[Danukem]

When I switch to a widescreen resolution (not sure about all of them, but one in particular), there's a model that does not display correctly. It is coded to be half-submerged (128 as part of its cstat), but in widescreen it appears fully above ground. This movie demonstrates:

http://deeperthought...escreen_bug.wmv

It may affect other models, but this is the first case that I noticed.

EDIT: The model displays correctly in Polymer.

[Plagman]

Does the model use CON-based rotation at all?

[Danukem]

Plagman, on Jan 17 2010, 11:09 PM, said:

Does the model use CON-based rotation at all?

It does this when it spawns:


cstator 642
sizeat 16 16
setactor[THISACTOR].pitch 1024
setactor[THISACTOR].mdzoff -5120



After that, the size and mdzoff continue to change, and of course it also spawns a lot of stuff etc but that's not relevant (it was necessary to change pitch and mdzoff to get the model oriented the way I wanted; it's the billiard ball model from the HRP).

[Plagman]

If you don't mess with the pitch and/or roll at all, does the model position and orientation look consistent between Polymost and Polymer? (regardless of it being correct for your purposes). If so, it sounds like bug 2867620. I don't know when I'll be able to look into it, though; I'm afraid Polymost bugs aren't very high on my list.

[Danukem]

Plagman, on Jan 17 2010, 11:21 PM, said:

If you don't mess with the pitch and/or roll at all, does the model position and orientation look consistent between Polymost and Polymer?

I presume so. At any rate, I was able to fix the problem by using zadd in the model definition instead of messing with it in the CON code.

[Franpa]

Is your intention to eventually discard the Polymost rendering engine plagman? Because that engine runs far faster then the Polymer engine.