Opened 14 years ago

Closed 12 years ago

Last modified 12 years ago

#3208 closed defect

INDY3: Minor Graphics Glitch (script/data bug)

Reported by: SF/mthreepwood Owned by:
Priority: low Component: Engine: SCUMM
Version: Keywords: original
Cc: Game: Indiana Jones 3


While in Castle Brunwald, in the room with the tapestry on the second floor, there is a pink line out of place. Saved game attached.

Win XP 0.10.0svn May 25

Indy3 VGA English

Ticket imported from: #1726606. Ticket imported from: bugs/3208.

Attachments (1)

indy3-vga.s08 (13.5 KB ) - added by SF/mthreepwood 14 years ago.
Saved Game

Download all attachments as: .zip

Change History (9)

by SF/mthreepwood, 14 years ago

Attachment: indy3-vga.s08 added

Saved Game

comment:1 by fingolfin, 14 years ago

Is this a regression, or did it already appear with 0.9.x / 0.8.x ?

comment:2 by SF/mthreepwood, 14 years ago

Yes, it existed in 0.8.x/0.9.x as well. I guess I just never noticed it before

comment:3 by fingolfin, 14 years ago

I wonder if this occurs in dosbox, too, hrm.

comment:4 by SF/thunderpeel2001, 14 years ago

Just tested this, it's in the original VGA game. It's not a bug with ScummVM.

comment:5 by fingolfin, 13 years ago

Priority: normallow
Summary: INDY3: Minor Graphics GlitchINDY3: Minor Graphics Glitch (script/data bug)

comment:6 by sev-, 12 years ago

This bugreport has been moved to Wiki to relevant Engine/TODO page. When the bug will be resolved, an appropriate message will be posted here and the bugreport link removed from Wiki

comment:7 by sev-, 12 years ago

Keywords: original added
Status: newclosed

comment:8 by SF/counting_pine, 12 years ago

Hi, I thought it might be an interesting exercise to take a look into this... The problem is in the resource file itself - Object 324 in room 23 (visible in room 135). The tapestry has a magenta line 8 pixels high in the bottom left corner.

(I've checked the English and German versions, and the resources are identical.)

The rest of that first column is identical to the wall, so if the first column can be skipped, that would effectively fix the problem. But I don't know whether SCUMM provides that capability, and if it does, whether SCUMMVM can access it from a high enough level.

There might be another possiblity as well: I think that the line's color might be the transparent color, so maybe it's possible this could be fixed by changing the way that strip is rendered. Unfortunately, it doesn't look like the strip drawing code gets to see the object ID, so that's probably not really going to be feasible. But maybe a simple resource patch can be done, I don't know ScummVM well enough to know whether resource patches are simple/acceptable, and I don't know the object format well enough to know if a simple patch is possible.

But in any case I guess the fix would have to be really trivial to be worth considering. Maybe on this info you can decide whether to fix or reject this bug report.

Note: See TracTickets for help on using tickets.