I haven't tested since 2.3.2 but I have definitely recieved the ignite damage from my Fireblast in the above scenario. The ignites from both crits get lumped into one. It seems to happen only when I get my fireblast off *immediately* after the fireball cast. I will see if it still happens in 2.3.2 and post some screencaps.

The ignite-delay is possibly occurring on all ignites, not just the initial one. I know with improved shadowbolt, the debuff often lags as much as a full second after I crit before the debuff appears (very annoying with nightfall--the second SB won't benefit). Your observation with ignite is basically the same thing. However, there is a qualitative differene between applying a new debuff, and changing one that's already there, so we need additional testing to find out whether ignite gets updated with lag time or not. I think ISB stacks get updated immediately, but then again they're a much simpler debuff since their local data is just stack size. Probably the best way to test this is proc an ignite, refresh an ignite, and look at the combat log delay between the second crit and the new ignite tick. But many times because of lag ><.

And Hatemonkey is right about attack animation. I've seen my missles do crazy loop de loop things while going towards a mob if the mob is large enough and their center of mass (or wherever the missle flies towards on the model) moves around enough during an attack animation. Thanks for reminding me. If the animation is sufficiently synchronized we're basically looking at Phase Of The Moon® theory for missle flight time.

Yes the animation goes crazy at times...but the question is...is that when the actual damage is done or even displayed to the player? I believe i remember time and time again when running away from an incoming bolt (shadow bolts doing BEM dailies come to mind) If you are just running on foot away from the missile, you will take the damage before the animation hits you. If you are on an epic mount, you can outrun the cast. (I'm really sure if I'm running out of range once it is cast or before it is cast) To me, it seems that the bolt simply has to get within a specific range of you before it's considered a hit....and not when the animation hits the player.

It's possible that i'm simply remembering this wrong, could anyone confirm losing hp before the graphic of the bolt hits? You would pretty much need a screen shot to do this.

Maybe, for simplicity's sake of coding it, the server calculates travel time at the time you finish casting your bolt, and sends the damage at the time it thinks its going to land (ie: when you see the damage appear on your combatlog). The bolt is fully drawn client-side, and could take longer to get to the target, but thats not affecting when you receive the damage dealt from the server.

Try to think of it this way. I doubt blizzard would literally 'track' all the bolts you cast and track their positioning. It seems too clunky to code for no real benefits. Additionally, constantly recalculating the travel time left server-side would also seem very overkill. I think they just calculate travel time in a straight line at cast time, and your UI is just not linked with that (ie: the drawing/animation of it).

One easy way to test this: Try casting a bolt, and have the target run away. Note if it takes more time to show on the combatlog (and denote the extra time to show it on the UI too). Then try the reverse -- if your target moves toward you, do you see the bolt land before the damage is shown ? (or rather, is the damage 'delayed' ?)

Based on old snowballing people as they jump on a FP, I do wonder a bit.

Actually thats a great idea. Using fall speed to test this. (I know that's not what you were saying, but you mentioning 'jumping' made me think of this)

Epic flight speed is faster than Fireball's travel time. I've kited mob fireballs around while bored, and I'm 95% certain that I took the damage after I landed (~3-4 minutes after the spell was cast). I'll check again tonight if nobody beats me to it.

One of our hunters early in BC outran a shadowbolt from Kazzak around Hellfire for about 25 minutes. But this is also a mob vs player situation, which may work differently.

I wouldn't be surprised if the animations aren't linked to what's actually happening on the server, as in you might think you're escaping the fireball on the client but fail to do so on the server, but you also may run fast enough on the server as well - depending on your lag and how soon/fast you're running away.

Also regarding the ignite bug that makes you lose damage with fireblast - wasn't that reported to lose either the fireblast or the fireball tick? Because according to this thread and the concepts all the theories/tests are based on, it seems as you should lose the fireblast's ignite every time. Is there a contradiction or is it simply not possible to lose fireball's ignite to the ignite bug (when doing a fireball+fireblast), as it'll always lose that of the fireblasts? In addition, does this mean that if I'm in melee range and have low enough spell dmg for fireblast to be a DPS increase, I will actually not suffer to the ignite bug or is the ignite delay be enough to create the bug even with 0 travel time (as 0 travel time + ignite delay could still be bigger than the time difference between the fireball and the fireblast).

Regarding "delays", I've been seeing similar things in the warrior thread regarding stuff related to shield block that seem to happen later than they should be. Another example of a delay is seal of blood as it seems to always deal its damage clearly after the melee hit even though they're supposed to happen at the same time (seal of righteousness though clearly doesn't have the same effect and does its damage on time).

I casted Fireball -> Fireblast in the screenshot below. The Fireblast impacts ~1s before the Fireball due to the travel time.



4868 critical damage, 1947 ignite damage. I did not lose the Fireblast damage.

How about, you get summoned, get someone to fireball you, accept the summon while the fireball is in mid air, see what happens ? That should give quite possibly the least debatable results.

EDIT: well, in theory we can assume reproducible rolling ignites from 2.5s fireballs. In practice, it never happens. Same for fireblasts really. There seems to be other elements in play. I did read many times over that fireball/fireblast was at least not always working. Same for rolling ignites, I have a good idea how to reproduce them, but in practice it isn't as easy because there seems to be external factors. If anything, I do expect that if I try 5 fireball/fireblast that one of them should give me the bug. I will know better this week end.

Did another test. Intentionally waited for the fireball animation before pressing fireblast. Result:


No Fireblast ignite damage.

I can reproduce both situations 100% of the time under the new casting system. To get the Fireblast damage to count, I just have to tap fireblast a couple times right before my fireball finishes casting. If I don't tap fireblast pre-emptively to fireballs completion, then the fireblast ignite damage will not register.

The only thing I can think of to explain it is that the fireblast must get processed before the fireball if you are quick enough. Maybe a spell processing heirarchy, instants processed before casts/channels when they arrive at essentially the same time?

Once, while world pvping on my hunter, I was running up to an alliance player with low hp hearthing. As I was running into range, I was mashing the arcane shot key, hoping to kill them before they got away. The hearth finished while my shot was in midair, and I watched in amazement as the little purple projectile took a 90 degree turn and rocketed off at top speed towards Ironforge (I assume). I've always wondered if that shot killed them...

If the animation IS linked to the server, then the shot would kill him as he arrived. Due to the animation behavior, that would seem to be the case,

I don't think there's any hierarchy going on, just some server-side delays in updating data fields before they are referenced by the next effect. If you Fireblast too soon:
*Fireball cast, checks for other ignites, sees none, calculates own ignite.
*Fireblast casts and lands, checks for other ignites, seens none, calculates own ignite and adds debuff.
*Fireball lands, fireblast ignite overriden with fireball ignite before getting a chance to tick.
Result: Lost ignite damage from the fireblast.

It's explainable with just the travel-time hypothesis. I think the key is whether or not the fireblast lands in between calculating, and applying, the fireball ignite. However, there does seem to be a contradiction between to events that occurred differently on the player end but were reported the same on the combat log. This gives some insight into the internal processing though. It looks to me that while there is a delay in processing the ignite damage, it's not as long as the delay before the ignite debuff actually appears. In this case I'm guessing it's so small that you only see it if your fireblast is effectively instant via a spell-queue.

Has anyone considered doing some tests with warriors spell reflect?

So say you duel a warrior and you cast a fireball and then cast fireblast immediately afterwards, which would get reflected? the fireblast or the fireball?

If the fireblast gets reflected then we know that travel time has an effect on all things as the dmg or spell effect is not applied until the missile makes contact.

If the fireball gets reflected before the fireblast then we know that spell order matters despite travel time.

If multiple spells land on a spell reflect or a player under the effects of grounding totem within a very small amount of time, they will all be absorbed / reflected - so that being buggy in itself makes it a poor test I would think.

I'll know for sure soon tonight. I know that before 2.3.2, fireball/fireblast did give the results i mentioned from time to time.

You can outrun a spell graphic forever on an epic air mount (some graphics even on an epic land mount), but in my experience the damage is always inflicted eventually, i.e. before the actual graphic hits. Presumably there is either a constant "time from cast completion to damage" or a maximum "time from cast completion to damage".

You might be able to test for this by using a G15 macro + a 1.5 second cast spell and examining damage timestamps... duel someone, have them mount up, then spam scorch while they fly in circles around you?

Ok well I made some tests and basically, I don't get very often ignite bugs with fireball/fireblast. Somewhere between 20 and 25% of the time.

The interesting part is that as you can see in the screenshots, master of elements is not linked with ignite (?). Sometimes you get 1x MOE for 2 ignites, sometimes you get 2. If that weren't enough, the ignite you get seems random too. Sometimes you get the fireblast ignite, sometimes you get fireball ignite. Something else to note; the order in the logs of fireball/fireblast is not consistent either. Maybe there is a link.

EDIT: and before someone asks, on the first screenshot I was using rank 4 fireball to not kill the mob.

Just for informational purposes here, you were at max Fire Blast range (26y) and tapping it just as the fireball finished?

I still feel like this spec isn't crazy viable. I think stacking Ignites has a chance, but as said above it's hard to build your gear specifically around rolling Ignites- aiming for spell critical, haste, and spell damage is near impossible.

Even if one were able to get the gear to support the spec, you're still basing a good slug of your damage on a chance of critting with nice timing, plus the fact that the damage is best when huge cooldowns such as Bloodlust/Heroism are up.

I don't think you can build a spec around rolling Ignite. Can it happen with a regular Fire spec? Sure. Should one aim for it? Nah.

Most if not all of the items you would choose to itemize for such a build are as good as if not better than their alternative non-haste items. Zhardoom + za ring + skull pop on bloodlust and you're already there. Add the bt haste ring, nimble thought and possibly the Zul'jin neck and you're almost at the icy veins number.

Building specifically for it is alot easier with a skull for sure. With that said - the numbers still aren't exact - as far as we're determined thus far you have a window of 100-200ms on any given fireball.