FYI absence_noticeability marker physically is there

[RJFerret]

I couldn't figure out why I couldn't jump/stand to access something critical anymore (an objective item), turns out it was because I'd added absence_noticeability to an object that had to be moved out of the way.

 

Apparently the invisible-to-player marker that is put there for AIs to notice when something is moved out of position was impeding jumping, and forcing a crouch if moving a higher object to stand upon.

 

Just something to keep in mind for the future!

[grayman]

Absence markers are non-solid.

 

Please file a bugtracker issue so we can investigate. Include a small test map showing the problem.

[RJFerret]

Are they? Doh. I was presuming there was a technical reason they couldn't be.

 

Part one done: http://bugs.thedarkmod.com/view.php?id=3663

 

Need to make a sample map yet.

[RJFerret]

http://bugs.thedarkmod.com/file_download.php?file_id=250&type=bug

 

Sample map.

 

Two side by side alcoves, each with a shelf, left one has absence_noticeability enabled, right one does not.

 

Withdraw each shelf (grab, pull back/out, drop on floor), observe you can freely jump on the right side compared to inhibited on left side.

[RJFerret]

Here's a follow up issue I just noticed today. For an item bound to another, the absence marker is spawned not where the item went missing, but back where it started on the map.

 

Should that be a new bug, feature request, or noted on this physical issue bug page (which is the only reason I realized, I physically bumped into the invisible object in another room!)

[grayman]

Specific example? I don't understand.

[RJFerret]

Sorry, here's an example, imagine a scroll bound to the hand of a writer standing by his desk. Since pick-pocketing from his hand would be very apparent, absence noticeability is set on.

 

His path leads off elsewhere, and the player takes the scroll when the writer is in another room, the AI won't react. Back by the desk, you'll bump into the physical invisible object suspended at hand height, where the scroll was at map start.

 

If an AI goes by the desk, or the writer returns, they'll react like something was stolen there.

 

Apparently, upon pick-pocketing, the absence marker is not put where the object actually is taken, but where the object was spawned at map start.

 

Better explanation?

[grayman]

Apparently, upon pick-pocketing, the absence marker is not put where the object actually is taken, but where the object was spawned at map start.

 

Yes, that's what happens. Apparently absence markers were designed for important items that don't move, the idea being that if they move away from their original position, AI should notice.

 

The absence marker is spawned as soon as the writer walks away from his original position. Since there's no feedback to the player, you won't notice the creation of the marker. If another AI approached the desk and saw the marker, he'd become alerted, even though the writer still might have the scroll in his possession elsewhere.

 

This is probably not the way to handle important moving items.

 

A script that keeps track of whether the scroll is bound to the writer seems a better solution for this situation.

[RJFerret]

The absence marker is spawned as soon as the writer walks away from his original position.

 

Actually it's not. Simply make the absence_location the carrying AI with a useful boundary setting. The marker isn't spawned until the item is taken away from the AI, but in the wrong spot.

 

PS: So not technically a bug, thanks for answering that, I'll post it as a separate feature request, since it doesn't sound related to the original issue's code so much.

 

PPS: http://bugs.thedarkmod.com/view.php?id=3671

Edited February 7, 2014 by RJFerret

[RJFerret]

A script that keeps track of whether the scroll is bound to the writer seems a better solution for this situation.

 

Easier than that, just attach a teleport to the AI, and port in an absence_marker triggered by a "player has item in possession" objective. (This also allows me to place it higher, to keep the physical bug away from the AI's pathing.)

[grayman]

Actually it's not. Simply make the absence_location the carrying AI with a useful boundary setting. The marker isn't spawned until the item is taken away from the AI, but in the wrong spot.

 

When the scroll is spawned, the code records the boundary box of the scroll.

 

Every 5 seconds, the code checks to see if the current boundary box of the scroll intersects the original boundary box of the scroll.

 

When the writer carrying the scroll walks away, the next of those checks will spawn the absence marker.

 

So, actually, it is.

[RJFerret]

There are spawnargs, the one I listed, absence_location, and one I referred to, absence_bounds_tolerance, which change that to the boundary of the entity specified, as well as an allowance.

 

So no, the boundary in my case was the AI's bounding box, plus 99. Completely irrelevant to the original object.

[grayman]

See my notes in the bugtracker issue.

[RJFerret]

So noted, and copied info here to there.