Shaman appears to slow enemy... but doesn't

[norbert]

I've noticed this twice before: sometimes, the Elvish Shaman slows one or more enemy units, but those enemy units aren't slowed at the start of the enemy's next turn. (Nope, the enemy units didn't level, nor did they have some kind of super power.) In a recent game (corrupt replay), my Shaman slowed several units in a row and died the same turn. The AI units were not slowed during their next turn.

[monochromatic]

Isn't slow reset at the beginning of a new turn? I mean, since the game was a survival, I assume the AI was player one. Player 2 slows AI units, but then on refresh of a new turn, AI units go back to normal. Or am I wrong?

[norbert]

Isn't slow reset at the beginning of a new turn? I mean, since the game was a survival, I assume the AI was player one. Player 2 slows AI units, but then on refresh of a new turn, AI units go back to normal. Or am I wrong?

Yes, you're right. What I meant was the enemy units weren't slowed after the enemy's turn, even though they 'were slowed'. We're currently in another Team Random Survival where it happened again:

<norbert_> it happened again in the game we're currently playing
<norbert_> enemy turn 11, Shaman at 8,1 against Cavalryman at 7,1
<norbert_> I heard the slowing sound and saw the attack
<norbert_> but the Cavalryman wasn't slowed
<norbert_> we're still in the game

[Edit: By the way, I've seen it happen in three different scenarios so far. In all of these cases we were playing with the default era.]

[psychic]

I guess it is going to be hard for the devs to do something based on your evidence.

[zookeeper]

Nothing in this thread gives me the impression that you're aware that slow disappears when the slowed unit ends its turn.

[Anonymissimus]

Isn't slow reset at the beginning of a new turn?

I think the intended behavior is that the unit is slowed from the point of being hit by a slow attack until the end of its (the unit's) next turn, thus affecting retaliation damage (standard strategy when attacking powerful enemy units, to make sure certain attacking valuable units have 100% chance to survive). No relation to game turns.
Judging from norbert's posts, the slow attack did at least not affect retaliation damage during the slowed unit's next turn.

[HomerJ]

Ok I see it this way:
If you are on the attack and slow an enemy unit it will be slowed for your turn, your allies turns (after your's) and it's own turn (enemy unit).
If you are defending and slow an enemy on retaliation, it will be slowed until it attack is over, (because it can not attack multiple times) the next action is end of turn of enemy unit which also ends the "slowed" status.

Conclusion: Slow affects enemy units until the end of their turn. Very easy and would explain the experienced behavior.

Greetz
HomerJ

[norbert]

I went on IRC, #wesnoth and #wesnoth-dev. There, I mentioned that in our next game (corrupt replay), I saw another Shaman slow attack 'fail'. Enemy turn 11, Elvish Shaman at 8,1 against Cavalryman at 7,1: she didn't slow the enemy unit at all, even though the ranged attack was successful and the slow sound played. I asked everyone to come see it for themselves and literally nobody joined our game. We played for at least 30 more minutes.

Nothing in this thread gives me the impression that you're aware that slow disappears when the slowed unit ends its turn.

Nothing gives me the impression that you really care. You saw me write you could join our game and you didn't. Now you claim I may not understand how slow works. The proof is in the replay I've linked just now. Is it corrupt? Too bad. There was proof and you didn't care. Observing a game fast and slowing down before turn 11 would not have taken more than a minute. I wrote the exact turn number, and the exact locations of the units. I know what I saw - 4 times now, in three different scenarios.

I guess it is going to be hard for the devs to do something based on your evidence.

You would think that replays are the best proof available. Oh wait, even better would've been if the developers could've joined a game in progress and someone would've mentioned the exact turn and location where it happened. Oh wait, I did that and they didnt' care. Not even enough to write something like: "I cannot join your game currently. Please wait an hour or so, so one of us can join it and see for ourselves."

[Edit: Sorry for the attitude and sarcasm in this post, I was pissed.]

[psychic]

How is venting your anger here going to solve anything?? The fact of the matter is that devs cannot do anything based on your current evidence.

[Alink]

Ok, indeed, when we talked on IRC I didn't observed your game (sorry I was busy coding wesnoth stuff in the middle of night and didn't want to start debugging some obscure bug before sleep time). But the next day, I did some work to make your replay plays correctly (it's a LUA bug, I think I have a valid fix but I need to see our LUA expert, silene, for the details). I can know watch your whole replay fine.

Anyway, the good news is that I don't see any bug. The shaman hit the dark adept at 7,2 and slow him, then later hit the cavalier at 7,1 do 2 damages, you hear a hit sound but no slow. The important detail is that it was not the shaman's turn. She was attacked by these units and only the adept used a ranged attack. The cavalier attacked with melee and the shaman must counter-attack with her *melee* weapon too which doesn't have slow special. (I suppose that you are aware of this rule about weapon's range). BTW, you clearly see the shaman moves and hits with her little staff and not staying fixed or using magical effects.

So, no bug, (except the LUA one in replay) but I hope that you now see why we don't drop everything at the first weird bug report and are sometimes a bit skeptical about user's knowledge

And please, be nice, remember that devs work for free but work when they want

[norbert]

Anyway, the good news is that I don't see any bug.

That is good news.

I still think I heard the "/data/core/sounds/slowed.wav" sound. I had just started this thread, so I was specifically paying attention to the Shaman's actions during that second match. I know that memory is a tenuous thing. Then again, it concerns short term memory and I've been playing Wesnoth for some time; I do recognize the slow sound effect, because it's very distinct. Can you send me the fixed replay of the second match or attach it to a post, because I would like to check if I can reproduce it.

[Anonymissimus]

It seems to me that you (norbert) thought of as a bug what's the intended behavior. If a unit attacks a shaman with ranged attack and is hit in retaliation, it is slowed (and sound plays), but the effect only lasts until the end of this turn of the unit. Following strikes by that unit (if there are any) are affected however.

[norbert]

the intended behavior [...] it is slowed (and sound plays), but the effect only lasts until the end of this turn of the unit

You're right, I didn't know/understand the intended behavior. I already kind of apologized for my attitude in my third post in this thread, but hereby do it again - especially to zookeeper. I assumed the period a unit is slowed always lasts until it's the next turn of the side who attacked when the slowing happened, and I was wrong.

What made it extra complicated (for me) is that I thought I heard a slow sound when the Cavalryman attacked the Shaman. I've asked Alink if he can provide a fixed replay of that second match, so I can take a look at it myself. I was probably wrong about that too; maybe it's the sleep deprivation.

[Alink]

The replay itself is fine, it's the engine which doesn't load LUA tags used in this scenario and thus breaks quickly. To play it correctly you need to patch your game. See bug #16243:
https://gna.org/bugs/index.php?16243
where I attached a possible fix (for source, you need to compile it), which seems to work fine.

I rechecked and the shaman hits with her staff.

Note the lack of slow sound in the image ;-)

shaman-slow.png (72.89 KiB) Viewed 3921 times

edit: btw, as I said, there is a slow attack on the adept, depending of your animation speed settings, the "Slowed" text could linger a bit or the sound can be played very close to the next fight.

[norbert]

edit: btw, as I said, there is a slow attack on the adept, depending of your animation speed settings, the "Slowed" text could linger a bit or the sound can be played very close to the next fight.

Ah... yes, that could explain it.
I always play with Accelerated Speed on max (16).
Thanks for adding the screenshot.