Error in Imp affecting call monster sprite

I must admit I've lost the plot since last fall, so bear with me...

re #1: no, the living is in the 1st slot, and the undead in the 2nd.  are you by chance using Dragonsbrethren's SrBehemoth fix, which puts a non-present living SrBehemoth in the 2nd formation's 1st slot so its data-dictated prize will be given after the 2nd battle?

...

but i reckon the first thing to do is test my theory by trying the Imping and battle transition without DB's patch applied, and see whether the problems go away or even decrease.

Huh. My investigation may well have been with Dragonsbrethren's fix applied at some point. Or I was making lazy assumptions 
However, I can confirm that the symptoms are the same even without Dragonsbretheren's patch applied. That is: if you apply Imp Skimp 1.4 on vanilla, then beat the living SrBehemoth while it is Imp'd, the undead SrBehemoth will be invisible, etc.

yeah, same here.  i get the invisible undead, and the cursor and any attack animations directed at it appear in the top right corner of the screen.  also, casting Slow on it crashes the game with a black screen (on this ancient version of Snes9x, anyway).

so much for my theory on DB's fix. :/

however, i also still see no evidence that the undead is sharing a formation slot with the living.  though i can't rule out Leet Sketcher's patch being fooled somehow into _thinking_ that they do.

when that's combined with a patch that is more vigilant about setting $62C2 flags and redrawing things, the graphical weirdness results!

Taking a guess, Leet Sketcher may have tried covering all bases by replacing every call to $257C, without necessarily knowing the context of each one.
I don't see any benefit to running the check when a new enemy appears, or when an enemy regenerates, so doing any more rocket surgery to accommodate those cases feels unnecessary at this point.
(Am I in the ballpark here?)

well, in the hypothetical case of old and new enemies sharing a formation slot: will a new enemy entering clear the $62C2 flag by default?  if so, then barring a monster with start-up Imp status, i guess undoing the call redirect would be safe.

however, even if the patch's changes are *unnecessary* in certain circumstances, i wish i had a damned clue why they're *harmful* for SrBehemoths.

if Dragonsbrethren's patch puts living SrBehemoth at coordinates matching or overlapping the undead, the latter could be compromised as a side effect of operating on the former.  in that case, moving them apart might dodge the problem, though i suspect oddities would still show up in the living's new spot.

DB's just got it at (0, 0) default coordinates (which get automatically changed to (192, 0) by the Back attack).  that does partially overlap the undead; its rear 2 tile columns, all but the bottom tile row.

doubt that'd be enough to explain invisibility, if his patch were even required for the bugginess in the first place.