2.0 Multiple stun/paralyze/prone/etc effects result in broken durations

[hugin7]

When crowd control effects are reapplied when an existing effect still has time remaining, duration is refreshed on the tool tips.  However, enemies start moving/attacking as if the effect has been removed.  This issue has been observed with "Prone" and "Paralysis" status effects.

[hugin7]

I've been able to reproduce this issue during almost every encounter with my crowd control-intensive party.  This bug is highly damaging to a CC-heavy play style, causing a lot of frustration when enemies unexpectedly wake up and damage the party when they are clearly under effects of crowd control.

[hugin7]

I just verified that the issue remains in the latest patch.  For example, hitting a pack of enemies with Wizard's Call to Slumber, waiting until a few seconds remain on the Prone effect, then refreshing Prone with Slicken results in the problem.  It seems that the enemies break free when the original effect wears off, ignoring the refreshed crowd control duration.

[hugin7]

Another example: use Cipher's Mental Binding to paralyze an enemy.  Reapplication prior to expiration results in getting attacked by an enemy with full-duration paralysis.

[hugin7]

Yet another example, with latest hotfixes installed: cast Wizard's Slicken to make enemy Prone.  When the duration of Prone is nearly over, apply Paralyzed via Cipher's Mental Binding.  Even though different effects are applied, the enemy still "comes to life" and attacks the party despite being Paralyzed successfully.

[hugin7]

The issue has not been resolved as of Patch 1.04.

[pacman]

Hmm.  I guess I assumed there was some sort of mechanism preventing you from chaining the same CC over and over.  It may be a bug, but I have to say I like it.

[hugin7]

Hmm.  I guess I assumed there was some sort of mechanism preventing you from chaining the same CC over and over.  It may be a bug, but I have to say I like it.

 

This is happening with different forms of CC.  Furthermore, if you accidentally hit the same enemy with multiple forms of CC, they can essentially become CC-immune prematurely.  For example, 12 sec paralysis + 1 sec prone can result in enemy running free after 1 sec... really?

 

Also, the tool tips show that the enemies are "Paralyzed," etc., so there is clearly a disconnect between the tool tips and what actually happens.  This is a poor way to handle CC, which creates an unpleasant experience, so I sincerely hope this isn't intentional.

[Zwiebelchen]

I think I recall reading that CCs have a priority, just like status effects like dazed, blinded, etc. ... a stronger debuff of the same group will overwrite the weaker debuff. So, for example, prone (which is considered stronger than paralyzed, as it causes a higher stat decrease than prone) will overwrite paralysis.

 

I think this is intended behaviour, albeit it's weird.

[AndreaColombo]

If that is the case, the weaker status should be effectively cancelled by the stronger one. Keeping the weaker one on display with its timer can become very misleading.

[Zwiebelchen]

If that is the case, the weaker status should be effectively cancelled by the stronger one. Keeping the weaker one on display with its timer can become very misleading.

 

True. But in that case it's more a visual bug, than a gameplay bug. Then again, I think the displayed behaviour should[/b] be the intended behaviour instead of the opposite.

Edited April 22, 2015 by Zwiebelchen

[Khalid the bear]

can confirm if you reaply paralysis with few seconds (or parts of seconds)remaining. the tooltip shows it renewed but the target starts moving around and attacking

[Khalid the bear]

http://forums.obsidian.net/topic/78277-104-paralysis-possibly-other-debuff-durations-really-messed-up-when-re-castingstacking/

i suggest a merge

[thelee]

http://forums.obsidian.net/topic/78277-104-paralysis-possibly-other-debuff-durations-really-messed-up-when-re-castingstacking/

 

i suggest a merge

 

Saw this thread get bumped and was just about to suggest the same thing.

 

And just to chime in -- it has nothing to do with debuff priorities or stronger debuffs overpowering weaker debuffs.

 

My issue, for example, is literally the exact same debuff re-applied over and over. People can use my saved game and try it out, you'll see enemies running around with Paralyzed debuff remaining.

Edited April 22, 2015 by thelee

[hugin7]

This is not related to debuff priorities.  Regardless of which debuffs are applied or their order, the following appears to happen:

 

1.  The game tracks all debuff durations in tool tips.

2.  The moment the first debuff that disables a unit expires, the game frees that unit without checking whether other disabling debuffs are still in progress.

3.  Despite the fact that the unit is freed, tool tip still shows remaining duration on existing disabling debuffs.

 

The issue can be reproduced easily with Prone + Paralyzed, Paralyzed + Prone, Paralyzed + Paralyzed, Prone + Prone, etc.

 

I do not see how this could possibly be intended behavior.  The issue presents itself as a classic bug in the algorithm in step 2 - the code needs to check whether a unit has any existing disabling debuffs before setting it free, rather than let it go the moment any disabling debuff expires.

 

Personally, I find this issue extremely unpleasant, since I run a CC-heavy party.  Out of all of the bugs I've encountered in this game, this is the one that I consider the highest priority - almost everything else can be avoided by reverting to an earlier save file.  I have had to adjust my play style just to avoid accidentally refreshing CC. 

 

I suspect that RNG-based CC, such as that of arbalest crits, could also interfere with stronger CC.  Having enemies you believe to have paralyzed for 13 sec suddenly wake up due to a random 1 sec Prone effect is a poor user experience.

 

I am also concerned that I've received no response from Obsidian on this bug after having it open for nearly 3 weeks.  I've received communication regarding more benign issues and a simple "we will add it to our bug database" would go a long way.

[thelee]

Do you know if this applies to _any_ debuff effect or just effects that disable enemy movement?

 

It would be really troublesome if it was any debuff effect as that means Chill Fog's short-duration blind effect may be sapping away alot of other debuff effects.

[hugin7]

Do you know if this applies to _any_ debuff effect or just effects that disable enemy movement?

 

It would be really troublesome if it was any debuff effect as that means Chill Fog's short-duration blind effect may be sapping away alot of other debuff effects.

 

Excellent question re: other debuffs.  I've been concerned about the same thing, though it's a bit harder to set up a test, since tool tips and combat logs might be out of line with actual in-game calculations.

[thelee]

it might be conceivable that it's not a calculation issue, but just a specific case where the game is too-eagerly clearing away movement-impairing effects as you say. like, maybe in my related bug, the Fampyrs are moving around but are still suffering the reflex/deflection/etc penalties of being Paralyzed.

 

Of course, only Obsidian would know, but they're mum about this (it would be nice just to acknowledge that this is in their bug tracker).

[thelee]

I've reported this before, but it's still an issue.

 

Simple way to recreate this: hit an enemy with two disabling afflictions with different durations.  Good easy example is a fighter's knock down (prone) with a cipher's mental binding (paralyze). If you have a couple of scrolls of paralyze, you can also try using the scroll once, waiting for some time to pass, and then using a second, so you overlap two paralyze durations.

 

When the *first* disabling affliction wears off, movement and action is restored, even though there's still a second disabling affliction with a remaining duration. This results in bizarre situations where you'll see enemies running around and attacking you even though they still have an active "Paralyze" affliction on them.

 

I strongly suspect that whatever boolean used to disable action/movement is being flipped when the first affliction wears off, even though there's still a second affliction active.

[Aarik D]

Hey thelee,

 

Could you please upload a save file along with an output log via dropbox and post the link here? Then I can take a look into this issue.

 

Thanks a bunch!

[AndreaColombo]

I'll provide a saved game with repro steps in early August when I'm back home if the OP cannot.

[thelee]

I don't have a direct saved game, but this is so consistently repro-able that I'll just find my way to a fight, save the game there, and write down specific steps.

 

I'll have one posted hopefully within a day.

[thelee]

Attached is a screenshot and included is a link to my saved game and the output log in my dropbox: https://www.dropbox.com/sh/eyovun0bxp0jawo/AABhxjXWDDBdU6-AdEHz-dhEa?dl=0

 

Note: I couldn't figure out which of the saved games was the one I wanted, so I included several. It's the one on top of Raedric's Hold right before you fight the Fampyr version of Raedric.

 

In the screenshot are several enemies who are still listed as being Paralyzed but are running around.

 

To reproduce:

 

1. start the fight

2. immediately select durance and use a scroll of paralysis to target all the enemies

3. after a few seconds, use the second scroll of paralysis to target all the enemies

4. almost *invariably*, you will notice some Fampyrs starting to move around while still having a Paralyze affliction listed on their tooltip.

 

 

 

[thelee]

Note: contrary to my original description, this is not as prevalent as I thought it was. I tried it in several situations but could not reproduce it. It might be a Fampyr or Scroll of Paralysis-specific issue. I'm not quite sure. But this Raedric's Hold case is definitely consistently reproducible. The other case that was consistently reproducible (but is long gone now) is the (optional) fight in the Endless Paths of Caed Nua at the end of level 8 that also coincidentally involved a Fampyr (and as a level has many other Fampyrs to fight).

 

I must've mostly noticed it because fights against Fampyrs are the ones where I'm most likely to use lots of Scrolls of Paralysis and other disabling mechanism, and am most noticably frustrated when they don't work (and my casters get dominated).

Edited July 24, 2015 by thelee

[Aarik D]

Hey thelee,

 

I have looked at all 3 of your save files and I was not able to reproduce the bug either. I did see however that when you knock a target prone and then paralyze them, they stand up when the prone status ends and resume their combat stance, however the target does not attack and their action bar is frozen for the duration of paralysis. That is just a really minor bug.

 

Let me know if this comes up again, 

 

Thank you and keep up the good work!

[thelee]

That's frustrating, since the screenshot very clearly shows the bug in action, and I'm willing to wager significant money that if I go home and load up those Raedric's Hold saves I will be able to repro it very quickly. Definitely repeatedly using scrolls of paralysis?

 

Come to think of it, I also had this issue while trying to fight the Master Below (but I don't have a save near there). Once again, very similar symptoms: repeated use of Scrolls of Paralysis, the dragon would get unstuck while still displaying a "Paralyze" affliction (which would frequently result in my death due to that super-damaging breath attack).