Glyph of Unleashed Lightning

[Endus]

More likely you just didnt adjust priority lists for movement

A finished priority list doesn't need to be adjusted for movement. Which leads me back to "these aren't polished priority lists and models, yet". You'll note, for instance, if you check the Elemental priority, there's a few entries that are flagged with "moving=1".

Never said they were perfectly polished, but they give general idea and alot of benefical stuff cant be simmed properly for other classes anyway, so even if every other spec suddenly lost 5k dps on the heavy movement for whatever reason, most of them would get it back instantly if we included the other stuff they have for movement

Except you can't do that, because you're taking sims, which exist to remove guesswork and estimates, and making guesses and estimates based on those values.

okay, resimmed default firemage on heavy, 10k iterations for quicker check, still 106,2k dps, will look into others.

That's the point. You originally stated 118.4k. A more than 12k discrepancy is a big deal; it's more than 10% off.

[Sarevokcz]

A finished priority list doesn't need to be adjusted for movement. Which leads me back to "these aren't polished priority lists and models, yet". You'll note, for instance, if you check the Elemental priority, there's a few entries that are flagged with "moving=1".

but the priority list is different. try running GoUL on heavy movement on default. then remove UE,moving=1 from priority, bam 600 dps

That's the point. You originally stated 118.4k. A more than 12k discrepancy is a big deal; it's more than 10% off.

No, as a matter of fact, the list includes default "stationary" specs with Presence of mind and "movement" specs with Scorch for all three specs. Both are there.

[Endus]

but the priority list is different. try running GoUL on heavy movement on default. then remove UE,moving=1 from priority, bam 600 dps

Which means the priority list isn't perfect.

Again, you're simply pointing out reasons the current sim doesn't do what you're trying to use it to do.

No, as a matter of fact, the list includes default "stationary" specs with Presence of mind and "movement" specs with Scorch for all three specs. Both are there.

There's the T14 (H) profiles, and there's "test" profiles. The latter aren't "movement" specs, they're likely being used to test changes to adjust the priority lists to properly account for movement. I'm not even sure it's really a "movement" spec, since the fire mage 509 test is using Invocation, which requires you to finish the channel to get the spellpower buff, and it's a 6s channel.

[Sarevokcz]

Which means the priority list isn't perfect.

Again, you're simply pointing out reasons the current sim doesn't do what you're trying to use it to do.

what, so now properly adjusting priority list is wrong? I made the obvious change in the priority list, removed UE, thus I used simcraft properly.

There's the T14 (H) profiles, and there's "test" profiles. The latter aren't "movement" specs, they're likely being used to test changes to adjust the priority lists to properly account for movement. I'm not even sure it's really a "movement" spec, since the fire mage 509 test is using Invocation, which requires you to finish the channel to get the spellpower buff, and it's a 6s channel.

I used t14H profile and switched one talent, and logically added scorch,moving=1 to the end of the list (since mage priority list doesnt have the if=talent.name.enabled) so it actually reflects what the mage would do. I dont just copypaste specs and press simulate without thought to priority list assuming its perfect.

[Endus]

what, so now properly adjusting priority list is wrong? I made the obvious change in the priority list, removed UE, thus I used simcraft properly.

Have you fully tested all other possible iterations, with every other likely flag set on each ability, to see how those changes affect the resulting DPS? No? Then you haven't been testing the priority. Did you test moving UE below LB on priority? Setting situational flags like "if FS is less than 3 ticks remaining"? If you aren't testing every change as thoroughly as you can, you're just fiddling. Which is fine, but you can't claim your results are definitive.

And the point was; you're seeing that there's issues with the profile. There's issues with other specs as well. Not necessarily with the priority, but the model as well.

I used t14H profile and switched one talent, and logically added scorch,moving=1 to the end of the list (since mage priority list doesnt have the if=talent.name.enabled) so it actually reflects what the mage would do. I dont just copypaste specs and press simulate without thought to priority list assuming its perfect.

Again, did you test a variety of other tweaks and changes to see how those changes affected the performance, or did you just make a couple tweaks and assume you'd fixed it?

Because the latter doesn't work. There's a lot of effort involved in tweaking priority lists, and it's not as simple as "just throw Scorch in there somewhere". And again; simulationcraft does not work well for direct 1:1 comparisons between classes in the first place, because it does not and cannot simulate actual combat. It's a tool. It does not give you a class balance list. That's not what it's for, so please stop trying to use it as such. The data does not work like that.

[Didrah]

So (sorry to interrupt) our best choices for glyphs are CL, FS, FET and UL am i right?

Which of these should never be moved (aka must have)? I know that UL and CL will be situational though.

[Endus]

So (sorry to interrupt) our best choices for glyphs are CL, FS, FET and UL am i right?

Which of these should never be moved (aka must have)? I know that UL and CL will be situational though.

The only one that would fit that description for raiding is Glyph of Flame Shock; if you can use Flame Shock and it'll get the full duration, it's always a DPS improvement. Glyph of Unleashed Lightning will be an improvement for any fight with significant movement, and likely a wash on most fights with relatively light movement; you probably only want to remove that as Elemental if you're standing in once place almost all the time. FET doesn't work out consistently enough to be worth using for the most part.

[Radux]

FET doesn't work out consistently enough to be worth using for the most part.

It should be noted, though, that this does have the possibility to work out as a dps increase pending on the duration of the encounter.

[jace5869]

I was thinking GoFE would be good for Spine, no?

[Radux]

I was thinking GoFE would be good for Spine, no?

Definitely possible.

[binkenstein]

FET doesn't work out consistently enough to be worth using for the most part.

It should be noted, though, that this does have the possibility to work out as a dps increase pending on the duration of the encounter.

I really need to sit down and review this, because I'm fairly certain that there's something I haven't considered here (mostly around how with the glyph FE syncs nicely with Ascendance and thusly all the buffs you'll pop for both)

[oblivionx]

The main problem with GoFE will be that point where you summon the ele and it doesn't do anything or runs the wrong way or the host of issues it has compounded by frequency of summoning it. I really don't know if they addressed it's issues, or should I say actually fixed it's issues.. they certainly have addressed it enough times and not fixed anything.

I was thinking GoFE would be good for Spine, no?

Man, if you can get the ele to actually attack the burning tendon... sure.

[Endus]

The main problem with GoFE will be that point where you summon the ele and it doesn't do anything or runs the wrong way or the host of issues it has compounded by frequency of summoning it. I really don't know if they addressed it's issues, or should I say actually fixed it's issues.. they certainly have addressed it enough times and not fixed anything.

If you mean the targeting, that's been fixed the entire beta.

[oblivionx]

If you mean the targeting, that's been fixed the entire beta.

Great deal, it actually works?

Because I still sit there in mumble saying "hey look at the fire elemental" a lot of times, because it's so borked it's funny as hell.

[Endus]

I really need to sit down and review this, because I'm fairly certain that there's something I haven't considered here (mostly around how with the glyph FE syncs nicely with Ascendance and thusly all the buffs you'll pop for both)

I was referring mostly to how the glyph's benefit affects uptime. I've run the math myself but getting it into a pretty graph seems pointless when you already did that work here which was the same pattern I was seeing, working it out myself;



There's two small windows, from 3:12 to 5:12 and from 9:12 to 10:24 (if I haven't screwed something up) where Glyphed FET has greater uptime than unglyphed. If you KNOW that your kill time will be in that range, glyphing is good. If it won't be, you'll have the same or better uptime without the glyph. The interesting zone to me is that 3:12 to 5:12 range; a lot of boss fights end up in there, so that'll be the range to watch. And, of course, fights that much with uptime due to bosses being untargetable or the like can muck all of this up.

I see what you're saying with Ascendance though; particularly with shortish fights in that first range, you might get a lot of oomph stacking CDs with the first summon and everything else, and then again with the second. Since guardians aren't "saving" stats, it's not about locking in Intellect levels any more, just normal CD stacking for multiplicative effect (and then spilling that benefit over to the FET, which doesn't feed back into the others like popping EM with War Banner and Stormlash Totem would).

Great deal, it actually works?

Because I still sit there in mumble saying "hey look at the fire elemental" a lot of times, because it's so borked it's funny as hell.

Yes, pretty much since the beta launched, the FET will swap targets within a second or so of you switching targets. It's the same AI that pets have with Assist on, we just can't change his behaviour like we could with a pet.

[Deleted]

Definitely possible.

The glyph has it's use. On a fight with more than 3.8 and less than 5 minutes length, you get an uptime of 72 seconds for fire elemental, 12 seconds more than the unglyphed version. I bet that this is a dps increase at least with the PE talent, especially for enhancer.

And of course it has it's strength that you can time it always with elemental mastery, signifcantly buffing it's damage. If you take PE and EM, your göyphed FE gets a nice boost the second time he's up - not only the first time.

[Deleted]

A couple of points about the Glyph of FE that I'm not 100% sure have been taken into account for current simulations on it.

1 - It scales actively with buffs now, so if you pop it unglyphed it will always have a fairly hefty chunk of its duration without any buffs. If you are using a spell power proc trinket for example, if you took the glyph and popped FE with the trinket proc each time, you would get a higher uptime from the proc on the FE with the glyph than without it

2 - The glyph spawns the FE at your feet again. Unless you are in melee range when you drop it, each drop requires a couple of seconds for the elemental to run to the boss, reducing the benefit of the Glyph. I don't know if they point outweighs the damage bonus from the glyph supplied from the first point or not.

[Sarevokcz]

Have you fully tested all other possible iterations, with every other likely flag set on each ability, to see how those changes affect the resulting DPS? No? Then you haven't been testing the priority. Did you test moving UE below LB on priority? Setting situational flags like "if FS is less than 3 ticks remaining"? If you aren't testing every change as thoroughly as you can, you're just fiddling. Which is fine, but you can't claim your results are definitive.

And the point was; you're seeing that there's issues with the profile. There's issues with other specs as well. Not necessarily with the priority, but the model as well.

Polished Live model and priority list works exactly like the beta one, its simply not suited for all tasks in one, you have to tweak it if you want to use it for different task/model. Case in point - the second you put GoUL in and sim heavy movement model, the difference between default priority list and one without UE,moving=1 is almost 1k dps

Again, did you test a variety of other tweaks and changes to see how those changes affected the performance, or did you just make a couple tweaks and assume you'd fixed it?

Because the latter doesn't work. There's a lot of effort involved in tweaking priority lists, and it's not as simple as "just throw Scorch in there somewhere". And again; simulationcraft does not work well for direct 1:1 comparisons between classes in the first place, because it does not and cannot simulate actual combat. It's a tool. It does not give you a class balance list. That's not what it's for, so please stop trying to use it as such. The data does not work like that.

1. Atleast I did tweak the specs to suit the simmed task, unlike you, who just selected heavy movement, doubleclicked default spec and let it sim. And really, It IS that simple to put Scorch at the bottom of priority list to get the general idea of the outcome.
2. The numbers arent obviously "perfect" or present exact reality or fight, I never claimed so, you just every single post accuse me of doing so. But if Bink can draw general idea of %s of total dps based on heavy movement model with tweaks to priority, glyphs or talents, so can anyone else who has basic understanding of the simcraft and classes and their talents he wants to tweak.
3. You claim here, that

Elemental Shaman is one of the classes with a lot of instants. You're making a false distinction based on faulty data. Our mobility is already going to be reasonable without GoUL, and with GoUL, it will be the best mobility of any caster in the game, by a fair margin.

and

And even if we are lower on a standing fight with zero movement or interaction with DPS, something we haven't seen since Patchwerk in Naxx, that doesn't make us unbalanced because, again, we have hands-down the best mobility of any caster in the game.

with ZERO evidence and ZERO examples. So please, elighten me, what DO you base your "facts" on? Again, alteast I put SOME work into tweaking all casters specs to sim properly for the given model (and there are holes that could be tweaked more obviously), while you didnt do ANYTHING AT ALL to back up your claims. Where is your fictionary model or fight that supports your claim? Where is atleast your anecdotal evidence?

Usually, you just come and say that "we dont have enough data to make educated estimate yet, wait for raidbots" when someone comes here and says we will underperform compared to others at some task or whatever and that the burden of proof is on him, since he made the claim.

Now you are doing the VERY SAME THING, so go ahead, post your proof. Until then, all your posts regarding this issue are invalid to me.

[Charge me Doctor]

with ZERO evidence and ZERO examples. So please, elighten me, what DO you base your "facts" on? Again, alteast I put SOME work into tweaking all casters specs to sim properly for the given model (and there are holes that could be tweaked more obviously), while you didnt do ANYTHING AT ALL to back up your claims. Where is your fictionary model or fight that supports your claim? Where is atleast your anecdotal evidence?

Usually, you just come and say that "we dont have enough data to make educated estimate yet, wait for raidbots" when someone comes here and says we will underperform compared to others at some task or whatever and that the burden of proof is on him, since he made the claim.

Now you are doing the VERY SAME THING, so go ahead, post your proof. Until then, all your posts regarding this issue are invalid to me.

If caster can continue casting his filler spell while moving, he is better on movement fights than any other caster who can't do that.
It is logical, he can use filler spell while moving, others can't, this makes ele shaman better.

[Sarevokcz]

If caster can continue casting his filler spell while moving, he is better on movement fights than any other caster who can't do that.
It is logical, he can use filler spell while moving, others can't, this makes ele shaman better.

No, it doesnt, thats the point. everyone has movement tools, I didnt even include some of them and the rough numbers show pretty clearly we arent leaps and bounds above everyone else, unless you assume some nonsense like 90% of time movement fight that never actually happens. Can you point me to some proof to back up your claim?