Covert Shinobi vs Hive Queen

Post Reply
stefan3iii
Posts: 319
Joined: Wed Feb 01, 2017 3:49 am

Covert Shinobi vs Hive Queen

Post by stefan3iii »

Had a strange issue where a covert Shinobi would be spotted near the edge of the queen's detection range while concealed. The tiles were not highlighted as though they'd reveal the Shinobi. I Alt - F4ed to reset the turn, because I thought that was some XCOM bullshit, and was able to repeat it several times.

Not sure if it's a UI issue (not displaying a tile as revealing), or if the hive queen's detection range is longer than normal.

This is 1.4, on a haven invasion mission. Hive Queen + 2 soldier pod.
tracktwo
Long War Dev
Posts: 241
Joined: Sat Jan 07, 2017 7:43 pm

Re: Covert Shinobi vs Hive Queen

Post by tracktwo »

I've heard a few play testers encountering similar things, and IIRC when they looked into it it turned out to be a tile that actually had a concealment breaking icon on it but it was obscured by some other "stuff" on the tile making it hard to see. We haven't changed the detection tile stuff in LW2 so if there is a bug here it's likely a vanilla one. If you can get a repro save file I can check it out, though.
stefan3iii
Posts: 319
Joined: Wed Feb 01, 2017 3:49 am

Re: Covert Shinobi vs Hive Queen

Post by stefan3iii »

Sorry don't have the save unfortunately. I'll add though that Gotcha Again was also not giving me any detection warnings, so I doubt it was just a VFX issue.
tracktwo
Long War Dev
Posts: 241
Joined: Sat Jan 07, 2017 7:43 pm

Re: Covert Shinobi vs Hive Queen

Post by tracktwo »

I've just reproed it myself. This looks to be a vanilla bug, not LW2-specific, as I've also repro'd it with a Sectopod.

Basically the bug is that the concealment breaking tile overlays and the gotcha again activation indicator use the distance from the unit's root tile, but detection breaking is determined with the distance from the soldier position to the nearest tile on the enemy. For almost all enemies, there is no difference, but for multi-tile enemies that means that the detection ranges are bigger than they appear when approaching from certain angles: for a 2x2 sectopod the detection tiles are correct on one side, and 1 tile too short on the opposite side. For 3x3's like the Hive Queen it's 2 tiles off on some sides (1.5 rounded up) as the root tile is a corner of the 3x3 box.

I'll need to think about how best to fix this, since a lot of the code that handles this stuff is in native code we can't easily modify.
stefan3iii
Posts: 319
Joined: Wed Feb 01, 2017 3:49 am

Re: Covert Shinobi vs Hive Queen

Post by stefan3iii »

Cool thanks for looking into it.
Post Reply