PRINTLINE: SAVE GAME DAMAGED
Moderator: Zyx
Forum rules
Please read the Forum rules and policies before posting. You may
to help finance the hosting costs of this forum.
Please read the Forum rules and policies before posting. You may

- Gambit37
- Should eat more pies
- Posts: 13773
- Joined: Wed May 31, 2000 1:57 pm
- Location: Location, Location
- Contact:
PRINTLINE: SAVE GAME DAMAGED
PRINTLINE: SAVE GAME DAMAGED
That's the error I get trying to use my first new CSBuild1.1 dungeon with CSB4Win. It's an empty, fully enlarged map, except for a few corridors and one champion portrait.
Is there a certain minimum of things I need to have in a dungeon for it work?
Also, I can't open this dungeon in CSBuild817 -- it tries all variations and then reports the SAVEGAME as damaged. It's a dungeon.dat file, not a savegame.
That's the error I get trying to use my first new CSBuild1.1 dungeon with CSB4Win. It's an empty, fully enlarged map, except for a few corridors and one champion portrait.
Is there a certain minimum of things I need to have in a dungeon for it work?
Also, I can't open this dungeon in CSBuild817 -- it tries all variations and then reports the SAVEGAME as damaged. It's a dungeon.dat file, not a savegame.
- Paul Stevens
- CSBwin Guru
- Posts: 4322
- Joined: Sun Apr 08, 2001 6:00 pm
- Location: Madison, Wisconsin, USA
Dungeons made with CSBuild 1.1 require the latest
CSBwin. There are new things in there that the old
CSBwin cannot understand.
The old CSBuild will not work either, for the same reason.
You made it with CSBuild 1.1. What version of CSBwin
did you try to use?
Edit: These things are supposed to be backward compatible.
Old dungeons can be opened with new versions. Not vise-versa.
CSBwin. There are new things in there that the old
CSBwin cannot understand.
The old CSBuild will not work either, for the same reason.
You made it with CSBuild 1.1. What version of CSBwin
did you try to use?
Edit: These things are supposed to be backward compatible.
Old dungeons can be opened with new versions. Not vise-versa.
- Gambit37
- Should eat more pies
- Posts: 13773
- Joined: Wed May 31, 2000 1:57 pm
- Location: Location, Location
- Contact:
I'm using CSBWin 9.6v5. If this is not the latest, which of those on your website should I use? What's the difference between a 'v' and a Plus? You have so many versions it's all rather confusing!
Code: Select all
CSBexe96.rar 984886 2003-08-04 18:03
CSBwin96Plus12.rar 840950 2004-03-10 00:14
CSBwin96Plus13.rar 820776 2004-03-22 18:09
CSBwin96Plus14.rar 820869 2004-03-26 20:03
CSBwin96v3.rar 219130 2003-12-28 19:28
CSBwin96v4.rar 219219 2004-01-05 16:02
CSBwin96v5.rar 219379 2004-01-17 19:30
- Paul Stevens
- CSBwin Guru
- Posts: 4322
- Joined: Sun Apr 08, 2001 6:00 pm
- Location: Madison, Wisconsin, USA
It does get confusing. But you folks keep asking
for something it ain't got. I'd be glad to slow down
if it is too much for you. Perhaps one new release
per quarter.
Go get CSBwin96Plus15.rar.
It is the latest. Now. Quick. More things are coming.
It has the changes to put both a decoration and
text on the same wall. Also two different texts
on the two sides of a convex corner. And some other
things to handle errors better. And some miscellaneous.
If that does not load your dungeon then something is very
much amiss and I would appreciate it if you could send
the dungeon to me. I have already seen one instance of
a dungeon built by CSBuild 1.1 that was unreadable. So
it is more than possible that you have another. A rather
ugly situation.
for something it ain't got. I'd be glad to slow down
if it is too much for you. Perhaps one new release
per quarter.
Go get CSBwin96Plus15.rar.
It is the latest. Now. Quick. More things are coming.
It has the changes to put both a decoration and
text on the same wall. Also two different texts
on the two sides of a convex corner. And some other
things to handle errors better. And some miscellaneous.
If that does not load your dungeon then something is very
much amiss and I would appreciate it if you could send
the dungeon to me. I have already seen one instance of
a dungeon built by CSBuild 1.1 that was unreadable. So
it is more than possible that you have another. A rather
ugly situation.
- Gambit37
- Should eat more pies
- Posts: 13773
- Joined: Wed May 31, 2000 1:57 pm
- Location: Location, Location
- Contact:
Ok, thanks for that -- it loads the dungeon fine. And great -- you've sorted out the text-on-a-graphic issue -- that's really cool! Looking forward to implementing that in my new dungeon...
I have noticed that there are some graphics issues with this .exe that don't exist in the 96v5. I have a modified graphics dat (including new sizes and scaling offsets) which displayed perfectly in 96v5, but has glitches in this new version. Any thoughts on what would cause that?
I have noticed that there are some graphics issues with this .exe that don't exist in the 96v5. I have a modified graphics dat (including new sizes and scaling offsets) which displayed perfectly in 96v5, but has glitches in this new version. Any thoughts on what would cause that?
- Paul Stevens
- CSBwin Guru
- Posts: 4322
- Joined: Sun Apr 08, 2001 6:00 pm
- Location: Madison, Wisconsin, USA
No idea about the graphics glitches. I don't remember
intentionally changing anything. Perhaps if I had more
explicit information about what is a 'graphics glitch'. Is
something being displayed upside down? It something
the wrong color? Is a Screamer Slice being displayed
where a dagger should be? Is there text on top of the
Champion portraits?
Can you send me your graphics file and dungeon
file and tell me how to see the 'glitch' for myself?
If I could compare the results in the two versions then
I could probably determine why they are different.
intentionally changing anything. Perhaps if I had more
explicit information about what is a 'graphics glitch'. Is
something being displayed upside down? It something
the wrong color? Is a Screamer Slice being displayed
where a dagger should be? Is there text on top of the
Champion portraits?
Can you send me your graphics file and dungeon
file and tell me how to see the 'glitch' for myself?
If I could compare the results in the two versions then
I could probably determine why they are different.
- Gambit37
- Should eat more pies
- Posts: 13773
- Joined: Wed May 31, 2000 1:57 pm
- Location: Location, Location
- Contact:
Hmm... I looked at this in more detail, and have discovered that the glitch occurs in both CSB4WIN 96v5 and CSB4WIN Plus15. This suggests that the error is in my custom graphics dat, rather than anything else that may have changed between program versions.
Consider this dungeon view:
I have a custom graphic on wall B. With the player at position 0, walls A,B and C are too far away to be seen. If I step forward to position 8, the graphic at B can now be seen. Looks perfect, and if I step to any of 1-9 it is drawn in the correct places with no glitches.
However, if Player is at Y, and steps forward, the graphic from B is glitched -- it's offset wrongly and has a black line on the right. It looks OK from all other positions after that, only 9 causes the problem.
Most strangely, if this glitch doesn't occur if the player starts at 0, steps forward, then steps back to 0 and then Y -- the graphic is perfect with no glitches.
This is possibly a quirk of the engine, or maybe something to do with a mistake in my understanding of Charlatan's notes on file graphics.dat structure 0558. I will look into it some more to see if I can work out what's up. In the mean time, you don't need to worry about it!
Consider this dungeon view:
Code: Select all
-----A-----B-----C----- <-- (distance 3)
| | | |
| 1 | 2 | 3 |
| | | |
----D-----E-----F---- <-- (distance 2)
| | | |
| 4 | 5 | 6 |
| | | |
---G-----H-----I--- <-- (distance 1)
| | | |
| 7 | 8 | 9 |
| | | |
---J-----K-----L---
| | | |
| X | 0 | Y | <-- Player
| | | |
-------
However, if Player is at Y, and steps forward, the graphic from B is glitched -- it's offset wrongly and has a black line on the right. It looks OK from all other positions after that, only 9 causes the problem.
Most strangely, if this glitch doesn't occur if the player starts at 0, steps forward, then steps back to 0 and then Y -- the graphic is perfect with no glitches.
This is possibly a quirk of the engine, or maybe something to do with a mistake in my understanding of Charlatan's notes on file graphics.dat structure 0558. I will look into it some more to see if I can work out what's up. In the mean time, you don't need to worry about it!

Hmm, I remember a subtle graphical glitch in ConfluxII to do with the statue - walking forwads, the distance 2 graphic sometimes had a black vertical line. Other times, it was fine. And as you found, if you walk forward then back the glitch disappeared. I only found it consistently on the statue graphic outside of the circus 99 door
If zyx or charlatan can shed any light on it, it may help with your own graphical glitch debugging
If zyx or charlatan can shed any light on it, it may help with your own graphical glitch debugging
- Paul Stevens
- CSBwin Guru
- Posts: 4322
- Joined: Sun Apr 08, 2001 6:00 pm
- Location: Madison, Wisconsin, USA
There is most certainly something wrong in the
runtime engine. If you can get the graphics file
and the dungeon to me, I will study it.
The answer could be that it is your error and it should
display incorrectly at all times. Or it may be that it should
display correctly at all times. But it certainly ought
not to depend on the history of the player's travels.
And if very well could be a problem with some other
graphic that is intefering with the display of this graphic.
And it could be your mistake that the runtime should
catch and complain about. The code almost always assumes
that the dungeon and the graphics files are perfect. It does
almost no checking. The program just barely fit into the
Atari and there simply was no room for data validation.
runtime engine. If you can get the graphics file
and the dungeon to me, I will study it.
The answer could be that it is your error and it should
display incorrectly at all times. Or it may be that it should
display correctly at all times. But it certainly ought
not to depend on the history of the player's travels.
And if very well could be a problem with some other
graphic that is intefering with the display of this graphic.
And it could be your mistake that the runtime should
catch and complain about. The code almost always assumes
that the dungeon and the graphics files are perfect. It does
almost no checking. The program just barely fit into the
Atari and there simply was no room for data validation.
Gambit, are you using a graphic size of type 7?
(like "lord order outside")
Several offsets and dimensions are wrongly calculated in the graphics.dat
Charlatan detected this error and fixed it for the conflux II graphics.dat
I'll search my notes to see if I can find the value to be changed.
But it would be best if Charlatan75 himself could see it.
(like "lord order outside")
Several offsets and dimensions are wrongly calculated in the graphics.dat
Charlatan detected this error and fixed it for the conflux II graphics.dat
I'll search my notes to see if I can find the value to be changed.
But it would be best if Charlatan75 himself could see it.
- Gambit37
- Should eat more pies
- Posts: 13773
- Joined: Wed May 31, 2000 1:57 pm
- Location: Location, Location
- Contact:
It's a combination -- I've changed the WOOD RING (group 0) so that it uses the offsets for group 6 -- but I've customsied my group 6 offsets to be slightly larger than the group 7 offsets (same width but higher so that items can cover the walls AND the floor/ceiling (for some special stuff I'm working on....)
I vaguely remember something about group 7 being wrong, but I used babelfish to translate the notes as I don't speak french. I will read through them again.
I vaguely remember something about group 7 being wrong, but I used babelfish to translate the notes as I don't speak french. I will read through them again.
- Paul Stevens
- CSBwin Guru
- Posts: 4322
- Joined: Sun Apr 08, 2001 6:00 pm
- Location: Madison, Wisconsin, USA
@Gambit37:
I will tell you what I found. Then I am going to quit looking
at the problem until I hear from you that the issue is not
resolved. I don't know much about this graphics file and
you can probably interpret this information better than I.
When you approach the decoration for the first time it
is shrunk to the necessary size and saved for future use.
If you approach it directly from the front then the parameters
for the shrinkage are:
Source width 0xa0
Source Height 0x86
destination width 0x4c
destination height 0x40
If you approach it from the right (so that it is to your left)
then everything is the same EXCEPT that destination width
becomes 0x49. Four pixels less. About the width of that
vertical bar.
I will tell you what I found. Then I am going to quit looking
at the problem until I hear from you that the issue is not
resolved. I don't know much about this graphics file and
you can probably interpret this information better than I.
When you approach the decoration for the first time it
is shrunk to the necessary size and saved for future use.
If you approach it directly from the front then the parameters
for the shrinkage are:
Source width 0xa0
Source Height 0x86
destination width 0x4c
destination height 0x40
If you approach it from the right (so that it is to your left)
then everything is the same EXCEPT that destination width
becomes 0x49. Four pixels less. About the width of that
vertical bar.
- Paul Stevens
- CSBwin Guru
- Posts: 4322
- Joined: Sun Apr 08, 2001 6:00 pm
- Location: Madison, Wisconsin, USA
No idea why they did that. It must have worked with their
graphics.
But I thought that you had changed these things. I did not
know that FTL had any wall graphics this size. I assumed that
you changed these parameters to fit YOUR graphics. So I
assumed it was an error on YOUR part. At any rate, if you
want it to work you have to fix it in the graphics.dat, no matter
whose mistake it is.
graphics.
But I thought that you had changed these things. I did not
know that FTL had any wall graphics this size. I assumed that
you changed these parameters to fit YOUR graphics. So I
assumed it was an error on YOUR part. At any rate, if you
want it to work you have to fix it in the graphics.dat, no matter
whose mistake it is.