castsequence lowers dps big time. 500dps loss

[DLoweinc]

How did this work?



I was wondering why my mage with a /castsequence reset=target/10 scorch, scorch, fireball, fireball, fireball, fireball, fireball, fireball

macro was SO far behind dps on my shaman and moonkin (who both have just /cast LB or /cast starfire macros)

i mean my shaman and moonkin are doing 700-800dps at lvl 64 and my mage is doing just under 400.... (my mage gets 1st gear priorities too)

Ok, worked out my castsequence for my mage.

/castsequence reset=target scorch, Fire Blast, Living Bomb, Fire Blast, Pyroblast, Fire Blast, Living Bomb, scorch, Fire blast, Pyroblast, Fire Blast, Living Bomb, scorch, Fire blast, Pyroblast, Fire Blast, Living Bomb, scorch, Fire blast, Pyroblast //repeat 
/castsequence reset=target scorch, null 
/cast Fireball

which should run as:

0 Scorch 
1.5 scorch 
3 fire blast 
4.5 living bomb 
6 fireball 
7.5 
9 fire blast 
10.5 pyroblast 
12 fireball 
13.5 
15 fire blast 
16.5 Living bomb 
18 scorch 
19.5 fireball 
21 
22.5 Fire Blast 
24 Pyroblast 
25.5 fireball 
27 
28.5 Living bomb 
30 scorch 
31.5 fireball 
33 
34.5 fire blast 
36 Pyroblast 
//repeat from 25.5

This assumes my crit rate is high enough to generate instant cast pyros 100% of the time (which its pretty darn close to). A Fireball gets inserted in to the sequence every time the castsequence lag should kick in after a scorch, as well as any point where the initial sequence is waiting on the fire blast cool down.

Edit: I have approximately a 90% chance to generate an instant cast pyroblast every sequence once the sequence gets running.

[Bigfish]

Well, my cast rotation is based on having a 50% crit rate, proccing hot streak, and getting a lot of instant cast pyroblasts. If you're looking for a more traditional fireball spam, I'd go with:

/castsequence reset=target scorch, null
/castsequence reset=target scorch,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, ,,,,,,,,,,,,,,,,,,,,,
/cast fireball

Ideally, it should get off the first two scorches, and then spam a few fireballs before refreshing scorch. But my macros suck before testing them.

By the way, the above macro (in my previous post) depends on fireball casting when fire blast is on cool down, which it in practice doesn't do, so scrap that whole thing.

[DLoweinc]

so I changed my /castsequences (to cast scorch on my mage, to cast insect swarm on my druid) to straight

/cast fireball and /cast starfire

holy cow the dps difference is insane.

even losing 7%-10% crit against targets doesn't matter, it is still a net dps gain. now i'm only lvl 65 atm so most things don't live longer than 20 seconds (even bosses).

but my guys are doing 1000-1200 dps on bosses at level 65 and that seems ridiculously high to me

[DLoweinc]

ut my guys are doing 1000-1200 dps on bosses at level 65 and that seems ridiculously high to me

That is really high for 65.

At 75, I'm doing about 1400 with my Shaman using a /castrandom LB, CL, LvB macro, and manually applying FS on bosses.

Ya that's what I was thinking. i only do that on bosses (and not always)

i can sustain 800dps during the entire instance which i think is pretty insane for that level. so I guess what I am really saying is "Thanks for telling me about /Castsequence Lowering dps!"

[Vicker]

I'm seeing very little difference. I get about 1100 dps on my 70 mage with his castsequence macro. Perhaps I just have a really good connection.

Edit: I do see the difference in the time it takes to start casting when I watch closely. With pyroblast and fireball in my rotation I spend a large amount of time charging up long casts, so it doesn't seem to be affecting me very much.

[neen]

I don't mean to resurrect this thread if it's a dead issue, but the error in just about all these tests is using dps as the metric. I know the original question is whether castsequence results in a lower dps, but the problem results in whether or not castsequence allows less casts over a span of time. So the proper metric has to be the number of casts over a span of time.

To test this ideally, I would set it up like someone else has posted and do 3 different tests:

1. /castsequence lb, lb, lb, lb, lb, lb
2. /castrandom lb, lb, lb, lb
3. /cast lb

Run each test 5-10 times for 1 minute (or however long is a reasonable time to continually cast) and see if there's a substantial difference in number of casts. One issue that could arise however is that even 1 extra cast over the span of a minute can result in the amount of dps loss the OP was talking about (~500). I don't think I'd be convinced this is a verified issue until a test along these lines has been completed, there's just too much randomness to using DPS to measure this.

Just for ease of use, flameshock/lavaburst macro makes my playing much much easier to manage and I would probably use this unless it was shown to be a huge profit to measure this otherwise. There is a hunter in my guild that had absolute atrocious dps, absolute last in any run. Someone set him up with a decent cast sequence macro to manage steadyshot for him and he was instantly competing for top dps on any run. Just pointing out there's lots of value to things that let you not have to think about the small stuff.

[Harem]

I posted earlier in this thread and it seems to have been ignored.

If there is a difference in DPS its because of a difference in when the next cast in the /castsequence is allowed to take place.

So...

We do the 3 tests Neen above mentions and do it for a few seconds more than a minute. Then we take the log apart and see how many times we were able to cast in EXACTLY 60 seconds.

This removes crits etc. and breaks it down to the only thing that could be causing a REAL difference in DPS which is the possibility of added delay/lag when using one method over the other.

[heffner]

That's exactly the test I did and posted already.

I have also been playing around with using a series of /castsequence events with commas. So far, this appears to resolve the problem as long as you have the proper reset timing in the macro. I am still trying to work this out though, but I am sure it's possible with at least two different sequential /castsequence lines in the macro.

[DLoweinc]

/Castsequence must wait for acknowledgement from the server of a completed cast before it casts the next in the sequence, where /cast doesn't care if the original spell went off or not.

this is the same reason that if you have a cooldown spell in a /castsequence that the sequence won't go past that spell while it is on cooldown.

here is an example.

on my mage i have this:

/use 13

/use 14

/cast combustion

/cast fireball

this allow me to spam this key and it will fireball until i run out of mana, works great. uses trinkets every time the cooldown is up, uses combustion each time the cooldown is up, and when they aren't, just spams fireball.

-----------------------

if i were to change it to:

/use 13

/use 14

/castsequence combustion, fireball

it would cast combustion and one fireball, then get stuck trying to cast combustion until the cooldown is up. it would not "bypass" the spell on cooldown. therefore, /castsequence waits for a successfull cast before moving to the next one in the list where as /cast does not check, it just keeps running down the line.

where this becomes a problem is that you can start casting a spell very briefly before the last one is finished (this is why mods like quartz show latency, so you can cast a bit early). with /cast you can cast as soon as the client believes the spell is done, but with /castsequence, it waits until the cast is finished, verifies from the server that the spell was successful, then moves to the next one. (i may have gotten my client/server operations mixed up as i don't know where the operations really lie, either on a server check or client check, but that really isn't the point and the logic is relatively sound).

i hope that answers some questions. switching to straight /cast macros really increased my dps by an order of magnitude.