You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I have procrastinated on reporting this one, although the screenshots are from the Zircon Engine, to the best of my knowledge this affects every DarkPlaces engine from 2014 forward as I initially thought this could be an engine bug only to find that every DarkPlaces seems affected. I procrastinated because I can work around this oddball issue by moving code to a different function or other ways, so this doesn't stop me from doing anything, but the disturbing nature of this issue is the concern ...
The offending code: float viewheight = getstati(STAT_VIEWHEIGHT);
(Because this code should not mess up rendering, something is somehow getting memory stomped .. mechanism unknown. And it only occurs if the "viewheight" variable is later read by QuakeC. So some Quake VM issue seems to be living here ...)
As a local variable that is read later in the function ...
(Notice the compass is rendered with a missing block).
local_getstati.MConverter.eu.webm
As a global variable ... there are no issues in the rendering.
global_getstati.MConverter.eu.webm
If the variable "viewheight" is a local variable and it is ALSO read later in the SAME function, it messes up my compass drawing with R_BeginPolygon.
I have procrastinated on reporting this one, although the screenshots are from the Zircon Engine, to the best of my knowledge this affects every DarkPlaces engine from 2014 forward as I initially thought this could be an engine bug only to find that every DarkPlaces seems affected. I procrastinated because I can work around this oddball issue by moving code to a different function or other ways, so this doesn't stop me from doing anything, but the disturbing nature of this issue is the concern ...
The offending code:
float viewheight = getstati(STAT_VIEWHEIGHT);
(Because this code should not mess up rendering, something is somehow getting memory stomped .. mechanism unknown. And it only occurs if the "viewheight" variable is later read by QuakeC. So some Quake VM issue seems to be living here ...)
As a local variable that is read later in the function ...
(Notice the compass is rendered with a missing block).
local_getstati.MConverter.eu.webm
As a global variable ... there are no issues in the rendering.
global_getstati.MConverter.eu.webm
If the variable "viewheight" is a local variable and it is ALSO read later in the SAME function, it messes up my compass drawing with R_BeginPolygon.
The drawing code looks like ...
The text was updated successfully, but these errors were encountered: