Rake bugged?

Face-rippin fun.

Moderator: Forum Administrators

User avatar
Bluewulf
Honored
Posts: 72
Joined: Tue Dec 10, 2013 4:19 pm

Rake bugged?

Post by Bluewulf » Wed Feb 25, 2015 3:56 am

I've spent a lot of time on target dummy today since 6.1 came out and seem to find my rake dropping very often with about 2 seconds left in its duration. I know its been known to happen every so often but its been happening pretty frequently now. Has anyone else noticed this so far?
Bluewulf (Redwolf/Bluewulf) - Illidan US

User avatar
aggixx
Exalted
Posts: 2292
Joined: Fri Nov 25, 2011 7:49 pm
Contact:

Re: Rake bugged?

Post by aggixx » Wed Feb 25, 2015 4:00 am

I hate to break it to you but its been like this since 6.0 came out :P

After clipping a bleed, if the remaining time between the 2nd to last tick and the bleed ending is below a certain amount then the bleed simply expires on the last full tick. This is true for all bleeds, but is mostly evident with Rake since it isn't nearly as desirable to ever clip the others.
ImageImage

User avatar
Bluewulf
Honored
Posts: 72
Joined: Tue Dec 10, 2013 4:19 pm

Re: Rake bugged?

Post by Bluewulf » Wed Feb 25, 2015 4:07 am

aggixx wrote:I hate to break it to you but its been like this since 6.0 came out :P

After clipping a bleed, if the remaining time between the 2nd to last tick and the bleed ending is below a certain amount then the bleed simply expires on the last full tick. This is true for all bleeds, but is mostly evident with Rake since it isn't nearly as desirable to ever clip the others.

Wow, dunno how I didn't notice it as much of a problem till now. I feel pretty silly :oops:

Thanks as always!
Bluewulf (Redwolf/Bluewulf) - Illidan US

User avatar
aggixx
Exalted
Posts: 2292
Joined: Fri Nov 25, 2011 7:49 pm
Contact:

Re: Rake bugged?

Post by aggixx » Wed Feb 25, 2015 4:32 am

Going back to replicating the behavior on a dummy it seems like it is more of an issue than it was previously. I used to not be able to produce cases where it faded before half of a full tick's duration remained but now I have no issues. Given the number of people I've seen report this I must wonder if they attempted to fix it but just ended up making it more apparent.
ImageImage

User avatar
Polihayse
Revered
Posts: 459
Joined: Wed Mar 12, 2014 10:46 am

Re: Rake bugged?

Post by Polihayse » Wed Feb 25, 2015 5:12 am

Yes, Bluewulf, I have definitely noticed rake falling off early much more often than it normally happens. I've had the same problem with rip recently too. I thought that the reason for rip falling off early had to do with rip extensions. I don't understand why it still happens.

The design of the bleed should be to fall off at 0 every time. If you cut off 1 second of rake by refreshing it at 5.5 seconds remaining it should still stay up until 0. The one thing that should change is that the final tick should tick for 66.7% of rake's current damage. This same mechanic applies to rip and thrash. Today, I had a rip fall off the target with 1.8 seconds remaining with the boss below 25% health. I hit ferocious bite and the game registered the input a fraction of a second after rip fell off and here I am with no energy and 8 seconds left on savage roar. I wanted to punch a hole in the wall. What is this threshold that you are talking about that determines whether a bleed falls off early? Does it change?

User avatar
aggixx
Exalted
Posts: 2292
Joined: Fri Nov 25, 2011 7:49 pm
Contact:

Re: Rake bugged?

Post by aggixx » Wed Feb 25, 2015 5:28 am

Just did some testing and a bunch of hurling numbers at the wall and seeing what sticks, and the "mechanic" now is that the bleed expires at time t where t is equal to the number of seconds since the last tick occurred prior to the most recent refresh. For example, if you refresh Rake 2 seconds after it ticks then it will fade when it has 2 seconds remaining. That final tick also deals exactly 50% of normal damage instead of full damage, and this occurs 100% of the time which is very different than the behavior in 6.0.3.

Together this means when any rake you have clipped expires you will always lose between 0.5 and 1.5 ticks worth of damage (less if you refresh soon after a tick, more if you refresh a while after).
ImageImage

User avatar
Polihayse
Revered
Posts: 459
Joined: Wed Mar 12, 2014 10:46 am

Re: Rake bugged?

Post by Polihayse » Wed Feb 25, 2015 5:35 am

Crap. I've been refreshing my rakes at like 12.1,9.1, or 6.1 seconds. Does this mean that it a dps loss to do this now, or should I just anticipate it falling off? That weakaura idea for tracking individual rake ticks within its duration seems to be much more appealing now. Maybe something as simple as a timer that counts from 0 to 3 and then back to 0 again when it hits 3 as long as rake stays on the target. At least I assume it is simple. I have no idea how to go about making it.

What do you think is the proper way to work with this new rake behavior to maximize dps?
Last edited by Polihayse on Wed Feb 25, 2015 5:39 am, edited 1 time in total.

User avatar
aggixx
Exalted
Posts: 2292
Joined: Fri Nov 25, 2011 7:49 pm
Contact:

Re: Rake bugged?

Post by aggixx » Wed Feb 25, 2015 5:39 am

I doubt the normal behavior is any significant DPS loss, the reality is you're just going to lose damage no matter what you do (unless you never let rake expire). If they haven't fixed it soon then I may take a look at what implications this has.

I can say that it is never worth letting a refreshed Imp Rake expire now, because the final tick will always deal half damage (aka the damage of a normal rake! what would've happened if you had just refreshed it) AND you're causing a downtime.
ImageImage

User avatar
Polihayse
Revered
Posts: 459
Joined: Wed Mar 12, 2014 10:46 am

Re: Rake bugged?

Post by Polihayse » Wed Feb 25, 2015 8:25 am

According to what you said, I came up with this:

Say you have a fresh rake up that you just applied. It lasts for 10.4 seconds until 4.6 is left on it. You refresh it at 4.6. This means that the total rake uptime is going to be:

15*2-(uptime_remaining-4.5)-time_after_last_tick
15*2-(4.6-4.5)-1.4=28.5 seconds

If you refresh it at 5.9 seconds, you get:

15*2-(5.9-4.5)-.1=28.5 seconds

In both situations, you would have the same outcome. As for the 50% rake tick, that only seems to happen if rake is refreshed above 4.5 seconds remaining. It behaves properly after 4.5 seconds remaining. There still is a dps loss even with a late refresh I think. I did a couple of tests that might help determine this.

I did a test where I recorded 2 applications of rake with the second at about 4.34 seconds remaining on the first rip. I recorded it and used a timer. These are the results:

Image

I was confused at first as to why the rake that falls off early looked like it was doing 12 more dps than the one that doesn't fall off. It seemed like the dps should be equal, but I soon realized error that I intuitively made. An important thing I forgot to include was the damage per energy. The DPE for the first one is 656.9 and the DPE for the second one is 688.6.

I tested out rake again. I applied rake and refreshed it 7 more times after 4.5 seconds left and before it fell off. It turns out that rake fell off at about 118.2. If half of these rakes where applied about .3 seconds after they fell off, then the uptime would be about 113.5 seconds out of a possible ~118.5 seconds. This is about 4.2% of the rake uptime missing with the same energy cost. This means that we lose more damage than we used to if we let rake fall off early.

ShmooDude
Exalted
Posts: 1055
Joined: Tue Feb 08, 2011 5:51 pm

Re: Rake bugged?

Post by ShmooDude » Wed Feb 25, 2015 9:13 am

This seems to apply to all DoTs, not just Rake. The behavior is inconsistent though as I could get non-form moonfire to work bug-free. Then I respeced to Moonkin to test those dots and back to feral spec and now the non-form moonfire was broken as well. The first time when moonfire was bug free, I was naked, but the second time I had gear on, so I don't know if that had anything to do with it. Servers went down before I could test it further.

User avatar
Polihayse
Revered
Posts: 459
Joined: Wed Mar 12, 2014 10:46 am

Re: Rake bugged?

Post by Polihayse » Wed Feb 25, 2015 12:09 pm

It appears that in a span of 2 minutes that it is less detrimental to let rake fall off for 3 seconds once than it is to let rake fall off 3 separate times for .3 seconds.

This new rake really bugs me. I find myself in situations very often where I have a rake up that is ticking for more than the next rake I am about to cast. Most of the time I pool my energy waiting near the end of rake's uptime for a potential weapon enchant or trinket proc. This was rewarding game play if done right. Now, I get slapped in the face every time this stupid dot falls off.

Kojiyama
Revered
Posts: 478
Joined: Mon Jan 19, 2015 1:47 pm

Re: Rake bugged?

Post by Kojiyama » Wed Feb 25, 2015 12:28 pm

ShmooDude wrote:This seems to apply to all DoTs, not just Rake. The behavior is inconsistent though as I could get non-form moonfire to work bug-free. Then I respeced to Moonkin to test those dots and back to feral spec and now the non-form moonfire was broken as well. The first time when moonfire was bug free, I was naked, but the second time I had gear on, so I don't know if that had anything to do with it. Servers went down before I could test it further.
I figured they would have implemented something like they did with HoTs and haste where partial ticks happened at the end based on the remaining time if the durations were offset due to clipping or haste. I guess they would not have looked at it for bleeds since they are not affected by haste.

(Or maybe this is an issue where the logic for partial ticks is kicking in when it shouldn't? e.g. the partial tick gets calculated as if your ticks were hasted, but since they aren't it falls off early without the increased damage? That's what the spreadsheet version would seem to suggest.)
Image

inseedious
Honored
Posts: 113
Joined: Sun Nov 09, 2014 10:15 am

Re: Rake bugged?

Post by inseedious » Wed Feb 25, 2015 12:51 pm

I really hope it is a bug, and not a DoT reworking. Even with proper fixes to DoT trackers, it would totally mess up our ability to apply Rake to build the 5th combo point...
Courier | Nemesis EU

User avatar
Polihayse
Revered
Posts: 459
Joined: Wed Mar 12, 2014 10:46 am

Re: Rake bugged?

Post by Polihayse » Wed Feb 25, 2015 1:42 pm

I made a post about it on the bug forum cause I have nothing better to do right now.

http://us.battle.net/wow/en/forum/topic ... 3?page=1#0

inseedious
Honored
Posts: 113
Joined: Sun Nov 09, 2014 10:15 am

Re: Rake bugged?

Post by inseedious » Wed Feb 25, 2015 3:04 pm

Nice post on battle.net forum. I will reply as soon as battle.net will be accessible again.
Courier | Nemesis EU

User avatar
aggixx
Exalted
Posts: 2292
Joined: Fri Nov 25, 2011 7:49 pm
Contact:

Re: Rake bugged?

Post by aggixx » Wed Feb 25, 2015 3:19 pm

Polihayse wrote:I was confused at first as to why the rake that falls off early looked like it was doing 12 more dps than the one that doesn't fall off. It seemed like the dps should be equal, but I soon realized error that I intuitively made. An important thing I forgot to include was the damage per energy. The DPE for the first one is 656.9 and the DPE for the second one is 688.6.
Yes, when I said 0.5 to 1.5 ticks I was talking in terms of damage for that cast, but that doesn't mean you lose that much damage total. 50% of the time that half tick is going to happen before the 1.5s mark which means the Rake actually did higher DPS than it would normally, if you can reapply quickly in that scenario the impact is minimal or possibly even beneficial: you're increase rake's DPS at the cost of raising its DPE.

Its a lot like the FB multistrike bug really, in regards to the DPS impact really isn't very large it can just be obnoxious/offputting to deal with.
ShmooDude wrote:This seems to apply to all DoTs, not just Rake. The behavior is inconsistent though as I could get non-form moonfire to work bug-free. Then I respeced to Moonkin to test those dots and back to feral spec and now the non-form moonfire was broken as well. The first time when moonfire was bug free, I was naked, but the second time I had gear on, so I don't know if that had anything to do with it. Servers went down before I could test it further.
It should only be bleeds. Hasted DoTs get partial ticks proportional to the time elapsed since the last tick vs the total tick time when they fade, I'd be shocked if they managed to break that too.
inseedious wrote:I really hope it is a bug, and not a DoT reworking. Even with proper fixes to DoT trackers, it would totally mess up our ability to apply Rake to build the 5th combo point...
You're overestimating the impact of this. The only time it is ever really an issue is when your Rake falls off and you're not there to refresh, then you lost a couple thousand damage. In normal play where most of your Rake refreshes happen at 4.5 to 3s left the impact is minimal or even beneficial.
ImageImage

User avatar
Polihayse
Revered
Posts: 459
Joined: Wed Mar 12, 2014 10:46 am

Re: Rake bugged?

Post by Polihayse » Wed Feb 25, 2015 3:28 pm

Do you think I am overreacting to the impact of this new design to bleeds? It is no longer possible to pool energy past 3 seconds remaining for potential trinket and ring procs without suffering significant losses in dps. I used this subtle trick pretty often and now I can't do it. This new change is also affecting rip. Attempting to refresh rip at less than 25% only to have it fall off at 1.8 seconds can have significant consequences and this problem should not exist.

User avatar
aggixx
Exalted
Posts: 2292
Joined: Fri Nov 25, 2011 7:49 pm
Contact:

Re: Rake bugged?

Post by aggixx » Wed Feb 25, 2015 3:33 pm

Polihayse wrote:Do you think I am overreacting to the impact of this new design to bleeds?
Well I wouldn't call it a new design since its a bug, but yes.
Polihayse wrote:It is no longer possible to pool energy past 3 seconds remaining for potential trinket and ring procs without suffering significant losses in dps. I used this subtle trick pretty often and now I can't do it. This new change is also affecting rip. Attempting to refresh rip at less than 25% only to have it fall off at 1.8 seconds can have significant consequences and this problem should not exist.
You can pool to 1.5s and then refresh there, if it fades any time sooner then you're going to gain damage from getting half a tick in less than half a tick's time and in this context you should be there watching it so you can reapply it immediately. Anything past 1.5s if it fades before you reapply you are losing damage.

Yes, you just have to be aggressive with the save FBs, same as it was in MoP (and Cata I guess) when the extensions were bugged. As long as you bite before 2s you should always be safe.
ImageImage

User avatar
Polihayse
Revered
Posts: 459
Joined: Wed Mar 12, 2014 10:46 am

Re: Rake bugged?

Post by Polihayse » Wed Feb 25, 2015 3:40 pm

I assumed it was a design since there seemed to be an attempted change to the way rake works. I wonder what they were trying to do by making these changes. I know that tweaking things in the game can cause other effects to happen, but I feel like a 5% buff to rake's damage should not cause rake to behave in this interesting way. I'm happy as long as this is a bug and it will be fixed soon.
Last edited by Polihayse on Wed Feb 25, 2015 3:43 pm, edited 1 time in total.

User avatar
aggixx
Exalted
Posts: 2292
Joined: Fri Nov 25, 2011 7:49 pm
Contact:

Re: Rake bugged?

Post by aggixx » Wed Feb 25, 2015 3:43 pm

Like I said, this is been an issue since launch, it just that it was much less apparent and functioned in a different way than it does now. They must've attempted to fix it and instead caused this behavior; it doesn't have anything to do with the Rake buff.
ImageImage

Flyx
Posts: 11
Joined: Wed Dec 31, 2014 8:02 am

Re: Rake bugged?

Post by Flyx » Wed Feb 25, 2015 3:50 pm

aggixx wrote:I doubt the normal behavior is any significant DPS loss, the reality is you're just going to lose damage no matter what you do (unless you never let rake expire). If they haven't fixed it soon then I may take a look at what implications this has.

I can say that it is never worth letting a refreshed Imp Rake expire now, because the final tick will always deal half damage (aka the damage of a normal rake! what would've happened if you had just refreshed it) AND you're causing a downtime.
The final tick does not always deal half damage. If you refresh it before pandemic it does always half damage, if you refresh during pandemic then it is % based on time still.

I've been messing around on a dummy for about an hour and have noticed the below.

Within Last Tick Refreshes
If you rake and refresh rake with 0.1 seconds till it expires, you will get a tick at 27 seconds and another at 27.1 seconds that is 3.33% of a normal ticks damage.
If you rake and refresh rake with 2.9 seconds till it expires, you will get a tick at 27 seconds and another at 29.9 seconds that is 96.66% of a normal ticks damage.

Within Pandemic Window Refreshes
If you rake and refresh rake with 3.1 seconds till it expires, you will get a tick at 27 seconds and another at 27.1 seconds that is 3.33% of a normal ticks damage.
If you rake and refresh rake with 4.4 seconds till it expires, you will get a tick at 27 seconds and another at 28.4 seconds that is 46.66% of a normal ticks damage.

Outside Pandemic Window Refreshes
The timer will refresh to 19.5 seconds. Your last tick will be at 3 seconds remaining - the time until next tick upon refresh that is 50% of a normal ticks damage. (e.g Refreshing when current rake has been on the target for 2.9 seconds will cause the refreshed rake to last 19.5 seconds with the last tick at 2.9 till expiry)


What I have not yet tested is the interaction between buffed rakes. If you refresh a BT rake with a non-BT rake with 0.1 until the next tick, is there a tick that is 2.9 seconds worth of BT rake and 0.1 sec worth of non-BT rake? EDIT: Just tested. When refreshing the new rake strength is applied to all of the ticks after refresh, much like pre-6.1.

I can't quite get my head around the implications of this yet though. It seems like you will do more DPS per rake by letting it drop off (up to 1.85%~) but it will cost you more energy. Is the DPE loss enough to offset the DPS increase, do they level out or is it a DPS nerf? It seems to me like this may be a DPS increase in disguise by allowing us to shift more energy into rake (higher damage than shred) by timing our refreshes. It also means we get that 1 energy from the 2 set bonus back a bit quicker.

Too much to think about.
Last edited by Flyx on Wed Feb 25, 2015 4:21 pm, edited 2 times in total.

User avatar
Polihayse
Revered
Posts: 459
Joined: Wed Mar 12, 2014 10:46 am

Re: Rake bugged?

Post by Polihayse » Wed Feb 25, 2015 4:12 pm

I wonder if amplifying this problem that ferals have had will cause blizzard to put more resources into fixing it for good. I have a feeling that they will change rake, rip and thrash back to the way it was behaving though. If bloody thrash were a talent, is there anything about the current behavior of bleeds that suggests its possible implementation eventually? Just a thought that popped into my head.

ShmooDude
Exalted
Posts: 1055
Joined: Tue Feb 08, 2011 5:51 pm

Re: Rake bugged?

Post by ShmooDude » Wed Feb 25, 2015 4:33 pm

aggixx wrote:It should only be bleeds. Hasted DoTs get partial ticks proportional to the time elapsed since the last tick vs the total tick time when they fade, I'd be shocked if they managed to break that too.
Its definitely not just bleeds. Try casting LI Moonfire twice in a row, it falls off with ~1.1 seconds left.

Its actually really weird.

No Gear

19.3 Seconds - Cast first moonfire
20.8 Seconds - Cast second moonfire - Should end at 20+0.3*20 = 26 after 20.8 or 46.8
21.3 - MF tick 1
23.3 - MF tick 2
...
45.3 - MF tick 13
47.3 - MF tick 14

Moonfire gained 0.5 seconds. All 14 ticks did full damage. Should have had 13.5 ticks over 27.5 seconds, instead got 14 ticks over 28 seconds.


Gear w/ 11.31% haste

40.5 Seconds - Cast first moonfire
41.9 Seconds - Cast second moonfire - Should end at 26 after 41.9 or 07.9
42.3 - MF tick 1
44.2 - MF tick 2
46.0 - MF tick 3
...
04.0 - MF tick 13
05.9 - MF tick 14
06.5 - MF tick 15

Moonfire lost ~1.4 seconds. The last tick did ~75% damage. Should have been ~15.03 ticks over 27.5 seconds, instead got 14.75 ticks over 26 seconds.

A single moonfire cast (no refresh) is 10 ticks w/ no haste and 11.06 ticks w/ 11.31% haste.


I've no idea what's actually going on in the background. Since the caster GCD is affected by haste, those 2 moonfire examples should have been in basically the same spot in the tick period (3/4 the way through) but exhibit different behavior. Rejuvenation doesn't seem to be affected though, but I think heals use significantly different code than damage spells

User avatar
Tinderhoof
Exalted
Posts: 2234
Joined: Mon May 24, 2010 10:21 pm

Re: Rake bugged?

Post by Tinderhoof » Wed Feb 25, 2015 4:54 pm

Polihayse wrote:If bloody thrash were a talent, is there anything about the current behavior of bleeds that suggests its possible implementation eventually? Just a thought that popped into my head.
I can say with a high level of certainty that this isn't going to happen in a WoD time frame. Who knows what future expansions will bring, I just don't think it will be this idea.

I think the idea itself was flashy looking because it would put out a lot of numbers, but it's a bit silly when you actually look at the mechanic. An AoE bleed that makes adds a second bleed to all targets. I mean wouldn't just buffing Thrash produce the same result with out adding an additional mechanic? One less thing to have go wrong like the above issues.

User avatar
Polihayse
Revered
Posts: 459
Joined: Wed Mar 12, 2014 10:46 am

Re: Rake bugged?

Post by Polihayse » Thu Feb 26, 2015 3:53 am

I did some more testing with rake. From this chart, it looks like rake has a pandemic threshold of 6 seconds rather than 4.5 seconds right now.

Image

Post Reply