Hunting and Creature Behavior Suggestions

[RogueGunslinger]

The post movement delay is the metric by which "speed of travel" is determined IG.

So yeah, that's what you're asking for.

No, the suggestion is to delay the NPC's reaction upon first detecting a PC. Not to delay it every time it moves into a new room during the chase.

I honestly don't think this is needed.

[musashi]

Ah, I misunderstood. I guess I'd be ok with that, I suppose, but only if it could somehow be coded in such a fashion that it's only the first movement the critter takes that has a pre-command delay.

Because if the mekillot sees you, waits a second or two, then runs in but you run out ... it's got post-command delay and pre-command delay again. And I think that would make it too easy to run away.

[Dalmeth]

Ah, I misunderstood. I guess I'd be ok with that, I suppose, but only if it could somehow be coded in such a fashion that it's only the first movement the critter takes that has a pre-command delay.

Because if the mekillot sees you, waits a second or two, then runs in but you run out ... it's got post-command delay and pre-command delay again. And I think that would make it too easy to run away.

I think you got confused between Synthesis's argument and mine, where he spoke of a greater movement delay on the first move, and I just wish for a more regular delay before movement on spotting something it should be aggressive to.

I tend to favor my method, because it allows for outright stupidity, where a person simply may not look north and see the mekillot.  As much as people say that you should be able to hear a giant beasty coming, the truth is people get distracted.  Some minimal level of awareness should be required.

I honestly don't think this is needed.

Perhaps a significant change isn't needed, but at the very least, we need to make the delay before an aggressive creature moves regular.  Last I checked, it could vary from a few seconds to almost instantly, and even using a macro to run can get you killed.

Movement speed is the only thing that will save you from some beasties that will kill you for doing nothing better than salting or hunting rittikki.  It isn't a matter of entering their territory or something else you could avoid, it's a matter of a random number generator not rolling against you.

Not to say there isn't a way you can be immune to all this, but it does require a particular style of play.  I'm open to allowing still further styles of play.

[Nyr]

Perhaps a significant change isn't needed, but at the very least, we need to make the delay before an aggressive creature moves regular.  Last I checked, it could vary from a few seconds to almost instantly, and even using a macro to run can get you killed.

This sounded familiar--I knew we'd covered this before, and I knew that I'd tested this and also responded to it before.

If you have specific evidence (with timestamps) that shows a mob entering a room and attacking within a half second, please submit a request with your time-stamped log.

All NPCs should have a delay before they can attack after moving, same as PCs.  If they don't, it's a bug and needs to be addressed.  But we don't have anything other than anecdotal evidence of this occurring.

I tested an NPC going through a series of scripted movement commands to attack a non-aggressive NPC a few rooms away.  I then switched out of the NPC and went to the room and waited to watch.

It eventually got to the room.

It did go through the same movement lag before attacking.

I tested an NPC group as well.

Three similar NPCs, all in the same room, were set up to follow one of these NPCs.  I named it Spanky.  One of the followers I named Bill.  I switched into Spanky and ran through movement commands to a non-aggressive NPC a few rooms away.  I switched out of Spanky and into Bill.  As soon as our group got to the target room, I hit "Kill target."  I had to wait.

Yeah, so far, nothing I'm finding does an instant attack with no movement lag.

Now working on nearby room aggro attack testing with multiple NPCs. 

Yes, it happens quickly. 

No, it is not instantaneous, and is (in fact) exactly comparable to the exact scenarios I described earlier.  It seems faster because you are going apeshit (appropriately) over reading the roomspam of things entering the room.

They all work the same, yes. Text-based games may transmit quickly over the interwebs (comparatively speaking, w/r/t graphical games), but we are talking about a matter of only a handful of seconds.  

Nyr ... test it with kryl ...  

Groups of kryl ... *shivers as the nightmares return*

Was just about to send in a request with this. Lost more than one PC... I will send it still.

I checked out a special attack that is set on a particular NPC that folks may be worried about with this change.   That special attack also will not fire off if the NPC is in a wait state (such as waiting from movement lag).  Note that there is not movement lag inside the room itself, so if (for instance) npcs are breaking off from a "pack" npc due to, say, another script...that is not considered "movement."

I don't mean carru by any of this testing, though.  That's intentional.

I can't really get more specific than this.

Yes, kryl are included in the "they all work the same way."

Gah, I keep skimming through my brain and coming up with more to say.

Yes, kryl are included in the "they all work the same way."

Well, there is one group of kryl set to follow a, "leader," so they may be a subtle modification that was overlooked with previous changes.

I checked them too

No, it is not instantaneous, and is (in fact) exactly comparable to the exact scenarios I described earlier.  It seems faster because you are going apeshit (appropriately) over reading the roomspam of things entering the room.

I'm not so sure about that....

I do recall having an AI agility character get beetle'd after loading my command queue down with "n n n n n n n n n n n n n n n n".

Was one of those situations where I was merrily riding northward, wild beetle pops in from the east, attacks, knocks me off my mount, I flee and then spam north and it caught up to me after a couple of rooms and eated mai character.

Though... This was sometime last year, or maybe even the year before, and it may have been caused by lag more than anything.

or it was running and you weren't (or your mount is slow as poop)

or it was running and you weren't (or your mount is slow as poop)

To be fair, yes, PCs are faster than most animals that could kill them instantly.  However, only a handful of mounts, which most PCs use, are as fast or faster than the player, and those mounts tend to be expensive or rare.

Though it is worth noting that one of the fastest mounts is also one of the most common.

Right.  That doesn't point to there being a problem with NPCs not having movement delay prior to attacking, though.  So far, I've yet to duplicate a single scenario mentioned, after multiple tests.  I will continue to try and find one, but so far people are reporting that this is intermittent, only happened once or twice with them, etc.--yet all scenarios are different, with different NPCs, NPC groups, and programs involved.  Even with that accounted for in testing, I am still unable to duplicate, which leads me to point towards connections, movement delay on the part of the PC, and player perception.  Even I was affected by the last bit during testing.  I had to use a timer to be sure that the delay was comparable; I was sure that I'd found something but it was approximately the same amount of delay.

[Synthesis]

Either something's wrong with your test that's giving you bad results, or the wait state on certain grouping mobs is near-zero.

[Nyr]

Either something's wrong with your test that's giving you bad results, or the wait state on certain grouping mobs is near-zero.

I don't see your results in the request queue with a time-stamped log, and I tested every single suggestion people had, so I'm going to go out on a limb and assume that the anecdotal evidence in your case does not trump the data.  We will need more direction than "I've got a feeling this is wrong."  Please submit a request and let us know what we have not looked at.

[musashi]

Oh yeah ...

Well ... well ... you're a neener.

[Maso]

I lost my 'Nyr'd' image. 

[Akaramu]

I lost my 'Nyr'd' image. 

Noes 

It shall be mourned.

[Nyr]

I lost my 'Nyr'd' image.  

just curious, but if I go back through your posts and find it, is it too meta for me to

[Synthesis]

I don't know what exactly the problem is with reproducing this issue from the staff side.  It reminds me of the first-round NPC double-attack bug that was ruled "not a bug" because supposedly PCs have the same chance of it occurring in their favor, but everyone knows this just ain't so.  I'd go and test it again to make sure to myself that it's still happening, but unfortunately I'm not in a position to be able to do that.

[Maso]

Too late, you did it anyway.

[Nyr]

I don't know what exactly the problem is with reproducing this issue from the staff side.  It reminds me of the first-round NPC double-attack bug that was ruled "not a bug" because supposedly PCs have the same chance of it occurring in their favor, but everyone knows this just ain't so.  I'd go and test it again to make sure to myself that it's still happening, but unfortunately I'm not in a position to be able to do that.

I'll take a look at this issue once more.

As for the main idea presented by this thread, I'm not really sure that's an "easy" code change.  Easy to discuss, sure...but ease of implementation should be left to our fantastic coding staff.

[Maso]

In the meantime, just have Storytellers watch all roaming PC's and send room echoes whenever something is within the range that should technically be visible but isn't.

[Nyr]

I don't know what exactly the problem is with reproducing this issue from the staff side.

Without getting into a lot of detail, it really is exactly what Morgenes and I were saying over and over:

We needed more data.  

I grabbed the results we had from our last testing sessions (about a year ago) and went back into testing, throwing things out there to try and make something stick.  Long story short, there was another script involved that was creating this problem, and this one required other variables in play to make it function (or malfunction).  Thanks to Morgenes, it has been fixed.

[Euphoric Nightmare]

opps...resolved

[Dalmeth]

Either something's wrong with your test that's giving you bad results, or the wait state on certain grouping mobs is near-zero.

I'm inclined to agree, but it's not with specific mobs.  It seems more like the move is a scheduled action, and sometimes, it's so quick after a pers0n enters the room, that you spend a whole second just realizing it happened.

It's not a regular thing, but the time in which a mob reacts to an entering PC is variable.

[Nyr]

Ok, so the issue we actually found isn't the issue we actually found?

[musashi]

Exactly. But not at all.

[Dalmeth]

Ok, so the issue we actually found isn't the issue we actually found?

No.  To say that sometimes aggressive NPCs attack instantly or that the response time to a particular NPC is variable is largely saying the same thing.

[Nyr]

Ok, so the issue we actually found isn't the issue we actually found?

No.  To say that sometimes aggressive NPCs attack instantly or that the response time to a particular NPC is variable is largely saying the same thing.

We're not saying that, though.  In fact, we've found that aggressive NPCs do not (and should not) attack instantly.  We found a bug in a script that allowed this to occur with a particular NPC type.  We fixed it.

[Maso]

The room echoes would still be cool though. More just for...coolness. And atmosphere. Worrying about a split second here are there and attack delays is for nerds. 

[evilcabbage]

I've experienced this double-attack bug, on the receiving end. It is not nice at all. I have yet to perform the double-attack, unless I'm wielding two weapons! So would I get a quadruple attack if the bug kicked in my favor?

[Nyr]

I've experienced this double-attack bug, on the receiving end. It is not nice at all. I have yet to perform the double-attack, unless I'm wielding two weapons! So would I get a quadruple attack if the bug kicked in my favor?

[Kryos]

My poor thread.

But as to the discussion at hand.  Nyr, in the case mentioned, at least as I've experienced it in limited fashion, you need to have two "aggro mobs" i.e., a player and some type of NPC that would be attacked by the 2nd NPC (the insta attacker) fighting.  The insta attacker moves into a bordering room.  Following its script, it then moves into the room with the fight and swings with nearly no delay.

In my case, this happened with the no-delay beast being a bahamet, if that's useful.  That being said, as advised, if I find myself believing this situation could arise again, I'll log it and use time stamping to give additional information.

On the other hand, any chance of seeing horrible death machines coming?