it's often said that over a certain amount of spelldamage, haste is better than pure spelldamage ; or, better, that they compensate together while going up , giving the better dps increase overall.
i don't remember exactly, but that spelldamage amount on wich we should switch to 1/1 haste-sd should be around 1200.

now, after all of this theorycraft : why into the spreadsheet, with a 5/5 t6 set and comparable stuff to complete the equip (1290 spelldamage raidbuffed without shaman or consumables), gemming for pure spelldamage (12sd gems) gives better results than putting 5haste/6damage gems ??

it seems quite a contraddiction to me.

gimme some hint, i'll soon have 4/4 t6 and i must choose how to gem my t6chest without trashing a ton of badges and golds in experiments ^^'

thanks in advance guys!

I just wanted to say thank you to all who posted to try to help me with my very low DPS on Brutallis. I have posted a link of last nights attempts and kill to show the improvement you were all able to impart on me.
Wow Web Stats

Looks like you're doing 1500+ for most attempts now. That's great to hear! Maybe try using AutoHotKey program that ef posted a few pages ago and see if you can really tweak it. I think you have gear potential to do 150-250 more dps with full consumables given your situation. (spriest/elemental shaman/6 drums/idol of raven goddess / keeping FF).

Also, with that gem tradeoff you are looking at converting 6 +spell, to 5 +haste. I think for pure SF spam (and no latency) the breakeven for that occurs at 1390 +spell. That means that at 1390 +spell, and zero existing +haste, the two gem choices are essentially equal.

You should be a little bit past 1390 with consumables, but your rotation is not pure SF (as Adoriele said), and the default latency in the spreadsheet is not zero. Perhaps you already have some existing +haste also.

Also consider that +haste "pulls ahead" of +spell slowly. Its not going to be a "big" improvement until your +spell is even higher. On the other hand, if you are ever seeing mana issues, +haste is a big loser since it doesn't help your DPM at all.

as a general rule of thumb, if you have decent gear 1 point of spell damage < 1 point in spell haste. if you're getting 45 spell haste by trading 20 spd that's definately an upgrade almost at any gear levels.

i can understand the general spd<spell haste ratio its just i feel liek when i go for haste i gimp my damage and vise versa, i cant seem to find that balance

Uhh, yeah. By definition if you gear for Haste you're gonna have less spell damage (given a certain iLvl). The point is to find out which provides more benefit or, more generally, which set of gear provides the greatest DPS. We can talk about the theory behind what determines DPS, and hopefully work towards more accurate models, but in the end it all comes down to what works best. And the spreadsheet is very good at figuring that out.

fully buffed, ofc.

Yea fully buffed sorry, we don't have an elemental shaman in the guild and I normally get shoved with 3/4 Mages and maybe a Shadow Priest, so I don't get extra spell damage in any other way. So is the 1077 and 1392 numbers based off buffed or unbuffed numbers? And by having 1177 Spell damage and 100 spell haste, am I working correcting and should I keep this 1:1 increase in stats?

Rules of thumb are rules of thumb. Yes. In an infinite mana situation, if all you do is spam Starfire, 1177 and 100 is the most optimum use of spell damage and haste. Good luck getting it with the gear selection we have, but yes, it's something to try and get close to. That said, you'll see that 1392 is the magic number for a 1:1.2 ratio, like what you get from gems. Yes, that small of a difference skyrockets the spell damage necessary. So if you're at 1177 and 0, adding 1 haste is better than adding 1 spell damage, but adding 40 spell damage is better than adding 30 haste, even though it pulls you further away from H = SD-1077.

Like I keep saying. Rules of thumb are good for estimation, but use the spreadsheet if you really want to know what will be best.

On your spreadsheet, Adoriele, I'm confused about something. I'm probably just reading the spreadsheet wrong, but it is saying that at 1139 spell damage, hit cap with 0 haste, a lionseye is equal to a crimson spinel. I was under the impression the spreadsheet would say that at 1392 spell damage. Should I be reading that number as how much 1 haste is worth in spell damage? Again, I assume I'm just reading it wrong, and I should be looking for the "equivalent # of crimson spinels" to say 1.2?